Patent Grant
7493237
This application is a Non-provisional application of Provisional Application Ser. No. 60/344,216 filed Dec. 21, 2001. Priority is claimed based on the aforesaid Provisional application Ser. No. 60/344,216 and based on Ser. No. 10/324,680, filed Dec. 20, 2002.
Not Applicable
This invention relates to monitoring of emissions and/or waste streams from a production facility.
The United States Environmental Protection Agency (EPA), prior to the present invention, has required monitoring and reporting on individual sources of actual or potential undesirable emissions of gaseous matter or liquid matter. These requirements have heretofore been satisfied by “end of the line” monitoring techniques. Heretofore, there has been no known method or system for the environmental monitoring and reporting of a combination of gaseous and liquid emissions from a production facility.
Of recent, the EPA combined air and water regulation applying to the pulp and paper industry, known as the Cluster Rule. This Cluster Rule was developed to minimize and control Hazardous Air Pollutant (HAP) emissions via direct air vents from non-condensable type gas systems (NCG) (referred to in the Rule as Low Volume High Concentration (LVHC) and High Volume Low concentration (HVLC systems), and from volatilization fro HAP bearing liquid streams originating in the pulping and evaporation processes. These liquid streams are produced from the condensation of relief or evaporation vapors in various direct and indirect condensing systems in the aforementioned areas. The Cluster Rule refers to these HAP bearing condensates as “named streams”.
The Cluster Rule is unique in the history of the industry as It is the first Rule to require monitoring of significant process parameters in the mill proper, and the first Rule to require the daily/continuous inventory of HAP9 produced in the mill proper. Most regulations look at final emissions on end-of-pipe treatment systems and their respective treatment efficiencies (eq. wastewater treatment basins, steam strippers, recovery boiler electrostatic precipitators etc). The industry was faced for the first time with monitoring AND reporting in-process activity as relates to HAP evolution, in addition to treatment. Many of these process areas were never monitored to this extent in the past and in many cases, no instrumentation was even present to track required parameters. Many new condensate collections systems had to be built with new piping to transport condensates from evaporator and pulping condensers to a main collection tank prior to delivery to one or more treatment devices. Operation parameters in the evaporators such as liquor flow, liquor solids, conductivity, condensate flow, temperature and valve positions along the delivery piping (to confirm actual collection) had to be installed and connected to the mill distributed control system (DCS) and process information (PI) systems. Digester systems required monitoring of chip meter rotation as an Indicator of pulp production, conductivity, condensate flow, temperature and valve position. Many of the Cluster Rule requirements did not provide instruction on the development of the monitoring and tracking systems, only the final goals.
On Apr. 15, 1998 the Environmental Protection Agency (EPA) promulgated the Cluster Rule for the pulp and paper industry. These rules establish the effluent guidelines and standards under the Clean Water Act and the national emission standards for EPA's designated hazardous air pollutants under the Clean Air Act and have a mill-wide effect on the affected International Paper mills.
The Clean Air Act Amendments of 1990 designated certain substances as hazardous air pollutants (HAPs) and required the industry to reduce HAPs using Maximum Achievable Control Technology (MACT) control measures. MACT means the best demonstrated control technology or practices used by similar sources of air toxics, defined by law as the average pollutant reduction achieved by the best-performing 12 percent of mills. The MACT regulation for the pulp and paper mills is codified in 40 CFR Part 63 Subpart S.
The regulation requires pulp and paper mills to control HAPs, using methanol and chlorine as surrogates in the mills' condensate, LVHC/HVLC and bleach plant systems, respectively.
This invention is directed to a method for the combined monitoring and reporting of actual and potential emissions of multiple phases of matter from a production facility which utilizes and/or generates such multiple phases of matter comprising the steps of:
This invention relates to an automated monitoring, record keeping and reporting system to comply with the regulation. The project objective is to comply with the requirements of these Cluster Rule components. This document was developed to establish the design specifications and programming methodology for this data collection system.
The purpose of this document is to describe the design of the record keeping and reporting system for condensate treatment using an aerated stabilization basin (ASB). The software is comprised of PI Data Archive software (which is used for automatic data collection from various process instrumentation and control systems) and Proficy software (which monitors and reports compliance based on the PI data and operator inputs). This documentation is directed toward system administrator level personnel but can be used for a basic understanding of how the system works.
The following sections describe the general configuration of the standard biological condensate treatment monitoring system. Deviations from the standard model, configuration listings for specific lines, and mill-specific details are contained within the appendices.
Foul condensate is collected in a central collection tank (Main Tank) from sources such as digesters, evaporators, and turpentine systems. For treatment in a biological system, the condensate is pumped through a hardpipe delivery system discharging below the surface of an aerated stabilization basin (ASB) (or some other device such as a UNOX system). In most cases, the flow from the Main Tank mixes with the remaining whole mill influent to create the total ASB influent flow. In a few cases, the total ASB Influent flow is equal to the hardpipe flow if the ASB is a dedicated condensate treatment system that receives no other wastewater. The metric used to determine ASB compliance is the Total ASB Influent soluble Chemical Oxygen Demand (sCOD) load relative to the basin processing capacity based on aeration horsepower (with the units of sCOD lbs/HP). sCOD is defined as the amount of oxygen required to oxidize all soluble compounds, both organic and inorganic, in water. sCOD is expressed in units of mg/l (ppm). Compliance is demonstrated by operating below the limit of sCOD lbs/HP determined in a Performance Test. Other measurements of ASB Influent Load such as to Total Organic Carbon (TOC) can be used in place of sCOD. (Specified as the alternative method in §63.463(j)2)
When the ASB treatment performance metric falls below the limit set in the performance test, the mill will respond in accordance with the SSM Plan and may retest to show compliance at this new parameter range with the result that no excess emission event occurred. (§63.453(p)) The monitoring system logs the potential Excess Emission (EE) event and corresponding operator responses to the event. The responses record the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) for the event. The report categorization specifies if the event is considered an allowable excess emission if the emission is due to a Startup, Shutdown, or Malfunction (SSM). The events are compiled by the system and reported to the state regulatory agency on a semi-annual basis or more frequently as required (§63.10).
For mills following this ASB Treatment methodology, a warning limit is attached to the 15 Day—MeOH Avg variable to warn the operator that MeOH collection is close to falling below the excess emission limit for condensate collection. If the methanol load remains lower than that collected and treated during the initial performance test, the facility may be required to raise the ASB efficiency (by lowering the sCOD lbs/HP target) following a required quarterly retest unless the methanol collection can be restored to original collection levels. The warning limit is specific to the mill based upon the biological treatment efficiency of the ASB at the sCOD lbs/HP upper limit for the ASB system. The value of the warning limit is calculated from the minimum fbio (fraction bio-degraded) that correlates to the sCOD lbs/HP upper limit, determined during a performance test; the limit is set to 11.1/fbio for bleached mills and 7.2/fbio for non-bleached mills. This warning notifies the operator to inspect and troubleshoot the condensate closed collection and treatment systems to insure compliance during the next quarterly performance test. Therefore the lower warning flag on collection may not result in an immediate excess emission for collection or treatment as long as the ASB continues to meet its initial performance test sCOD lbs/HP target. However if methanol collection levels are not restored by the quarterly test, excess emissions could be recorded indefinitely (on a daily basis) until the ASB efficiency is increased or collection restored. (§63.446(e) & (p))
In addition to capturing and categorizing EE events, the monitoring system also captures and records failures (downtime) of the Continuous Monitoring System (CMS). All Condensate Treatment ASB CMS events are manually triggered and are 24 hours in duration. This event is summarized and reported to the state in a semi-annual CMS performance report or more frequently as required. The report categorization specifies if the event is considered allowable based on the specific regulations. (§63.8(c)2, §63.8(c)8 and §63.10)
In addition to monitoring and recording the above, the monitoring system records and displays operating parameters (on the ASB Treatment Autolog) to insure that the ASB is running under normal operating conditions. These operating parameters are used with specification limits applied to notify the operator (through color coding) to take whatever action is necessary to restore the ASB to normal operating conditions. The parameters are used for display only and do not create any events. The sample location for the operating parameters will vary by mill, but the standard operating parameters for all ASB's are; sCOD, dissolved oxygen (DO), dissolved oxygen uptake rate (DOUR), mixed liquor suspended solids (MLVSS), and specific oxygen uptake rate (SOUR).
The Total Influent Load to the ASB is monitored in three ways:
The sCOD load is calculated by multiplying the total daily ASB influent (Gals) by the sCOD (ppm) with appropriate factors to convert the result into lbs/day delivered to the ASB. Aerator horsepower is the product of an aerator horsepower factor (a mill may have several different factors if they maintain different types of aerators) and the number of aerators of each type in service. Both Total ASB Influent flow and sCOD may require multiple calculations to first determine the contribution of the hardpipe and whole mill influent. The total sCOD (lbs/day) inlet load is divided by the total aerator horsepower (HP/day) to determine the sCOD lbs/HP for the day, or:
Proficy calculates the total sCOD lbs/day, the total aerator HP/day, and the sCOD lbs/HP ratio once an operator manually enters the type and number of aerators (and/or blower systems) running, a daily sCOD test(s), and the Total ASB Influent Flow (note: at certain mills Total ASB Influent flow may be automatically entered from PI as the sum of the whole mill influent and hard pipe flows). The parameters required to calculate sCOD lbs/day are the Continuous Monitoring System (CMS) parameters for ASB treatment.
Proficy compares the sCOD lbs/HP against a upper specification warning limit established during a Performance Test to determine if a potential EE event has occurred. The duration of a potential EE event is 24 hours. Performance Tests, conducted quarterly, relate the sCOD lbs/HP ratio to a minimum required ASB MeOH removal efficiency (fbio). A sCOD lbs/HP value greater than the warning limit indicates the ASB is outside of the operating range established during the Performance Test. This indicates that the ASB is potentially overloaded and the ASB removal efficiency may be less than required for compliance.
When the potential EE event is created, the mill must respond in accordance with the SSM Plan and may retest to show compliance at this new parameter range with the result that no excess emission event occurred. The Proficy software logs the potential EE event and corresponding operator responses to the event. The responses record the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) for the event. The report categorization specifies if the event is considered an allowable excess emission if the emission is due to a Startup, Shutdown, or Malfunction (SSM). A comment is required to be entered in Proficy whenever a potential EE event occurs.1 The events are compiled by the system and reported to the state regulatory agency on a semi-annual basis or more frequently as required. 1 This is accomplished by forcing an operator to enter comment on the Trouble reason code in the Proficy downtime event.
For mills following this ASB Treatment methodology, a warning limit (the Proficy lower user specification limit) is attached to the 15 Day—MeOH Avg variable to warn the operator that MeOH collection is close to falling below the excess emission limit (the Proficy lower warning specification limit) for condensate collection. If the methanol load remains lower than that collected and treated during the initial performance test, the facility may be required to raise the ASB efficiency (by lowering the sCOD lbs/HP target) following a required quarterly retest unless the methanol collection can be restored to original collection levels. The Proficy lower user specification limit is specific to the mill based upon the biological treatment efficiency of the ASB at the sCOD lbs/HP upper limit in Proficy for the ASB system. The value of the warning limit (Proficy lower user specification limit) is calculated from the minimum fbio (fraction bio-degraded) that correlates to the sCOD lbs/HP upper limit, determined during a performance test; the limit is set to 11.1/fbio for bleached mills and 7.2/fbio for non-bleached mills. This warning notifies the operator to inspect and troubleshoot the condensate closed collection and treatment systems to insure compliance during the next quarterly performance test. Therefore the lower warning flag on collection may not result in an immediate excess emission for collection or treatment as long as the ASB continues to meet its initial performance test sCOD lbs/HP target. However if methanol collection levels are not restored by the quarterly test, excess emissions could be recorded indefinitely (on a daily basis) until the ASB efficiency is increased or collection restored.
In addition to capturing and categorizing events, the Proficy system also captures and records failures (downtime) of the Continuous Monitoring System (CMS). All Condensate Treatment ASB CMS events are manually triggered and are 24 hours in duration. This event is summarized and reported to the state in a semi-annual CMS performance report or more frequently as required. The report categorization specifies if the event is considered allowable based on the specific regulations.
In addition to monitoring and recording the above, Proficy records and displays operating parameters (on the ASB Treatment Autolog) to insure that the ASB is running under normal operating conditions. These operating parameters are used with specification limits applied to notify the operator (through color coding) to take whatever action is necessary to restore the ASB to normal operating conditions. The parameters are used for display only and do not create any events. The sample location for the operating parameters will vary by mill, but the standard operating parameters for all ASB's are; sCOD, dissolved oxygen (DO), dissolved oxygen uptake rate (DOUR), mixed liquor suspended solids (MLVSS), and specific oxygen uptake rate (SOUR).
Table-1 gives the process inputs typically required for ASB systems, their engineering units, data source, and corresponding Proficy variable names.
The percent treatment minimum limit (Minimum % Treatment) reflects the fbio (fraction bio-degraded) that correlates to the maximum sCOD lbs/HP ratio (Total LB COD/HP) that was measured during any performance test (initial or quarterly). This maximum ratio (displayed on the Max sCOD lbs/HP Upper Limit Autolog variable) is the Proficy upper warning specification limit attached to the variable Total sCOD lbs/HP (see table 2 below).
Additionally each mill may define mill specific operating variables to be monitored in addition to those specified above. User Specification limits for the operating parameters are listed in the specification limits table in Section V. Table-2 lists typical calculated variables for the system and a brief description of each.
The ASB Treatment Data Flow is depicted in
The ASB basin is considered to be running anytime that the Condensate Collection system is operating. Consequently the ASB potential to emit status (PTE status) is equivalent to the Condensate Collection potential to emit. Whenever the Condensate Collection system is shutdown for a majority of the day (>80% of the potential runtime or 4.8 hours in a 24 hour period) the ASB is also considered shutdown. See the section below (Condensate System Shutdown) for a detailed explanation of how this is indicated within the system.
The total reporting minutes of ASB operation, reported to the appropriate regulatory authority on a semi-annual or more frequent basis as required, correspond to the total source operating minutes of the Condensate Collection system.
The whole mill influent flow and hard pipe flow (if separate streams exist) going into the ASB are sampled and analyzed daily for sCOD. The sCOD load (Calculated ASB Influent sCOD Load) is the sum of the two streams' sCODs (Total ASB Influent sCOD) multiplied by their daily total flows (Total ASB Influent Flow). Some mills have two sCOD loading (one from condensate sources and one from mill influent sources) implying that the Calculated ASB Influent sCOD Load will be the sum of the products of the sCOD and flows from each source for the day.
A separate mill-specific Autolog will be designed to calculate the total aeration horsepower, Total Aeration HP.2 For each type of aerator, the number of aerators in operation will be multiplied by their respective horsepower to calculate the total horsepower for that specific aerator type. The total horsepower's for all types of aerators in operation are then summed to calculate the total aeration horsepower (Total Aeration HP). 2 At some mills this will be directly incorporated into the main ASB autolog sheet.
This value (Total sCOD lbs/HP) is an estimate of the sCOD load relative to the processing capacity of the basin and is calculated by dividing the ASB influent sCOD load (Calculated ASB Influent sCOD Load) by the total aeration horsepower (Total Aeration HP).
An excess emission event is generated under the following conditions:
If the mill SSM plan allows for retesting of the ASB at the higher sCOD lbs/HP ratio and the testing of the ASB determines that the sCOD lbs/HP ratio resulted in maintaining the removal efficiency, the mill may report the event as No Excess Emission.
If the parameter value is exceeded and the SSM plan allows for it, the mill may chose to run a performance test to show compliance at this new parameter range. If the removal efficiency was maintained the event may be reported as No Excess Emission. A comment in Proficy is required whenever this condition occurs. All ASB Treatment EE events are 24-hours in duration.
A reportable 24-hour CMS downtime event is created whenever the operator or environmental contact chooses the “Bad Data—24-Hr CMS” selection on the pull-down menu of the ASB Treatment Data Quality (CMS) variable. Manually selecting this option results in the creation of a 24-hour CMS event. A 24-hour CMS event results whenever one of the following parameters (required to determine sCOD lbs/HP) cannot be determined for the day:
All ASB Treatment CMS events are 24-hours in duration.
Whenever the condensate system has been shutdown for greater than 80% of the day the operator or environmental contact should indicate the shutdown by selecting the “Shutdown” selection from the ASB Treatment Data Quality (CMS) variable.
Guidelines for Use of Manual Pull-Down Selections
Typically, all inputs to the standard ASB treatment model are manual entries (with the possible exception of the ASB inlet flow); therefore PI tags are not required.
Standard Proficy Model
The Proficy model consists of input variables, calculated variables, stored procedures, and Visual Basic scripts (VB scripts). Variables and associated parameters for a typical ASB treatment plant and descriptions of the stored procedures and the VB scripts are included below. Complete listings of the Stored Procedures can be found in following Table 3 herein below.
3The sampling offset is determined by the mill-specific start of day time. The offset value is the number of minutes from midnight to the mill start of day.
Specification Limits
Proficy has upper and lower specification limits that can be defined for every variable: entry limits, user limits, warning limits, and reject limits. The following descriptions define how Proficy uses these limits to trigger events and inform operators of impending events:
4The sampling offset is determined by the mill-specific start of day time. The offset value is the number of minutes from midnight to the mill start of day.
User Limits
Warning Limits
Reject Limits
Entry Limits
All manually entered operating parameters have Upper and Lower Entry specification limits.
Calculations
ASB Treatment EE Events
Type: Stored Procedure—spLocal_ASBTreatmentEvents
ASB Treatment CMS Events
Type: Stored Procedure—spLocal_ASBTreatmentCMS
Calc (A*8.34*B/1000000)
Type: Equation
Upper Warning Spec Limit
Type: Equation
Calc (A/B)
Type: Equation
Calc (A*B)
Type: Equation
Returns the product of the two inputs, A and B.
Calc (A/B*1000)
Type: Equation
Variables:
spLocal_ASBTreatmentCMS
/*
Procedure Name: spLocal_ASBTreatmentCMS
Copyright (C) 2001, International Paper Company
Process Management Application Group
Revision History:
General Description:
This stored procedure creates a 24-hour downtime event triggered by a manual input from the operator.
Triggers:
Variables:
The purpose of this document is to describe the design of the record-keeping and reporting system for the Condensate Collection system. The software is comprised of PI Data Archive software (which is used for automatic data collection from various process instrumentation and control systems) and Proficy software (which monitors and reports compliance based on the PI data and operator inputs). This documentation is directed toward system administrator level personnel but is useful for gaining a basic understanding of how the system works.
The following sections describe the general configuration of the standard condensate collection monitoring system. Deviations from the standard model, configuration listings for specific lines, and mill-specific details are contained within the appendices.
Cluster Rule regulations require that affected sites maintain continuous compliance with one of the following options for condensate collection:
Sites must obtain regulatory agency approval for their proposed method of continuous compliance and the continuous monitoring system (CMS). This document details IP's primary approach for continuous compliance using the lb/ton method referenced herein as the “Main Tank” or “Main Tank Collection” method.
The continuous monitoring system (CMS) is operated to measure the quantity of methanol (MeOH) collected in the main condensate collection tank relative to pulp production. The regulatory requirement (§63.446(c)3) for compliance is to collect a minimum quantity of methanol per oven dried ton of pulp produced at the digester (7.2 lbs/ODTP for a non-bleached mill and 11.1 lbs/ODTP for a bleached mill). The lbs/ODTP collected in the main tank is calculated over an averaging period (e.g. fifteen-days). The collection quantity is derived from three primary process variables:
The data for pulp production, condensate flow and MeOH concentration are collected on a daily basis. Regulatory requirements for reduction of monitoring data are defined in §63.8(g), which requires four or more data points equally spaced over each 1-hour period. We are using daily totals of pulp production and condensate flow to match the collection period of the daily composite sample, which is used to determine the average daily MeOH concentration. Because there are rather large variances in these process values on a day to day basis, a 15-day rolling average is used to determine the lbs/ODTP value for excess emission reporting.
The monitoring system logs all Excess Emission (EE) events and operator responses to those events, on a daily basis. The responses recorded by the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) for the event. The report categorization specifies if the event is considered an allowable excess emission due to Startup, Shutdown, and Malfunction (SSM) provisions, as required in §63.6(e)3(iii). The events are compiled by the system and reported to the state regulatory agency on a semi-annual basis or more frequently as required (§63.10).
In addition to capturing and categorizing EE and bypass events, the monitoring system also captures and records failures (downtime) of Continuous Monitoring System (CMS) devices, referred to as CMS events. CMS out of control conditions are defined in §63.8(c)7. Condensate collection CMS parameters include the MeOH Concentration, pulp production measurement (ODTP) and the daily total condensate flow. The monitoring system records these CMS events on a daily basis, along with the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) for the event, as required by §63.8(c)8. These events are summarized and reported to the state in a semi-annual CMS performance report or more frequently as required (§63.10).
Program Design
The data for pulp production and condensate flow is collected and archived by the PI system and made available to the Proficy system as daily totals. MeOH concentration data is received automatically, through a file transfer from the testing lab, or manually entered (as a fixed value or manual override) into Proficy. If the MeOH concentration is relatively stable, a fixed (factor) value for the concentration may be used in place of the lab daily analysis when approved by the appropriate regulatory authority.
At the beginning of each mill day, Proficy computes the relative MeOH collection rate (lbs MeOH/ODTP) over a 15-day window by dividing the 15-day collected MeOH total by the 15-day pulp production total (using only days and values exhibiting good data quality). This 15-day average lbs/ODTP collected is compared against the lbs per ODTP required for compliance to determine if an Excess Emission (EE) event has occurred. EE events are captured and recorded by the system whenever the calculated 15 Day lbs/ODTP of MeOH falls below the required minimum. Since this is a daily calculation, when this occurs the system records 24-hours of EE.
The Proficy software logs all EE events and operator responses to those events. The operator responses determine the Trouble, Cause, Correction (response), and Report Code (report categorization) for the event. The report categorization specifies if the event is considered an allowable excess emission due to Startup, Shutdown, and Malfunction (SSM) provisions. The events are compiled by the system and reported to the state regulatory agency on a semi-annual basis or more frequently as required.
Proficy also monitors for “bypass events” from the condensate closed collection system. A bypass event occurs when a portion of the condensate flow is diverted away from the collection system while the area is in a running state (i.e., the potential to emit HAPS [PTE] existed). Diverts are typically a result of flow diversion to sewer due to high conductivity or vessel overflow due to a malfunction—although other reasons for diverts exist. Proficy records the duration of the bypass events along with the operator responses to those events. The operator responses determine the Trouble, Cause, Correction (response), and Report Code. Bypass event reports are maintained by the mill to help categorize excess emission events (and as supporting documentation for Leak Detection and Repair (LDR) record keeping).
In addition to capturing and categorizing EE and bypass events, the Proficy system also captures and records failures (downtime) of Continuous Monitoring System (CMS) devices, referred to as CMS events. Condensate collection CMS parameters include the MeOH Concentration, pulp production measurement (ODTP determined from a chip meter or digester blows) and the daily total condensate flow. Whenever data for any of the parameters fails to meet preset criteria (out of range, poor instrument signal quality, flatline signal, or missing MeOH lab test results) the system suspends all calculations until intervention by an operator or the environmental contact. Intervention is made by either entering manual data or by selecting from a pull-down menu indicating that the system received Bad Data (creating a 24-hour CMS event and removing the day from the 15-day MeOH average calculation) or was Shutdown for greater than 80% of the production day (removing the day from the calculation but not creating a CMS event). Fields exist in the system to accommodate the manual data entry of the CMS parameters (using methods allowed by the state regulatory agency as a back up for instrumentation failures), resulting in no CMS event even when failures in automatic data collection occur.
CMS events are created manually when an operator or environmental contact determines that one or more of the CMS parameters have failed to obtain sufficient data to compute Daily MeOH collection for a 24-hour period. The individual creates the 24-hour CMS event by selecting “Bad Data” from the pull down menu on the Main Tank Proficy Autolog sheet. The system records the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) for the event. These events are summarized and reported to the state in a semi-annual CMS performance report or more frequently as required. Again, the report categorization specifies if the event is considered allowable based on the specific regulations.
Additionally Proficy provides a selection on the pull-down menu to indicate that the condensate sources were shutdown for more than 80% of the production day (i.e., operational for <4.8 hours). As with the CMS “Bad Data” selection, this has the effect of removing the day from the 15 day MeOH average calculation. Details of this process and guidelines on utilizing the menu selections are explained in detail below.
Table-6 provides the minimum required process inputs, their engineering units, associated PI tags (typical), and corresponding Proficy variable names. Italicized text represents mill-specific information.
Proficy also calculates, and periodically writes to PI, the data shown in Table-7 or 8:
Table-9 lists typical Proficy variables for the system and a brief description of each.
The following paragraphs describe the interrelationship between the PI and Proficy variables and how they work together to complete the calculation of the 15 day MeOH collection average.
Digesters and Evaporators PTE State
A performance equation calculates an individual area's potential to emit (PTE) status each minute in PI. The performance equation logic returns a state of “CanEmit” when condensate is present in the area's condensate collection system. This is normally during the period from startup of the area (digester or evaporator) until a mill specific period after the area stops operating and methanol has been cleared from the system. The digester area PTE-state (CR-Dig_PTE.STAT) is calculated each minute based upon mill specific criteria (such as chip meter feed or extraction flows for a continuous digester). Similarly, the evaporator area PTE-state (CR-Evap_PTE.STAT) is calculated each minute and is based upon mill specific criteria (typically steam or liquor flow).
Condensate System PTE State
The Condensate system's potential to emit (PTE) is determined in PI using a performance equation, CR-Cond_PTE.STAT which is calculated every minute. The equation logic returns a state of “CanEmit” when either the digester area or evaporator area has a potential to emit status of “CanEmit”. When both areas have a PTE status of “CanNotEmit” the condensate PTE tag returns a state of “CanNotEmit”.
Condensate Daily Downtime Counter
At the start of each mill day, a PI performance equation, CR-Cond_Down.Day, totals the “CanNotEmit” time for the condensate system (CR-Cond_PTE.Stat) over the previous 24-hour period. This value is read by Proficy and is used for both the daily display and daily calculation of condensate runtime (“CanEmit” for the daily period). The daily runtime minutes are kept in Proficy and are used to compute the total runtime minutes for the reporting period.
The Evaporator Area PTE, Digester Area PTE, overall Condensate System PTE and Daily Downtime data flow is depicted in FIG. I-A.
Pulp Production Filtered Tag and Percent Good
For every new snapshot value for the raw DCS PI tag, a PI performance equation, CR-pulp_production.Filt, filters the raw DCS tag for bad data quality or non-running status (PTE status of “CanNotEmit”). The check for a flat-lined signal is not required since most pulp production totals are calculated from the chip meter speed or the blow counter which are generally static values. If the PTE status is in a “CanEmit” state the value of the tag is compared against upper and lower reject limits (maintained in Proficy and written periodically to PI). If the tag is within the limits the raw value is archived; if the tag is outside the limits the text string “BAD” is archived instead. When the PTE status is “CanNotEmit” a value of 0 is archived representing no additional pulp production for the minute.
At the millday rollover, a PI performance equation CR-pulp_production.PctGd, calculates the percentage of time that the CR-pulp_production.Filt tag had a valid numerical value over the previous mill day (1440 minutes). The CR-pulp_production.PctGd tag is read by Proficy and displayed on an Autolog sheet to help explain missing data and for monitoring by operators and the environmental contact.
Daily pulp production data flow is depicted in
Daily Pulp Production
At the start of each mill day a PI totalizer tag, CR-pulp_production.DAY, performs a time-weighted total of the digester pulp production rate filtered tag (CR-pulp_production.Filt, ODTP/min) over the previous 24-hour period. Only production rate values while the digester area's PTE status is “CanEmit” are included in the total.
Proficy reads the pulp production daily total and stores the value in the variable Daily—PI Digester Tons. As long as 80% of the daily runtime minutes5 for pulp production experienced good data quality, the PI system will extrapolate a production total based upon 100% of the runtime minutes. The operator can also manually enter a pulp production value (Daily—Manual Digester Tons) to override an incorrect or missing PI value in the calculation of the daily and 15 Day—Digester Tons. 5 The current implementation uses a totalizer period which is 24-hrs for the daily runtime.
Daily pulp production data flow is depicted in
For every new snapshot value for the raw DCS PI tag, a PI performance equation, CR-cond_collection.Filt, examines the raw DCS tag for bad data quality, a flat-lined signal, or non-running status (PTE status of “CanNotEmit”). If (1) the PTE status is in a “CanEmit” state, (2) the difference between the maximum value of the raw tag for the past three hours and the minimum value of the raw tag for the past three hours is greater than zero, and (3) the raw value is within upper and lower data quality limits the raw value is archived by the filter tag; if the value of the tag is outside the limits or the maximum value minus the minimum value over the three hour period is zero a value of “BAD” is archived by the tag instead. If the PTE status is “CanNotEmit” a value of 0 is archived representing no flow for the minute. At the millday rollover, a PI performance equation CR-cond_collection.PctGd calculates the percentage of time that the CR-cond_collection.Filt tag had a valid numerical value over the previous mill day (1440 minutes). The CR-cond_collection.PctGd tag is read by Proficy and displayed on an Autolog sheet to help explain missing data and for monitoring by operators and the environmental contact.
At the end of each mill day a PI totalizer tag, CR-cond_collection.DAY, calculates a time-weighted totalized flow out of the main collection tank (GPM) over the previous 24-hour period. Proficy reads the condensate daily total and stores the value in the variable Daily—PI Main Tank Totalized Flow. As long as 80% of the daily runtime minutes experienced good flow meter data quality, the PI system will extrapolate the flow total based upon 100% of the runtime minutes. The operator can also manually enter a flow value for the day (Daily—Manual Main Tank Totalized Flow) that will override an incorrect or missing PI value for daily flow. This value (and the Daily—MeOH Conc. Used for Avg value—see below) is used to calculate the daily collected MeOH (Daily—MeOH Collected). Daily collected MeOH is used in the calculation of 15 day collected MeOH (15 Day—MeOH Collected).
Daily condensate data flow is depicted in
MeOH Concentration
MeOH concentration is determined by lab analysis of samples taken from the main collection tank. The CRC lab analysis uses File Transfer Protocol (FTP) to automatically enter the lab determined MeOH concentration into the Main Tank Autolog variable Daily—Lab MeOH Conc for the period (mill day) from which the sample was taken (and applies to). For other labs, the daily concentration must be manually entered by the mill. Alternatively a second Autolog variable, Fixed MeOH Conc, can be used in place of the Daily—Lab MeOH Conc if the mill and state regulatory agency agree upon an approach to calculate and verify a fixed MeOH factor, referred to as the Fixed MeOH Conc (Fixed MeOH Concentration). A pull down selection (Concentration Method) is used to select between the use of the Daily—Lab MeOH Conc and the Fixed MeOH Conc. The Fixed MeOH Conc is a manually entered, repeating Autolog variable and is used whenever the pull down selection is set to FIXED CONC. The calculation of Daily MeOH Avg (lbs/ODTP) will immediately occur once the daily tons produced (ODTP) and daily totalized flows are entered in the system (either manually or automatically from PI data). Since this is a mill specific averaging period, the system administrator, in concert with the environmental contact, is responsible to manually update the value of Fixed MeOH Conc to accurately reflect the most current fixed factor MeOH concentration whenever the factor value changes (and in accordance with the regulatory agency agreed upon requirements). If the Concentration Method pull down is set to DAILY SAMPLE, the system will wait until a lab concentration is available in the Daily—LAB MeOH Conc field to compute the Daily MeOH Avg (lbs/ODTP).
A third variable, Daily—Manual MeOH Conc, is available for the environmental contact to enter a manual concentration that will override the automatically entered value (either the Daily—LAB MeOH Conc [if Concentration Method is set to DAILY SAMPLE] or the Fixed MeOH Conc [if Concentration Method is set to FIXED CONC]) in case of an incorrect or missing concentration. Either the automatic or manually entered concentration (if entered) is copied into a fourth variable, Daily—MeOH Conc Used for Avg. The value initially is set to the automatically entered value (Daily—LAB MeOH Conc or Fixed MeOH Conc). The value updates when:
Whenever the value in this variable changes, the system will re-compute the Daily MeOH Avg (lbs/ODTP) and affected 15 day averages using the new value.
Security will be applied to the variables Fixed MeOH Conc and the Concentration Method selection field to prevent anyone except the designated person from modifying the method used (Daily or Fixed) or change the value of the repeating fixed concentration. This is usually accomplished by the security on the autolog display.
Daily MeOH concentration data flow is depicted in
Fifteen-day totals for collected pounds MeOH (15 Day—MeOH Collected) and pulp production (15 Day—Digester Tons) are calculated in Proficy from the respective daily values. The calculation looks at the data over the last 30-days and sums the most recent 15 daily values where the corresponding data quality is good (as specified by the data quality flag, Condensate Collection Data Quality (CMS)). Fifteen values are required before a total is calculated. The 15-day average MeOH, 15 Day—MeOH Avg (lbs/ODTP), is calculated by dividing the 15-day collected MeOH total (15 Day—MeOH Collected) by the 15-day pulp production total (15 Day—Digester Tons).
For mills following the ASB Treatment methodology, a warning limit (the Proficy lower user specification limit) is attached to the 15 Day—MeOH Avg variable to warn the operator that MeOH Collection is close to falling below the excess emission limit (the Proficy lower warning specification limit) for condensate collection. The Proficy lower user specification limit is specific to the mill based upon the biological treatment efficiency of the ASB at the sCOD/HP upper limit in Proficy for the ASB system. The value of the warning limit (Proficy lower user specification limit) is calculated from the minimum fbio (fraction bio-degraded) that correlates to the sCOD/HP upper limit, determined during a performance test; the limit is set to 11.1/fbio for bleached mills and 7.2/fbio for non-bleached mills. This warning notifies the operator to inspect and troubleshoot the condensate closed collection and treatment systems to insure compliance during the next quarterly performance test.
Data flow for 15-day totals is depicted in
Condensate Collection System EE
A main tank condensate collection EE event is created whenever the 15 Day—MeOH Avg (lbs/ODTP) is less than its lower warning specification limit configured in Proficy. The event duration is 24-hours. Data flow for condensate system EE is depicted in
Bypasses of the condensate closed collection system are monitored by PI. A typical bypass indicator is the state of a two-way divert valve (Open/Closed) or the state of a tank overflow indicator (Overflow/NotOverflow). For divert valves, a PI performance equation, CR-devicename.Divert, returns a value of “Collect” when flow through the device is directed toward the main condensate collection tank and returns a value of “Divert” when flow through the device is diverted from the main collection tank (while the device's area—digesters, evaporators or both—has a PTE status of “CanEmit”). These performance equations are calculated every minute. Bypass events are monitored for Leak Detection and Repair reporting and may contribute to an EE event if the 15-day average MeOH Lbs/ODTP collected at the main tank falls below the lower warning specification limit.
Proficy monitors these tags using Proficy downtime model 200 with up to a 15 minute filter. Any PI value other than “Collect” begins a Bypass event. The Event ends when the PI value returns to “Collect”.
Bypass Event data flow is depicted in
Condensate Data Quality Indicator Events
For Data Quality indicator events, Proficy monitors the data quality status for the main tank flow meter and each digester production indicator (blow counters are usually exempt) using Proficy downtime Model-200 with a mill specific delay filter. PI performance equations, CR-devicename.DQ, return a value of “Bad” when the instrument readings are outside the mill-specified instrument range while the respective area has a PTE status of “CanNotEmit” as indicated by the associated filtered (.Filt) tag; otherwise, the returned value is “Good”. Whenever Proficy reads any value from PI other than “Good,” a Data Quality Indicator event is started. The event ends when the PI value returns to “Good.” These events are not reportable to the state and are used for diagnostic troubleshooting of the closed condensate collection system.
Missing MeOH concentration data due to problems with the sample or the lab test are captured with manual downtime events in Proficy. This event is not reportable to the state and is used for diagnostic troubleshooting of the closed collection system.
A reportable, 24-hour CMS downtime event is created whenever the operator sets the Condensate Collection Data Quality (CMS) pull-down selection to a value of “Bad Data—24 Hr CMS”. This selection will be chosen when the MeOH Concentration, daily flow total, or daily digester production (ODTP) cannot be determined for the day. The operator will use the reasons assigned to the Data Quality Indicator events for the day to assign the appropriate reasons to the 24-hour CMS event. If the mill is using a fixed MeOH concentration factor (Concentration Method set to FIXED CONC), the absence of a daily MeOH concentration will no longer result in a reportable CMS event.
Whenever the condensate system sources (digesters and evaporators) have been shutdown for a majority of the day (<20% of the potential runtime minutes or <4.8 hours per production day) the operator can manually select the option Shutdown from the pull-down selection on the Condensate Collection Data Quality (CMS) variable. This has the effect of eliminating the shutdown day data from use in computing subsequent 15-day rolling MeOH collection averages.
When one or more of the required values to compute MeOH collection are missing, Proficy will suspend MeOH calculations until the values are available or operator intervention (through manual entry of a value or manual selection regarding data quality) is made. The table below reflects the appropriate action under differing runtime conditions and/or data quality conditions.
Standard PI Model
Tag Name Specifications
All Cluster Rule PI tags will begin with “CR-”.
Digital State Set Specifications
The following are the minimum required digital state sets in PI to support the Cluster Rule Bleach Plant model.
Scan Class Specifications
The following scan classes must be available in PI. Note, the scan class number will vary from mill to mill.
A one minute scan class offset 0 seconds from midnight;
A twenty-four hour scan class offset to the start of mill day.
Examples of the scan class syntax are as follows:
/f=00:01:00, 00:00:00 (alternately /f=00:01:00, 0)
/f=24:00:00, 07:00:00 (alternately /f=24:00:00, 25200) for mill day at 07:00 am
PI Totalizer Configuration
PI Totalizer tags are used to calculate daily totals from flow meters and production rate tags. In order to properly account for potential to emit status and percent good limits for the source tag, the following procedures are used to configure these totalizers:
Tables 10-1 and 10-2 provide tag configuration examples of performance equations for a typical condensate collection model. Tables 10-3 and 10-4 provide tag configuration examples of totalizers for a typical condensate collection model. Table 10-5 gives exception and compressions attribute standards for raw DCS PI tags.
6Running Status tag logic is to be defined so that any error conditions will default to the value of “Running” (final clause is else “Running”)
7If delaytime is not required, running status logic is used in the PTE tag and the running status tag is not needed. PTE Status tag logic is defined so that the default value is “CanNotEmit” (final clause is else “CanNotEmit”)
8Daily downtime tag logic, Y+7H refers to 7:00 am yesterday and T+7H refers to 7:00 am today for a mill-day rollover of 7:00 am (adjust for mill's actual rollover)
9Data Quality (CMS) tag logic is defined so that any error conditions will default to the value of “BAD” (final clause is else “BAD”)
10Bypass Divert tag logic is defined so that any error conditions will default to the value of “Divert” (final clause is else “Divert”)
11Lower User Specification Limit is written to PI ONLY when ASB Treatment method is used.
CR-Dig.Stat/CR-Evaps.Stat
cond_collection_tag.Filt
cond_collection_tag.DQ
pulp_production.PctGd
cond_collection_tag.PctGd
CR-Dig.Stat/CR-Evaps.Stat
cond_collection_tag.Filt
cond_collection_tag.DQ
pulp_production.PctGd
cond_collection_tag.PctGd
pulp_production.DAY
cond_collection.DAY
1Totalizes values for yesterday's MILL day. Example shows offset for mill day rollover at 7:00 am.
Standard Proficy Model
The Proficy model consists of input variables (PI inputs), calculated variables, stored procedures, and Visual Basic scripts (VB scripts). Variables for a typical Condensate Main Tank Collection system and descriptions of the stored procedures and the VB scripts are included below. Complete listings of the Stored Procedures can be found herein.
1The sampling offset is determined based upon the mill-specific start of day time. The offset value is the number of minutes from midnight to the mill start of day. Example shows mill day start at 7:00 am.
2Values written to PI
1The sampling offset is determined based upon the mill-specific start of day time. The offset value is the number of minutes from midnight to the mill start of day. Example shows mill day start at 7:00 am.
Specification Limits
Proficy has upper and lower specification limits that can be defined for every variable: entry limits, user limits, warning limits, and reject limits. The following descriptions define how Proficy uses these limits to trigger events and inform operators of impending events:
MeOH Test Data File Transfer
MeOH concentration test results are imported from the testing lab host via file transfer. New test data is stored in a delimited ASCII file in a specified folder on the testing lab host. Proficy's FTP engine, at a specified frequency, looks for new files with a name matching a specified mask in the designated folder on the remote host. When the FTP engine detects a new file, the file is moved from the host to the folder “\Proficy\Incoming” folder on the Proficy server. Similarly, Proficy import Model-79 continuously monitors “\Proficy\Incoming” every minute for a new data file. If a new file is found, the data is parsed and transferred as inputs to the stored procedure spLocal_CRCEvent_Data where it is processed. The data file structure consists of four fields: the data source ID (e.g., example: CR-AU-MT-HP-IN where AU=mill Id), date, timestamp, and test result. The stored procedure runs once for every record in the file, creates a production event (event number format—mmddhhmmss from the data's date/time), and records the data in the TESTS table while retaining the data's relationship to the event number. If processing is successful, the file is moved to the folder “\Proficy\Processed” and the file name appended with a timestamp designating the processing date/time. If processing is unsuccessful, the file is moved to the folder “\Proficy\UnProcessed” and timestamped.
Calculations
15 Day Total
Type: Stored Procedure—spLocal—15 DayTotal
This procedure looks at daily data for a specified variable (either Daily—MeOH Collected (Lbs) or Daily—Digester Tons Used for Avg (ODTP)) over the last 30-days and sums the most recent 15 daily values where the data quality is good (as specified by the data quality flag, Condensate Collection Data Quality (CMS)). Fifteen values are required before a total is calculated. Values with a timestamp that is not the mill-day rollover are excluded from the calculation.
Condensate EE Events
Type: Stored Procedure—spLocal_CondEvents
This procedure reads the value of the dependent variable (15 Day—MeOH Avg (Lbs/ODTP)) and compares it to the variable's specification limit, as specified in the calculation inputs (LW, LR, etc.). If this value is outside of the limit, then a 24-hr downtime event is created (appended if a contiguous event exists) on the variable's unit. The following table lists the possible status messages and their definition. This status message is displayed on the Autolog display as the variable Condensate EE Events.
Condensate CMS Events
Type: Stored Procedure—spLocal_CondCMSEvents This stored procedure creates a 24-hour downtime event that is triggered by a manual input from the operator (via Condensate Collection Data Quality (CMS) which is configured as the dependent variable). The following table lists the possible status messages and their definition. This status message is displayed on the Autolog display as the variable Condensate CMS Events.
Calc (A*8.34*B/1000000)
Type: Equation
Calculates the MeOH quantity (Lbs) from the MeOH concentration (A in ppm) and the condensate collection total (B in gals).
ManualUpdate
Type: Stored Procedure—spLocal_ManualUpdate
Fixed or LAB MeOH Conc
Type: Stored Procedure—spLocal_MeOHConcSelect
GetMeOHLabData
Type: Stored Procedure—spLocal_GetMeOHLabData
Uptime (Daily)
Type: Equation
Lower Warning Spec Limit
Type: Equation
Lower User Spec Limit
Type: Equation
Calc (A/B)
Type: Equation
AutoLog Displays
THIS IS THE BEGINNING OF lvhc hvlc
The purpose of this document is to describe the design of the record keeping and reporting system for the collection and destruction of Low Volume/High Concentration (LVHC) and High Volume/Low Concentration (HVLC) gases. The software is comprised of PI Data Archive software (which is used for automatic data collection from various process instrumentation and control systems) and Proficy software (which uses the data collected by PI in conjunction with manual inputs and business rules to monitor and report on the performance of the LVHC/HVLC collection and destruction system). This documentation is directed toward system administrator level personnel.
The following sections describe the general configuration of the standard LVHC/HVLC monitoring system. Deviations from the standard model, configuration listings for specific areas, and mill-specific details are contained in the appendices.
Low Volume/High Concentration (LVHC) and High Volume/Low Concentration gases from regulated sources (e.g., blow tanks, blow heat recovery, turpentine system, stripper off gas, diffusion washers, etc.) are collected by a closed vent system and treated by incineration in one or more of the following systems:
(a) Thermal oxidizer (incinerator),
(b) Power Boiler,
(c) Lime Kiln, or
(d) Flare.
The Proficy system is used to track both Excess Emission (EE) and Continuous Monitoring System (CMS) DOWNTIME events. Excess emission events occur whenever LVHC/HVLC gases (also referred to as Non-Condensable Gases) are vented to the atmosphere, when gases are inadequately treated, and when no destruction device is operating while gases are being produced. PI monitors the state of each potential emission source (e.g., vent valves, rupture disks, relief valves, loop seals, etc.) while accounting for the area's Potential to Emit status and triggers Proficy to record an event anytime gases are vented. The recorded event includes the event start time, end time and duration.12 12 As required by 40CFR §63.10(c). The regulations provide a non-SSM excess emissions allowance of 1% of operating time for the reporting period for LVHC systems before a violation is recorded (4% for HVLC systems) as stated in §63.443 (e).
PI tags also monitor the state of all destruction devices. The PI tags trigger instantaneous excess emission events in Proficy whenever any individual destruction device stops operating while regulated gas is directed to it (as determined by mill-specific process input signals). Destruction device excess emissions are recorded by the system whenever
The Proficy system also captures and records failures (downtime) of Continuous Monitoring System (CMS) devices, referred to as CMS events. LVHC/HVLC CMS events are created only for applicable destruction devices (thermal oxidizers) and only when the device is in use as a destruction device. The system records failures whenever the data signal
The CMS events are summarized individually for each applicable control device and reported separately to the state in a semi-annual CMS performance report or more frequently as required.
The Proficy software logs all excess emission and CMS events and operator responses to those events. The responses record the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) for the event. The report categorization specifies if the event is considered an allowable excess emission or CMS occurrence as the event may be allowed due to Startup, Shutdown, and Malfunction (SSM) provisions. The events for LVHC collection/treatment and HVLC collection/treatment are compiled separately by the system and reported separately to the state regulatory agency on a semi-annual basis or more frequently as required.
The PI system also calculates and makes available to Proficy a “Daily Down Time” which is the time that the processes capable of producing regulated HAPs are not operating. Proficy, in turn calculates the process uptime. The total Daily Uptime for the reporting period becomes the denominator in determining if the mill has exceeded the excess emission allowance for the reporting period.
Additionally, the PI system calculates the time each day that each LVHC CMS device (thermal oxidizer) is not used to treat gases (Daily Downtime). Proficy uses this daily calculation to calculate the thermal oxidizer uptime, which becomes the denominator in determining if the mill has exceeded the CMS allowance for the reporting period.14 14 As required by 40CFR §63.454(b)(11)-(12) and §63.10(c)-(e). HVLC and LVHC CMS downtime is calculated and reported as a percentage of source runtime.
Events and TCC answers are recorded within the Proficy system. On a periodic or scheduled basis, mill environmental personnel can run reports listing the events (start time, end time, and duration) and their TCC answers, summarizing the total duration of all events by specific report code, and calculating excess emissions and CMS downtime against the allowances. The reports are run from Microsoft Excel using an Excel VBA add-in specifically written and designed to generate environmental reports which meet the regulatory reporting requirements.15 15 The regulatory record keeping and reporting requirements are codified in 40 CFR §63.6(e)(3), §63.8(c)(1), and §63.10.
For HVLC and LVHC reporting simultaneous excess emission events answered with different report codes are allotted time in the report summaries according to the following report hierarchy:16 16 The report code hierarchy is from top to bottom; that is, if one event is categorized Other Known Causes and a simultaneous event is categorized Process Problems, the event time is allocated and summarized as towards Other Known Causes.
1. Other Unknown Causes
Additionally a sixth report code, No Excess Emission, eliminates an event from inclusion in the report categorization hierarchy and indicates that the event was recorded by the system in error. When this report code is utilized, the user must have appropriate documentation that the event was created in error and that no excess emission occurred. The single event will be excluded from the report summarization but concurrent events, either unanswered or with different report codes will be included in the report summary.
For excess emission events that contain incomplete or missing TCC answers, the report system allocates the event time to either Other Unknown Causes (in the case that there was no simultaneous event answered) or to the report code category of simultaneous events following the hierarchy above.17 17 When the only existing simultaneous event is answered No Excess Emission, the unanswered event is categorized as Other Unknown Causes for the purposes of report code summarization.
For HVLC and LVHC reporting, simultaneous CMS events answered with different report codes are allotted time according to the following report hierarchy:
Additionally a sixth report code, No Monitor Downtime, eliminates an event from inclusion in the report categorization hierarchy and indicates that the event was recorded by the system in error. When this report code is utilized, the user must have appropriate documentation that the event was created in error and that monitoring of the thermal oxidizer was maintained. Unanswered events (or events with incomplete answers resulting in a missing report code) are categorized as Other Unknown Causes from a report summarization standpoint.
Tables-16, 17, & 18 give PI tag naming conventions and description for typical variables used in the standard model.
The following sections describe in detail how the Proficy /LVHC/HVLC model triggers EE and CMS events.
Mill Area State—PTE
Each area that produces regulated gases has a potential to emit (PTE) performance equation tag in PI. The performance equation logic returns a state of “CanEmit” when HAPS are present, and can potentially be emitted when a vent valve is opened. This is normally during the period from startup of the area until a mill specified period after the area stops running, and regulated gases have been cleared from all areas of the system. The area PTE state is calculated each minute based upon mill specified, site specific criteria such as flow, motor running state or pump running state.
A block diagram of potential to emit and daily downtime/uptime data flow is depicted in
Vent Source EE
For each mill area that can produce regulated gases, PI receives raw DCS states for all of the possible emission points. Generally these are digital tags that give the state of the valve (open or closed) or rupture disk (if the source is a modulating valve, PI receives an analog value from the DCS that represents % open). An event-based performance equation, CR-devicename.VENT, determines when an emission point is venting to the atmosphere while the mill area has a potential to emit. This PI performance equation returns a value of “Vent” or “NotVent.”
Vent Source EE events are created by Proficy using the Proficy downtime model 200. Proficy monitors the digital tag, CR-devicename.VENT, for the fault value of “Vent”. Whenever the value enters the fault state (or remains in the fault state for a period longer than a specified filter time), an EE event is recorded by the system.
A block diagram of the vent data flow is depicted in
Main Vent Filtering (Optional)
Main vent filtering is an additional configuration to each “.Vent” tag that attempts to reduce the number of events that operators must answer when a system-wide event occurs. All upstream vents points are filtered out in PI whenever the Main Vent tag has a value of “Vent.” The main vent is defined as the last vent before the gases are routed to the destruction devices (vents at the destruction devices are not main vents).
A PI performance equation, CR-MainVent.Filt, outputs “Venting” when the main vent is “Venting” and continues to output “Venting” for a mill determined time after the main vent returns to “NotVenting.” This delay is intended to give operators time to close the upstream vents after they have closed the main vent. Each upstream vent tag is set to “NotVenting” whenever the CR-MainVent.Filt tag has the value of “Venting”.
Destruction Device Treating LVHC/HVLC Gases and EE
For destruction devices, a PI performance equation, CR-devicename.TREAT, determines if the device is accepting gases by verifying the correct operating conditions (operating flow, motor running, operating pressure or operating temperature), and that the appropriate valves are in position for LVHC/HVLC gases to be directed to the device. For thermal oxidizer devices (incinerators), where a burner management system (BMS) is connected to PI, the preferred running indicator is the BMS “Ready to Accept Gases” tag. The destruction device treating status is required for thermal oxidizer devices to determine running time for the CMS device on the reports. For other types of destruction devices, the device treating status is for other purposes. A block diagram of destruction device treatment status data flow is depicted in
Destruction Device CMS
Mills that utilize a thermal oxidizer as a destruction device for LVHC/HVLC gases must monitor the temperature of the incinerator and report CMS downtime whenever the incinerator is in use and the flame temperature sensor can not be read by PI. A PI performance equation, CR-devicename.DQ, calculates the value “Bad” when the temperature is out of range, or is in an error state, and the incinerator is selected for treatment; otherwise the value “Good” is calculated.
Proficy uses the Proficy downtime model 200 to monitor CR-devicename.DQ for the fault state, “Bad”. Whenever the PI tag value “Bad” is detected (or remains for a period longer than a specified filter time), a CMS event is recorded by the system.
A block diagram of destruction device EE and CMS data flow is depicted in
CMS Runtime Counter
At the start of each mill day, a PI performance equation, CR-Incin_Down.Day, totals the “NotTreating” time for CR-Incin.TREAT over the previous 24-hour period. This value is read by Proficy and is used for both the daily display and daily calculation of Incinerator runtime (“Treating” for the daily period). The daily runtime minutes are kept in Proficy and used to compute the total incinerator treating runtime minutes for the reporting period.
LVHC/HVLC PTE Downtime Counter
Every minute, a PI performance equation, CR-LVHC_PTE.STAT/CR-HVLC_PTE.STAT, looks at each mill area PTE tag, CR-millarea_PTE.STAT. If any one mill area tag has a value of “CanEmit”, the equation returns the digital state “CanEmit”. If all of the mill area tags have a value of “CanNotEmit”, the equation returns the digital state “CanNotEmit”. At the start of each mill day, another PI performance equation, CR-LVHC_Down.Day/CR-VLC_Down.Day, totals the “CanNotEmit” time for CR-LVHC_PTE.STAT/CR-HVLC_PTE.STAT over the previous 24-hour period. This value is read by Proficy and is used for both the daily display and daily calculation of LVHC/HVLC runtime (“CanEmit” for the daily period). The daily runtime minutes are kept in Proficy and used to compute the total runtime minutes for the reporting period.
A block diagram of potential to emit and daily downtime/uptime data flow is depicted in
Tag Name Specifications
Digital State Set Specifications
The following are the minimum required digital state sets in PI to support the Cluster Rule
Scan Class Specifications
The following scan classes must be available in PI. Note, the scan class number will vary from mill to mill.
Examples of the scan class syntax are as follows
Tables 19 and 20 provide tag configuration examples for a typical LVHC/HVLC model. Table 21 contains exception and compression statistic requirements for underlying DCS PI tags.
18Define logic so that else “Running” (or “Treating”) is the final clause, so that any error conditions will result in a default value of “Running” (or “Treating”)
19Define logic so that (else “Vent”) is the final clause, so that any error conditions will result in a default value of “Vent”
20Define logic so that (else “Bad”) is the final clause, so that any error conditions will result in a default value of “Bad”.
21Where possible, use event scheduled PE tags for EE and DQ tags. This will help guarantee that PE calculations are performed shortly after the underlying process values change. For time based PE tags, take care in assigning scan classes so that undue delays are not incurred waiting for multiple passes through the PE scans.
22Y+7H refers to 7:00 AM Yesterday & T+7H refers to 7:00 AM today (used when the mill start of day = 7:00)
1. VI. Standard Proficy Model
The Proficy model consists of input variables (PI inputs), calculated variables and equations. Variables for a standard LVHC/HVLC collection system and descriptions of EE and CMS event logic are included below.
4201
1The sampling offset is determined based upon the mill Start of Day time. The offset value is the number of minutes from midnight to the mill start of day. In this example the start of day is 7:00 AM (as there are 420 minutes from midnight until 7:00 AM).
EE Event Logic
Emission source EE events are generated whenever a vent valve, rupture disk or other valve type opens to the atmosphere while the gas source(s) is operating as determined by PI. Likewise, destruction device EE events are generated whenever the device is not running and regulated gases are being directed to the device. Overlapping intervals from two or more EE events is counted as a single event for the duration of the overlap. Overlapping events are resolved at report creation by the report package and not by the Proficy or PI models. EE events are detected using downtime model-200, with an optional, mill specific delay filter.
CMS Event Logic
CMS downtime events are monitored only for particular destruction devices. Typically, the trigger is a PI tag that tests thermal oxidizer device temperature data quality. CMS events are detected using downtime model-200, with an optional, mill specific delay filter.
Describe interaction between DCS/PI/Proficy as a part of the general overview. See Figures below.
The purpose of this document is to describe the design of the Continuous Monitoring System for foul-condensate treatment monitoring operations. Specifically, the design of the monitoring system for a Steam Stripper column adhering to the 92% compliance option, as specified in §63.446(e)3, is addressed by this document.
The software is comprised of PI Data Archive software (which is used for automatic data collection from various process instrumentation and control systems) and Proficy software (which monitors and reports compliance based on the PI data and operator inputs). This documentation is directed toward system administrator level personnel but can be used as a basic understanding of how the system works.
The following sections describe the general configuration of a standard steam stripper monitoring system following the 92% efficiency option. Deviations from the standard model, configuration listings for specific lines, and mill-specific details are contained within the appendices.
Foul condensate is collected in a central collection tank (also referred to as a Main Foul Condensate Collection Tank or Stripper Feed Tank) from sources such as digesters, evaporators, and turpentine systems. From this tank, the condensate is usually heated in a stripper condensate pre-heater heat exchanger using hot, stripped condensate, before being fed to the steam stripper column. Strippers that operate at a vacuum and low temperatures may not have a pre-heater.
Typically, low pressure steam is used to strip the MeOH out of the foul condensate however strippers can also use medium pressure steam, steam generated from a condensate re-boiler, or evaporator vapor to strip the MeOH from the condensate. The steam flow carries the vaporized MeOH out of the column to a reflux condenser that condenses most of the water vapor out of the MeOH/water vapor stream leaving the column. The concentrated MeOH vapor is often called stripper off gas (SOG). Hydrogen sulfide and other total reduced sulfur (TRS) compounds will be stripped along with the MeOH and are found in high concentrations in the SOG. The SOG is sent to an incinerator, boiler, or kiln where it is incinerated for disposal. The stripped condensate is collected in the bottom of the steam stripper and usually sent through the condensate preheater to heat the incoming condensate to within about 20° F. of the stripper column operating temperature. After exiting the pre-heater, the stripped condensate is either sent to a sewer or is used back in the process.
Most steam stripper pre-heaters are designed to heat the inlet condensate temperature to within 20° F. of the outlet stripped condensate temperature. If the feed temperature is colder than design, more steam is consumed to preheat the condensate in the column, leaving less steam to actually strip (e.g. reducing the effective stripping steam). In this case, the total stripping steam required to accomplish the same degree of methanol removal should be increased to offset the portion of applied steam needed to further heat the colder incoming condensate. In general, the main reason why the inlet and outlet temperature gap widens over time is due to fouling of the condensate pre-heater.
One of the treatment options for a steam stripper is to remove or strip 92% of the MeOH in the condensate entering the stripper. The efficiency of a steam stripper to remove MeOH correlates to the ratio of effective steam flow to condensate flow in the stripper. The minimum effective steam ratio to maintain a minimum 92% MeOH removal efficiency is established by the mill during a Performance Test and used as a lower limit to determine excess emissions events.
Typically, 0.2 lbs of low pressure steam is needed for stripping to achieve 92% methanol removal per pound of foul condensate. This ratio can be expressed as a percentage, such as 20%. For a given condensate flow, inlet temperature, and MeOH concentration; the steam to foul condensate feed flow ratio is fairly constant to achieve a specific methanol removal. Some of the steam fed to the stripping column is condensed to heat the incoming foul condensate to the boiling temperature at the stripper operating pressure. The heating occurs quickly in the first feed tray of the column. About 0.001 pound of steam is needed to heat 1 pound of condensate by 1° F., or about 0.02 pounds of steam (0.02%) to heat the foul condensate 20° F. The steam that is actually doing the work to strip the methanol out of the condensate is referred to as effective steam. Every 10° F. drop in feed temperature takes 1% off the effective steam flow ratio. The effect steam ratio needed to get greater than 92% removal is approximately 0.18.
Five parameters are required to compute the effective steam ratio, which includes the three parameters required by §63.453(g):
Effective Steam Ratio (ESR) is computed as the ratio of effective steam flow divided by the foul condensate flow, or:
For example, assuming:
yields an effective steam ratio of
As the effective steam flow ratio drops below its target, the operator can either increase steam flow to get the effective steam back up to its target level, or can reduce flow to the stripper at the same steam flow to restore the effective steam flow ratio target. The later method may result in slowing back production, or may risk sewering too much condensate per the collection requirements.
These variables are collected and archived by the PI system and made available to the Proficy system to analyze against specific criteria to determine if an Excess Emission (EE) event has occurred. Two types of excess emission events can occur during the operation of a 92% Steam Stripper system: a low 3-hour rolling average stripper efficiency event (3-hour rolling average excess emission event) and a stripper bypass event (stripper excess emission bypass event). Excess emission events for steam stripper treatment are recorded by the system whenever:
The steam stripper system has a 10% allowance against condensate runtime for all excess emission events, as specified in §63.446(g).
The steam stripper treatment system has the potential to emit pollutants whenever the condensate sources are operating. Therefore the potential to emit runtime for the steam stripper system corresponds to the condensate collection operating time reported to the state regulatory agency on a semi-annual basis or more frequently as required.
The PI system computes an effective steam flow and effective steam ratio every minute (CR-SS_EFFSteam.Filt and CR-SS_ESRatio.Filt) from the four parameters above (using 1000 as an enthalpy constant). The effective steam flow calculation flow is clamped at zero in the PI tag (CR-SS_ESRATIO.RAW). Additionally every fifteen minutes the related PI tag (CR-SS_ESRatio.Filt) compute what percentage of time the data quality of the effective steam ratio was good over the fifteen minute interval.
The following table 22 gives an overview of the minimum required process inputs, their engineering units, associated PI tags, and corresponding Proficy variable names.
23To complete the effective steam ratio calculation Condensate Feed Flow must be expressed in lbs/hr. To convert condensate flow to lbs/hr, multiply the flow rate (in gal/min) by 8.35 * 60.
Data quality limits for the Bottom Temperature, Feed Temperature, Condensate Flow, and Steam Flow are maintained in PI. These data quality limits are used by a PI performance equation to determine if the PI process value has “Good” or “Bad” signal quality and contribute to the overall data quality of the Effective Steam Ratio calculation. Anytime that the data quality of the four parameters results in a failure of the system to reliably calculate an effective steam ratio for the fifteen minute interval, the system records a Continuous Monitoring System (CMS) event (explained in detail below).
The following sections describe in detail how the PI/Proficy steam stripper model computes effective steam and triggers Steam Stripper EE and CMS events.
a) Steam Stripper PTE and Total Runtime
In general, the steam stripper treatment system has a potential to emit pollutants whenever the condensate collection system or steam stripper column is operating. Specifically the steam stripper treatment system has three distinct potential to emit (PTE) conditions. First the steam stripper has potential to emit pollutants whenever it is operating (usually determined by a minimum flow on a flow meter and a “Running”/“NotRunning” indicator on the stripper). Under these conditions emissions occur whenever the 3 hour rolling average of stripper efficiency falls below 92%. Secondly steam stripper treatment emissions can occur whenever the condensate system is operating while the stripper is not operating. Under this condition overflows of the collection tank or foul condensate diverts upstream of the collection tank are considered steam stripper treatment excess emissions. Lastly emissions can occur if the foul condensate is present in the stripper feed tank (indicated by tank level) irrespective of stripper or condensate system operating status. In this case, emissions occur if the foul condensate is pumped out of the feed tank and towards a non-treated collection point (such as to sewer or through the column when steam is not present).
The steam stripper system has a 10% emission allowance against source (condensate collection) operating time for all types of emissions. The total number of runtime minutes used to calculate this emission allowance corresponds to the total number of runtime minutes for the condensate collection system over the same period of time.
The steam stripper is considered to be treating when the following three conditions are met:
On a daily basis, a PI tag (CR-SS_Treat.Day) computes the total number of minutes that the steam stripper was down during the previous production day. This value is used by Proficy to compute the daily total number of runtime minutes of the steam stripper system. These daily totals are used by the reporting system to compute the total number of steam stripper runtime minutes over the reporting period.
b) Computing the Effective Steam Ratio
The effective steam ratio is computed in PI each minute based upon the process data for the four process parameters above from the mill DCS system (and assuming a constant of 1000 for enthalpy). The raw value for each of the four parameters necessary to compute effective steam is first validated in PI against upper and lower specification limits (maintained in PI). If the value is within range (and not flat-lined), PI records the value into an intermediate variable (CR-SS-FeedSteam.FILT, CR-SS-CondFlow.FILT, CR-SS-BottomTemp.FILT, and CR-SS-CondTemp.FILT); if the value is out of range the intermediate variable records “Bad” instead. These PI performance equations are event based (calculated every time a new value enters the PI snapshot) so that data buffered in the PI interface will be captured and used regardless of its PI archive status. If all four process values exhibit good data quality, PI uses the intermediate values to compute an effective steam flow for the minute. PI stores this value in the PI tag CR-SS_EFFSTEAMFilt. It is possible for the effective steam flow calculation to have a negative result during times of stripper upsets. Since negative values are theoretically impossible and can cause long periods of low three hour averages, the PI tag, CR-SS_ESRATIO.RAW clamps the effective steam ratio to 0 whenever the tag CR-SS_EFFSTEAMFilt has a negative value. If all four process values exhibit good data quality, PI outputs the value from CR-SS_ESRATIO.RAW to Proficy.
c) Steam Stripper CMS Events
When the steam stripper system is running, failures to calculate the effective steam ratio of the stripper result in Continuous Monitoring System (CMS) events. Every fifteen minutes, Proficy computes a time-weighted average of the effective steam ratio calculation (CR-SS-SS_Ratio.Filt) over the previous fifteen minutes and stores the value in the variable Effective Steam Ratio—15 Min PI Avg. At the same time Proficy examines the CR-SS_ESRatio.PctGd tag to determine if CR-SS_ESRatio.Filt maintained “Good” data quality during at least 50% of the fifteen-minute period. If so, the computed fifteen-minute average is copied into the Proficy variable Effective Steam—15 Min Qualified Avg. If any fifteen-minute period fails to meet the 50% criteria while the SS CMS Treating Status (15 min) value is Treating, Proficy instead creates a fifteen-minute steam stripper CMS downtime event (via a stored procedure calculation) for the period or it appends fifteen-minutes to an existing CMS event (if a contiguous CMS event already exists).
Every fifteen-minutes Proficy also reads and displays the data quality results (% Good) for each parameter required to complete the effective steam ratio calculation. These data-quality values assist the operator in determining which of the four signal(s) was (were) responsible if the effective steam ratio average could not be calculated (resulting in the CMS event). These values are displayed on the steam stripper display for diagnostic purposes but do not, by themselves, create CMS events.
The Proficy software logs all Steam Stripper CMS events and operator responses to those events. The responses record the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) of the event. The events are compiled, measured against the stripper operating time for the reporting period, and reported to the state regulatory agency on a semi-annual basis or more frequently as required.
d) Steam Stripper Excess Emission Events
Two types of excess emission events can occur during the operation of a 92% Steam Stripper system: a 3-hour rolling average excess emission event and a stripper bypass excess emission event.
Once per hour, Proficy examines all fifteen-minute qualified averages (Effective Steam Ratio—15 Min Qualified Avg) during the previous three-hour period. If greater than 50% of the averages exist and have good data quality, Proficy computes a 3-hour rolling average effective steam ratio (Effective Steam Ratio—3 HR Avg.) from all fifteen minute averages exhibiting Good data quality. This computed three hour average is compared against a lower limit (lower specification warning limit on the Effective Steam—3 HR Avg. variable) and if the value falls below the limit and the SS EE Treating Status (1 Hr) value is Treating, Proficy creates a one-hour Excess Emission downtime event or, in the case that a previous contiguous excess emission event existed, it appends one-hour to the existing event (via a stored procedure calculation). The value is also written back to the PI tag CR-SS_ESRatio.3H for trending within the mill.
No manual entry of steam stripper effective steam ratio is available in the system.
The Proficy software logs all Steam Stripper Rolling Average EE events and operator responses to those events. The responses record the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) of the event. All report categorizations except No Excess Emission are totaled and reported to the state regulatory agency on a semi-annual basis, or more frequently as required, against the 10% steam stripper emission allowance.
In addition to 3-hour rolling average excess emission events, a steam stripper column also incurs excess emissions if condensate bypasses the stripper column prior to or without treatment while the condensate collection system is operating or during stripper downtime if previously collected condensate is diverted to a non-treated collection point (such as sewer).
The five types of PI calculations used to monitor steam stripper bypass excess emissions are described below. All of the following PI performance equations are evaluated at least once a minute and are monitored by the Proficy system using model 200 (with a mill specific filter applied). Proficy creates an (EE) event for each minute that the PI performance equations' value is E.
When the tank level is greater than a maximum while the condensate system is operating and the stripper is not running, the performance equation records the minute as a steam stripper bypass excess emission.
When a bypass value located after the tank outlet and before the condensate collection boundary (flow meter) is open (bypassing) while the condensate system is operating and the stripper is not running, a performance equation records the minute as a steam stripper bypass excess emission.
When the condensate system is operating, the stripper is not running, the main collection tank is not overflowing and all bypasses after the main collection tank outlet are not bypassing, a performance equation checks for any overflows or diverts upstream of the main collection tank. If any upstream diverts occur under the described conditions, the performance equation records the minute as a steam stripper bypass excess emission.
When the stripper is not running, a PI performance equation examines the state of any bypass valves past the collection tank (flow meter) but prior to the stripper column to determine if collected condensate is being directed away from the steam stripper column, which is recorded as a steam stripper bypass excess emission.
When the condensate system is operating and the stripper is not running, a performance equation monitors the foul condensate feed flow to the column. If the condensate feed flow is greater than a minimum value, the performance equation records the minute as a steam stripper excess emission.
The Proficy software logs all Steam Stripper EE Bypass events and operator responses to those events. The responses record the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) of the event. The steam stripper system has a 10% allowance against the overall condensate system runtime period. Events categorized as No Excess Emission are excluded from this calculation however all other report codes are included in it. The events are compiled and reported to the state regulatory agency on a semi-annual basis or more frequently as required.
Should a bypass event occur simultaneously during the period when the steam stripper three-hour effective steam rolling average falls below the minimum effective steam limit, only one hour of excess emissions will be reported by the reporting system. That is, in any 24-hour period, there can be no more than 24 hours of total stream stripper excess emissions.
Table-23 gives the process inputs required for a typical steam stripper system, their engineering units, data source, and corresponding Proficy variable names.
Table-24 lists typical calculated variables for the system and a brief description of each.
e) Tag Name Specifications
f) Digital State Set Specifications
The following are the minimum required digital state sets in PI to support the Cluster Rule Steam Stripper 92% model.
g) Scan Class Specifications
The following scan classes must be available in PI. Note, the actual scan class number will vary by location.
h) PI Tag Configuration Specification
The following tables 25 to 29 provide the typical PI tags (and their configuration) required for a Steam Stripper Treatment system following the 92% treatment methodology and standard exception and compression attribute values for mill specific DCS PI tags.
24The Condensate PTE tag (CR-Cond-PTE.STAT) is available and displayed in the Condensate Collection system. A unique tag for Steam Stripper treatment is not required.
25This example assumes that the raw flow is expressed in M-gpm (1000's gal/min). Subsequent calculations require that the units of condensate flow (gpm) and feed steam rate (lbs/hr) match. To convert the condensate flow (in gpm) into lbs/hr, multiply the flow by 8.35*60. If flow is expressed in M-gpm, the conversion factor is further divided by 1000.
2Calculated in Proficy and written periodically to PI.
3Maintained in Proficy as a Specification Limit and written periodically from Proficy to PI
The Proficy model consists of input variables, calculated variables, stored procedures, and Visual Basic scripts (VB scripts). Variables and associated parameters for a typical 92% steam stripper treatment system and descriptions of the stored procedures and the VB scripts are included below. Complete listings of the Stored Procedures can be found herein below.
42026
26The sampling offset is determined based upon the mill Start of Day time. The offset value is the number of minutes from midnight to the mill start of day. In this example the start of day is 7:00 AM (as there are 420 minutes from midnight until 7:00 AM).
4201
Proficy Calculations
Steam Stripper Rolling Average Excess Emission events are created in one hour increments using the stored procedure spLocal_SSTreatmentEvents as described below. Bypass events (Treatment Bypass, Overflow bypass, Flow Emissions and Upstream Bypass) EE events are created using Proficy's downtime model 200 with a 61-second (or other mill specific) filter applied. The PI tags, CR-SS_Overflow.EE, CR-SS_CondDivert.EE, CR-SS_FlowEmissions.EE and CR-SS_Bypass.EE, trigger the start of an event whenever their state changes from OK (the normal running state) to EE (the fault state). The event ends when the state changes back to OK. If the state returns to OK within the filter period the state changes are ignored and no event is created in Proficy.
CMS event logic
CMS events for the Steam Stripper treatment system are created in fifteen minute intervals as described below in the stored procedure spLocal_SSTreatmentEvents. There are no CMS events associated with collection tank overflows or treatment bypass valves.
SS Qualified 15 Min Avg
Type: VBScript
The inputs to this script are the SS CMS Treating Status (15 Min), the 15 minute raw PI average of Effective Steam ratio, the percent good value for the effective steam calculation over the fifteen minute window, and a lower reject specification limit attached to the percent good variable. This script is triggered by time (based on the sample interval for the variable—normally 15 minutes) or an input value change. This script filters the 15 minute average based on the 15 minute average Treating status and the percent good value for the average. If the percent good value is greater than required (lower reject limit) and the Treating status is Treating, the script outputs the average value for the period. If the Treating status is NotTreating or the percent good value is less than required, this script outputs a null value.
SS Qualified 15 Min Avg Status
Type: VBScript
The inputs to this script are the effective steam 15 minute percent good value, the lower warning limit for percent good, the SS CMS Treating Status (15 Min) and the raw PI effective steam ratio 15 minute average. This script is triggered by time (based on the sample interval for the inputs) or an input value change. This script outputs the status of the Eff Steam Ratio 15 Min Avg (Status) for display on the Autolog display. If the Treating status is NotTreating, this script outputs Unit Down. If the percent good value is greater than 50% and the Treating status is Treating, this script outputs OK. If the percent good value is less than 50% and the Treating status is Treating, the script outputs Bad Val.
SS 15 Min Avg/Status Reassembly
Type: VBScript
The inputs to this script are the Eff Steam Ratio 15 Min Avg and the Eff Steam Ratio 15 min Avg (Status). This script is triggered by time (based on the sample interval for the inputs) or by an input value change. This script combines the two inputs into one string value based on the string value of the Eff Steam Ratio 15 Min Avg (Status). If the Eff Steam Ratio 15 Min Avg (Status) is OK, this script outputs the Eff Steam Ratio 15 min Avg. If the Eff Steam Ratio 15 min Avg (Status) is Unit Down or Bad Val, this script outputs Unit Down or Bad Val.
spLocal_StripperRollingAvg_wAvgPTE
Type: Stored Procedure
This procedure has two inputs, the percent good value for effective steam ratio and the SS EE Treating Status (1 Hr), and one dependant variable, the reassembled 15 min avg/status for the effective steam ratio. This procedure calculates a 3 hour moving average of the dependant variable every hour from a minimum number of samples over the 3 hour interval. The requirement for a good average is that there must be more than 50% good samples. “Good” samples consist of valid numeric values taken while the EE Treating Status (1 Hr) is Treating and the percent good value is greater than 50%, as determined by the 15 min avg/status reassembly VB script. Values of Bad Val, Unit Down and NULL are excluded from the moving average. The triggers for this procedure are time (based on the sample interval for the variable), value change for the dependant variable or value change for the input variable.
spLocal_StripperRollingAvgStatus_wAvgPTE
Type: Stored Procedure
This procedure has four inputs (the percent good value for the effective steam ratio, the lower warning limit for this variable, the effective steam ratio 3 Hr Rolling Avg and the SS EE Treating Status (1 Hr)) and one dependant variable (the reassembled 15 min avg/status for the effective steam ratio). This procedure generates a status string to compliment the 3 hour moving average calculation, spLocal_StripperRollingAvg_wAvgPte. The following table 30 shows the possible outputs for this procedure and the sample types required to generate them.
The triggers for this procedure are time (based on the sample interval for the variable), value change for the dependant variable or value change for the input variable.
spLocal_SSTreatmentEvents
Type: Stored Procedure
This procedure is used to create CMS and EE events for the effective steam ratio. The inputs variables and dependant variables for both CMS and EE are shown in the following table 31.
This procedure tests for CMS or EE events by comparing the dependant variable value against a lower warning specification limit as specified in the calculation input. If the value is above the lower warning limit, a downtime event with duration as specified in the inputs is created. If an event exists for the previous time interval, the duration is appended to the existing event and the event end time is updated. The triggers for this procedure are time (based on the sample interval for the variable), value change for the dependant variable or value change for the input variable.
The purpose of this document is to describe the design of the Continuous Emissions Monitoring System for Bleach Plant Scrubber monitoring operations. The software is comprised of PI Data Archive software (which is used for automatic data collection from various process instrumentation and control systems) and Proficy software (which uses the data collected by PI in conjunction with manual inputs and business rules to monitor and report on the performance of the scrubbing process). This documentation is directed toward system administrator level personnel.
The following sections describe the general configuration of the standard bleach plant monitoring system. Deviations from the standard model, configuration listings for specific lines, and mill-specific details are contained within the appendices.
Cluster Rule regulations require that a continuous monitoring system (CMS) be operated to measure the following parameters for each bleach plant gas scrubber:
The data for these variables are collected and archived by the PI system and made available to the Proficy system to analyze against specific criteria to determine if an Excess Emission (EE) event has occurred. Excess Emission events, are recorded by the system when the bleach plant has the potential to emit (PTE) pollutants, as defined in PI, and one of the three monitored parameters does not meet the specified operating criteria. A bleach plant has the potential to emit pollutants when it is running or has been shutdown for less than a specified period of time (typically one hour) as defined by each facility. The potential to emit corresponds to the total source operating time reported to the state regulatory agency on a semi-annual basis or more frequently as required.
The Proficy software logs all events and operator responses to those events. The responses record the operator determined Trouble, Cause, Correction (response), and Report Code (report categorization) for the event. The report categorization specifies if the event is considered an excess emission, as the emission may be allowed due to Startup, Shutdown, and Malfunction (SSM) provisions. The events are compiled by the system and reported to the state regulatory agency on a semi-annual basis or more frequently as required. In addition to capturing and categorizing events, the Proficy system also captures and records failures (downtime) of Continuous Monitoring System (CMS) devices, referred to as CMS events. The system records failures whenever the validity of the data is suspect or out of range. These are also summarized and reported to the state in a semi-annual CMS performance report or more frequently as required. Again, the report categorization specifies if the event is considered allowable based on the specific regulations.
The following table 32 gives an overview of the minimum required process inputs, their engineering units, associated PI tags, and corresponding Proficy variable names.
Proficy also maintains, and periodically writes to PI, the specification limits (upper data quality limit, lower data quality limit, and excess emission limits) for the pH/ORP and recirculation flows. The data quality limits are used by PI to determine if the PI data has “Good” or “Bad” data quality while the excess emission limit is used by Proficy to determine when excess emission events occur.
The following sections describe in detail how the Proficy bleach plant model triggers EE and CMS events.
Bleach Plant State—PTE
The bleach plant's potential to emit (PTE) is determined in PI using a performance equation. The performance equation logic returns a state of “CanEmit” during the period from startup of the bleach plant until a mill specified period after the bleach plant stops running. The bleach plant run-state is calculated each minute based upon mill specified criteria—typically CLO2 flow, motor running state, or pump running state.
A PI performance equation translates the digital PTE status into a numeric value with 0=“CanNotEmit” and 100=“CanEmit”. Proficy averages this numeric PTE value of a specified time period and compares the average to a mill specified limit (usually 50) to determine if the bleach plant had a potential to emit over the desired time period. The Proficy variable, BP % Time CanEmit (15 min), averages the numeric PTE status over the previous 15 minute period. If this average is greater than or equal to the lower warning specification limit for this variable, the status is “CanEmit”. If the average is less than the lower warning specification limit, the status is “CanNotEmit”. This PTE status is used by Proficy to qualify the 15 min pH (or ORP) and recirculation flow 15 minute averages and to filter out CMS events when the status is “CanNotEmit”. The Proficy variable, BP % Time CanEmit (1 hr), averages the numeric PTE status over the previous hour. If this average is greater than the lower warning specification limit for this variable, the status is “CanEmit”. If the average is less than or equal to the lower warning specification limit, the status is “CanNotEmit”. This PTE status is used by Proficy to qualify the three hour rolling average.Bleach Plant PTE Counter
At the start of each mill day, a PI performance equation totals the “CanNotEmit” time over the previous 24-hour period. This value is read by Proficy and is used for both daily display and daily calculation of bleach plant runtime (“CanEmit” for the daily period). The daily runtime minutes are kept in Proficy and used to compute the total runtime minutes for the reporting period.
Recirculation Flow
Bleach Plant scrubber recirculation flow is read by PI from the mill DCS system. The raw value is first validated in PI against the upper and lower specification limits provided by Proficy. If the value is within range PI records the value in an intermediate variable (CR-BP-ScrubRecirc.FILT); If the value is out of range the intermediate variable records “BAD” instead. This PI performance equation is event based (calculated every time a new value enters the PI snapshot) so that data buffered in the PI interface will be captured and used regardless of its PI archive status.
Every 15 minutes, Proficy uses the filtered values to calculate a flow average over the 15-minute interval. Values marked “BAD” by PI are excluded from the calculated average.
The PI system also calculates a data quality metric that provides Proficy with the information it needs to determine whether the measurement of the recirculation flow is reliable. The metric is determined within PI by examining the percentage of time over the 15-minute interval that the recirculation flow data has maintained “good” data quality. This same calculation tests for a flat-lined signal over an extended period of time and calculates a “% Good” of zero if the signal value has remained unchanged. Proficy samples this “%-Good” value every 15-minutes and generates a 15-minute CMS downtime event (via a stored procedure calculation) whenever the percentage falls below 50% within the 15 minute period.
Recirculation flow EE events are triggered based upon a 3 hour rolling average calculation performed within Proficy. Once per hour, a stored procedure (spLocal_BleachRollingAvg) averages the previous twelve 15 Minute Averages for flow rate over the previous 3-hour window (3 Hr Rolling Avg). If the 3-hour average value is less than the lower warning specification limit configured in Proficy, a I-hour EE event is generated by the stored procedure “spLocal_BleachEvents”. This 3-hour rolling average calculation excludes averages within periods that reflect a “%-Good” less than 50%, that had no Potential To Emit, and that contained NULL values. Therefore for a 3 Hour Average to be calculated and an EE Event to be created, a minimum of seven valid 15 Minute Averages (>50%, or 7/12) must be present in the 3 hour window.
Each 15-minute flow average (“Qualified 15 Min Avg”) is accompanied by a corresponding status message (“Qualified 15 Min Avg Status”) that is set to “OK” upon successful calculation of the average. Similarly the 3-hour rolling average has an equivalent variable (“3 Hr Rolling Avg Status”) that provides the status regarding calculation of the 3 hour rolling average. The status messages and their meanings are summarized in the tables 33 and 34 below.
A block diagram of scrubber recirculation data flow is depicted in
pH/ORP
The monitoring of pH/ORP is exactly analogous to that for recirculation flow except that a manually entered pH or ORP value can override the Proficy calculated 15 minute average. Additionally ORP measurements are compared to an upper warning specification limit as opposed to a lower warning limit specification used for pH and recirculation flow.
A block diagram of scrubber pH/ORP monitoring data flow is depicted in
Scrubber Fan
Scrubber fan running status is determined within PI and communicated to Proficy through the use of a digital signal. Within PI, running status is determined by either comparing the scrubber fan amps to a minimum limit, by examining the differential pressure across the fan to be greater than a minimum limit, or by examining the scrubber motor status from the DCS (through the use of status from a zero speed switch or equivalent digital signal).
Scrubber fan EE and CMS events are created by using the Proficy downtime model 200. The Proficy model is typically configured with a 61 second filter (to eliminate signal noise) in conjunction with a PI performance equation to act as the event trigger.
For excess emissions calculation, the PI tag CR-BP-ScrubFan.EE returns the digital state “EE” whenever the PI logic determines that the fan is not running while the system is in a “CanEmit” state; otherwise the equation returns the value “OK”. To determine CMS downtime, a second PI performance equation (CR-BP-ScrubFan.DQ) verifies that the fan amp value (or fan running switch status) is within range (or has a valid state) and returns the value “Good.” If these conditions are not met, (and the PTE state of “CanEmit” exists) the equation instead returns the value “Bad.”
Proficy monitors the two digital tags CR-BP-ScrubFan.EE and CR-BP-ScrubFan.DQ for the fault values of “EE” and “Bad” respectively. If either value remains in the fault state for longer than the filter time, an EE or CMS event is recorded by the system.
A block diagram of scrubber fan monitoring data flow is depicted in
Tag Name Specifications
All Cluster Rule PI tags will begin with “CR-”.
For locations with multiple bleach lines, each line will be differentiated by CR-BPx, where x represents the mill naming convention. For example, Franklin will use CR-BPE for the E-Line and Augusta will use CR-BP1 for #1 Bleach Plant.
Digital State Set Specifications
The following are the minimum required digital state sets in PI to support the Cluster Rule Bleach Plant model.
Scan Class Specifications
The following scan classes must be available in PI. Note, the scan class number will vary from mill to mill.
A one minute scan class offset 0 seconds from midnight;
A fifteen minute scan class offset 0 seconds from midnight;
A twenty-four hour scan class offset to the start of mill day.
Examples of the scan class syntax are as follows:
/f=00:01:00, 00:00:00 (alternately /f=00:01:00, 0)
/f=00:15:00, 00:00:00 (alternately /f=00:15:00, 0)
/f=24:00:00, 07:00:00 (alternately /f=24:00:00, 25200) for mill day at 07:00 am
PI Tag Configuration Specification
The following tables provide tag configuration examples for a typical bleach plant model and the standard compression and exception attribute settings for the underlying mill tags.
’ ‘*−60M’, ‘*’, “
”)) then PrevVal(‘CR-BP-PTE.STAT’,
’ ‘*−60M’, ‘*’, “
”) > 0 then “CanEmit” else “CanNotEmit”
’ <> “
”) and (‘CR-BP-PTE.STAT’ = “CanEmit”) then “EE” else “OK”
, if (‘
’ > ‘CR-BP-ScrubRecirc.LLL’) and
’ < ‘CR-BP-ScrubRecirc.HHL’) then ‘
’ else “BAD”
, if (‘
’ > ‘CR-BP-ScrubPH.LLL’) and (‘
’ < ‘CR-
’ else “BAD”
’ <> “
” and ‘
’ <> “
”) and (‘CR-BP-PTE.STAT’ <>
The Proficy model consists of input variables (PI inputs), calculated variables, stored procedures, and Visual Basic scripts (VB scripts). Variables for a typical bleach plant (monitoring pH) and descriptions of the stored procedures and the VB scripts are included below. Complete listings of the Stored Procedures can be found herein below.
1This example monitors pH of the effluent. When ORP (Oxygen Reduction Potential) of the effluent is monitored instead of pH, the pH Lower Limit is replaced by an ORP Upper Limit.
2The sampling offset is determined based upon the mill Start of Day time. The offset value is the number of minutes from midnight to the mill start of day. In this example the start of day is 7:00 AM (as there are 420 minutes from midnight until 7:00 AM).
1This example monitors pH of the effluent. When ORP (Oxygen Reduction Potential) of the effluent is monitored instead of pH, the pH Lower Limit is replaced by an ORP Upper Limit.
pH, ORP and flow CMS events are created from the stored procedure, BleachEvents, as described below. Scrubber fan CMS events are created using Proficy's downtime model 200 with a 61-second filter applied. The PI tag, CR-BP-ScrubFan.DQ, triggers the start of an event whenever its state changes from Good (the normal running state) to Bad (the fault state). The event ends when the state changes back to Good. If the state returns to Good within one minute, the change is ignored and an event is not created.
pH, ORP and flow EE events are created from the stored procedure, BleachEvents, as described below. Scrubber fan EE events are created using Proficy's downtime model 200 with a 61-second filter applied. The PI tag, CR-BP-ScrubFan.DQ, triggers the start of an event whenever its state changes from OK (the normal running state) to EE (the fault state). The event ends when the state changes back to OK. If the state returns to OK within one minute, the change is ignored and an event is not created
VB Script Descriptions
BP PTE CMS Status (15 min)
The inputs to this script are the BP % Time CanEmit (15 min) and the lower warning limit for BP % Time CanEmit (15 min). This script is triggered by time (based on the sample interval for the variable) or an input value change. This script compares the 15 min average numeric PTE value to its lower warning limit. If the % Time CanEmit (15 min) value is less than the lower warning limit (usually 50), the PTE status is CanNotEmit. If the % Time CanEmit (15 min) value is greater than or equal to the lower warning limit, the PTE status is CanEmit.
BP PTE EE Status (1 Hr)
The inputs to this script are the BP % Time CanEmit (1 Hr) and the lower warning limit for BP % Time CanEmit (1 Hr). This script is triggered by time (based on the sample interval for the variable) or an input value change. This script compares the 1 hr average numeric PTE value to its lower warning limit. If the % Time CanEmit (1 hr) value is less than or equal to the lower warning limit (usually 50), the PTE status is CanNotEmit. If the % Time CanEmit (1 Hr) value is greater than the lower warning limit, the PTE status is CanEmit.
Qualified 15 Min Avg
The inputs to this script are the BP CMS PTE Status (15 min), the 15 minute raw PI average for pH, ORP or flow, the percent good value for pH, ORP or flow and the lower warning limit for percent good. This script is triggered by time (based on the sample interval for the variable) or an input value change. This script filters the 15 minute average (pH, ORP or recirculation flow) based on the PTE status or the percent good value for the average. If the percent good value is greater than 50% and the PTE status is CanEmit, this script outputs the average value. If the PTE status is CanNotEmit or the percent good value is less than 50%, this script outputs a null value.
Qualified 15 Min Avg Status
The inputs to this script are the BP CMS PTE Status (15 min), the 15 minute raw PI average for pH, ORP or flow, the percent good value for pH, ORP or flow and the lower warning limit for percent good. This script is triggered by time (based on the sample interval for the inputs) or an input value change. This script outputs the status of the Qualified 15 minute Average (pH, ORP or recirculation flow) for display on the Autolog display. If the PTE status is CanNotEmit, this script outputs Unit Down. If the percent good value is greater than 50% and the PTE status is CanEmit, this script outputs OK. If the percent good value is less than 50% and the PTE status is CanEmit, the script outputs Bad Val.
15 Min Avg/Status Reassembly
The inputs to this script are the Qualified 15 min Avg and the Qualified 15 min Avg Status. This script is triggered by time (based on the sample interval for the inputs) or an input value change. This script combines the two inputs into one string value based on the string value of the Qualified 15 Min Avg Status. If the Qualified 15 Min Avg Status is OK, this script outputs the Qualified 15 min Avg. If the Qualified 15 min Avg Status is Unit Down or Bad Val, this script outputs Unit Down or Bad Val.
Stored Procedure Descriptions
ManualUpdate
This procedure has one input, the 15 min raw PI avg for pH or ORP and one dependant variable, the manually entered value for pH or ORP. This procedure performs a signal selection between a manually entered value and another variable. If the dependant variable value (the manually entered value) is NULL, the output is the value of the input variable (the 15 min raw PI avg). Otherwise, the output is set to the value of the dependant variable. The triggers for this procedure are time (based on the sample interval for the variable), value change for the dependant variable or value change for the input variable.
BleachRollingAvg with AvgPTE
This procedure has two inputs, the percent good value for pH, ORP or flow and the BP EE PTE Status (1 Hr), and one dependant variable, the reassembled 15 min avg/status for pH, ORP or flow. This procedure calculates a 3 hour moving average of the dependant variable every hour from a minimum number of samples over the 3 hour interval if the average PTE status over the last hour is CanEmit. The requirement for a good average is that there must be more than 50% good samples. “Good” samples consist of valid numeric values taken while the PTE status is CanEmit and the percent good value is greater than 50%, as determined by the 15 min avg/status reassembly VB script. Values of Bad Val, Unit Down and NULL are excluded from the moving average. The triggers for this procedure are time (based on the sample interval for the variable), value change for the dependant variable or value change for the input variable.
BleachRollingAvgStatus with AvgPTE
This procedure has three inputs, the percent good value for pH, ORP or flow, the three hour rolling avg for pH, ORP or flow and the BP Ee PTE Status (1 Hr), and one dependant variable, the reassembled 15 min avg/status for pH, ORP or flow. This procedure generates a status string to compliment the 3 hour moving average calculation, BleachRollingAvg. The following table 40 shows the possible outputs for this procedure and the sample types required to generate them.
The triggers for this procedure are time (based on the sample interval for the variable), value change for the dependant variable or value change for the input variable.
BleachEvents
This procedure is used to create CMS and EE events for pH, ORP and flow. This procedure has one input, the BP EE/CMS PTE Status (1 Hr/15 Min), and one dependant variable, the 3 hr rolling avg for pH, ORP or flow. This procedure tests for CMS or EE events when the average PTE status if CanEmit by comparing the dependant variable value against upper or lower specification limits as specified in the calculation inputs. If the value is above (below) the upper (lower) specification limit, a downtime event with duration as specified in the inputs is created. If an events exists for the previous time interval, the duration is appended to the existing event and the event end time is updated. The triggers for this procedure are time (based on the sample interval for the variable), value change for the dependant variable or value change for the input variable.
Stored Procedure Listings
SpLocal_BleachEvents
/*
Procedure Name: spLocal_BleachEvents
General Description:
This procedure tests for CMS or EE events by comparing the dependant variable value against upper or lower specification limits as specified in the calculation inputs. If the value is above (below) the upper (lower) specification limit, a downtime event with duration as
specified in the inputs is created. If and event exists for the previous time interval, the duration
is appended to the existing event and the event end time is updated.
The “Potential to Emit” (PTE), if configured for CMS events, is also taken into account.
Triggers:
1. Time (based on sample interval for variable)
2. Dependant variable value change
3. Input value change
In order for the calculation to execute, non-optional calculation input values cannot be NULL.
Inputs and Depedencies:
Type: Status message (string)
Variables:
1. Described in body of code.
Tables Modified:
1. Timed_Event_Details.
SpLocal_BleachRollingAvg
/*
Procedure Name: spLocal_BleachRollingAvg
Copyright (C) 2001, International Paper Company
Revision History:
General Description:
This procedure calculates a 3-hour moving average of the dependent variable (typically ph, ORP or recirculation flow) value every hour from a minimum number of samples over the 3-hour interval.
Currently, the requirement is that there must be more than 50% good samples in order for the average to be calculated. “Good” samples consist of valid numeric values taken while there was potential to emit (PTE) and where the data validity, as determined by the %-Good PI variable, is good. Null values and values where the %-Good requirement is not met are excluded from the moving average. Typically, for the standard model, this procedure calculates the average of the 15-minute ph, ORP, or recirculation flow values over the last 3-hours.
Triggers:
1. Time (based on sample interval for variable)
2. Dependant variable value change
3. Input value change
In order for the calculation to execute, non-optional calculation input values cannot be NULL.
Inputs and Dependencies:
SpLocal_BleachRollingAvgStatus
/*
Procedure Name: spLocal_B leach RollingAvgStatus
Copyright (C) 2001, International Paper Company
Revision History:
General Description:
This procedure generates a status message to compliment the 3-hour moving average calculation result.
Triggers:
1. Time (based on sample interval for variable)
2. Dependant variable value change
3. Input value change
In order for the calculation to execute, non-optional calculation input values cannot be NULL.
Inputs and Depedencies:
Type: Status message (string)
Variables:
1. Described in body of code.
Tables Modified:
1. N/A
SpLocal_ManualUpdate
/*
Procedure Name: spLocal_ManualUpdate
Copyright (C) 2001, International Paper Company
Revision History:
General Description:
This procedure performs a signal selection between a manually entered value and another variable. If the dependant variable value (the manually entered value) is NULL, the output is the value of the input variable (the PI value). Otherwise, the output is set to the value of the dependant variable.
Triggers:
1. Time (based on sample interval for variable)
2. Dependant variable value change (the manually entered value)
3. Input value change (the PI variable)
In order for the calculation to execute, non-optional calculation input values cannot be NULL.
Inputs and Dependencies:
1. Requires configuration of the dependant variable which is the manually entered value.
2. Inputs described in body of code.
Outputs:
SpLocal_MillDay
/*
Procedure Name: spLocal_MillDay
Copyright (C) 2001, International Paper Company
Process Management Application Group
Revision History:
General Description:
This procedure calculates a date string for display that coincides with the mill day. The time at which the mill day begins is hard-coded within this procedure (see comments below).
Triggers:
1. Time (based on sample interval for variable)
Inputs and Dependencies:
1. Inputs described in body of code.
Outputs:
1. Date string for the mill day.
Variables:
1. Described in body of code.
Tables Modified:
1. N/A
| Number | Name | Date | Kind |
|---|---|---|---|
| 4901245 | Olson et al. | Feb 1990 | A |
| 5257530 | Beattie et al. | Nov 1993 | A |
| 5564283 | Yano et al. | Oct 1996 | A |
| 5930993 | Kammann et al. | Aug 1999 | A |
| 6385558 | Schlemm | May 2002 | B1 |
| 6701255 | Batug et al. | Mar 2004 | B2 |
| 20020152037 | Sunshine et al. | Oct 2002 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20060161388 A1 | Jul 2006 | US |
