Patent Grant
7130703
The present invention relates generally to process control and safety systems used in process plants and, more particularly, to a system having a voter logic block with operational and maintenance override functionality.
Process control systems, like those used in chemical, petroleum or other processes, typically include one or more process controllers communicatively coupled to at least one host or operator work station and to one or more field devices via analog, digital or combined analog/digital buses or lines. The field devices, which may be, for example valves, valve positioners, switches and transmitters (e.g., temperature, pressure and flow rate sensors), perform functions within the process plant such as opening or closing valves and measuring process variables. The process controllers receive signals indicative of process measurements made by the field devices and/or other information pertaining to the field devices, use this information to implement control routines and then generate control signals which are sent over the buses or lines to the field devices to control the operation of the process. Information from the field devices and the controllers is typically made available to one or more applications executed by the operator work station to enable an operator to perform any desired function with respect to the process, such as configuring the process, viewing the current state of the process, modifying the operation of the process, etc.
Furthermore, in many processes, a separate safety system is provided to detect significant safety related problems within the process plant and to automatically close valves, remove power from devices, switch flows within the plant, etc. When a problem occurs which might result in or lead to a serious hazard in the plant, such as a spill of toxic chemicals, an explosion, etc. These safety systems typically have one or more separate controllers apart from the standard process control controllers, called logic solvers, which are connected to safety field devices via separate buses or communication lines installed within the process plant. The logic solvers use the safety field devices to detect process conditions associated with significant events, such as the position of certain safety switches or shutdown valves, overflows or underflows in the process, the operation of important power generation or control devices, the operation of fault detection devices, etc. to thereby detect “events” within the process plant. When an event is detected, the safety controller takes some action to limit the detrimental nature of the event, such as closing valves, turning devices off, removing power from sections of the plant, etc. Generally, these actions include switching safety devices into a tripped or“safe” mode of operation which is designed to prevent a serious or hazardous condition within the process plant.
It is common practice in safety instrumented systems to use redundant input devices such as transmitters and switches in order to detect events within the system to provide greater safety integrity or process variable measurement availability. In such systems, it is sometimes necessary to provide voting logic functionality in the shutdown logic to determine whether the process condition is acceptable or dangerous based on the redundant inputs. While such voting logic is fairly strait-forward, in that it typically needs merely to determine a majority vote of the inputs to determine if an event condition has occurred, these voting systems have no effective overrides. However, it is frequently desirable in both safety systems and in some process control systems to be able to override the output of the voting function to, for example, prevent operation of the shut down system during startup of the process control system, enable maintenance personnel to perform maintenance operations on one or more of the input devices, allow selected process conditions to be ignored temporarily, etc.
While in the past, configuration or safety engineers have sometimes manually programmed voting logic into the safety system controllers using different programming languages, this programming step unfortunately has been tedious, time consuming and fraught with errors, which can be serious because a failure of the safety system to operate properly can lead to serious injury or even death on the part of plant personnel and to the destruction of potentially millions of dollars of equipment and material within a plant. In general, some useful features which are not easily integrated into known voting logic functionality include maintenance bypass of selected inputs to the voting logic system, startup bypasses, startup and/or trip delay features, etc.
A safety system within a process plant uses one or more voter function blocks that can be easily integrated into a function block diagram programming environment to implement the voting logic specified by a user along with various maintenance overrides and bypass features. Such a voter function block, which is easy to create, use, test, debug and document, includes one or input limit detection units which detect if an associated input has reached a specified limit indicating a condition within the process plant, an input bypass unit that may be set by a user to override the consideration of that input in the voting function, a voting function logic entity that determines if a trip condition exists based on the values of the valid inputs, and an inhibit block that may used to override the output of the voter logic block during, for example, a startup or other operational condition. The voter function block may be communicatively connected to other function blocks, such as to analog or digital input function blocks, analog or digital output function blocks, control function blocks, cause and effect function blocks which implement cause and effect logic, etc. to implement the voter logic as part of a larger process control or safety system strategy. In one case, the voter function block may be used to detect the presence of an event within a process safety system based on multiple measurements of a process variable made by redundant measurement or sensor devices.
The voter function block describe herein is easy to create, as, in its basic form, it only requires the configuration or safety engineer to supply an indication of the number of inputs to be analyzed, the type of voter logic to be used and the overrides or bypass features to be used to define the desired operation of the voter function block. This voter function block is also easy to integrate into a controller or logic solver which uses function block logic as the voter function block can be integrated in the same manner as any other function block, by interconnecting inputs and outputs of the voter function block to other function blocks or elements within the control strategy. As a result, this voter function block is also easy to document, test and debug. Moreover, this voter function block may provide additional functionality not normally provided in safety systems, such as providing overrides or bypass features to be used during runtime as well as during maintenance and startup procedures.
Referring now to
Generally speaking, each of the nodes 18 and 20 of the process plant 10 includes both process control system devices and safety system devices connected together via a bus structure that may be provided on a backplane into which the different devices are attached. The node 18 is illustrated in
Likewise, the node 18 includes one or more safety system logic solvers 50, 52, while the node 20 includes safety system logic solvers 54 and 56. Each of the logic solvers 50–56 is an I/O device having a processor 57 that executes safety logic modules 58 stored in a memory 79 and is communicatively connected to provide control signals to and/or receive signals from safety system field devices 60 and 62. Additionally, each of the nodes 18 and 20 includes at least one message propagation device (MPD) 70 or 72, which are communicatively coupled to each other via a ring type bus connection 74 (only part of which is illustrated in
The process controllers 24 and 26, which may be, by way of example only, DeltaV™ controllers sold by Fisher-Rosemount Systems, Inc., or any other desired type of process controllers are programmed to provide process control functionality (using what are commonly referred to as control modules) using the I/O devices 28, 30 and 32 (for the controller 24), the I/O devices 34 and 36 (for the controller 26) and the field devices 40 and 42. In particular, each of the controllers 24 and 26 implements or oversees one or more process control routines stored therein or otherwise associated therewith and communicates with the field devices 40 and 42 and the workstations 14 to control the process 10 or a portion of the process 10 in any desired manner. The field devices 40 and 42 may be any desired types of field devices, such as sensors, valves, transmitters, positioners, etc., and may conform to any desired open, proprietary or other communication or programming protocol including, for example, the HART or the 4–20 ma protocol (as illustrated for the field devices 40), any fieldbus protocol such as the FOUNDATIONS® Fieldbus protocol (as illustrated for the field devices 42), or the CAN, Profibus, the AS-Interface protocols, to name but a few. Similarly, the I/O devices 28–36 may be any known types of process control I/O devices using any appropriate communication protocol(s).
The safety logic solvers 50–56 of
A common backplane 76 (indicated by a dashed line through the controllers 24, 26, the I/O devices 28–36, the safety logic solvers 50–56 and the MPDs 70 and 72) is used in each of the nodes 18 and 20 to connect the controllers 24 and 26 to the process control I/O cards 28, 30 and 32 or 34 and 36, as well as to the safety logic solvers 50, 52, 54 or 56 and to the MPDs 70 or 72. The controllers 24 and 26 are also communicatively coupled to, and operate as a bus arbitrator for the bus 22, to enable each of the I/O devices 28–36, the logic solvers 50–56 and the MPDs 70 and 72 to communicate with any of the workstations 16 via the bus 22.
As will be understood, each of the workstations 16 includes a processor 77 and a memory 78 that stores one or more configuration and/or viewing applications adapted to be executed on the processor 78. A configuration application 80 and a viewing application 82 are illustrated in an exploded view in
Conversely, the viewing application 82 may be used to provide one or more displays to a user, such as to a process control operator, a safety operator, etc., which includes information about the state of the process control system 12 and the safety system 14 either in separate views or in the same view, if so desired. For example, the viewing application 82 may be an alarm display application that receives and displays indications of alarms to an operator. If desired, such an alarm viewing application may take the form as disclosed in U.S. Pat. No. 5,768,119 entitled “Process Control System Including Alarm Priority Adjustment” and U.S. patent application Ser. No. 09/707,580 entitled “Integrated Alarm Display in a Process Control Network,” both of which are assigned to the assignee of this patent and are hereby expressly incorporated by reference herein. It will be understood, however, that the alarm display or alarm banner of these patents may receive and display alarms from both the process control system 12 and the safety system 14 in an integrated alarm display as the alarms from both systems 12 and 14 will be sent to the operator work station 14 executing the alarm display application and will be recognizable as alarms from different devices. Likewise, an operator may deal with safety alarms displayed in an alarm banner in the same manner as process control alarms. For example, the operator or user may acknowledge safety alarms, turn off safety alarms, etc. using the alarm display, which will send messages to the appropriate process controller 24, 26 within the safety system 14 using communications over the bus 22 and the backplane 76 to take the corresponding action with respect to the safety alarm. In a similar manner, other viewing applications may display information or data from both the process control system 12 and the safety system 14 as these systems may use the same types and kinds of parameters, security and referencing so that any data from one of the systems 12 and 14 can be integrated into a display or view traditionally provided for a process control system.
The diagnostic application 84 may be used to implement diagnostics or maintenance programs within the process control and safety system of the plant 10. Such diagnostic applications, which may perform any desired types of diagnostics or maintenance procedures, such as running process and valve tests, startup procedures, etc., may provide overrides to one or more voter function blocks (described below) used within the process plant 10 to prevent operation of the safety system based on inputs from one or more devices effected by the diagnostic procedures.
In any event, the applications 80, 82 and 84, as well as any other applications may send separate configuration and other signals to and may receive data from each of the process controllers 24 and 26 as well as from each of the safety system logic solvers 50–56. These signals may include process-level messages related to controlling the operational parameters of the process field devices 40 and 42, and may include safety-level messages related to controlling the operational parameters of the safety-related field devices 60 and 62. While the safety logic solvers 50–56 may be programmed to recognize both the process-level messages and the safety-level messages, the safety logic solvers 50–56 are capable of distinguishing between the two types of messages and will not be capable of being programmed or effected by process-level configuration signals. In one example, the programming messages sent to the process control system devices may include certain fields or addresses which are recognized by the safety system devices and which prevent those signals from being used to program the safety system devices.
If desired, the safety logic solvers 50–56 may employ the same or a different hardware or software design as compared to the hardware and software design used for the process control I/O cards 28–36. The use of alternate technologies for the devices within the process control system 12 and devices within the safety system 14 may minimize or eliminate common cause hardware or software failures. Furthermore, the safety system devices, including the logic solvers 50–56 may employ any desired isolation and security techniques to reduce or eliminate the chances of unauthorized changes being made to the safety-related functions implemented thereby. For example, the safety logic solvers 50–56 and the configuration application 80 may require a person with a particular authority level or a person located at a particular work station to make changes to the safety modules within the logic solvers 50–56, with this authority level or location being different from the authority or access level or location needed to make changes to the process control functions performed by the controllers 24 and 26 and the I/O devices 28–36. In this case, only those persons designated within the safety software or located at workstations authorized to make changes to the safety system 14 have authorization to alter safety-related functions, which minimizes the chances of corruption to the operation of the safety system 14. As will be understood, to implement such security, the processors within the safety logic solvers 50–56 assess the incoming messages for proper form and security and operate as gatekeepers on changes being made to the safety-level control modules 58 executed within the safety logic solvers 50–56.
As will be understood, the use of the backplane 76 in each of the nodes 18 and 20 enables the safety logic solvers 50 and 52 and the safety logic solvers 54 and 56 to communicate locally with one another to coordinate safety functions implemented by each of these devices, to communicate data to one another, or to perform other integrated functions. On the other hand, the MPDs 70 and 72 operate to enable portions of the safety system 14 that are disposed at vastly different locations of the plant 10 to still communicate with one another to provide coordinated safety operation at different nodes of the process plant 10. In particular, the MPDs 70 and 72 in conjunction with the bus 74 enable the safety logic solvers associated with different nodes 18 and 20 of the process plant 10 to be communicatively cascaded together to allow for the cascading of safety-related functions within the process plant 10 according to an assigned priority. Alternatively, two or more safety-related functions at different locations within the process plant 10 may be interlocked or interconnected without having to run a dedicated line to individual safety field devices within the separate areas or nodes of the plant 10. In other words, the use of the MPDs 70 and 72 and the bus 74 enables a safety engineer to design and configure a safety system 14 that is distributed in nature throughout the process plant 10 but that has different components thereof communicatively interconnected to enable the disparate safety related hardware to communicate with each other as required. This feature also provides scalability of the safety system 14 in that it enables additional safety logic solvers to be added to the safety system 14 as they are needed or as new process control nodes are added to the process plant 10.
It will be understood that the logic solvers 50–56 may be programmed to perform control activities with respect to the safety devices 60 and 62, using a function block programming paradigm. In particular, as illustrated in an expanded view of one of the safety control modules 58a (stored in the memory 79) of the logic solver 54, a safety control module may include a set of communicatively interconnected function blocks that can be created and downloaded to the logic solver 54 for implementation during operation of the process 10. As illustrated in
While the expanded view of the safety control module 58a of
The outputs of the trip limit check blocks 95 and the pre-limit check blocks 96 (which may be, for example, digital signals that are set to a high valve when the limits or the pre-limits are met in the blocks 95 and 96) are each delivered to one of a set of input bypass inhibit blocks 98a, 98b and 98c. The input bypass inhibit blocks 98 perform input inhibits on the individual inputs IN1, IN2 and IN3 so that one or more of these inputs may be inhibited, that is, not used within the voter function block 94 to determine if a trip condition exists or if a pre-trip alarm condition exists. Each of the input bypass inhibit blocks 98 provides an output for the associated trip limit condition to a trip voter logic block 100a and provides an output for the associated pre-limit condition to a pre-trip voter logic block 100b. The voter logic blocks 100a and 100b perform voter logic as described in more detail below to determine if, based on the inputs thereto, a trip condition or a pre-trip alarm condition exists.
The trip voter logic block 100a and the pre-trip voter logic block 100b provide a trip signal and a pre-trip alarm signal (when these conditions are determined to exist), respectively, to a startup inhibit block 102 which may inhibit the voter function block 94 from providing any trip signal or pre-trip alarm signal output during, for example, a startup or other performance or run-time procedure in which it is desirable to inhibit the operation of the voter function block 94. The startup inhibit block 102 develops a trip output signal (labeled Out) determined as a result of the operation of the trip voter logic block 100a and the startup inhibit block logic and additionally develops a Pre_out signal determined as a result of the operation of the pre-trip voter logic block 100b and the startup inhibit block logic. The Out signal may be used to drive the operation of a shutdown procedure within the safety system 14 of
The voter function block 94 may include a set of parameters, some of which are indicated in
The voter function block 94 also has an internal trip type configuration parameter named Trip_Type which defines the normal and tripped state values associated with the inputs and/or outputs of the voter function block 94. For example, when the voter function block 94 is configured as ‘De-energized to Trip’ (which may be the default value), the normal operating value of the output is one and the tripped state value is zero. Conversely, when the voter function block 94 is configured as ‘Energized to Trip,’ the normal operating value is zero and tripped state value is one. This initial determination is made at the trip limit check blocks 95a, 95b and 95c and at the pre-limit check blocks 96a, 96b and 96c, which correspond to the inputs IN1, IN2, and IN3, respectively. A detect type (Detect_Type) parameter may be used to determine whether the comparison to the trip limit is to be a greater than (high limit) comparison or to be a less than (low limit) comparison. This comparison occurs at the appropriate trip limit check blocks 95 and the pre-limit check blocks 96 to determine if the input signals have reached the predetermined limits.
As will be understood, the outputs of the trip limit check blocks 95 will each indicate whether a trip is indicated by a corresponding one of the inputs IN1, IN2 and/or IN3. As discussed above, a maintenance override or bypass can be applied by the input bypass inhibit blocks 98 for each of the individual inputs IN1, N2 and IN3 to prevent those inputs from being used in the voting logic applied by the voter logic blocks 100. This bypass feature is desirable when, for example, maintenance is being performed on a transmitter or other field device providing the input signal to the voter function block 94. When using voting logic that determines a trip output based on multiple inputs, maintenance bypasses are not always a necessity because a single false vote to trip (that may result due to maintenance activities on the sensor providing the input) will not necessarily result in a trip. However, this bypass functionality is desirable to prevent false trips during maintenance activities and may be needed in some voter logic, such as in a one out of two voter logic scheme in which the existence of even a single trip signal from redundant sensors will result in a trip.
When one of the input bypass inhibit blocks 98 causes an input to be bypassed, the bypassed input will not be used by the voter logic blocks 100a and 100b to develop a trip signal or a pre-trip alarm signal, even if the input value exceeds the limits specified by the trip limit or the pre-trip limit parameters. To enable bypassing, a bypass permit (Bypass_Permit) parameter can first be enabled to control whether bypassing of the inputs is to be allowed in the first place. Generally speaking, bypassing on inputs will be allowed if the Bypass_Permit parameter is set or enabled while input bypassing will not be allowed if the Bypass_Permit parameter is unset or not enabled. While a single Bypass_Permit parameter may be applicable for all of the bypass inhibit blocks 98, a separate bypass permit may be set for each of input bypass inhibit blocks 98a, 98b, 98c.
If the Bypass_Permit parameter is set or enabled, then a BYPASSx parameter may be used to cause one or more of the bypass inhibit blocks 98 to operate to inhibit the use of an associated one of the inputs IN1, IN2 or IN3. The x in the BYPASSx parameter indicates which one of the inputs IN1, IN2 or IN3 is to be disabled. If desired, more than one input may be inhibited at any particular time or the voter function block 94 may be configured to allow only one input to be inhibited at a time. The Bypass_Permit and the BYPASSx parameters may be set or issued in any desired manner, such as by an operator display button on an operator or maintenance screen, a physical keyswitch, a discrete input into the safety module, by a configuration, control, display or diagnostic application or in any other manner. Of course, if the use of a bypass permit is not needed in any particular implementation of the voter function block 94, the default value of the Bypass_Permit parameter can be set to be enabled at configuration of the voter function block 94.
A bypass timeout (Bypass_Timeout) parameter may be used to set the amount of time after which a bypass for one of the blocks 98 is set that the bypass will automatically expire. In this case, each of the input bypass inhibit blocks 98 may include as one of a set of timers 110 a bypass timer that is set to the Bypass_Timeout parameter value and that may be counted down at the start of the bypass. In this case, the input bypass inhibit blocks 98 may inhibit the use of the associated input until the BYPASSx is turned off or until the bypass timer reaches zero. As will be understood, bypass timers are used to assure that bypasses are removed after a predetermined amount of time.
If desired, the input bypass inhibit blocks 98 may also be configured to provide a reminder alarm to a user, such as an operator, safety engineer, technician, etc. to remind or notify the user that a bypass timeout is imminent. If bypasses are configured to disappear or be unenabled upon a bypass timeout, notification can be sent to a user or other operator in advance of the timeout by setting a reminder time (REMINDER_TIME) parameter to some non-zero value. In this case, if the bypass timer is non-zero but is less than the reminder time parameter and any bypassed input is voting to trip, the reminder alarm can be activated to provide an alarm to the user to indicate that a shut down may occur upon expiration of the bypass timer, which is imminent. If there are no bypassed inputs voting to trip, the alarm does not need to be activated, although it may still be activated. It will be understood, however, that even when the bypass timeout alarm is active, a trip is not necessarily imminent because there may not be enough other inputs voting to trip to cause the trip voter logic block 100a to produce a trip signal.
In one embodiment, the bypass timer is re-armed only when the first bypass is timed-out. However, the bypass timer may be a write-able parameter so that, after notification that a timeout is about to occur, the bypass timer can be incremented using an operator display button (or some other suitable technique) to extend the bypass time. Such a feature enables a user to extend the bypass time when, for example, a maintenance procedure is still being performed on the field device providing the bypassed input to the voter function block 94. Alternatively, the notification of bypass timeout may be for indication purposes only when, for example, a bypass is not to be unenabled when the bypass timer times out. In this case the reminder alarm may be set to be active when the bypass timer times out, even if the reminder time parameter is set to zero. However, if the reminder time parameter is non-zero, the reminder will still occur prior to timeout (if the input is voting to trip). The reminder alarms and bypass alarms may be acknowledged or non-acknowledged alarms.
The voting logic performed by the voter logic blocks 100a and 100b is preferably configured as an “M out of N” logic function. According to this functionality, M inputs must vote to trip out of the total of N inputs. For example, the voter function block 94 can be configured as a two out of three (2oo3) voter, which means that two of the three inputs must meet the trip limit before the voter logic block 100a output is set to the trip state value and two out of three of the inputs must meet the pre-trip limit before the pre-trip voter logic block 100b is set to a pre-trip alarm value. The N value in the “M out of N” function is determined from the number uninhibited inputs while the M value is determined based on an internal parameter of the block called number to trip (NUM_TO_TRIP), whose default value may be set to any desired value equal to or less than N at configuration. Common voting schemes may include, for example, two out of three, (2oo3), one out of two (1oo2), two out of two (2oo2), etc. However, any other voting logic may be used. Because of the other features of the block 94, the voter function block 94 may also be used for single transmitter applications such as in a one out of one (1oo1) voter function logic situation.
Generally speaking, 1oo2 or 1oo1 voting schemes will need a maintenance bypass function because disabling even one of the transmitters in a maimer that causes a detected trip condition at the input of the voter function block 94 for that transmitter during maintenance activities will necessarily result in a trip condition being set by the voter logic block 100a. However, voter function blocks configured to require multiple votes to trip can still benefit from a bypass function for more predictable behavior during maintenance procedures.
Bypassing one of the inputs IN1,IN2 or IN3 may effect the voter logic blocks 100a and 100b in one of two manners. It may either cause the number of inputs needed to determine a trip condition (or a pre-trip alarm condition) to be reduced by one or it may cause this number of inputs to remain the same. For example, when the voter logic block 100a is configured as a 2oo3 voter logic block and one of the inputs IN1 IN2 or IN3 is bypassed, the voting scheme may then become a 1oo2 voting scheme, meaning that the number of inputs needed to vote to trip is reduced by one (along with the number of possible inputs). Optionally, the 2oo3 voting scheme may be changed to a 2oo2 voting scheme when a selected input is bypassed, meaning that the number of inputs needed to vote to trip stays the same (even though the number of possible inputs is reduced by one). A bypass options parameter may be used to specify whether the actual number required to trip is to be reduced by one or not when an input is bypassed.
The default behavior of the input bypass inhibit blocks 98 may be configured to allow only one input to be bypassed at a time. This functionality may be enforced by a writecheck feature, which prevents a second input from being bypassed. Optionally, multiple inputs can be bypassed simultaneously. If desired, the BYPASSx parameter may have an additional writecheck, which requires the bypass permit BYPASS_PERMIT parameter to be true or set before the BYPASSx parameter may be set.
After voting is performed at the trip voter logic block 100a, according to the selected M out of N voting scheme, a trip-delay-on time parameter TRIP_DELAY_ON may be applied so that the voted trip condition must be active for a configurable period of time (the default value of which may be set at zero seconds) before the OUT signal changes to the tripped state value. In a similar manner, a trip-delay-off time parameter TRIP_DELAY_OFF (the default value of which may be set at zero seconds) may be applied to delay the time at which the OUT signal is returned to the normal state value when the vote to trip condition clears, i.e., when the trip voter logic block 100a determines that a trip condition does not exist based on the inputs thereto. Of course, the trip-delay-on time parameter and the trip-delay-off time parameter may have different and any desired values, and may be applied to one or both of the Out signal produced by the trip voter logic block 100a and the Pre-out alarm signal produced by the pre-trip voter logic block 100b. If desired, the trip-delay-on time and the trip-delay-off time periods may be independently configurable for the trip voter logic block 100a and the pre-trip voter logic block 100b and may be tracked by one of the timers 110.
As indicated above, the startup inhibit block 102 provides for startup or other operational override functionality. It may, for example, be necessary to override the output of the voter function block 94 to force the Out signal to be in the normal state for a brief period of time during startup or other temporary operational situations. This startup inhibit functionality may be used, for example, to inactivate a standing trip demand generated by the voter function block 94 because the process or a relevant portion thereof is in a shutdown state, to thereby allow the process startup procedure to proceed to the point where the process values being provided at the inputs of the voter function block 94 are no longer at values that indicate that a trip should be initiated.
In one example, the startup inhibit block 102 may include default behavior such that, on receiving an indication of a startup, which may be indicated by setting a Startup parameter, the startup inhibit block 102 forces the Out signal and, if desired, the Pre_out signal to the normal state value for a configurable period of time defined by a startup delay (STARTUP_DELAY) parameter. The startup inhibit block 102 may include a startup countdown timer as one of the timers 110 which is set to the value specified by the startup delay parameter and which starts to countdown upon receiving the startup indication via the startup parameter. When the countdown timer times out, the trip voter logic block 100a and the pre-trip voter logic block 100b resume normal trip detection. The startup inhibit block 102 may be configured such that a subsequent setting of the startup parameter will not affect the startup time while the startup timer is timing down. Optionally, it is possible to allow each new setting of the startup parameter to re-arm the startup timer so that a pending trip on timeout can be avoided.
Similar to the input bypass inhibit blocks 98, the startup inhibit block 102 may have a reminder function which may be turned on by, for example, setting a bypass parameter. This reminder functionality operates for startup bypasses in essentially the same manner as it operates for input bypasses (maintenance bypasses). Thus, when the startup timer is greater than zero but less than a configurable reminder time (REMINDER_TIME) parameter (which may be set at configuration) and there are enough votes to trip, a reminder alarm condition becomes active indicating that the bypass is about to expire which will result in a shutdown based on the values of the inputs IN1, IN2 and IN3.
If desired, the startup timer may additionally or alternatively automatically expire when the inputs have stabilized, that is, when there have not been enough votes to trip for a configurable period of time. This stable time may be tracked by a stable timer which may be one of the timers 10 and which may detect when the output of the voter logic block 100a is stable at, for example, a non-trip or normal value for the specified period of time. In this case, while the startup timer is timing down, the stable timer may time up whenever there are not enough votes to trip and may reset whenever the trip votes meet or exceed the number required to trip. If the stable timer reaches the configured stable time value, the startup timer is reset to zero and the normal trip detection functionality resumes. Of course, the stable timer does not reset at the end of the startup time period, but is reset at the beginning of a startup and at any time during the startup inhibit period when there are enough trip votes.
Alternatively, the startup bypass time need not be based on a fixed time period or on values of the inputs IN1, IN2 and IN3 to the voter function block 94 but can, instead, be based on the occurrence or nonoccurrence of an event. In this case the startup bypass ends when a startup reset parameter is set or becomes set or true, which may occur upon the detection of the event. In this manner, the startup bypass may be tied to the existence or non-existence of an event of indeterminable time length.
If desired, the status of the inputs IN1, IN2 and/or IN3 may be used to influence the behavior of the voter function block 94 and this status behavior may be set using a status options parameter. As will be understood, in may systems, such in HART and Fieldbus systems, transmitters or other field devices will send a status signal along with a process variable signal or a process value, wherein the status signal indicates the status of the transmitter itself. Such status signals may indicate that the transmitter is in a normal or good state, or in an abnormal state, such as a bad or other undesirable state which may cause the value of the process variable being sent by the transmitter to be of questionable nature. Thus, the status of the inputs signals provided to the IN1,IN2 and IN3 inputs of the voter function block 94 may be determined and used to effect the voting scheme or the manner in which the inputs will be used in the voting scheme.
If desired, the voting schemes used by the blocks 100 may be set so that one failed transmitter (i.e., one input having a bad status) will not automatically initiate a trip when other transmitters are available to indicate a valid value of the process variable being measured. When considering the status of the input signals, one option is to always use the value of the input IN1, IN2 or IN3 regardless of the status of the input. In this manner, a hardware failure will not necessarily cause a shutdown and time will be allowed for repair. Another option is to treat a bad status on an input as if the input were bypassed, which prevents the input from voting to trip in the same manner as described above with respect to the input bypass inhibit blocks 98. A third option is to automatically consider the input as a vote to trip if the status of the input is bad. This may be configured as the default option, which provides the highest level of safety for 1ooX voting schemes.
Of course, the use of the status of the inputs to the voter function block 94 may be treated the same or differently in each of the trip voter logic block 100a and the pre-trip voter logic block 100b. If desired, the status of the Out signal and the Pre_out signal may be set as Good unless all non-bypassed inputs have a bad status, in which case, the status of the Out and the Pre-out signals may be set as Bad. If desired, when any non-bypassed input has a bad status, an alarm condition parameter indicating a bad input may be set by the voter function block 94.
The voter function block 94 enters the trip inhibited state 138 from any of the other states whenever a startup bypass is active or when it is not possible to trip because there are not enough inputs participating in the voting scheme, either due to one or more of the inputs being bypassed or one or more of the inputs having a bad status and a bad status is treated such that the input will not be used in the vote. The voter function block 94 may enter any other state from the trip inhibited state 138 based on the setting of the delay parameters (the trip-delay-on time parameter and the trip-delay-off time parameter) as well as the inputs to the voter function block 94 when the inhibit condition disappears.
As will be understood from the state diagram 130, the voter function block 94 will generally be in the normal state 140 when the inputs to the voter function block 94 are at values such that the voting scheme used in the trip voter logic block 100a of
The voter function block 94 will remain in the voted to trip-delayed state 136 during the delay period, as set by the trip-delay-on time parameter, which may be configured by a user, a configuration engineer, etc. When the trip-delay-on time period times out (and is not reset by the user) and the trip condition still exists, the voter function block 94 will enter the tripped state 132 and set the Out signal to the tripped value. However, if the trip condition goes away during the vote to trip-delayed state 136, the voter function block 94 will return to the normal state 140 without effecting the Out signal.
When in the tripped state 132, the voter function block 94 sets the Out signal to the tripped state and holds it in that state until the voter function block 94 enters either the trip inhibited state 138 or the normal state 140. If a tripped condition disappears because, for example, one or more of the inputs to the voter function block 94 changes, thereby causing the trip voter logic to detect a vote not to trip condition, the voter function block 94 will go directly to the normal state 140 if trip-delay-off time period is not set, or will enter the voted normal-delayed state 134 if the trip-delay-off parameter is set to some non-zero value. The voter function block 94 will remain in the voted normal-delayed state 134 until the first of the trip-delay-off timer times out (at which time the voter function block 94 will enter the normal state 140), a trip inhibited condition arises (at which time the voter function block 94 will enter the trip inhibited state 138) or a voted to trip condition reappears based on a change in the inputs (at which time the voter function block 94 will re-enter the tripped state 132).
While the state diagram 130 of
Furthermore, if desired, the voter function block 94 may provide one or more status variables or signals that indicate the trip and pre-trip state in which the voter function block 94 is disposed. A typical status value for these signals may be ‘Normal’ and less commonly ‘Tripped’. However, as illustrated in the state diagram 130 of
If desired, the process control system 12 may be configured to capture and store information in an event chronicle regarding when inputs are bypassed or when a bypass is removed based on the signals and parameters associated with the voter function block 94. In many cases, these events may be recorded inherently because the bypass parameter is written directly to the voter function block 94. In some cases, bypasses and bypass permits are wired from physical keyswitches and special writechecking/event logging may be required. In practice, bypass permitting may be performed for a group of voter function blocks or for an individual voter function block. When bypass timeouts are used, special event logging may be used to capture the fact that the block has removed the bypasses. Of course, the voter function block 94 may generate any desired event records (by creating event record signals) to be sent to and stored in, for example, the configuration or historical database 21 of
As can be seen, the use of the voter function block 94 as described herein makes it easy to create a function block having a known format and thus, once created, needs only to be programmed with an appropriate configuration parameter settings to provide proper operation. Implementation of the voter function block within a function block programming environment is easy because the voter function block can be created and communicatively connected to other function blocks in any known or desired manner. Likewise, debugging of the voter function block also becomes easier than voter logic created in other programming languages because the voter function block typically only needs to be debugged within the context of how it is connected within the programming environment and the configuration parameters that are sent to it. Still further, documentation of the voter function block is easy because it is a typical function block having standard documentation changed only by the individual parameter settings provided thereto.
While the voter function block 94, which is an analog voter function block, is described in detail herein, it will be understood that digital voter function blocks may operate in a similar manner. However, in a digital voter function block (which processes digital input signals), the limit check blocks or detection units 95 will merely operate to detect the value of the digital signal as being a logical one or a logical zero, wherein one of these states is defined as meeting a trip condition and the other as not meeting a trip condition. Additionally, the pre-limit check blocks or detection units 96 may not be needed or possible in digital voter function blocks as the digital input signals typically only take on one of two states and cannot be compared against some other limit between those two states. However, if desired, the outputs of the trip limit check blocks 95 may be provided to a second voter logic block (that uses a more relaxed voting scheme than the voting scheme used by the trip voter logic block 100a) to determine a pre-trip alarm. Additionally, while the voter function blocks 92 and 94 are described herein as including five and three inputs, respectively, any other number of inputs could be used instead.
While
While the voter function blocks are described herein using state machine diagrams, these diagrams are only intended to describe the voter logic and bypass functionality. It will be understood that state machines need not be used and, if used, can be implemented in any form such as via hardware or software written, in any programming language. To be such a state machine, an element, such as a software program, routine, object, etc. need only cause the function block to transition between states as explained or defined herein (or other states) or as represented by the outputs of the function block to thereby cause the transition of the output from a normal to a tripped state or vice-versa.
When implemented, any of the elements described herein, including the inhibit blocks, voter logic blocks, state machines, signal connections, etc. may be implemented in software stored in any computer readable memory such as on a magnetic disk, a laser or optical disk, or other storage medium, in a RAM or ROM of a computer or process or, etc. Signals and signal lines described herein can take any form, including actual wires, data registers, memory locations, etc. This software may take any form, including application software executed on a general purpose computer or process or hard coded software burned into, for example, an application specific integrated circuit (ASIC), an EPROM, EEPROM, or any other firmware device. Likewise, this software may be delivered to a user, a process plant, an operator work station, a controller, a logic solver or any other computing device using any known or desired delivery method including, for example, on a computer readable disk or other transportable computer storage mechanism or over a communication channel such as a telephone line, the Internet, the World Wide Web, any other local area network or wide area network, etc. (which delivery is viewed as being the same as or interchangeable with providing such software via a transportable storage medium). Furthermore, this software may be provided directly without modulation or encryption or may be modulated and/or encrypted using any suitable modulation carrier wave and/or encryption technique before being transmitted over a communication channel.
Of course, the voter function blocks described herein can be implemented using any external process control communication protocol (besides a Fieldbus protocol or a DeltaV protocol) and may be used to communicate with any type of function block including any function block that is similar to or the same as any of the different function blocks specifically identified by or supported by the Fieldbus protocol. Moreover, while the voter function blocks in one embodiment hereof are described as Fieldbus “function blocks,” it is noted that the use of the expression “function block” herein is not limited to what the Fieldbus protocol identifies as a function block but, instead, includes any other type of block, program, hardware, firmware, etc., entity associated with any type of control system and/or communication protocol that can be used to implement some process control routine functionality or that has a predefined setup or protocol for providing information or data to other such function blocks. Thus, while function blocks typically take the form of objects within object oriented programming environment, this need not be case and can, instead, be other logical units used to perform particular control (including input and output) functions within a process plant or control environment using any desired programming structure or paradigm.
Thus, while the present invention has been described with reference to specific examples, which are intended to be illustrative only and not to be limiting of the invention, it will be apparent to those of ordinary skill in the art that changes, additions or deletions may be made to the disclosed embodiments without departing from the spirit and scope of the invention.
| Number | Name | Date | Kind |
|---|---|---|---|
| 4427620 | Cook | Jan 1984 | A |
| 4448033 | Briccetti | May 1984 | A |
| 5768119 | Havekost et al. | Jun 1998 | A |
| 5970430 | Burns et al. | Oct 1999 | A |
| 6014612 | Larson et al. | Jan 2000 | A |
| 6049578 | Senechal et al. | Apr 2000 | A |
| 6155282 | Zachary et al. | Dec 2000 | A |
| 6186167 | Grumstrup et al. | Feb 2001 | B1 |
| 6292523 | Senechal et al. | Sep 2001 | B1 |
| 6510351 | Blevins et al. | Jan 2003 | B1 |
| 6721609 | Wojsznis et al. | Apr 2004 | B1 |
| 6774786 | Havekost et al. | Aug 2004 | B1 |
| 6850973 | Larson et al. | Feb 2005 | B1 |
| 6898468 | Ott et al. | May 2005 | B1 |
| 7010450 | Law et al. | Mar 2006 | B1 |
| 20020052673 | Seong et al. | May 2002 | A1 |
| 20020091451 | Summers et al. | Jul 2002 | A1 |
| 20020130846 | Nixon et al. | Sep 2002 | A1 |
| 20020145515 | Snowbarger et al. | Oct 2002 | A1 |
| 20030062494 | Snowbarger et al. | Apr 2003 | A1 |
| 20040193290 | Ott et al. | Sep 2004 | A1 |
| 20040199351 | Ott et al. | Oct 2004 | A1 |
| 20050097194 | Eisenbeis et al. | May 2005 | A1 |
| Number | Date | Country |
|---|---|---|
| 1 091 199 | Apr 2001 | EA |
| 2 274 929 | Aug 1994 | GB |
| WO-0038040 | Jun 2000 | WO |
| WO-03098172 | Nov 2003 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20040255013 A1 | Dec 2004 | US |
