Patent Application
20050087558
The present invention relates to a system and method for determining the dispensing efficiency of fuel dispensers and/or fuel dispensing points in a service station environment to determine if the fuel dispensers/fuel dispensing points contain a blockage and/or performance problem that affects flow rate.
Service stations are comprised of a plurality of fuel dispensers that dispense fuel to motor vehicles A conventional exemplary fueling environment 10 is illustrated in
The central building 12 need not be centrally located within the fueling environment 10, but rather is the focus of the fueling environment 10, and may house a convenience store 18 and/or a quick serve restaurant (QSR) 20 therein. Both the convenience store 18 and the quick serve restaurant 20 may include a point-of-sale 22, 24, respectively. The central building 12 may further house a site controller (SC) 26, which in an exemplary embodiment may be the G-SITE® sold by Gilbarco Inc. of Greensboro, N.C. The site controller 26 may control the authorization of dispensing events and other conventional activities as is well understood. The site controller 26 may be incorporated into a point-of-sale, such as point of sale 22, if needed or desired. Further, the site controller 26 may have an off-site communication link 28 allowing communication with a remote location for credit/debit card authorization, content provision, reporting purposes or the like, as needed or desired. The off-site communication link 28 may be routed through the Public Switched Telephone Network (PSTN), the Internet, both, or the like, as needed or desired.
The car wash 14 may have a point-of-sale 30 associated therewith that communicates with the site controller 26 for inventory and/or sales purposes. The car wash 14 alternatively may be a stand-alone unit. Note that the car wash 14, the convenience store 18, and the quick serve restaurant 20 are all optional and need not be present in a given fueling environment.
The fueling islands 16 may have one or more fuel dispensers 32 positioned thereon. Each fuel dispenser 32 may have one or more fuel dispensing points. The term “dispensing point” can be used interchangeably with fuel dispenser 32 for the purposes of this application. A dispensing point 32 is a delivery point for fuel. The fuel dispensers 32 may be, for example, the ECLIPSE® or ENCORE®) sold by Gilbarco Inc. of Greensboro, N.C. The fuel dispensers 32 are in electronic communication with the site controller 26 through a LAN or the like.
The fueling environment 10 also has one or more underground storage tanks 34 adapted to hold fuel therein As such, the underground storage tank 34 may be a double-walled tank. Further, each underground storage tank 34 may include a liquid level sensor or other sensor 35 positioned therein. The sensors 35 may report to a tank monitor (TM) 36 associated therewith. The tank monitor 36 may communicate with the fuel dispensers 32 (either through the site controller 26 or directly, as needed or desired) to determine amounts of fuel dispensed, and compare fuel dispensed to current levels of fuel within the underground storage tanks 34 to determine if the underground storage tanks 34 are leaking. In a typical installation, the tank monitor 36 is also positioned in the central building 12, and may be proximate to the site controller 26.
The tank monitor 36 may communicate with the site controller 26 and further may have an off-site communication link 38 for leak detection reporting, inventory reporting, or the like, which may take the form of a PSTN, the Internet, both, or the like. As used herein, the tank monitor 36 and the site controller 26 are site communicators to the extent that they allow off-site communication and report site data to a remote location. The site controller 26 and the tank monitor 36 are typically two separate devices in a service station environment.
In addition to the various conventional communication links between the elements of the fueling environment 10, there are conventional fluid connections to distribute fuel about the fueling environment as illustrated in
Pipes 42 connect the underground storage tanks 34 to the fuel dispensers 32. Pipes 42 may be arranged in a main conduit 44 and branch conduit 46 configuration, where the main conduit 44 carries the fuel that is pumped by a fuel pump, such as a submersible turbine pump (not shown) for example, from the underground storage tanks 34 to the branch conduits 46, and the branch conduits 46 connect to the fuel dispensers 32. Typically, the pipes 42 are double-walled pipes comprising an inner conduit and an outer conduit. Fuel flows in the inner conduit to the fuel dispensers, and the outer conduit insulates the environment from leaks in the inner conduit. For a better explanation of such pipes and concerns about how they are connected, reference is made to Chapter B13 of PIPING HANDBOOK, 7th edition, copyright 2000, published by McGraw-Hill, which is hereby incorporated by reference.
As better illustrated in
After the fuel leaves the shear valve 48, the fuel typically passes through a flow control valve 49 located inline to the fuel supply piping 47. The flow control valve 49 may be used to control the flow of fuel into the fuel dispenser 32. The flow control valve 49 may be a two stage valve so that the fuel dispenser 32 controls the flow of fuel in a slow mode at the beginning of a dispensing event and at the end of the transaction (in the case of a prepaid fuel transaction), and a fast mode for fueling during steady state after slow flow mode is completed.
After the fuel leaves the flow control valve 49 in the fuel supply piping 47, the fuel may encounter a filter 50 to filter out any contaminants in the fuel before the fuel reaches the flow meter 52 that is typically located on the outlet side of the filter 50. The filter 50 helps to prevent contaminates from passing to the fuel flow meter 52 and the customer's fuel tank. Contaminates can cause a fuel flow meter 52 to malfunction and/or become un-calibrated if the meter 52 is a positive displacement meter, since the contaminate can scrub the internal housing of the meter 52 and increase the volume of the meter 52. If a filter 50 becomes clogged or blocked in any way, either wholly or partially, this will impede the flow of fuel from the fuel dispenser 32 and thereby reduce the maximum throughput/flow rate of the fuel dispenser 32. The maximum throughput of the fuel dispenser 32 is the maximum flow rate at which the fuel dispenser 32 can deliver fuel to a vehicle if no blockages or performance problems exist.
The filter 50 is changed periodically by service personnel during service visits, and is typically replaced at periodic intervals or when a fuel dispenser 32 is noticeably not delivering fuel at a fast enough flow rate. Because the filter 50 is changed in this manner, a fuel dispenser 32 may encounter unusual and unintended low flow rates for a period of time before they are noticed by the station operators and/or before service personnel replace such filters 50 during periodic service visits. There are also other components of a fuel dispenser 32 in addition to the filter 50 than may cause a fuel dispenser 32 to not deliver fuel at the intended flow rate, such as a defective or blocked valve 48, meter 52, hose 58, nozzle 60, or any other component in the fuel supply line 47 of the fuel dispenser 32.
After the fuel leaves the filter 50, the fuel enters into the fuel flow meter 52 to measure the amount of volumetric flow of fuel. The amount of volumetric flow of fuel is communicated to a controller 54 in the fuel dispenser 32 via a pulse signal line 56 from the fuel flow meter 52. The controller 54 typically transforms the pulses from the pulse signal line 56 into the total number of gallons dispensed and the total dollar amount charged to the customer, which is then typically displayed on LCD displays (not shown) on the fuel dispenser 32 visible to the customer. Note that the flow control valve 49 discussed above may be located on either the inlet or outlet side of the fuel flow meter 52.
After the fuel leaves the fuel flow meter 52, the fuel is delivered to the fuel supply piping 47 on the outlet side of the fuel flow meter 52 where it then reaches a hose 58. The hose 58 is coupled to a nozzle 60. The customer controls the flow of fuel from the hose 58 and nozzle 60 by engaging a nozzle handle (not shown) on the nozzle 60 as is well known.
If there is any blockage, either partially or wholly, in the fuel supply piping 47 within the fuel dispenser 32 or any components located inline to the fuel supply piping 47, the fuel cannot be delivered by the fuel dispenser 32 to a vehicle at the maximum throughput or flow rate that the fuel dispenser 32 would be capable of performing if no blockage existed. A blockage in the fuel supply piping 47 can occur within the piping 47 itself or as a result of a blockage in any of the components that are located inline to the fuel supply piping 47, including but not limited to the shear valve 48, the flow control valve 49, the filter 50, the fuel flow meter 52, the hose 58, and the nozzle 60. Also, if the submersible turbine pump that pumps fuel from the underground storage tank 34 to the fuel dispensers 32 is suffering from reduced performance and/or pumping rate, this may result in fuel dispensers 32 not delivering the maximum throughput or flow rate of fuel.
Any decline in the submersible turbine pump performance, a blockage in the fuel supply piping 47, or a blockage in components located inline to the fuel supply piping 47 may cause the fuel dispenser 32 to either not deliver fuel at all or at a reduced rate, thereby reducing the throughput efficiency of the fuel dispenser 32 and possibly requiring a customer to spend more time refueling a vehicle. The customer may be frustrated and therefore not visit the same service station for his or her fueling needs. The reduced throughput of the fuel dispenser 32 may also cause other customers to wait longer for a fueling position thereby resulting in lost revenue in terms of lost opportunity revenues. If the fuel dispenser 32 throughput efficiency can be measured and then compared against a normal throughput in an automated manner, fuel dispenser 32 throughput problems can be detected shortly after their occurrence to allow a station operator and/or service personnel to remedy the problem more quickly.
Until the present invention, one method known for monitoring the throughput efficiency of a fuel dispenser 32 is to calculate the flow rate of the fuel dispenser 32. The flow rate is the amount of fuel delivered by the fuel dispenser 32, as measured by the fuel flow meter 32, over the period of time that the fuel was flowing. For example, if a fuel dispenser 32 delivers ten gallons of fuel to a vehicle in a two minute dispensing transaction, the flow rate of the fuel dispenser 32 is five gallons per minute. The fuel dispenser 32 may determine the flow rate by dividing the volume of fuel dispensed, as measured by the fuel flow meter 52, by time, or the flow rate may be determined manually by dividing the volume of fuel delivered as indicated by the fuel dispenser 32 volume display by time. However, with these techniques, several issues can occur which will inaccurately reduce the measured flow rate from the true maximum flow rate capability of the fuel dispenser 32. For example, the nozzle may not be fully engaged during the entire dispensing event thereby reducing the volume throughput and also the calculated flow rate. If the fuel dispenser 32 were to start a timer when performing a flow rate calculation based on the activation and deactivation of the fuel dispenser 32, the timer may start before fuel flow begins thereby causing the time factor in the flow rate calculation to include what is known as “dead time.”
When the customer desires to end the dispensing event, the customer will disengage the nozzle 60 handle (labeled as “Flow End”) and then deactivate the fuel dispenser 32. This deactivation causes a “Dispense End” message to occur. This message is received by the tank monitor 36, the site controller 26, and/or another control system, and indicates the ending time of the dispensing event, the fueling point number or name, and the total amount and/or running totalizer amount of fuel dispensed. The time between disengaging the nozzle 60 handle and deactivating the fuel dispenser 32 is also “dead time.” As you can see in
Therefore, there exists a need to determine if a fuel dispenser 32 has a performance and/or blockage issue that is preventing the fuel dispenser 32 from dispensing the maximum flow rate possible even though the commonly available information from a dispensing event messages includes dead time and/or time of purposefully reduced dispensing flow rates.
The present invention relates to a system and method for determining the dispensing throughput of fuel dispensers in a service station environment using commonly available dispensing event information wherein the dead time and flow rate variability included in the information of the dispensing event is reduced and/or eliminated.
The present invention calculates the maximum dispensing efficiency of a fuel dispenser using the dispensing event information even though the dispensing event information includes dead time and/or purposefully reduced dispensing rates by a customer or due to automated prepay transaction flow reduction. A control system receives the dispensing event information for fuel dispensers and calculates what is known as a “maximum dispensing efficiency curve.” From this maximum dispensing efficiency curve, the control system can determine the maximum possible flow rate of a dispensing point, the minimum amount of “dead time,” of a dispensing point, or both, called the “maximum dispensing efficiency.” The “maximum dispensing efficiency” calculation is used to detect the difference between true blockages and/or performance issues versus reduced flow rates caused by other means, such as the customer varying the flow rate via the nozzle, nozzle snaps, or performance problems with the fuel pump used to pump fuel from an underground storage tank to a fuel dispenser.
In one embodiment, the “best of bins” mathematical technique is used to determine the maximum dispensing efficiency curve from a sample set of volume and time pair measurements for a dispensing point. Each volume and time pair measurement is comprised of the volume of fuel dispensed over the measured amount of time for one dispensing event. The slope of the maximum dispensing efficiency curve and/or the minimum “dead time” is calculated to arrive at a maximum dispensing efficiency for the dispensing point. This maximum dispensing efficiency can be further analyzed to determine if the dispensing point contains a true blockage and/or performance problem.
In another embodiment, an “iterative fit” mathematical technique is used to determine the maximum dispensing efficiency curve from the sample set of volume and time pair measurements for a dispensing point. The slope of the maximum dispensing efficiency curve is calculated to arrive at a maximum dispensing efficiency for the dispensing point. This maximum dispensing efficiency can be further analyzed to determine if the dispensing point contains a true blockage and/or performance problem.
In another embodiment, a “Hough” mathematical technique is used to determine the maximum dispensing efficiency curve, and may be used as a pre-filtering technique for the other mathematical techniques of determining the maximum dispensing efficiency curve. The slope of the maximum dispensing efficiency curve and/or the minimum “dead time” is calculated to arrive at a maximum dispensing efficiency for the dispensing point. This maximum dispensing efficiency can be further analyzed to determine if the dispensing point contains a true blockage and/or performance problem.
If the control system determines that the maximum dispensing efficiency for a dispensing point is less that it should be, this is a result of a blockage and/or performance problem at the fuel dispenser, since the maximum dispensing efficiency cureve has essentially removed the inclusion of “dead time” from the calculation. In this instance, the control system can generate an alarm, send a message to a site controller and/or tank monitor, notify an operator and/or service personnel, and/or send a message to an off-site system.
The control system may use a number of techniques for determining if the maximum dispensing efficiency of a dispensing point indicates a blockage or performance problem. The control system may compare the maximum dispensing efficiency of a dispensing point to a threshold value stored in memory or calculated in real time according to a formula, The control system may compare the maximum dispensing efficiency of a dispensing point to all other maximum dispensing efficiencies for all other dispensing points. The control system may compare the currently calculated maximum dispensing efficiency of a dispensing point to past calculated maximum dispensing efficiencies for the dispensing point to determine if an anomaly exists.
Those skilled in the art will appreciate the scope of the present invention and realize additional aspects thereof after reading the following detailed description of the invention in association with the accompanying drawing figures.
The accompanying drawing figures incorporated in and forming a part of this specification illustrate several aspects of the invention, and together with the description serve to explain the principles of the invention.
The embodiments set forth below represent the necessary information to enable those skilled in the art to practice the invention and illustrate the best mode of practicing the invention. Upon reading the following description in light of the accompanying drawing figures, those skilled in the art will understand the concepts of the invention and will recognize applications of these concepts not particularly addressed herein. It should be understood that these concepts and applications fall within the scope of the disclosure and the accompanying claims.
After enough volume and time pair measurements have been made, a control system that receives the dispensing events for fuel dispensers 32, such as the site controller 26 or tank monitor 32 for example, calculates what is known as a maximum dispensing efficiency curve 62. From this maximum dispensing efficiency curve 62, the control system can determine the maximum possible flow rate of a fuel dispenser 32 called the “maximum dispensing effidency.” In turn, calculation of the maximum possible flow rate of a fuel dispenser 32 also allows the determination of the minimum possible dead time of a fuel dispenser 32 since a volume and time pair measurement for a fuel dispenser 32 using the dispensing events will always include some amount of “dead time” and the volume and time pairs that represent this maximum dispensing efficiency will have the minimum possible “dead time.” Thus, the maximum dispensing efficiency as used herein can mean the maximum possible flow rate of the dispensing point 32, the minimum amount of “dead time” for the dispensing point 32, or both. The “maximum dispensing efficiency” calculation is used to detect the difference between true blockages and/or performance issues versus reduced flow rates caused by other means, such as the customer varying the flow rate via the nozzle 60, nozzle 60 snaps, or performance problems with the fuel pump used to pump fuel from the underground storage tank 34 to a fuel dispenser 32.
As illustrated in
Note that since the measured volume and time pairs calculated for dispensing events at a fuel dispenser 32 are based on the volume of fuel measured by the fuel flow meter 52, the maximum dispensing efficiency can be determined for each fuel flow meter 52 that is present in a fuel dispenser 32 independently for fuel dispensers 32 that contain more than one fuel flow meter 52. Depending on configuration of the fuel dispenser 32, the fuel dispenser 32 may be capable of dispensing fuel to a vehicle at more than one “dispensing point.” A “dispensing point” is present for each point at which fuel can be delivered from a fuel dispenser 32. For example, in the case of a three-product fuel dispenser 32 that is not a blending fuel dispenser, the fuel dispenser 32 will have three separate fuel flow meters 52—one for each of the three different grades of fuel. The fuel will either be delivered to its own dedicated separate hose 58 and nozzle 60, or to a single hose 58 and nozzle 60 that is coupled to each fuel flow meter 52.
In the above three hose 58 and nozzle 60 example, there are three dispensing points where a maximum dispensing efficiency can be calculated for each dispensing point independently. In the above one hose 58 example, there are still three fuel flow meters 52, but only one hose 58 and nozzle 60. This configuration only has one dispensing point, but three maximum dispensing efficiencies can still be calculated since there are three fuel flow meters 52. If the blockage is present in the hose 58 of such a fuel dispenser 32, all three maximum dispensing efficiencies calculated for each fuel flow meter 58 will be affected. If the blockage or performance problem is present before the fuel supply lines 47 from each of the fuel flow meters 52 are coupled to the single hose 58 and nozzle 60, then only the maximum dispensing efficiency for the fuel flow meter 52 with the blockage or performance problem will be affected.
If a fuel dispenser 32 has the capability of determining flow rates of its dispensing events, there is an alternative method of determining and recording volume and time pair measurements (block 102 in
Determining volume and time pair measurements from this alternative embodiment is still useful in determining the maximum dispensing efficiency of a suffering from a performance problem, or if individual dispenses were performed at lower flow rates due to human or other cause, the flow rate calculated by the fuel dispenser 32 will be less than optimal and hence the volume and time pair measurement deduced from the calculated flow rate and volume information will represent a less than optimal dispensing event efficiency; however the maximum dispensing efficiency then calculated will represent the maximum attainable flow rate.
In summary, the present invention has the ability to determine a blockage and/or performance problem in a fuel dispenser 32 on a dispensing point by dispensing point basis. The application will refer to fuel dispenser 32 and dispensing point 32 interchangeably hereafter since the determination of the maximum dispensing efficiency is based on the dispensing point 32 of which a fuel dispenser 32 may have one or more.
After enough volume and time pair measurements for dispensing events at a dispensing point 32 have been accumulated and recorded, the control system determines the maximum dispensing efficiency of the dispensing point 32 (block 106). Each of these plurality of volume and time pair measurements for dispensing events of a dispensing point 32 can be represented in two-dimensional table of volume of fuel versus dispensing times, as illustrated in
The slope and time axis intercept of this maximum dispensing efficiency curve 62 (8.3 GPM as illustrated in
As shown in
After the control system determines the maximum dispensing efficiency for a dispensing point 32, the control system stores this calculation for future analysis to detect if a blockage and/or performance issue exists within the dispensing point 32 (block 108). This process repeats as illustrated in
In
After the initial maximum dispensing efficiency curve 62 is determined, the control system determines boundary lines on each side of the initial maximum dispensing efficiency curve 62 based on the statistical variability in the volume and time pair measurements to determine all of the volume and time pair measurements that fit within the boundaries. The process of finding the best line fit to the volume and time pair measurements is then again repeated, but only using the events that fit within the previously determined boundaries and excluding all others. This process is repeated iteratively until one of several limits is reached. One limit goal is when the line fits the remaining points well based on the standard deviation of the residuals. Another limit could be to stop iterations when the slope of each successive fitted line stops changing by a determined significant amount. Yet another limit could be to stop iterations when the standard deviation of the residuals of each successive fitted line stops changing by a determined amount. After the iterative process is finished by reaching one of the limits defined, the maximum dispensing efficiency of the dispensing point 32 is determined as the slope of the finalized maximum dispensing efficiency curve 62 and the minimum dead time is determined as the time axis intercept.
R=V/(T−d)
Note that a point in (T, V) space actually maps to an infinite number of points in (d, R) space (it maps to a hyperbola). The time runs along the X-axis in both spaces, and Volume (in (T, V) space) and Rate (in (d, R) space) run along the Y-axis The Hough transform limits the solution space with minimum and maximum values for d and R (block 164), then partitions it into N×M rectangular regions (bins) (block 166). The center of each bin is a distinct point (dc. Rc). Each point (dc, Rc) has a vote counter assigned to it (block 168). In this example, the minimum and maximum values of the solution space are set by the physical system being modeled, and are usually on the order of Rε(0 gpm, 20 gpm) and dε(0 seconds, 30 seconds). Usually, N is chosen so that the bins are 1 to 5 seconds wide, and M is chosen so that the bins are 0.1 to 1.0 GPM tall. These configuration parameters are configurable, and can change for different applications, such as diesel dispensers instead of gasoline dispensers, etc.
At this point, there are actually two different implementations of the Hough algorithm that may be used in this embodiment called the “Time Hough” and the “Rate Hough.” For the “Time Hough” transform, the control system takes each point in the dispensing event volume and time space (T, V), iterates through all the valid values for “dc,” maps the valid values to the Hough space (dc, Rc), and increments the vote counter at the location. For the “Rate Hough” transform, the control system takes each point in the dispensing event volume and time space (T, V), iterates through all the valid values for “Rc” (Rate Hough), maps the valid values to the Hough space (dc, Rc), and increments the vote counter at the location. In either case of the “Time Hough” transform or the “Rate Hough” transform, the control system determines the bin with the highest vote count and chooses this bin as the solution, and all the points in (T, V) space that voted for that bin by the control system are selected as the points on the maximum dispensing efficiency curve 62 (block 170), and the process ends (block 172).
In an alternate of this embodiment, the pair of bins (adjacent in ‘d’ for “Time Hough,” and ‘R’ for “Rate Hough”) with the highest combined vote count is selected by the control system Also, the control system may use the described “Hough” transforms as a filter to the volume and time pair measurements, rather than to obtain the maximum dispensing efficiency curve 62. After the volume and time pair measurements are filtered via the points selected from one of the aforementioned “Hough” transforms, the remaining volume and time pair measurements selected by the filtering are fed to a standard least-squared-error fit straight line algorithm, or any of the aforementioned techniques of fitting a line to volume and time pair measurements to determine the maximum dispensing efficiency curve 62.
It is also possible to provide pre-filtering to the volume and time pair measurements before such measurements are processed by a “Hough” transform in order to provide better data for the “Hough” Transform. The technique is known as a “Binning Algorithm,” and may be used as a pre-processor on the volume and time pair measurements before a “Hough” transform is performed or before any of the previously described techniques for fitting a line through the volume and time pair measurements is made.
The binning algorithm can take on three forms according to the present invention: “Volume Binning,” “Time Binning,” and “Volume/Time Binning.” The Volume Binning algorithm works by creating a series of bins representing ranges of dispensed volume in volume and time pair measurements (T, V) space. The control system then distributes all of the available volume and time pair measurements for dispensing events into these bins, and selects from each bin the dispensing event with the lowest time (T) value. The “Time Binning” algorithm works by creating a series of bins representing ranges of time (T) in the volume and time pair measurements (T, V) space. The control system then distributes all the available dispensing events into these bins, and selects from each bin the dispensing event with the highest volume (V) value. The Time/Volume Binning algorithm works by creating the union of points returned by the Volume Binning and Time Binning algorithms. This algorithm attempts to ameliorate the limitations of one algorithm by the other. After a binning algorithm is performed on the volume and time pair measurements, any of the aforementioned line fitting techniques may be used to determine the maximum dispensing efficiency curve 62.
Now that the maximum dispensing efficiency of a dispensing point 32 can be calculated, the control system can analyze the maximum dispensing efficiency of a dispensing point 32 to determine if the dispensing point 32 is experiencing a blockage or performance problem since the dead time in such calculation has theoretically been eliminated for all practical purposes. If the control system determines that the maximum dispensing efficiency of the dispensing point 32 is not as expected, the control system can take automated measures on its own to trigger an investigation of the dispensing point 32 so that any problems can be alleviated quickly and without having to wait until a service station operator or service personnel recognizes the problem manually or via customer complaints on slow dispensing point 32 throughput.
If the maximum dispensing efficiency for the dispensing point 32 was not significantly lower than the threshold value in decision 204, the control system next determines if the maximum dispensing efficiency is significantly higher than the threshold value (decision 210). The threshold value in this instance is selected such that a positive answer to decision 210 means that the maximum dispensing efficiency calculated is higher than possible and therefore an error condition exists that should be logged and/or reported via an alarm (block 212). If the answer to decision 210 is negative, this means that the maximum dispensing efficiency was not either greater than normal or lower than normal and thus no error or alarm conditions exists—i.e. a blockage or performance problem does not exist.
The process in
Those skilled in the art will recognize improvements and modifications to the preferred embodiments of the present invention. All such improvements and modifications are considered within the scope of the concepts disclosed herein and the claims that follow.
