WO2009055967A1 - Real-time model validation - Google Patents

Real-time model validation Download PDF

Info

Publication number
WO2009055967A1
WO2009055967A1 PCT/CN2007/003100 CN2007003100W WO2009055967A1 WO 2009055967 A1 WO2009055967 A1 WO 2009055967A1 CN 2007003100 W CN2007003100 W CN 2007003100W WO 2009055967 A1 WO2009055967 A1 WO 2009055967A1
Authority
WO
WIPO (PCT)
Prior art keywords
real time
embedded
values
simulated
output values
Prior art date
Application number
PCT/CN2007/003100
Other languages
French (fr)
Inventor
Danqing Zhang
Chenzhou Ye
Ming Ge
Guantien Tan
Original Assignee
Honeywell International Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Honeywell International Inc. filed Critical Honeywell International Inc.
Priority to PCT/CN2007/003100 priority Critical patent/WO2009055967A1/en
Publication of WO2009055967A1 publication Critical patent/WO2009055967A1/en

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B17/00Systems involving the use of models or simulators of said systems
    • G05B17/02Systems involving the use of models or simulators of said systems electric

Definitions

  • SIS Supervisory information systems
  • a typical SIS may contain at least one integrated and flexible real-time database and provide tools for process monitoring and management of a power generation system.
  • An SIS may also perform economic performance calculations and analysis, optimal operation scheduling and guidance, equipment health management and fault diagnosis.
  • the tools may be modularized software packages, each of which reads data from the real-time database, performs some calculations, and writes the calculations back to the real-time database. Most of the tools employ one or several mechanistic or statistical models, or hybrids of them to perform special purpose calculations. If the estimation or prediction by the models deviates significantly from reality, the final results of the affected tools can be meaningless, misleading, or sometimes dangerous.
  • FIG. 1 is a block diagram of a Supervisory information system (SIS) according to an example embodiment.
  • FIG. 2 illustrates the relation between the optimizer and the simulation model both embedded in an optimization tool of SIS
  • FIG. 3 is time graph illustrating a measured boiler efficiency, an estimated boiler efficiency to optimized settings over time by the SIS of FIG.1 according to an example embodiment.
  • FIG. 4 is a block diagram representing how graph lines corresponding to real time performance and the performance estimations to current conditions and optimized settings are generated.
  • FIG. 5 is time graph illustrating a measured boiler efficiency, an estimated boiler efficiency over time by the SIS of FIG.1 , and a predicted boiler efficiency with modified simulation parameters according to an example embodiment with the feature of online model validation.
  • FIG. 6 is a block diagram of an example computer system for executing code corresponding to the SIS and other functions.
  • the software may consist of computer executable instructions stored on computer readable media such as memory or other type of storage devices.
  • computer readable media is also used to represent any means by which the computer readable instructions may be received by the computer, such as by tangible media and different forms of wired or wireless transmissions.
  • modules which are software, hardware, firmware or any combination thereof. Multiple functions may be performed in one or more modules as desired, and the embodiments described are merely examples.
  • the software may be executed on a digital signal processor, ASIC, microprocessor, or other type of processor operating on a computer system, such as a personal computer, server or other computer system.
  • a supervisory information system (SIS) illustrated generally at 100 in FIG. 1 may be used to manage, monitor and optimize the overall performance of a power plant 1 10 (coal fired, combined cycle, etc.).
  • SIS 100 may include multiple software modules or tools such as 120, 125, 130, and 135 to calculate and monitor performance of the power plant 1 10 at both component level and system level, and give optimization recommendations to different parts of the power plant.
  • An SIS manager 1 15 maybe used to monitor and coordinate the work of all these tools.
  • the tools retrieve the data from database 140 through or not through SIS manager, perform some kind of calculation, monitoring , or optimization and then send the calculation results to a visualization panel 150 (can be a common component shared by all SIS tools as displayed in the example or a dedicated one attached to every tool independently) and meanwhile save calculation results back to database 140 through or not through SIS manager 1 15.
  • a visualization panel 150 can be a common component shared by all SIS tools as displayed in the example or a dedicated one attached to every tool independently
  • features have been added to both the tool side and the side of visualization panel 150.
  • Most of the tools employ one or several mechanistic or statistical models or hybrids of them to perform special purpose calculations or optimizations.
  • a load dispatch tool of SIS needs one or several models to accurately estimate the fuel consumption rate of every generating unit under given various operating conditions so that it can generate a power generation plan that can fulfill the load demand with the minimum amount of fuel.
  • a combustion optimization tool may use one or several models to estimate the efficiency of a target boiler under all possible conditions so that it can find the optimal settings after trying all these conditions.
  • An example optimization process 200 is illustrated in FIG. 2. Given a current set of working conditions X at 205, an optimizer 210 will generate a set of possible changes to X; send a modified set of conditions X' at 215 to a simulation model M at 220 which is supposed to closely mimic the behavior of a target system or component S to see if X' 215 can bring better performance.
  • an estimated resulting performance M(X') at 225 the optimizer 210 will generate another set of new settings in some heuristic way and send it to the simulation again to evaluate the resulting performance. The iteration will be terminated if the result is good enough or some other criteria have been fulfilled, resulting in an out of optimized settings or optimal changes at 225.
  • performance of these types of tools relies heavily on the reliability or validation of these embedded models. If the estimation or prediction of the models deviates significantly from the reality, the final results of the affected tools can be meaningless, or misleading, or sometimes dangerous.
  • the time series graph 500 may be stored on a tangible computer readable medium such as a memory device.
  • Graph 500 may also be displayed by display 155 to provide a user the ability to check, in an online or real time mode, validation of the model which is embedded in an optimization system as a fundamental component to simulate output of a specific real system under various conditions. If real time data of both the actual output of the target system and the simulated output of the embedded model are available, they are both illustrated simultaneously in the same time series graph so that the discrepancy between them over a long enough and up to now period can be visually displayed.
  • the time range in one embodiment may cover at least one complete running cycle of the system, so that recent response of the model under various conditions can be observed. Based on the graph, the user can easily judge whether the model embedded in the optimization system is still valid and furthermore make the right decision about whether to adopt the optimized setting or not.
  • FIG. 4 illustrates a trend graph visualization of performance in real time at 410, an optimized performance at 415 and an estimated simulation model performance at 420. In embodiments without the feature provided by this invention, the simulation model performance may not be illustrated. In one embodiment with the feature provided by this invention, the current settings at 425 will be sent to the simulation model 220 to get the estimation from the model for current settings. The estimation will be displayed on trend graph together with the real time data and estimated resulting performance.
  • Graph 300 is a time series graph with a time range from some past time to current time in one embodiment.
  • Line 320 corresponds to the change of actual boiler efficiency, the real data, while line 310 depicts the change of estimated boiler efficiency if the optimized settings recommended by the combustion optimization system take effect.
  • the estimated efficiency is about 1% higher than the real one, which makes the optimized settings quite attractive, the user has no idea about the reliability of the result without knowing the validation of the model embedded in the optimization system at current time.
  • a time series graph is utilized to simultaneously illustrate the change of both the system output and a simulated output from an embedded model in a real time mode so that discrepancies between the real output and model output over a long enough period extending up to real time in one embodiment, may be visually displayed.
  • the time range in one embodiment covers at least one complete running cycle of the system, so that recent response of the model under various conditions may be observed. Based on the graph, a user may easily judge whether the model embedded in the optimization system is still valid and furthermore make the right decision about weather to adopt the optimized setting or not.
  • the validation of the model can be considered as high at current time; otherwise the validation of the model is considered as poor, which may inform the user to adjust some part of the model or rebuild the whole model to make it adapt to the reality.
  • the model may be updated in an automated manner.
  • S represents the real system (for example a boiler)
  • X represents one or more inputs (set points of manipulated variables of the boiler) to the system
  • Y represents the true response (boiler efficiency) of the system given X . This fact can be expressed in the following equation:
  • the embedded module or tool has stored real time values in database 140. If sufficient real time values of both X and Y is available from database 140, the visualization tool 150 retrieves the values corresponding to the desired embedded tool, and creates a time series graph to simultaneously illustrate the change of both Y and Y in a real time mode so that the discrepancy between Y and Y over a long enough and up to now period can be visually reflected.
  • the time range should cover at least one complete running cycle of the system if there is one, so that recent response of the model under various conditions can be observed.
  • each pair of variables (one from Y together with its counterpart from Y ) can either be displayed in the same graph or displayed in respective graphs. If the discrepancy is small enough even when the working conditions change, the validation of the model can be considered as high at a current time; otherwise the validation of the model is considered as poor, which may motivate the user to adjust some part of the model or rebuild the whole model to make it adapt to the reality.
  • a mechanism can be added to update the model in an automatic way, such as incrementally changing parameters of one or more functions as a result of the difference between simulated and real outputs over time. Neural networks or other expert systems may be used to further tune models.
  • FIG. 5 illustrates a time series graph 500 with a time range from some past time to current time.
  • the line 510 represents the change of actual boiler efficiency or Y while line 520 represents the change of estimated boiler efficiency if the optimized settings recommended by a combustion optimization system take effect.
  • the combustion optimization system may be an expert system that may be part of the SIS that calculates settings to optimize performance of a system being modeled.
  • An additional line 530 represents the boiler efficiency predicted by the model (model output or Y ) given actual settings (inputs to the model) are added. Note line 530 line 520 originate from the same model embedded in the optimization system. During the whole time range displayed in the graph, the discrepancy between line
  • model validation indices such as Expected Value / Variance / Standard Deviation of the model deviation and the linear coefficients between model output and actual value may also be visualized in further embodiments on the screen.
  • FIG. 6 A block diagram of a computer system that executes programming for performing the above algorithm is shown in FIG. 6.
  • a general computing device in the form of a computer 610 may include a processing unit 602, memory 604, removable storage 612, and non-removable storage 614.
  • Memory 604 may include volatile memory 606 and non-volatile memory 608.
  • Computer 610 may include — or have access to a computing environment that includes - a variety of computer- readable media, such as volatile memory 606 and non-volatile memory 608, removable storage 612 and non-removable storage 614.
  • Computer storage includes random access memory (RAM), read only memory (ROM), erasable programmable read-only memory (EPROM) & electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD ROM), Digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium capable of storing computer-readable instructions.
  • Computer 610 may include or have access to a computing environment that includes input 616, output 618, and a communication connection 620. The computer may operate in a networked environment using a communication connection to connect to one or more remote computers.
  • the remote computer may include a personal computer (PC), server, router, network PC, a peer device or other common network node, or the like.
  • the communication connection may include a Local Area Network (LAN), a Wide Area Network (WAN) or other networks.
  • LAN Local Area Network
  • WAN Wide Area Network
  • Computer-readable instructions stored on a computer-readable medium are executable by the processing unit 602 of the computer 610.
  • a hard drive, CD-ROM, and RAM are some examples of articles including a computer- readable medium.
  • Plant contain a trend graph simultaneously displaying real time performance and the performance estimations to current conditions and optimized settings.
  • the estimations are made by the simulation model used in the optimization tool for a system or component.
  • the trend graph displays data of a long enough duration to cover at least one complete cycle of the target system or component.
  • the trend graph may help a user to judge whether the model embedded in the optimization tool is still valid and furthermore make the right decision about whether to adopt the optimized setting or not.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

A system and method include optimization tools that interface with at least one embedded model corresponding to a component or process in the plant that simulates expected output values as a function of real time input values. A visualizer receives the simulated expected output values and the real time output values and provides a time series graph signal on a computer readable medium that may be displayed to illustrate discrepancies between the simulated expected output values and the real time output values.

Description

Real-Time Model Validation
Background
[0001] Supervisory information systems (SIS) manage, monitor and optimize overall performance of power plants. SIS is considered by power generating companies as one of the major solutions to further improve plant-wide operating efficiency. A typical SIS may contain at least one integrated and flexible real-time database and provide tools for process monitoring and management of a power generation system. An SIS may also perform economic performance calculations and analysis, optimal operation scheduling and guidance, equipment health management and fault diagnosis. [0002] The tools may be modularized software packages, each of which reads data from the real-time database, performs some calculations, and writes the calculations back to the real-time database. Most of the tools employ one or several mechanistic or statistical models, or hybrids of them to perform special purpose calculations. If the estimation or prediction by the models deviates significantly from reality, the final results of the affected tools can be meaningless, misleading, or sometimes dangerous.
[0003] Although testing the models with historical data can be used to evaluate the model validation in history, SIS designs provide no effective methods to verify the validation of embedded models in real-time or online mode. SIS users are almost blind to the validity of models at the current time, which can adversely affect the reliability of an SIS.
Brief Description of the Drawings
[0004] FIG. 1 is a block diagram of a Supervisory information system (SIS) according to an example embodiment. [0005] FIG. 2 illustrates the relation between the optimizer and the simulation model both embedded in an optimization tool of SIS [0006] FIG. 3 is time graph illustrating a measured boiler efficiency, an estimated boiler efficiency to optimized settings over time by the SIS of FIG.1 according to an example embodiment.
[0007] FIG. 4 is a block diagram representing how graph lines corresponding to real time performance and the performance estimations to current conditions and optimized settings are generated.
[0008] FIG. 5 is time graph illustrating a measured boiler efficiency, an estimated boiler efficiency over time by the SIS of FIG.1 , and a predicted boiler efficiency with modified simulation parameters according to an example embodiment with the feature of online model validation.
[0009] FIG. 6 is a block diagram of an example computer system for executing code corresponding to the SIS and other functions.
Detailed Description [0010] In the following description, reference is made to the accompanying drawings that form a part hereof, and in which is shown by way of illustration specific embodiments which may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized and that structural, logical and electrical changes may be made without departing from the scope of the present invention. The following description of example embodiments is, therefore, not to be taken in a limited sense, and the scope of the present invention is defined by the appended claims. [0011] The functions or algorithms described herein may be implemented in software or a combination of software and human implemented procedures in one embodiment. The software may consist of computer executable instructions stored on computer readable media such as memory or other type of storage devices. The term "computer readable media" is also used to represent any means by which the computer readable instructions may be received by the computer, such as by tangible media and different forms of wired or wireless transmissions. Further, such functions correspond to modules, which are software, hardware, firmware or any combination thereof. Multiple functions may be performed in one or more modules as desired, and the embodiments described are merely examples. The software may be executed on a digital signal processor, ASIC, microprocessor, or other type of processor operating on a computer system, such as a personal computer, server or other computer system.
[0012] A supervisory information system (SIS) illustrated generally at 100 in FIG. 1 may be used to manage, monitor and optimize the overall performance of a power plant 1 10 (coal fired, combined cycle, etc.). SIS 100 may include multiple software modules or tools such as 120, 125, 130, and 135 to calculate and monitor performance of the power plant 1 10 at both component level and system level, and give optimization recommendations to different parts of the power plant. An SIS manager 1 15 maybe used to monitor and coordinate the work of all these tools. The tools retrieve the data from database 140 through or not through SIS manager, perform some kind of calculation, monitoring , or optimization and then send the calculation results to a visualization panel 150 (can be a common component shared by all SIS tools as displayed in the example or a dedicated one attached to every tool independently) and meanwhile save calculation results back to database 140 through or not through SIS manager 1 15. In one embodiment, features have been added to both the tool side and the side of visualization panel 150. [0013] Most of the tools employ one or several mechanistic or statistical models or hybrids of them to perform special purpose calculations or optimizations. For example, a load dispatch tool of SIS needs one or several models to accurately estimate the fuel consumption rate of every generating unit under given various operating conditions so that it can generate a power generation plan that can fulfill the load demand with the minimum amount of fuel. A combustion optimization tool may use one or several models to estimate the efficiency of a target boiler under all possible conditions so that it can find the optimal settings after trying all these conditions. [0014] An example optimization process 200 is illustrated in FIG. 2. Given a current set of working conditions X at 205, an optimizer 210 will generate a set of possible changes to X; send a modified set of conditions X' at 215 to a simulation model M at 220 which is supposed to closely mimic the behavior of a target system or component S to see if X' 215 can bring better performance. Based on the simulation result, an estimated resulting performance M(X') at 225, the optimizer 210 will generate another set of new settings in some heuristic way and send it to the simulation again to evaluate the resulting performance. The iteration will be terminated if the result is good enough or some other criteria have been fulfilled, resulting in an out of optimized settings or optimal changes at 225. [0015] Needless to say that performance of these types of tools relies heavily on the reliability or validation of these embedded models. If the estimation or prediction of the models deviates significantly from the reality, the final results of the affected tools can be meaningless, or misleading, or sometimes dangerous. [0016] Although testing the models with historical data, which is a common practice, can evaluate the model validation in history, SlS designs provide no effective and direct methods to monitor and verify the validation of embedded models in real-time or online mode. SIS users are almost blind to the validation of models at current time, which can be detrimental to the reliability of SIS. [0017] The visualization tool creates a signal embodying a time series graph
500 such as that shown in FIG. 5, that includes at least one estimate 530 as well as a real time value 510 from a target system or component. The time series graph 500 may be stored on a tangible computer readable medium such as a memory device. Graph 500 may also be displayed by display 155 to provide a user the ability to check, in an online or real time mode, validation of the model which is embedded in an optimization system as a fundamental component to simulate output of a specific real system under various conditions. If real time data of both the actual output of the target system and the simulated output of the embedded model are available, they are both illustrated simultaneously in the same time series graph so that the discrepancy between them over a long enough and up to now period can be visually displayed. The time range in one embodiment may cover at least one complete running cycle of the system, so that recent response of the model under various conditions can be observed. Based on the graph, the user can easily judge whether the model embedded in the optimization system is still valid and furthermore make the right decision about whether to adopt the optimized setting or not. [0018] FIG. 4 illustrates a trend graph visualization of performance in real time at 410, an optimized performance at 415 and an estimated simulation model performance at 420. In embodiments without the feature provided by this invention, the simulation model performance may not be illustrated. In one embodiment with the feature provided by this invention, the current settings at 425 will be sent to the simulation model 220 to get the estimation from the model for current settings. The estimation will be displayed on trend graph together with the real time data and estimated resulting performance.
[0019] Referring to graph 300 in FIG. 3, the values are representative of combustion optimization in one embodiment. Graph 300 is a time series graph with a time range from some past time to current time in one embodiment. Line 320 corresponds to the change of actual boiler efficiency, the real data, while line 310 depicts the change of estimated boiler efficiency if the optimized settings recommended by the combustion optimization system take effect. Although the estimated efficiency is about 1% higher than the real one, which makes the optimized settings quite attractive, the user has no idea about the reliability of the result without knowing the validation of the model embedded in the optimization system at current time. Using a real time value of both a system input and output, a time series graph is utilized to simultaneously illustrate the change of both the system output and a simulated output from an embedded model in a real time mode so that discrepancies between the real output and model output over a long enough period extending up to real time in one embodiment, may be visually displayed. [0020] The time range in one embodiment covers at least one complete running cycle of the system, so that recent response of the model under various conditions may be observed. Based on the graph, a user may easily judge whether the model embedded in the optimization system is still valid and furthermore make the right decision about weather to adopt the optimized setting or not. If the discrepancy is by far small enough even when the working conditions change, the validation of the model can be considered as high at current time; otherwise the validation of the model is considered as poor, which may inform the user to adjust some part of the model or rebuild the whole model to make it adapt to the reality. In some embodiments, the model may be updated in an automated manner. [0021] In one example, the following method is utilized for real-time model validation. Suppose S represents the real system (for example a boiler), X represents one or more inputs (set points of manipulated variables of the boiler) to the system, and Y represents the true response (boiler efficiency) of the system given X . This fact can be expressed in the following equation:
Y = S(X) [0022] An embedded model M is setup to mimic the action of S . When the given input is X , its response is Y , a simulated response, which should be equal to Y if M is precise enough. This fact can be expressed by:
Y = M(X)
[0023] In the real applications, it is almost impossible to keep Y = Y , the actual response equal to the simulated response, so the discrepancy between them over a long enough period until the present time can be used to evaluate the consistency between the model and the reality. If the discrepancy is by far small enough even when the working conditions changes, the validation of the model can be considered as high at the current time; otherwise the validation of the model is considered as poor, which may require the user adjust some part of the model or rebuild the whole model to make it adapt to the reality.
[0024] In one embodiment, the embedded module or tool has stored real time values in database 140. If sufficient real time values of both X and Y is available from database 140, the visualization tool 150 retrieves the values corresponding to the desired embedded tool, and creates a time series graph to simultaneously illustrate the change of both Y and Y in a real time mode so that the discrepancy between Y and Y over a long enough and up to now period can be visually reflected. In one embodiment, the time range should cover at least one complete running cycle of the system if there is one, so that recent response of the model under various conditions can be observed.
[0025] Based on the graph, whether displayed on a display, printed, or displayed in some other manner, the user can easily judge whether the model embedded in the optimization system is still valid and furthermore make the right decision about whether to adopt an optimized setting or not. Note that if Y or
Y contains more than one variable, each pair of variables (one from Y together with its counterpart from Y ) can either be displayed in the same graph or displayed in respective graphs. If the discrepancy is small enough even when the working conditions change, the validation of the model can be considered as high at a current time; otherwise the validation of the model is considered as poor, which may motivate the user to adjust some part of the model or rebuild the whole model to make it adapt to the reality. [0026] In one embodiment, a mechanism can be added to update the model in an automatic way, such as incrementally changing parameters of one or more functions as a result of the difference between simulated and real outputs over time. Neural networks or other expert systems may be used to further tune models. Embedded models may be very different from each other and may be highly dependent on the entity that they are modeling. The type of tuning used may vary significantly between the models. That is one the characteristics of embedded systems that make them difficult to validate. The present SIS provides a means by which the embedded models may be tested for validation without the need to fully understand how each model works. By utilizing the outputs of the process and the simulation results of the model, each model may be tested in a similar manner without having the build additional validation functionality into each model. [0027] In a further example utilizing combustion optimization, FIG. 5 illustrates a time series graph 500 with a time range from some past time to current time. The line 510 represents the change of actual boiler efficiency or Y while line 520 represents the change of estimated boiler efficiency if the optimized settings recommended by a combustion optimization system take effect. The combustion optimization system may be an expert system that may be part of the SIS that calculates settings to optimize performance of a system being modeled. An additional line 530 represents the boiler efficiency predicted by the model (model output or Y ) given actual settings (inputs to the model) are added. Note line 530 line 520 originate from the same model embedded in the optimization system. During the whole time range displayed in the graph, the discrepancy between line
510 (Y ) and line 530 ( Y ) is small enough (less than 0.25%), which indicates a high consistency between the model and the reality at current time. Compared with the small discrepancy, the estimated efficiency after optimization is a significant improvement to the true value. Based on this observation, the user can have a high chance to make a right decision about whether to take the optimized settings or not at this time.
[0028] Note that in some cases when the noise is large enough to disturb the observation and the decision of operators, the original data should be pre-processed before shown on the screen. Additional model validation indices, such as Expected Value / Variance / Standard Deviation of the model deviation and the linear coefficients between model output and actual value may also be visualized in further embodiments on the screen.
[0029] A block diagram of a computer system that executes programming for performing the above algorithm is shown in FIG. 6. A general computing device in the form of a computer 610, may include a processing unit 602, memory 604, removable storage 612, and non-removable storage 614. Memory 604 may include volatile memory 606 and non-volatile memory 608. Computer 610 may include — or have access to a computing environment that includes - a variety of computer- readable media, such as volatile memory 606 and non-volatile memory 608, removable storage 612 and non-removable storage 614. Computer storage includes random access memory (RAM), read only memory (ROM), erasable programmable read-only memory (EPROM) & electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD ROM), Digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium capable of storing computer-readable instructions. Computer 610 may include or have access to a computing environment that includes input 616, output 618, and a communication connection 620. The computer may operate in a networked environment using a communication connection to connect to one or more remote computers. The remote computer may include a personal computer (PC), server, router, network PC, a peer device or other common network node, or the like. The communication connection may include a Local Area Network (LAN), a Wide Area Network (WAN) or other networks. [0030] Computer-readable instructions stored on a computer-readable medium are executable by the processing unit 602 of the computer 610. A hard drive, CD-ROM, and RAM are some examples of articles including a computer- readable medium. Conclusion
[0031] Optimization tools of a Supervisory Information System of Power
Plant contain a trend graph simultaneously displaying real time performance and the performance estimations to current conditions and optimized settings. The estimations are made by the simulation model used in the optimization tool for a system or component. The trend graph displays data of a long enough duration to cover at least one complete cycle of the target system or component. The trend graph may help a user to judge whether the model embedded in the optimization tool is still valid and furthermore make the right decision about whether to adopt the optimized setting or not.
[0032] The Abstract is provided to comply with 37 CF. R. § 1.72(b) to allow the reader to quickly ascertain the nature and gist of the technical disclosure. The Abstract is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims.

Claims

1. A system comprising: a computer system having memory containing computer programs for running at least one embedded model corresponding to a component or process in the plant that simulates expected output values as a function of real time input values; and a visualizer that receives the simulated output values corresponding to current conditions and optimized settings and the real time output values and provides a time series graph signal on a computer readable medium that may be displayed to illustrate discrepancies between the simulated expected output values and the real time output values.
2. The system of claim 1 wherein the at least one embedded model is coupled to a real time database for reading and writing data to the real time database.
3. The system of claim 1 wherein the one or more embedded models are independent of each other.
4. The system of claim 1 wherein the industrial plant comprises a power plant.
5. The system of claim 4 wherein at least one embedded model corresponds to fuel consumption of a boiler or combustion efficiency of the boiler.
6. The system of claim 1 wherein the time series graph signal includes a line representative of real time output values and a line representative of simulated real time output values for the current input values.
7. The system of claim 6 and further comprising an optimization system that provides a different set of input values to optimize performance, and wherein the graph signal includes a line representative of simulated output values for the different set of input values.
8. The system of claim 1 wherein the time series graph includes time series values extending at least a complete running cycle of the component or process.
9. A system comprising: a plurality of embedded models which simulate multiple components within a power plant and access a database to read and store data related to inputs to the components and outputs from the simulation of the component; and means for creating and storing an output signal from which time series data corresponding to actual component output and simulated component output from one or more embedded models can be visually compared by a user.
10. The system of claim 9 wherein the at least one embedded model is coupled to the real time database for reading and writing data to the real time database.
1 1. The system of claim 9 wherein one or more embedded models are independent of each other.
12. The system of claim 9 wherein at least one embedded model corresponds to fuel consumption of a boiler or combustion efficiency of the boiler.
13. The system of claim 9 wherein the time series data corresponds to a time series graph that includes a line representative of real time output values and a line representative of simulated real time output values for the current input values.
14. The system of claim 13 and further comprising an optimization system that provides a different set of input values to optimize performance, and wherein the time series graph includes a line representative of simulated output values for the different set of input values.
15. The system of claim 13 wherein the time series graph includes time series values extending at least a complete running cycle of the component or process.
16. The system of claim 9 wherein the SIS comprises functionality to manage, monitor and optimize the overall performance of the plant
17. A method comprising: operating multiple embedded systems that simulate various components of a power plant; interfacing the multiple embedded models with a database that stores inputs to the components of the plant, real time performance of the components of the plant and simulated performance from the multiple embedded models; and generating a signal for storage or display that is representative of a graph illustrating time series data corresponding to at least one actual output of a component and a simulated output of the component from at least one of the embedded models.
18. The method of claim 17 wherein one or more embedded models are independent of each other.
19. The method of claim 17 wherein at least one embedded models simulates fuel consumption of a boiler or combustion efficiency of the boiler.
20. The method of claim 17 and further comprising: calculating optimization input values to optimize performance of a component, and wherein the graph includes a line representative of simulated output values for the optimization input values.
PCT/CN2007/003100 2007-10-31 2007-10-31 Real-time model validation WO2009055967A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2007/003100 WO2009055967A1 (en) 2007-10-31 2007-10-31 Real-time model validation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2007/003100 WO2009055967A1 (en) 2007-10-31 2007-10-31 Real-time model validation

Publications (1)

Publication Number Publication Date
WO2009055967A1 true WO2009055967A1 (en) 2009-05-07

Family

ID=40590513

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2007/003100 WO2009055967A1 (en) 2007-10-31 2007-10-31 Real-time model validation

Country Status (1)

Country Link
WO (1) WO2009055967A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104199433A (en) * 2014-09-26 2014-12-10 胡景宗 Centralized controlling auxiliary pre-warning system in heat-engine plant
US20170193371A1 (en) * 2015-12-31 2017-07-06 Cisco Technology, Inc. Predictive analytics with stream database
CN107831741A (en) * 2017-10-18 2018-03-23 上海华电电力发展有限公司 New time series data read method for power plant
CN113777957A (en) * 2021-09-28 2021-12-10 天津华能杨柳青热电有限责任公司 Three-dimensional visual simulation system of liquid slag discharge boiler

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996002025A1 (en) * 1994-07-08 1996-01-25 Siemens Aktiengesellschaft Guide system for a power station
CN1694025A (en) * 2005-04-28 2005-11-09 南京科远控制工程有限公司 Automatic control system based on artificial intelligence for heat-engine plant
CN1801018A (en) * 2005-11-11 2006-07-12 南京科远控制工程有限公司 Interface method and apparatus for plant-level monitoring system and decentralized control system for power plant

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996002025A1 (en) * 1994-07-08 1996-01-25 Siemens Aktiengesellschaft Guide system for a power station
CN1694025A (en) * 2005-04-28 2005-11-09 南京科远控制工程有限公司 Automatic control system based on artificial intelligence for heat-engine plant
CN1801018A (en) * 2005-11-11 2006-07-12 南京科远控制工程有限公司 Interface method and apparatus for plant-level monitoring system and decentralized control system for power plant

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104199433A (en) * 2014-09-26 2014-12-10 胡景宗 Centralized controlling auxiliary pre-warning system in heat-engine plant
US20170193371A1 (en) * 2015-12-31 2017-07-06 Cisco Technology, Inc. Predictive analytics with stream database
CN107831741A (en) * 2017-10-18 2018-03-23 上海华电电力发展有限公司 New time series data read method for power plant
CN113777957A (en) * 2021-09-28 2021-12-10 天津华能杨柳青热电有限责任公司 Three-dimensional visual simulation system of liquid slag discharge boiler

Similar Documents

Publication Publication Date Title
JP6626910B2 (en) Computer mounting method, computer system and computer device
US20240152571A1 (en) Dynamic outlier bias reduction system and method
KR102169561B1 (en) Data cleansing system and method for inferring feed composition
US8321187B2 (en) Process simulation utilizing component-specific consumption data
CN103353753B (en) For the self diagnosis Process control loop of process device
US20090138306A1 (en) Facility risk assessment systems and methods
EP2500853A2 (en) Graphical language for optimization and use
CN105844050A (en) Time correlation based replacement time calculation method for numerical control machine tool components
CN102693262A (en) Method of determining the influence of a variable in a phenomenon
CN115562225A (en) Industrial robot operation and maintenance management method and device, computer equipment and storage medium
EP2339418A1 (en) Method and device for enhancing production facility performances
CN101561904A (en) Process data-based method and system for determining cost of software project
WO2009055967A1 (en) Real-time model validation
CN106776331A (en) The assessment method of software maturity, device and system
Friederich et al. A Framework for Validating Data-Driven Discrete-Event Simulation Models of Cyber-Physical Production Systems
CN113792421B (en) TPM equipment management data processing system and method based on digital twinning
EP4224380A1 (en) Maintenance simulation device and maintenance simulation method
Raman et al. Learning framework for maturing architecture design decisions for evolving complex SoS
US20230065835A1 (en) Information processing device, evaluation method, and computer-readable recording medium
JP2011123187A (en) Operation simulator
US20230061033A1 (en) Information processing device, calculation method, and computer-readable recording medium
Yahyatabar et al. A multi-stage stochastic programming for condition-based maintenance with proportional hazards model
CN117634855B (en) Project risk decision method, system, equipment and medium based on self-adaptive simulation
Huseby et al. Bayesian Methods and Expert Opinions Applied to System Reliability Estimation and Management
Yamany et al. A multi-agent framework for building an automatic operational profile

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 07816713

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 07816713

Country of ref document: EP

Kind code of ref document: A1