Petri Nets modeling method and device integrating aging failure dynamic model and fault diagnosis
By using Petri Nets modeling and FMEA analysis, combined with digital twin technology and the HIL platform, the shortcomings in modeling and control strategy verification of hydrogen fuel cell systems were addressed. This enabled high-precision fuel cell model construction and fault diagnosis, improving the system's durability and energy management reliability.
Patent Information
- Application Number
- CN202511450334.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-11
- Publication Date
- 2026-01-23
AI Technical Summary
Existing technologies struggle to achieve high-precision system-level modeling and real-time control of hydrogen fuel cells, particularly in system-level excitation and interpretation of fault modes such as stack aging, membrane electrode failure, and plate corrosion. Furthermore, the general HIL platform lacks a dedicated physical model for fuel cells and the ability to inject faults, resulting in insufficient verification of control strategies.
A high-precision fuel cell model was constructed using the Petri Nets modeling method, combined with FMEA analysis and digital twin technology. Fault spectrum was generated through functional decomposition and signal transmission analysis, and simulation verification was performed using the RCP and HIL platforms to realize the verification of the full life cycle control strategy of the fuel cell system.
It significantly improves the efficiency and reliability of control strategy verification for fuel cell systems, shortens the development cycle, enhances system durability and the reliability of vehicle energy management, and enables accurate diagnosis and prediction of fuel cell-specific faults.
Smart Images
Figure CN121389553A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application relates to the technical field of fuel cell modeling, and in particular to a Petri Nets modeling method and device integrating an aging failure dynamic model and fault diagnosis. BACKGROUND
[0002] Initially, hydrogen fuel cells are regarded as one of the core powers for realizing green low carbon due to zero emissions, high efficiency and fast low-temperature starting capability. However, the complex processes such as electrochemical reaction, multiphase flow, heat-mass coupling and material aging in the internal of the electric pile exist in the cross-scale and cross-physical domain, so that the system-level accurate modeling and real-time control become the key bottleneck for industrialization landing. Foreign countries started early in the field of HFC (Hydrogen Fuel Cell) digital twinning, and research institutions in the United States and Japan have carried out virtual prototype research on hydrogen energy systems since around 2010, and have applied the technology to design optimization, parameter calibration and fault prediction; domestic universities such as Tsinghua University and Zhejiang University have followed up and initially established a fuel cell system model, but there is still a gap with the international advanced level in terms of model dimension, dynamic accuracy and aging-failure coupling description.
[0003] In the test verification link, the ECU-TEST launched by Tracetronic Company is widely used by vehicle manufacturers due to its good cross-platform characteristics, but it lacks sufficient coverage of professional test scenarios for HFC, and cannot perform system-level excitation and interpretation on specific fault modes such as stack aging, membrane electrode failure and pole corrosion. At the same time, the existing rapid control prototype (RCP) and hardware-in-the-loop (HIL) platforms are mostly targeted at general power domain controllers, and lack dedicated physical models, fault injection and aging-failure dynamic libraries for fuel cell controllers (FCCU), which leads to insufficient verification of control strategies in the whole life cycle of "health-decline-failure", and makes it difficult to support online energy management and fault prediction requirements of long-life and high-reliability commercial vehicle working conditions.
[0004] In view of this, the present application is proposed. SUMMARY
[0005] The application provides a Petri Nets modeling method and device integrating an aging failure dynamic model and fault diagnosis, which can at least partially improve the above problems.
[0006] To achieve the above object, the application adopts the following technical scheme: A Petri Nets modeling method integrating an aging failure dynamic model and fault diagnosis, comprising: determining the development requirements and indexes of the fuel cell system, defining the function requirements of the FCCU and performing signal transmission analysis, obtaining corresponding data, performing FMEA analysis on the failure modes based on the corresponding data, and obtaining the system fault spectrum; Based on the system failure spectrum, the Petri Nets modeling method is adopted for construction processing to obtain an initial fuel cell model; Obtain working condition data, and simulate the initial fuel cell model according to the working condition data until the simulation result meets a preset condition to obtain a final fuel cell model.
[0007] The application further provides a Petri Nets modeling device integrating an aging failure dynamic model and fault diagnosis, which comprises: A failure spectrum generation unit is configured to determine development requirements and indexes of a fuel cell system, define FCCU function requirements and perform signal transmission analysis, obtain corresponding data, perform FMEA analysis on failure modes based on the corresponding data, and obtain a system failure spectrum; A construction unit is configured to, based on the system failure spectrum, adopt the Petri Nets modeling method for construction processing to obtain an initial fuel cell model; A simulation verification unit is configured to obtain working condition data, simulate the initial fuel cell model according to the working condition data until the simulation result meets a preset condition, and obtain a final fuel cell model.
[0008] In summary, the application proposes an integrated solution of “high-precision digital twin + function-fault mapping + hardware-in-loop” to solve the industry pain points of insufficient verification of hydrogen fuel cell controllers (FCCU) in the whole life cycle of “health-decline-failure”, lack of aging-failure coupling mechanism, and inability of general HIL platforms to inject fuel cell-specific faults. Through cross-scale modeling, the stack electrochemistry, flow-mass transfer, thermal management, and aging-failure dynamics are unified and packaged into a real-time updateable twin body. The function-signal-fault full spectrum mapping is established by the Petri Net graphical method, and the fault injection sequence under any decline depth is automatically generated. Combined with RCP and real sensor load, the FCCU special HIL bench is constructed to realize the precision, response, and fault tolerance verification of the control strategy under real physical excitation. The solution fills the gap of domestic hydrogen fuel cell special virtual prototype and test platform, significantly shortens the development cycle, improves the system durability and vehicle energy management reliability, and has outstanding industrial promotion value.
[0009] Specifically, through automatic testing, the functions, safety, performance, etc. of the FCCU controller are comprehensively tested. At the same time, the digital twin technology is applied to the hydrogen fuel cell system, the existing platform experimental test data are combined, the twin model identification and construction are carried out through the digital twin technology, the internal data are mined, the relationship between each variable of the hydrogen fuel cell and the system is analyzed, and fault diagnosis and prediction are carried out. The Petri Nets modeling method integrating the aging failure dynamic model and the fault diagnosis is used for hydrogen fuel cell FCCU software system development. Based on the MATLAB / Simulink tool kit, the hydrogen fuel cell system FCCU control model is built, and the controlled object model is built according to the bench parameters, the off-line simulation and code generation are downloaded to the RCP (rapid control prototype) for HIL simulation, and the accuracy and effectiveness of the designed control strategy are verified. BRIEF DESCRIPTION OF DRAWINGS
[0010] Figure 1 is a flowchart of the Petri Nets modeling method integrating the aging failure dynamic model and the fault diagnosis provided by the first embodiment of the present application; Figure 2 is a technical roadmap of the hydrogen fuel cell FCCU software system development provided by the embodiment of the present application; Figure 3 is a FMEA fault analysis example diagram provided by the embodiment of the present application; Figure 4 is a Petri Nets graphical analysis example diagram provided by the embodiment of the present application; Figure 5 is a schematic diagram of a zero-dimensional model of a PEMFC stack provided by the embodiment of the present application; Figure 6 is a schematic diagram of a fluid simulation model of a PEMFC stack provided by the embodiment of the present application; Figure 7 is a schematic diagram of an energy management strategy model provided by the embodiment of the present application; Figure 8 is a schematic diagram of an HIL bench carrying the FCCU system provided by the embodiment of the present application; Figure 9 is a module schematic diagram of the Petri Nets modeling device integrating the aging failure dynamic model and the fault diagnosis provided by the second embodiment of the present application. DETAILED DESCRIPTION
[0011] In order to make the purpose, technical scheme and advantages of the present application clearer, further detailed description will be made below in combination with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.
[0012] Virtual simulation test model is high in complexity and difficulty. The hydrogen fuel cell is a complex system involving physical and chemical processes, and has characteristics such as multiple inputs and outputs, strong nonlinearity and coupling. In order to study the nature and working characteristics of the hydrogen fuel cell, the first problem to be solved is to establish a fuel cell system model according to the fuel cell reaction mechanism and the structure of each subsystem accessory. However, the physical and chemical processes of the hydrogen fuel cell are complex, and multiple processes such as hydrogen and oxygen delivery, electrode reaction, and electrolyte transport need to be considered. In addition, the performance of the hydrogen fuel cell is affected by many factors such as temperature, humidity, gas flow, etc. Therefore, the present project plans to establish a multi-time scale hydrogen fuel cell system model to reflect the working characteristics of the hydrogen fuel cell from micro and macro aspects, ensure the effectiveness of the model, and improve the simulation calculation accuracy of the model.
[0013] Reference Figure 1 、 Figure 2 As shown in FIG. 1, the first embodiment of the present application discloses a Petri Nets modeling method integrating aging failure dynamic model and fault diagnosis, which can be executed by a Petri Nets modeling device integrating aging failure dynamic model and fault diagnosis (hereinafter referred to as modeling device), in particular, by one or more processors in the modeling device to implement the following method: S1, determine the development requirements and indexes of the fuel cell system, define the functional requirements of the FCCU and analyze the signal transmission, obtain corresponding data, perform FMEA analysis on the failure modes based on the corresponding data, and obtain the system fault spectrum; Specifically, step S1 further comprises: using a functional decomposition method to define and decompose the functional requirements of the fuel cell system, wherein based on the working conditions and functional requirements of the fuel cell system, the fuel cell system is combined as a multi-functional module, and the multi-functional module is decomposed into a plurality of layers of sub-functions and interfaces until the sub-functions and interfaces correspond to component control signals, so as to realize one-to-one correspondence between signals and functions; Based on the corresponding data, the FMEA analysis method is used to decompose and connect the failure modes, wherein for different failure modes, the failure components, failure mechanisms, failure causes and corresponding signals are decomposed and connected to obtain the system fault spectrum.
[0014] In the present embodiment, based on the working conditions and requirements of the fuel cell system, the FCCU functional requirement definition and control scheme design are carried out, the development requirements and indexes of the FCCU are clarified, the functional definition and signal transmission analysis are carried out by using the functional decomposition method, the failure modes are analyzed by using the FMEA analysis method, and then the system running state, functional connection and working mode conversion can be simulated and verified by using the Petri Nets graphical simulation method. The FCCU functional requirements and internal relations are explained by taking multiple FCCU products as an example.
[0015] Specifically, the fuel cell system is regarded as a multifunctional module combination, which is decomposed into several layers of sub-functions and interfaces layer by layer until each sub-function and interface corresponds to a control signal at the component level, thereby completing the one-to-one correspondence of signals and functions. On this basis, the above-mentioned corresponding data is used to carry out FMEA analysis on different failure modes: for each failure mode, the failure components, failure mechanisms, failure causes and their corresponding control signals are decomposed and associated, all failure paths are systematically sorted and connected, and finally a complete system failure spectrum is formed. The failure spectrum uses signals as the link to tightly couple functional requirements and potential failures, providing direct and traceable data sources for subsequent fault injection in the digital twin model and failure scenario reproduction on the HIL bench, avoiding the defect that the traditional FMEA is disconnected from the actual control logic, and significantly improving the verification efficiency and pertinence of the control strategy in the whole life cycle of "health-decline-failure". The FMEA example is shown in Figure 3 .
[0016] S2, based on the system failure spectrum, a Petri Nets modeling method is used for construction and processing to obtain an initial fuel cell model; Specifically, step S2 further includes: based on the system failure spectrum, MATLAB&Simulink modeling and simulation software is used for construction of the fuel cell model, wherein a fuel cell stack voltage zero-dimensional model is used as the basis to extend the flow of the air inlet system, the flow of the internal reaction gas and water of the stack, and the mass transfer, solid-liquid-gas phase coupling multi-physical field and thermal management inside the gas flow channel. A Petri Nets modeling method is used to integrate the aging and failure dynamic model and fault diagnosis to obtain an initial fuel cell model, wherein the initial fuel cell model includes a hydrogen system model, an air system model, a stack model, a thermal management model, and a flow model. The stack zero-dimensional model is shown in Figure 5 , and the stack fluid simulation model is shown in Figure 6 .
[0017] In this embodiment, a three-dimensional model of the fuel cell system controlled object is constructed, which reveals the mechanisms including the working principle of the fuel cell system, aging and failure, internal reaction gas flow in the stack, mass transfer inside the gas flow channel, anode and cathode air inlet system flow, heat transfer and thermal management, multi-physical field coupling, etc. A high-precision, multi-dimensional, and composite function physical model of the fuel cell system is constructed. Based on the energy management strategy of the vehicle-mounted fuel cell, a fuel cell system control scheme is developed, a fuel cell system control model is constructed using classical control and fuzzy control methods, the output power is taken as the control object, the BOP system control parameters are optimized, and high-speed, accurate, and high-fault-tolerant data real-time acquisition, storage, processing, operation, communication, and fault diagnosis functions are realized.
[0018] Specifically, based on the system fault spectrum obtained in step S1, the Petri Nets modeling method is used for construction and processing to obtain an initial fuel cell model. In specific implementation, relying on the MATLAB / Simulink modeling and simulation software, first, a zero-dimensional model of fuel cell stack voltage is taken as the core, and the flow of the intake system, the flow of the reaction gas and water inside the stack, and the mass transfer process inside the gas flow channel are simultaneously expanded in the periphery of the zero-dimensional model. Further, solid-liquid-gas phase coupling multi-physical field and thermal management are introduced to make the stack model and auxiliary system in the same coupling calculation layer. Then, through the Petri Nets modeling method, the aging and failure dynamic model and fault diagnosis logic are embedded in the above-mentioned sub-models in the form of "place-transition", the token migration mechanism of the health state and the failure state is realized, and the initial fuel cell model which can automatically switch the model parameters according to the signal condition is formed. The initial fuel cell model is composed of a hydrogen system model, an air system model, a stack model, a thermal management model, and a flow model. The data interaction between each sub-model is completed through the Simulink signal line, ensuring that each "function-signal-failure" record in the fault spectrum can find the corresponding calculation node and trigger path in the model, thereby providing a direct and executable model carrier for subsequent RCP rapid prototyping and HIL bench real-time fault injection, avoiding the tedious operation of manually replacing the parameter set for health-failure switching in the traditional modeling method, and significantly improving the failure scene reproduction efficiency and the pertinence of control strategy verification. Among them, the Petri Nets graphical simulation method is used for simulation analysis and verification of the function relationship and demand transmission. Based on the fuel cell power calculation model, the system working state is simulated and calculated based on function decomposition and system fault graph, and the system working state is controlled by different control signals to realize MIL simulation analysis and verification. The Petri Nets graphical analysis is as shown in Figure 4
[0019] Among them, by using the Petri Nets graphical simulation method, the system running state, function relationship and working mode conversion can be simulated and verified. This method can intuitively represent various states and transition relationships of the system, which helps to find potential logical problems and optimize control strategies, thereby improving the efficiency and coverage of automated testing. For example: "using the Petri Nets graphical simulation method to simulate and verify the system running state, function relationship and working mode conversion". For example, the Petri Nets graph can be used to represent the start-up, operation, shutdown and other states of the fuel cell system, as well as the transition conditions between states, and through simulation verification, whether the system response under various working conditions meets the expected requirements.
[0020] Furthermore, by integrating aging and failure dynamic models with fault diagnosis using the Petri Nets modeling method, a more comprehensive description of the system's dynamic behavior and failure modes can be achieved. This method can simulate the performance degradation process of the system over time, improving the accuracy and predictability of fault diagnosis, thereby enhancing the functionality and practicality of the digital twin system. For example, the Petri Nets model can be used to describe the aging process of the membrane electrode assembly in a fuel cell, including factors such as decreased catalyst activity and reduced membrane conductivity. By comparing this data with real-time monitoring data, the health status of the system can be accurately determined, the remaining service life predicted, and timely maintenance warnings issued. This study develops an energy management strategy for an on-board hydrogen fuel cell system. It designs the energy transfer and power control between the fuel cell, power battery, and drive motor at the vehicle level. An energy management strategy model is established in MATLAB / Simulink based on FCCU control signals and transfer methods. Using classical and fuzzy control methods, the system control accuracy is optimized through algorithms under hardware and software boundary conditions, improving the system's fault tolerance and response rate. This achieves accurate, real-time, and high-speed response from the FCCU software system. The energy management strategy model and classical control examples are shown below. Figure 7 As shown.
[0021] S3: Acquire operating condition data, and perform simulation processing on the initial fuel cell model based on the operating condition data until the simulation results meet the preset conditions, thus obtaining the final fuel cell model.
[0022] Specifically, step S3 further includes: acquiring operating condition data collected from the actual vehicle, performing model-in-the-loop offline simulation processing on the initial fuel cell model in the MATLAB environment based on the operating condition data, obtaining simulation results, and comparing the simulation results with preset measured data to obtain comparison results, so as to verify the accuracy, function and speed of the model; The control algorithm of the model is converted into RCP-recognizable code by the compilation software, and then loaded into the RCP application layer of the selected preset commercial development platform to realize the control of the system prototype. The actual controlled object is tested by controlling the system prototype and the test results are obtained. HIL simulation of the FCCU software system was performed to verify the controller's function, response rate, and control accuracy, in accordance with the FCCU functional requirements definition and decomposition, and the verification results were obtained. Based on the comparison results, test results, and verification results, a judgment is made. When it is determined that the model meets the preset conditions, it indicates that the model meets the requirements, and the final fuel cell model is obtained.
[0023] In this embodiment, based on the physical model of the controlled object and the control strategy, a model is established by using computer-aided modeling and analysis software such as MATLAB / Simulink, model in the loop (MIL) offline simulation is performed, and the functions and accuracy of the physical model and the control model are verified. Rapid control prototype (RCP) loading and verification are continued. By using a commercial development platform such as rapid controller prototype, the control strategy model is automatically coded and downloaded to the platform, and the prototype of the control system is quickly realized. The actual controlled object is tested by the control system prototype, the control effect of the control strategy is verified, and the control parameters are optimized online. Finally, the FCCU software system is simulated by HIL. By externally connecting temperature sensors, humidity sensors, back pressure valves, pressure reducing valves, proportional valves and other controlled objects, the control function, response rate and control accuracy of the controller are verified according to the definition and decomposition of the FCCU functional requirement.
[0024] Specifically, for FCV (Fuel cell vehicle), first, the working condition data in the actual vehicle running process is obtained by CAN recorder and vehicle data acquisition terminal, covering current, voltage, temperature, pressure, flow and other key parameters, and is imported into MATLAB environment. Based on the working condition data, the initial fuel cell model is processed by model in the loop (MIL) offline simulation, and the simulation result is obtained, wherein the input parameters and signals of the model are changed by manual input, given working condition and other methods. The simulation result is compared with the preset measured data to evaluate the performance of the model in terms of accuracy, functional response and calculation rate, ensure that the model output is highly consistent with the actual system behavior, and thus verify the reliability and applicability of the model.
[0025] Subsequently, the control algorithm in the model is automatically converted into a code recognizable by the RCP platform by using a compiling software, and is downloaded to the application layer of the selected commercial rapid control prototype (RCP) development platform to build a system prototype. The actual controlled object (such as a stack, an air compressor, a back pressure valve, etc.) is tested in real time by the prototype, the test result is obtained, and the execution effect and stability of the control strategy on the real hardware are further verified.
[0026] On this basis, the FCCU software system is connected to a hardware in the loop (HIL) simulation platform for hardware in the loop simulation, such as Figure 8As shown, according to the pre-defined and decomposed FCCU functional requirements, comprehensive verification of controller functions, response rates and control accuracy is carried out, and verification results are obtained. The HIL platform reproduces various operating and fault conditions by simulating sensor signals and actuator feedback, ensuring that the control performance of the controller under different conditions meets the design requirements. That is, by externally connecting temperature sensors, humidity sensors, back pressure valves, pressure reducing valves, proportional valves and other controlled objects, based on the definition and decomposition of FCCU controller functional requirements, the sensor signals and controller input signals are changed by manual input, changing the sensor environment, program control and other methods, the FCCU functional requirements are verified, and the controller functions, response rates and control accuracy are verified.
[0027] Finally, the comparison results, test results and verification results are judged. When the model meets the preset conditions in terms of accuracy, function implementation, response rate and control accuracy, it means that the model has engineering application ability, that is, it can be determined as the final fuel cell model. The embodiment realizes the three-step closed loop of "offline simulation-RCP test-HIL verification", which not only effectively improves the reliability of the model and the efficiency of the controller development, but also significantly reduces the frequency of real vehicle testing and development cost, and provides a solid guarantee for the precise deployment and reliable operation of the fuel cell system control strategy.
[0028] In summary, the present application proposes a set of "function-fault spectrum mapping-multi-physical field coupling modeling-offline simulation-RCP rapid prototype-HIL hardware-in-the-loop" closed-loop verification methods around the hydrogen fuel cell controller (FCCU) in the whole life cycle, which solves the industry pain points such as "health-decline-failure" verification deficiency, aging mechanism loss, and general test platform unable to inject fuel cell specific faults. Through function decomposition and FMEA analysis, each potential failure is accurately traced back to the calibrated control signal, forming a system fault spectrum that is homologous to the Simulink model. Based on the spectrum, a zero-dimensional stack-one-dimensional flow-multiphase thermal coupling Petri Net model is constructed, which can complete the token migration and parameter self-update in the same simulation step, and realize the second-level reproduction of "failure scenarios" in virtual space. Then, relying on real vehicle working condition data, MIL offline simulation is carried out, and the simulation output is compared with the measured data to ensure the model accuracy. Then, through automatic code generation, the control strategy is downloaded to the RCP, and a control prototype that can drive real stacks, valves, pumps and sensors is quickly built to complete the function and response test under real load. Finally, the HIL bench is connected, the FCCU function definition is verified, and the control accuracy and fault tolerance of the controller under extreme and declining conditions are verified. After three-step closed-loop iteration to the preset index, the final fuel cell model is locked, which can be directly used for mass production of FCCU software release. This method reproduces the "flooded, membrane dry, and plate corrosion" faults in the laboratory stage in advance without increasing additional hardware costs, shortens the control strategy iteration cycle, reduces the vehicle-level calibration sample, and significantly reduces the development cost and risk. At the same time, the model and fault spectrum are stored homologously, and when the design changes or new failure modes appear, the database can be updated to automatically synchronize to the virtual prototype, realizing "one modeling, continuous evolution", and providing a replicable engineering path for long-life and high-reliability operation of fuel cells.
[0029] Please refer to Figure 9 The second embodiment of the present application provides a Petri Nets modeling device integrating aging failure dynamic model and fault diagnosis, which comprises: A fault spectrum generation unit 101 is configured to determine the development requirements and indexes of the fuel cell system, define the function requirements of the FCCU and perform signal transmission analysis, obtain corresponding data, perform FMEA analysis on the failure modes based on the corresponding data, and obtain a system fault spectrum; A construction unit 102 is configured to construct and process based on the system fault spectrum using a Petri Nets modeling method to obtain an initial fuel cell model; A simulation verification unit 103 is configured to obtain working condition data, and perform simulation processing on the initial fuel cell model according to the working condition data until the simulation result meets the preset condition to obtain a final fuel cell model.
[0030] The failure spectrum generating unit 101 is specifically configured to define and decompose the functional requirements of the fuel cell system by using a functional decomposition method, wherein the fuel cell system is combined as a multi-functional module based on the working conditions and the functional requirements of the fuel cell system, and the multi-functional module is decomposed into a plurality of layers of sub-functions and interfaces until the sub-functions and interfaces correspond to component control signals, so as to realize one-to-one correspondence between signals and functions. Based on the corresponding data, the FMEA analysis method is used to decompose and connect the failure modes, wherein the failure components, failure mechanisms, failure causes and corresponding signals are decomposed and connected for different failure modes, to obtain the system failure spectrum.
[0031] The construction unit 102 is specifically configured to construct a fuel cell model by using MATLAB&Simulink modeling and simulation software based on the system failure spectrum, wherein the fuel cell stack voltage zero-dimensional model is used as the basis to extend the flow of the air intake system, the flow of the internal reaction gas and water of the stack, the mass transfer inside the gas flow channel, and the solid-liquid-gas phase coupling multi-physical field and thermal management. The Petri Nets modeling method is used to integrate the aging and failure dynamic model and the fault diagnosis to obtain an initial fuel cell model, wherein the initial fuel cell model includes a hydrogen system model, an air system model, a stack model, a thermal management model and a flow model.
[0032] The simulation verification unit 103 is specifically configured to obtain working condition data collected by an actual vehicle, perform model-in-the-loop offline simulation processing on the initial fuel cell model in the matlab environment based on the working condition data, obtain simulation results, compare the simulation results with preset measured data, obtain comparison results, and verify the accuracy, function and rate of the model. The control algorithm of the model is converted into RCP recognizable code by compiling software, and is loaded into the RCP application layer of the selected preset commercial development platform to realize control of the system prototype, and the actual controlled object is tested by the control system prototype to obtain test results. The HIL simulation is performed on the FCCU software system, the FCCU functional requirement definition and decomposition are used as a benchmark to verify the function, response rate and control accuracy of the controller, and verification results are obtained. Based on the comparison results, the test results and the verification results, it is judged that when the model meets the preset conditions, it is indicated that the model meets the requirements, and the final fuel cell model is obtained.
[0033] The above describes the preferred embodiments of the present application. It should be noted that those skilled in the art can make some improvements and refinements without departing from the principles of the present application, and these improvements and refinements are also considered within the protection scope of the present application.
Claims
1. A Petri Nets modeling method integrating dynamic model of aging failure and fault diagnosis, characterized in that, The application relates to a fuel cell system modeling method. The application comprises the following steps: determining development requirements and indexes of a fuel cell system, defining and analyzing signal transmission of FCCU function requirements, obtaining corresponding data, performing FMEA analysis on failure modes based on the corresponding data, and obtaining a system failure spectrum; based on the system failure spectrum, an initial fuel cell model is obtained by using a Petri Nets modeling method; 2. The Petri Nets modeling method integrating aging failure dynamic model and fault diagnosis according to claim 1, characterized in that, obtaining working condition data, simulating the initial fuel cell model according to the working condition data until the simulation result meets preset conditions, and obtaining a final fuel cell model. The application comprises the following steps: using a function decomposition method to define and decompose the function requirements of the fuel cell system, wherein, based on the working conditions and function requirements of the fuel cell system, the fuel cell system is regarded as a combination of multiple function modules, and the multiple function modules are decomposed into a plurality of layers of sub-functions and interfaces until the sub-functions and interfaces correspond to component control signals, so that one-to-one correspondence between signals and functions is realized; 3. The Petri Nets modeling method integrating aging failure dynamic model and fault diagnosis according to claim 1, characterized in that, based on the corresponding data, the failure modes are decomposed and connected by using an FMEA analysis method, wherein, for different failure modes, the failure components, failure mechanisms, failure causes and corresponding signals are decomposed and connected to obtain a system failure spectrum. based on the system failure spectrum, an initial fuel cell model is obtained by using a Petri Nets modeling method, and the specific steps are as follows: based on the system failure spectrum, a fuel cell model is constructed by using MATLAB&Simulink modeling simulation software, wherein, based on a fuel cell stack voltage zero-dimensional model, an air inlet system flow, internal reaction gas and water flow of the stack, internal mass transfer of the gas flow channel, solid-liquid-gas phase coupling multi-physical field and thermal management are extended; 4. The Petri Nets modeling method integrating aging failure dynamic model and fault diagnosis according to claim 1, characterized in that, an initial fuel cell model is obtained by using a Petri Nets modeling method to integrate an aging and failure dynamic model and fault diagnosis, wherein, the initial fuel cell model comprises a hydrogen system model, an air system model, a stack model, a thermal management model and a flow model. working condition data is obtained, the initial fuel cell model is simulated according to the working condition data until the simulation result meets preset conditions, and a final fuel cell model is obtained, and the specific steps are as follows: working condition data collected by an actual vehicle is obtained, the initial fuel cell model is simulated in a loop offline in a MATLAB environment based on the working condition data, a simulation result is obtained, the simulation result is compared with preset measured data, a comparison result is obtained, and the accuracy, function and rate of the model are verified; a control algorithm of the model is converted into RCP recognizable code by using compiling software, the RCP recognizable code is loaded to an RCP application layer of a preset commercial development platform, control of a system prototype is realized, an actual controlled object is tested by using the control system prototype, and a test result is obtained. HIL simulation is performed on the FCCU software system, the FCCU function requirement definition and decomposition are compared, the function, response rate and control accuracy of the controller are verified, and a verification result is obtained; Based on the comparison result, the test result and the verification result, when it is judged that the model meets the preset condition, it is indicated that the model meets the requirement, and the final fuel cell model is obtained.
5. A Petri Nets modeling device integrating dynamic model of aging failure and fault diagnosis, characterized in that, It comprises: a failure spectrum generation unit configured to determine the development requirements and indexes of the fuel cell system, perform definition and signal transmission analysis on the FCCU function requirement, obtain corresponding data, perform FMEA analysis on the failure mode based on the corresponding data, and obtain a system failure spectrum; a construction unit configured to construct and process the initial fuel cell model based on the system failure spectrum and using the Petri Nets modeling method; a simulation verification unit configured to obtain working condition data, perform simulation processing on the initial fuel cell model according to the working condition data, until the simulation result meets the preset condition, and obtain the final fuel cell model.
6. The Petri Nets modeling device integrating aging failure dynamic model and fault diagnosis according to claim 5, characterized in that, The failure spectrum generation unit is specifically configured to: define and decompose the function requirement of the fuel cell system using a function decomposition method, wherein the fuel cell system is combined as a multi-functional module based on the working conditions and function requirements of the fuel cell system, and the multi-functional module is decomposed into a plurality of layers of sub-functions and interfaces until the sub-functions and interfaces correspond to component control signals, so as to realize one-to-one correspondence between signals and functions; based on the corresponding data, the FMEA analysis method is used to decompose and connect the failure modes, wherein the failure components, failure mechanisms, failure causes and corresponding signals are decomposed and connected for different failure modes, and a system failure spectrum is obtained.
7. The Petri Nets modeling device integrating aging failure dynamic model and fault diagnosis according to claim 5, characterized in that, The construction unit is specifically configured to: based on the system failure spectrum, the MATLAB&Simulink modeling and simulation software is used to construct the fuel cell model, wherein the fuel cell stack voltage zero-dimensional model is used as the basis, and the flow of the inlet air system, the flow of the internal reaction gas and water in the stack, the mass transfer in the gas flow channel, the solid-liquid-gas phase coupling multi-physical field and thermal management are extended; the Petri Nets modeling method is used to integrate the aging and failure dynamic model and fault diagnosis, and an initial fuel cell model is constructed, wherein the initial fuel cell model comprises a hydrogen system model, an air system model, a stack model, a thermal management model and a flow model.
8. The Petri Nets modeling device integrating aging failure dynamic model and fault diagnosis according to claim 5, characterized in that, The simulation verification unit is specifically configured to: obtain working condition data collected from an actual vehicle, perform model-in-the-loop offline simulation processing on the initial fuel cell model in the matlab environment based on the working condition data, obtain a simulation result, compare the simulation result with the preset measured data, obtain a comparison result, and verify the accuracy, function and rate of the model; the control algorithm of the model is converted into RCP recognizable code through compiling software, and is loaded into the RCP application layer of the selected preset commercial development platform, the control of the system prototype is realized, and the actual controlled object is tested through the control system prototype, and a test result is obtained. HIL simulation is performed on the FCCU software system, and the FCCU function requirement definition and decomposition are compared, the function, response rate and control accuracy of the controller are verified, and the verification result is obtained; Based on the comparison result, the test result and the verification result, when it is judged that the model meets the preset condition, it is indicated that the model meets the requirement, and the final fuel cell model is obtained.