Self-adaptive control method, device and equipment for fuel cell and medium

Abstract

The invention relates to a fuel cell self-adaptive control method, device and equipment and a medium, and belongs to the technical field of fuel cells, the fuel cell self-adaptive control method monitors the running state and load power demand of a fuel cell in real time to determine the current running condition mode of the fuel cell; activating a control strategy corresponding to the current operation condition mode in a preset multi-mode control strategy, and performing self-tuning on key parameters of the control strategy based on a pre-trained extreme learning machine to obtain an original control quantity; and the historical control quantity of the control strategy corresponding to the operation condition mode of the fuel cell at the previous moment is obtained, and the weighted sum of the original control quantity and the historical control quantity is taken as the final control quantity at the current moment, so that the energy efficiency of the fuel cell is improved.

Description

Technical Field

[0001] This invention relates to the field of fuel cell technology, and in particular to an adaptive control method, device, equipment, and medium for fuel cells. Background Technology

[0002] Proton exchange membrane fuel cells (PEMFCs) are a highly efficient and clean energy conversion device that has shown broad application prospects in transportation, stationary power generation and other fields. A fuel cell system is a complex nonlinear system involving the coupling of multiple physical fields such as electrochemistry, thermodynamics and fluid mechanics. Its operating performance, efficiency and lifespan are highly dependent on the stability and optimization of the operating conditions.

[0003] Control methods for fuel cell systems can be categorized into traditional control methods and advanced intelligent control methods, such as classic PID control, fuzzy logic control, neural network control, and model predictive control (MPC). PID controller parameters are typically fixed, but the time-varying and nonlinear characteristics of fuel cell systems make it difficult for fixed-parameter PID controllers to maintain optimal control performance under all operating conditions, severely limiting the efficiency improvement of fuel cell systems and the lifespan of core components. Fuzzy control performance heavily relies on rule bases developed by expert experience, and the adaptive adjustment capability of the rules is limited. Neural network control has strong learning capabilities, but requires a large amount of data for training, and its real-time performance and reliability face challenges in the vehicle environment. Although model predictive control can be optimized, its performance is highly dependent on the accuracy of the model.

[0004] Therefore, existing fuel cell system control methods lack the ability to effectively adapt to changes in dynamic characteristics of fuel cell systems under all operating conditions and throughout their entire life cycle, making it difficult to coordinate and handle specific control requirements under different operating modes, thereby reducing the energy efficiency of fuel cells. Summary of the Invention

[0005] In view of this, it is necessary to provide an adaptive control method, device, equipment and medium for fuel cells to solve the technical problem that the control of fuel cells is difficult to coordinately handle the specific control requirements under different operating modes, which reduces the energy efficiency of fuel cells.

[0006] To address the aforementioned problems, in a first aspect, the present invention provides an adaptive control method for a fuel cell, comprising: Real-time monitoring of the fuel cell's operating status and load power requirements is used to determine the fuel cell's current operating mode. In the preset multimodal control strategy, the control strategy corresponding to the current operating condition mode is activated, and the key parameters of the control strategy are self-tuned based on the pre-trained extreme learning machine to obtain the original control quantity. Obtain the historical control values ​​of the control strategy corresponding to the previous operating mode of the fuel cell, and use the weighted sum of the original control value and the historical control value as the final control value at the current moment.

[0007] In one possible implementation, the parameters of the operating state include total voltage, total current, hydrogen inlet and outlet pressure, hydrogen inlet and outlet temperature, air inlet and outlet pressure, air inlet and outlet temperature, circulating water inlet and outlet temperature, humidifier internal temperature, voltage of each individual cell in the fuel cell stack, and ambient temperature.

[0008] In one possible implementation, the current operating condition mode includes cold start mode, hot start mode, idling operation mode, low load steady state mode, high load steady state mode, dynamic loading mode, and dynamic load reduction mode.

[0009] In one possible implementation, the pre-trained extreme learning machine performs self-tuning on the key parameters of the control strategy to obtain the original control quantity, including: Obtain the parameters of the current operating condition mode, input the parameters of the operating state into the pre-trained extreme learning machine, and obtain the prediction key parameters of the control strategy; The key parameters of the control strategy are weighted based on the predicted key parameters, and the control parameters of the control strategy are updated based on the weighted key parameters to obtain the original control quantity.

[0010] In one possible implementation, the control strategy includes a model predictive control strategy, a PID control strategy, and a fuzzy logic control strategy; the key parameters include the proportional, integral, and derivative coefficients of the control strategy, as well as the prediction time domain and the control time domain.

[0011] In one possible implementation, obtaining the historical control values ​​of the control strategy corresponding to the previous operating mode of the fuel cell, and using the weighted sum of the original control value and the historical control value as the final control value at the current moment, includes: During the switching of operating mode of fuel cell, when the current operating mode of fuel cell is the same as the operating mode of the previous moment, a first weight is determined based on the current operating mode or the operating mode of the previous moment. The original control quantity and the historical control quantity are weighted and fused based on the first weight to obtain the final control quantity of the control strategy corresponding to the current operating mode. When the current operating mode of the fuel cell is different from the operating mode at the previous moment, a second weight is determined based on the operating mode at the previous moment. The original control quantity and the historical control quantity are weighted and fused based on the second weight to obtain the final control quantity of the control strategy corresponding to the current operating mode. The first weight and the second weight are different.

[0012] In one possible implementation, the control quantities include electric heater power, air compressor speed, inlet air throttle opening, outlet air throttle opening, hydrogen proportional valve opening, hydrogen tailpipe exhaust time, hydrogen tailpipe exhaust frequency, hydrogen circulation pump speed, circulating water pump speed, and radiator speed.

[0013] In a second aspect, the present invention also provides a fuel cell adaptive control device, comprising: The current mode determination module is used to obtain the current operating status and load power demand of the fuel cell in order to determine the current operating mode of the fuel cell. The parameter self-tuning module is used to activate the control strategy corresponding to the current operating condition mode in the preset multimodal control strategy, and to self-tun the key parameters of the control strategy based on the pre-trained extreme learning machine to obtain the original control quantity. The adaptive control module is used to obtain the historical control quantity of the control strategy corresponding to the previous operating mode of the fuel cell, and to use the weighted sum of the original control quantity and the historical control quantity as the final control quantity at the current moment.

[0014] Thirdly, the present invention also provides a fuel cell control device, comprising: a processor and a memory; The memory stores a computer-readable program that can be executed by the processor; When the processor executes the computer-readable program, it implements the steps in the fuel cell adaptive control method as described above.

[0015] Fourthly, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instructions, which, when executed by a processor, can implement the steps of the fuel cell adaptive control method described in any one of the above-mentioned method items.

[0016] The beneficial effects of this invention are: real-time monitoring of the fuel cell's operating status and load power demand to determine the current operating mode of the fuel cell; activation of the control strategy corresponding to the current operating mode in a preset multimodal control strategy, and self-tuning of the key parameters of the control strategy based on a pre-trained extreme learning machine to obtain the original control quantity; acquisition of the historical control quantity of the control strategy corresponding to the fuel cell's previous operating mode, and using the weighted sum of the original and historical control quantities as the final control quantity at the current moment; intelligent switching of the control strategy is achieved by activating the control strategy corresponding to the current operating mode in the multimodal control strategy; self-tuning of the key parameters of the control strategy through a pre-trained extreme learning machine to achieve parameter adjustment of the current control strategy; and smooth processing of control commands during the switching process of different modal control strategies, generating the final fuel cell system control command, avoiding abrupt changes in the final control quantity, effectively solving the adaptive control of the fuel cell under various typical operating conditions such as start-up, idling, dynamic loading, and steady-state operation, and improving the energy efficiency of the fuel cell. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of an embodiment of the adaptive control method for fuel cells provided by the present invention; Figure 2 A schematic diagram of the structure of the adaptive control method for fuel cells provided by the present invention; Figure 3 This is a schematic diagram of an embodiment of the adaptive control device for fuel cells provided by the present invention; Figure 4 This is a schematic diagram of an embodiment of the fuel cell control device provided by the present invention. Detailed Implementation

[0019] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0020] In this document, the term "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] This invention discloses an adaptive control method, apparatus, device, and medium for fuel cells, which can be used in a computer. The method, apparatus, or computer-readable storage medium involved in this invention can be integrated with the aforementioned apparatus or be relatively independent.

[0022] One specific embodiment of the present invention discloses an adaptive control method for a fuel cell, which can be executed by a computer, specifically by one or more processors of the computer. For example... Figure 1 As shown, the adaptive control method for fuel cells includes: S101. Obtain the current operating status and load power demand of the fuel cell to determine the current operating mode of the fuel cell; It should be noted that by monitoring the operating status and load power demand of the fuel cell, the current operating mode of the fuel cell system can be identified.

[0023] S102. Activate the control strategy corresponding to the current operating mode in the preset multimodal control strategy, and self-tun the key parameters of the control strategy based on the pre-trained extreme learning machine to obtain the original control quantity. It should be noted that the key parameters of the control strategy are self-tuned by the pre-trained extreme learning machine, thereby realizing the parameter adjustment of the current control strategy.

[0024] S103. Obtain the historical control quantity of the control strategy corresponding to the previous operating mode of the fuel cell, and use the weighted sum of the original control quantity and the historical control quantity as the final control quantity at the current moment.

[0025] It should be noted that smoothing the original control quantity and historical control quantity enables smooth processing of control commands during the switching process of different modal control strategies, generating the final fuel cell control command and avoiding abrupt changes in the final control quantity.

[0026] In some embodiments, in step S101, the current operating status and load power demand of the fuel cell are obtained to determine the current operating mode of the fuel cell. The fuel cell is monitored in real time, and its operating status parameters are collected in real time. These parameters include total voltage, total current, hydrogen inlet and outlet pressures, hydrogen inlet and outlet temperatures, air inlet and outlet pressures, air inlet and outlet temperatures, circulating water inlet and outlet temperatures, humidifier internal temperature, voltage of each individual cell in the fuel cell stack, and ambient temperature. The load power demand changes are obtained through these operating status parameters, thereby identifying the current operating mode of the fuel cell. These current operating modes include cold start mode, hot start mode, idling mode, low-load steady-state mode, high-load steady-state mode, dynamic loading mode, and dynamic unloading mode. For a schematic diagram of the fuel cell adaptive control structure, please refer to [link to schematic diagram]. Figure 2 ,like Figure 2 As shown, the multi-mode controller library contains different controllers corresponding to cold start mode, hot start mode, idling operation mode, low load steady-state mode, high load steady-state mode, dynamic loading mode, and dynamic unloading mode. Different control strategies exist in the controllers. The control strategy for dynamic loading mode prioritizes ensuring a rapid response of cathode oxygen concentration. The control strategy for idling operation mode aims to minimize aging stress. The control strategy for high load steady-state mode aims to minimize the single-cell voltage threshold.

[0027] In some embodiments, in step S102, a control strategy corresponding to the current operating condition mode is activated in a preset multimodal control strategy, and the key parameters of the control strategy are self-tuned based on a pre-trained extreme learning machine to obtain the original control quantity; according to the current operating condition mode of the fuel cell, a control strategy corresponding to the current operating condition mode is activated in the preset multimodal control strategy, the control strategy including model predictive control strategy, PID control strategy and fuzzy logic control strategy, the key parameters of which include the proportional, integral and derivative coefficients of the control strategy as well as the prediction time domain and the control time domain; the activation process of the multimodal control strategy can be that when the rate of change of the load demand power exceeds a preset threshold, When the fuel cell is determined to have entered a dynamic loading mode, the corresponding control strategy for the dynamic loading mode is activated, and the parameter tuning process of the control strategy is triggered. When the rate of decrease in load demand power exceeds a preset threshold, the control strategy continues to operate. When the fuel cell system enters a dynamic load shedding mode, the corresponding control strategy and its parameter tuning process are activated. The parameters of the current operating condition mode are obtained and input into a pre-trained extreme learning machine to obtain the predicted key parameters of the control strategy. The key parameters of the control strategy are weighted based on the predicted key parameters, and the control parameters of the control strategy are updated based on the weighted key parameters to obtain the original control quantity. The process of parameter self-tuning using an Extreme Learning Machine (ELM) is as follows: The ELM consists of an input layer, a hidden layer, and an output layer. During the offline learning phase, the total voltage, total current, hydrogen inlet and outlet pressures and temperatures, air inlet and outlet pressures and temperatures, circulating water inlet and outlet temperatures, humidifier internal temperature, minimum single-cell voltage of the fuel cell stack, and ambient temperature under different fuel cell operating conditions are used as inputs to train the ELM. After feature extraction and transformation of the input data through the hidden layer, the output layer outputs the proportional gain, integral gain, and derivative gain of the controller under the historical optimal operating data, as well as the predicted time-domain parameters and control time-domain parameters. In other words, the inputs are the historical state parameters of the fuel cell system, the output voltage, and the output current data sequence; the outputs are the proportional gain, integral gain, and derivative gain of the activated controller. The proportional, integral, and derivative coefficients, as well as parameters in the prediction and control time domains, are used to train the limit learning machine (ELM) for each controller in the multimodal controller library, resulting in a trained ELM. The controller parameters are the parameters of the control strategies in each controller. During self-tuning, the total voltage, total current, hydrogen inlet and outlet pressures, hydrogen inlet and outlet temperatures, air inlet and outlet pressures, air inlet and outlet temperatures, circulating water inlet and outlet temperatures, humidifier internal temperature, minimum single-cell voltage of the fuel cell stack, and ambient temperature are obtained from the current mode of the actual fuel cell system and input into the trained ELM. This yields the predicted optimal proportional, integral, and derivative coefficients, as well as parameters in the prediction and control time domains. These parameters are then weighted and calculated with the existing control parameters of the activated controller to obtain the initial control parameters and control variables for the fuel cell in this mode.

[0028] In some embodiments, in step S103, the historical control quantity of the control strategy corresponding to the previous operating mode of the fuel cell is obtained, and the weighted sum of the original control quantity and the historical control quantity is used as the final control quantity at the current moment to achieve adaptive control of the fuel cell. During the switching of the operating mode of the fuel cell, different weights are derived using a fuzzy switching algorithm according to the mode to which the fuel cell belongs in the previous operating mode. When the current operating mode of the fuel cell is the same as the operating mode in the previous moment, a first weight is determined based on the current operating mode or the operating mode in the previous moment. The original control quantity and the historical control quantity are then weighted and fused based on the first weight. In the current operating mode or the previous operating mode, a first weight is set through a fuzzy switching algorithm. Based on the first weight, a weighted fusion algorithm is used to weight and fuse the original control quantity and the historical control quantity to obtain the final control quantity of the control strategy corresponding to the current operating mode. When the current operating mode is the same as the previous operating mode, the weight coefficient of the original control quantity corresponding to the current operating mode is set to 0.9 and the weight coefficient of the historical control quantity corresponding to the previous operating mode is set to 0.1 through the fuzzy switching algorithm. According to the first weight, the original control quantity and the historical control quantity are weighted and fused to obtain the final control quantity of the control strategy corresponding to the current operating mode. When the current operating mode of the fuel cell is different from the operating mode at the previous moment, a second weight is determined based on the operating mode at the previous moment. The original control quantity and the historical control quantity are weighted and fused based on the second weight to obtain the final control quantity of the control strategy corresponding to the current operating mode. The first weight and the second weight are different. Specifically, when the operating modes are different and the operating mode at the previous moment is the cold start mode, the weight coefficient of the original control quantity corresponding to the current operating mode is set to 0.1 and the weight coefficient of the historical control quantity corresponding to the operating mode at the previous moment is set to 0.9 through the fuzzy switching algorithm. The original control quantity and the historical control quantity are weighted and fused to obtain the final control quantity of the control strategy corresponding to the current operating mode, that is, the final control quantity at the current moment. When the operating conditions are different and the operating condition at the previous moment is the dynamic loading mode, the weight coefficient of the original control quantity corresponding to the current operating condition mode is set to 0.2 and the weight coefficient of the historical control quantity corresponding to the operating condition mode at the previous moment is set to 0.8 through the fuzzy switching algorithm. The original control quantity and the historical control quantity are weighted and fused to obtain the final control quantity at the current moment. When the operating conditions are different and the previous operating condition was a large load steady-state mode, the weight coefficient of the original control quantity corresponding to the current operating condition is set to 0.3 and the weight coefficient of the historical control quantity corresponding to the previous operating condition is set to 0.7 through the fuzzy switching algorithm. The original control quantity and the historical control quantity are weighted and fused to obtain the final control quantity at the current moment. When the operating conditions are different and the operating condition at the previous moment is the dynamic load reduction mode, the weight coefficient of the original control quantity corresponding to the current operating condition mode is set to 0.8 and the weight coefficient of the historical control quantity corresponding to the operating condition mode at the previous moment is set to 0.2 through the fuzzy switching algorithm. Based on the determined weights, the original control quantity and the historical control quantity are weighted and fused to obtain the final control quantity at the current moment. When the operating conditions are different and the previous operating condition was a hot start mode, an idle operation mode, or a low-load steady-state mode, the fuzzy switching algorithm sets the weight coefficient of the original control quantity corresponding to the current operating condition mode to 0.85, and the weight coefficient of the historical control quantity corresponding to the previous operating condition mode to 0.15. Based on the determined weights, the original control quantity and the historical control quantity are weighted and fused using a weighted fusion algorithm to obtain the control quantity of the control strategy corresponding to the current operating condition mode, that is, to obtain the final control quantity of the control strategy corresponding to the current operating condition mode. Figure 2 As shown, the multimodal controller library includes multimodal control strategies corresponding to different operating conditions. During the switching of operating conditions, the dynamic coordinator smooths the original control quantity of the current operating condition mode and the historical control quantity of the previous operating condition mode. The dynamic coordinator uses fuzzy switching and weighted fusion algorithms to smooth the control quantity of the control strategy during the switching of operating conditions. That is, the dynamic coordinator mixes the output commands of the controller to be deactivated and the controller to be activated during the transition phase of the operating condition switching to avoid abrupt changes in the control quantity. The control quantity includes electric heater power, air compressor speed, inlet air throttle opening, outlet air throttle opening, hydrogen proportional valve opening, hydrogen tail valve exhaust time, hydrogen tail valve exhaust frequency, hydrogen circulation pump speed, circulating water pump speed, and radiator speed. The fuel cell is controlled according to the final control quantity of the control strategy corresponding to the current operating condition mode to achieve adaptive control of the fuel cell.

[0029] In summary, the adaptive control method for fuel cells provided by this invention monitors the operating status and load power demand of the fuel cell in real time to determine the current operating mode of the fuel cell; activates the control strategy corresponding to the current operating mode in a preset multimodal control strategy, and performs self-tuning on the key parameters of the control strategy based on a pre-trained extreme learning machine to obtain the original control quantity; obtains the historical control quantity of the control strategy corresponding to the previous operating mode of the fuel cell, and uses the weighted sum of the original control quantity and the historical control quantity as the final control quantity at the current moment to achieve adaptive control of the fuel cell and improve the energy efficiency of the fuel cell.

[0030] To better implement the fuel cell adaptive control method in the embodiments of the present invention, based on the fuel cell adaptive control method, correspondingly, as follows: Figure 3 As shown, this embodiment of the invention also provides a fuel cell adaptive control device, the fuel cell adaptive control device 300 including: The current mode determination module 301 is used to obtain the current operating status and load power demand of the fuel cell in order to determine the current operating mode of the fuel cell. The parameter self-tuning module 302 is used to activate the control strategy corresponding to the current operating mode in the preset multimodal control strategy, and to self-tun the key parameters of the control strategy based on the pre-trained extreme learning machine to obtain the original control quantity. The adaptive control module 303 is used to obtain the historical control quantity of the control strategy corresponding to the previous operating mode of the fuel cell, and to use the weighted sum of the original control quantity and the historical control quantity as the final control quantity at the current moment.

[0031] like Figure 4 As shown, the present invention also provides a fuel cell control device 400, which can be a computing device such as a mobile terminal, desktop computer, laptop, handheld computer, or server. The fuel cell control device 400 includes a processor 401, a memory 402, and a display 403. Figure 4 Only some components of the fuel cell control device 400 are shown; however, it should be understood that implementation of all shown components is not required, and more or fewer components may be implemented instead.

[0032] In some embodiments, the memory 402 may be an internal storage unit of the fuel cell control device 400, such as a hard disk or memory of the fuel cell control device 400. In other embodiments, the memory 402 may be an external storage device of the fuel cell control device 400, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the fuel cell control device 400. Furthermore, the memory 402 may include both internal and external storage units of the fuel cell control device 400. The memory 402 is used to store application software and various types of data installed on the fuel cell control device 400, such as program code installed on the fuel cell control device 400. The memory 402 can also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 402 stores a fuel cell adaptive control program, which can be executed by the processor 401 to implement the fuel cell adaptive control method of the various embodiments of the present invention.

[0033] In some embodiments, processor 401 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 402 or process data, such as a fuel cell adaptive control method.

[0034] In some embodiments, display 403 may be an LED display, a liquid crystal display, a touch-screen liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 403 is used to display identification information of the fuel cell adaptive control program and to display a visual user interface. Components 401-403 of the fuel cell control device 400 communicate with each other via a system bus.

[0035] In some embodiments, when the processor 401 executes the fuel cell adaptive control program in the memory 402, it implements the various steps in the fuel cell adaptive control method as described in the above embodiments. Since the fuel cell adaptive control method has been described in detail above, it will not be repeated here.

[0036] Accordingly, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instruction, which, when executed by a processor, can implement the steps or functions of the fuel cell adaptive control method provided in the above-described method embodiments.

[0037] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0038] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive control method for a fuel cell, characterized in that, include: Obtain the current operating status and load power demand of the fuel cell to determine the current operating mode of the fuel cell; In the preset multimodal control strategy, the control strategy corresponding to the current operating condition mode is activated, and the key parameters of the control strategy are self-tuned based on the pre-trained extreme learning machine to obtain the original control quantity. Obtain the historical control values ​​of the control strategy corresponding to the previous operating mode of the fuel cell, and use the weighted sum of the original control value and the historical control value as the final control value at the current moment.

2. The adaptive control method for fuel cells according to claim 1, characterized in that, The parameters of the operating status include total voltage, total current, hydrogen inlet and outlet pressure, hydrogen inlet and outlet temperature, air inlet and outlet pressure, air inlet and outlet temperature, circulating water inlet and outlet temperature, humidifier internal temperature, voltage of each individual cell in the fuel cell stack, and ambient temperature.

3. The adaptive control method for fuel cells according to claim 2, characterized in that, The current operating conditions include cold start mode, hot start mode, idling mode, low load steady state mode, high load steady state mode, dynamic loading mode, and dynamic load reduction mode.

4. The adaptive control method for fuel cells according to claim 3, characterized in that, The pre-trained extreme learning machine performs self-tuning on the key parameters of the control strategy to obtain the original control input, including: Obtain the parameters of the current operating condition mode, input the parameters of the operating state into the pre-trained extreme learning machine, and obtain the prediction key parameters of the control strategy; The key parameters of the control strategy are weighted based on the predicted key parameters, and the control parameters of the control strategy are updated based on the weighted key parameters to obtain the original control quantity.

5. The adaptive control method for fuel cells according to claim 4, characterized in that, The control strategies include model predictive control, PID control, and fuzzy logic control; the key parameters include the proportional, integral, and derivative coefficients of the control strategy, as well as the prediction time domain and control time domain.

6. The adaptive control method for fuel cells according to claim 4, characterized in that, The process of acquiring historical control values ​​corresponding to the previous operating mode of the fuel cell, and using the weighted sum of the original control value and the historical control value as the final control value at the current moment, includes: During the switching of operating mode of fuel cell, when the current operating mode of fuel cell is the same as the operating mode of the previous moment, a first weight is determined based on the current operating mode or the operating mode of the previous moment. The original control quantity and the historical control quantity are weighted and fused based on the first weight to obtain the final control quantity of the control strategy corresponding to the current operating mode. When the current operating mode of the fuel cell is different from the operating mode at the previous moment, a second weight is determined based on the operating mode at the previous moment, and the original control quantity and the historical control quantity are weighted and fused based on the second weight to obtain the final control quantity of the control strategy corresponding to the current operating mode. The first weight and the second weight are different.

7. The adaptive control method for fuel cells according to claim 6, characterized in that, The control parameters include electric heater power, air compressor speed, inlet air throttle opening, outlet air throttle opening, hydrogen proportional valve opening, hydrogen tailpipe exhaust time, hydrogen tailpipe exhaust frequency, hydrogen circulation pump speed, circulating water pump speed, and radiator speed.

8. A fuel cell adaptive control device, characterized in that, include: The current mode determination module is used to obtain the current operating status and load power demand of the fuel cell in order to determine the current operating mode of the fuel cell. The parameter self-tuning module is used to activate the control strategy corresponding to the current operating condition mode in the preset multimodal control strategy, and to self-tun the key parameters of the control strategy based on the pre-trained extreme learning machine to obtain the original control quantity. The adaptive control module is used to obtain the historical control quantity of the control strategy corresponding to the previous operating mode of the fuel cell, and to use the weighted sum of the original control quantity and the historical control quantity as the final control quantity at the current moment.

9. A fuel cell control device, characterized in that, Including memory and processor; The memory stores a computer-readable program that can be executed by the processor; When the processor executes the computer-readable program, it implements the steps of the fuel cell adaptive control method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the fuel cell adaptive control method according to any one of claims 1-7.

Images