Method and device for double-layer optimal economic predictive control of a membrane exchange fuel cell

Abstract

This invention discloses a two-layer optimized economic predictive control method and apparatus for proton exchange membrane fuel cells. The method includes establishing a mechanistic model of the proton exchange membrane fuel cell system, determining the system's control variables, output variables, and external current disturbance signals; an upper real-time optimization layer constructs a steady-state economic index function based on the current external current disturbance signal and the principle of maximizing the system's net output power, solves for the optimal steady-state operating point, and obtains the optimal output variable setpoint; a lower real-time control layer receives the optimal output variable setpoint, solves for the optimal control variable sequence based on the proton exchange membrane fuel cell prediction model, and adjusts the system's control variables to make the system's actual output variable track the optimal output variable setpoint. This invention solves the problem of low system net efficiency caused by prior art that only focuses on the performance of the fuel cell stack while neglecting the power consumption of auxiliary equipment, thus improving the overall operating economy and energy utilization rate of the system.

Description

Technical Field

[0001] This invention belongs to the field of fuel cell control technology, specifically relating to a method and apparatus for optimizing economic forecasting and control of a bilayer exchange membrane fuel cell. Background Technology

[0002] Maximizing the net output efficiency of a proton exchange membrane fuel cell (PEMFC) system depends not only on the precise control of the physical states of the stack, such as temperature, humidity, and pressure, but also on reducing the power consumption of auxiliary systems. Due to the coupling effect of thermodynamic and electrochemical characteristics, the system faces a complex trade-off mechanism: increasing the reactant inlet pressure enhances the stack's electrochemical output but leads to a significant increase in air compressor energy consumption; similarly, maintaining a higher stack operating temperature, while beneficial to the reaction rate, significantly increases the energy consumption for humidification and the cooling load on the cooling fan.

[0003] In existing research on PEMFC control strategies, the vast majority of work focuses on optimization control at the stack level. The core objective is typically to maintain the stack's output voltage, membrane humidity, or operating temperature at specific setpoints to prevent membrane drying, flooding, or overheating, thereby maximizing the stack's output power. However, this control logic, which only focuses on the performance of individual stack cells, has significant limitations at the system level. Existing control strategies often assume that increasing reactant gas pressure or cooling flow rate is always beneficial, as this improves the stack's output performance. Most current control methods fail to directly optimize "net system power" (i.e., stack output power minus the power consumption of all auxiliary equipment). This results in a suboptimal energy efficiency (economic) state, even though the system performs well at the electrochemical level. Furthermore, PEMFC systems exhibit strong nonlinearity, large time lag, and multivariate coupling characteristics. A single control layer struggles to simultaneously handle steady-state operating point optimization based on economic indicators and rapid dynamic tracking in response to load disturbances. In summary, existing technologies generally focus solely on battery stack performance while neglecting the power consumption of auxiliary equipment, leading to low system net efficiency and hindering the improvement of overall system operational economy and energy utilization. Therefore, conducting research addressing these issues has significant theoretical and practical value. Summary of the Invention

[0004] Therefore, the technical problem to be solved by the present invention is to overcome the technical defects of the prior art that only focuses on the performance of the battery stack and ignores the power consumption of auxiliary equipment, resulting in low net system efficiency, and to propose a method and device for optimizing economic prediction and control of a double layer of exchange membrane fuel cell.

[0005] To achieve the above objectives, the technical solution adopted by this invention is: a dual-layer optimized economic predictive control method for exchange membrane fuel cells, wherein the method employs a dual-layer control structure comprising an upper real-time optimization layer and a lower real-time control layer, and the method includes the following steps: Step S1: Establish a mechanism model of the proton exchange membrane fuel cell system and determine the control variables, output variables and external current disturbance signals of the proton exchange membrane fuel cell system. Step S2: The upper real-time optimization layer constructs a steady-state economic index function based on the current external current disturbance signal and the principle of maximizing the net output power of the proton exchange membrane fuel cell system. It then uses the particle swarm optimization algorithm to solve for the optimal steady-state operating point under a given load and obtains the optimal output variable setpoints. The optimal output variable setpoints include the fuel cell stack output voltage, cathode-side relative humidity, and fuel cell stack operating temperature. Step S3: The lower real-time control layer receives the optimal output variable setpoint calculated by the upper real-time optimization layer, solves the finite-time domain optimal control variable sequence based on the constructed proton exchange membrane fuel cell prediction model, and adjusts the control variables of the proton exchange membrane fuel cell system to make the actual output variable of the proton exchange membrane fuel cell system track the optimal output variable setpoint.

[0006] In one embodiment of the present invention, the calculation formula for maximizing the net output power of the proton exchange membrane fuel cell system in step S2 includes: ; In the formula, For the output power of the proton exchange membrane fuel cell system, This refers to the operating power of the air compressor. The operating power of the circulating water pump. The operating power of the humidifier. This refers to the operating power of the cooling fan.

[0007] In one embodiment of the present invention, step S2, the method for constructing a steady-state economic index function based on the principle of maximizing the net output power of the proton exchange membrane fuel cell system, includes: The steady-state economic performance index function J integrates economic objectives and safety constraint objectives, and its expression is as follows: ; In the formula, The optimal net output power of the proton exchange membrane fuel cell system, i.e., the economic objective, is to achieve this. This represents the absolute value of the error between the steady-state relative humidity on the cathode side and the optimized range. This represents the absolute value of the error between the steady-state operating temperature of the battery stack and the optimized operating temperature range. and These represent the weighting coefficients of the corresponding performance indicators.

[0008] In one embodiment of the present invention, step S2, the method for solving the optimal steady-state operating point under a given load using a particle swarm optimization algorithm, includes: The optimization algorithm employs particle swarm optimization, and the specific solution steps include: Step S21: Given an external current, initialize a particle swarm that satisfies the control variable constraints, including particle position and velocity. Step S22: Substitute the N groups of particles under the current iteration into the mechanism model of the proton exchange membrane fuel cell system and simulate to the steady state point to obtain the steady state output value of the system under the current parameters; Step S23: Substitute the steady-state output value obtained in step S22 into the preset steady-state economic index function for calculation to obtain the optimization index corresponding to the current particle and its corresponding steady-state state. Step S24: Compare the current particle's optimization index with the global optimum, and sequentially determine and update the individual optimum and the global optimum for the current iteration; after the current iteration is completed, the algorithm updates the particle position and velocity vector for the next iteration based on the latest extreme value information; Step S25: Determine whether the maximum number of iterations has been reached. If so, output the steady-state output value corresponding to the global optimal solution as the optimal output variable setting value of the lower real-time control layer.

[0009] In one embodiment of the present invention, in step S3, the method for constructing the prediction model includes: Locally linearized state-space models are established at different operating points. The membership degree of each local model is calculated by the change of external current signal. The local models are then weighted and fused to obtain the current prediction model. ; ; In the formula, The number of local models. For membership function, To switch variables, , , For the first The system matrix of each local model, C is the output matrix. Let k be the system state vector at time k. Let k be the system output vector at time k. Let be the system state vector at time k+1.

[0010] In one embodiment of the present invention, step S3, the method for solving the finite-time domain optimal control variable sequence based on the constructed proton exchange membrane fuel cell prediction model, includes: The lower-level real-time control layer, based on the constructed prediction model and combined with the current state estimate, solves the optimal control problem in the finite time domain at each sampling time, obtaining the optimal control variable sequence for a preset number of future steps. The finite time domain optimization problem is as follows: ; In the formula, The objective function value, Set the optimal output variable value for time k+j. The output variable of the proton exchange membrane fuel cell system at time k+j is predicted at time k. The optimal control variable sequence is calculated for time k+j. and These are the output error weight matrix and the control weight matrix, respectively. To control the increment of the variable.

[0011] In one embodiment of the present invention, step S3, a method for adjusting the control variables of the proton exchange membrane fuel cell system to make the actual output variable of the proton exchange membrane fuel cell system track the optimal output variable set value, includes: The lower real-time control layer extracts only the first control variable from the calculated optimal control variable sequence and applies the control variable to the corresponding actuator of the proton exchange membrane fuel cell system to adjust the control variables of the proton exchange membrane fuel cell system in real time. The control variables include air compressor voltage, cooling water mass flow rate and humidifier power. When entering the next sampling time, the optimal control variable sequence is solved repeatedly and the control variables of the proton exchange membrane fuel cell system are adjusted so that the actual output variable of the proton exchange membrane fuel cell system continuously tracks the optimal output variable set value until the deviation between the actual output of the proton exchange membrane fuel cell system and the optimal output variable set value converges to the preset allowable error range.

[0012] Based on the same inventive concept, the present invention also provides a dual-layer optimized economic forecasting and control device for exchange membrane fuel cells. The device employs a dual-layer control structure comprising an upper real-time optimization layer and a lower real-time control layer. The device includes: The system model building module is used to build a mechanistic model of the proton exchange membrane fuel cell system and determine the control variables, output variables and external current disturbance signals of the proton exchange membrane fuel cell system. The upper-level real-time optimization module is used to construct a steady-state economic index function based on the principle of maximizing the net output power of the proton exchange membrane fuel cell system according to the current external current disturbance signal, and to use the particle swarm optimization algorithm to solve for the optimal steady-state operating point under a given load, thereby obtaining the optimal output variable setpoints, wherein the optimal output variable setpoints include the fuel cell stack output voltage, cathode-side relative humidity, and fuel cell stack operating temperature. The lower-level real-time control module receives the optimal output variable setpoint calculated by the upper-level real-time optimization layer, solves the finite-time domain optimal control variable sequence based on the constructed proton exchange membrane fuel cell prediction model, and adjusts the control variables of the proton exchange membrane fuel cell system to make the actual output variable of the proton exchange membrane fuel cell system track the optimal output variable setpoint.

[0013] Based on the same inventive concept, the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method described above.

[0014] Based on the same inventive concept, the present invention also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described above.

[0015] The beneficial effects of this invention are as follows: By constructing a two-layer control structure including an upper real-time optimization layer (RTO) and a lower real-time control layer (MPC), this invention solves the problem of low net system efficiency caused by focusing only on the performance of the fuel cell stack while ignoring the power consumption of auxiliary equipment in the prior art. It maximizes the net output power of the proton exchange membrane fuel cell system under variable load conditions, thereby improving the overall operating economy and energy utilization of the system. Attached Figure Description

[0016] Figure 1 This is a schematic flowchart of the bilayer optimized economic forecasting and control method for exchange membrane fuel cells according to the present invention.

[0017] Figure 2 This is a diagram of the RTO-MPC dual-layer control structure of the proton exchange membrane fuel cell system of the present invention.

[0018] Figure 3 This is a schematic diagram of the 3×3 mechanism model of the proton exchange membrane fuel cell system of the present invention.

[0019] Figure 4 The result of particle swarm optimization under external current is shown in the figure.

[0020] Figure 5This diagram illustrates the optimal control variable sequence for achieving optimal net power under different external current conditions. The optimal control variable sequence includes the battery stack output voltage, cathode-side relative humidity, and battery stack temperature.

[0021] Figure 6 This is a schematic diagram of an external current step signal.

[0022] Figure 7 The output variable curve of RTO-MPC for a proton exchange membrane fuel cell system under varying external current is shown.

[0023] Figure 8 The graph shows the RTO-MPC control variables of a proton exchange membrane fuel cell system under varying external current. Detailed Implementation

[0024] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0025] like Figure 1 As shown in the figure, this invention provides a two-layer optimized economic predictive control method for exchange membrane fuel cells. The method adopts a two-layer control structure including an upper real-time optimization layer and a lower real-time control layer. The specific method includes the following steps: Step S1: Establish a mechanism model of the proton exchange membrane fuel cell system and determine the control variables, output variables and external current disturbance signals of the proton exchange membrane fuel cell system. Step S2: The upper real-time optimization layer constructs a steady-state economic index function based on the current external current disturbance signal and the principle of maximizing the net output power of the proton exchange membrane fuel cell system. It then uses the particle swarm optimization algorithm to solve for the optimal steady-state operating point under a given load and obtains the optimal output variable setpoints. The optimal output variable setpoints include the fuel cell stack output voltage, cathode-side relative humidity, and fuel cell stack operating temperature. Step S3: The lower real-time control layer receives the optimal output variable setpoint calculated by the upper real-time optimization layer, solves the finite-time domain optimal control variable sequence based on the constructed proton exchange membrane fuel cell prediction model, and adjusts the control variables of the proton exchange membrane fuel cell system to make the actual output variable of the proton exchange membrane fuel cell system track the optimal output variable setpoint.

[0026] This invention solves the problem of low net system efficiency caused by focusing only on the performance of the fuel cell stack while ignoring the power consumption of auxiliary equipment in the prior art by constructing a two-layer control structure including an upper real-time optimization layer (RTO) and a lower real-time control layer (MPC). It maximizes the net output power of the proton exchange membrane fuel cell system under variable load conditions, thereby improving the overall operating economy and energy utilization of the system.

[0027] In step S1, as follows Figure 3 As shown, a mechanistic model of a proton exchange membrane fuel cell system is established, the control variables, output variables, and external current disturbance signals of the proton exchange membrane fuel cell system are determined, and the net output power Powfc is defined as the optimization objective, and its calculation formula is as follows: ; In the formula, For the output power of the proton exchange membrane fuel cell system, This refers to the operating power of the air compressor. The operating power of the circulating water pump. The operating power of the humidifier. This refers to the operating power of the cooling fan.

[0028] Considering that the control systems of the auxiliary equipment are all in operation, this embodiment directly calculates the working power of each auxiliary equipment based on the control variables.

[0029] In step S2, such as Figure 2 As shown, the upper Real-Time Optimization (RTO) layer is designed as follows: This layer is responsible for finding the economic optimization of the system's steady-state operating point, and specifically includes the following: 1. Establish a steady-state economic performance index function J, which integrates economic objectives and safety constraints. The primary objective is to maximize the net output power of the proton exchange membrane fuel cell system. Penalty terms are introduced for the relative humidity on the cathode side (0.85~0.95) and the stack operating temperature (353~359 K) as soft constraints. The expression for the steady-state economic performance index function J is: ; In the formula, The optimal net output power of the proton exchange membrane fuel cell system, i.e., the economic objective, is to achieve this. This represents the absolute value of the error between the steady-state relative humidity on the cathode side and the optimized range. This represents the absolute value of the error between the steady-state operating temperature of the battery stack and the optimized operating temperature range. and These represent the weighting coefficients of the corresponding performance indicators.

[0030] Using the Particle Swarm Optimization (PSO) algorithm, under a given external load current disturbance, the algorithm searches online for the optimal combination of control variables that optimizes index J. It outputs the optimal battery stack output voltage, optimal cathode-side relative humidity, and optimal battery stack operating temperature, which serve as the trajectory of the optimal output variable setpoints for the lower-level real-time control layer. Specifically, as follows... Figure 4 As shown, Figure 4 The horizontal axis represents the iteration count M (25 iterations are shown), and the vertical axis represents the three core control variables that need to be optimized in the system model. The multiple colored lines in the figure represent the trajectories of multiple "particles" (i.e., different parameter combinations) during the optimization process. The optimization algorithm uses particle swarm optimization, and the specific solution steps include: Step S21: Given an external current, initialize a particle swarm that satisfies the control variable constraints, including particle position and velocity. Step S22: Substitute the N groups of particles under the current iteration into the mechanism model of the proton exchange membrane fuel cell system and simulate to the steady state point to obtain the steady state output value of the system under the current parameters; Step S23: Substitute the steady-state output value obtained in step S22 into the preset steady-state economic index function for calculation to obtain the optimization index corresponding to the current particle and its corresponding steady-state state. Step S24: Compare the current particle's optimization index with the global optimum, and sequentially determine and update the individual optimum and the global optimum for the current iteration; after the current iteration is completed, the algorithm updates the particle position and velocity vector for the next iteration based on the latest extreme value information; Step S25: Determine whether the maximum number of iterations has been reached. If so, output the steady-state output value corresponding to the global optimal solution as the optimal output variable setting value of the lower real-time control layer.

[0031] In step S3, such as Figure 2 As shown, the lower real-time control layer (MPC) is designed as follows: This layer is responsible for the rapid tracking control of the upper-layer optimized settings through multi-model prediction and optimization control algorithms, specifically including the following: 1. Multi-model prediction model: To address the nonlinear characteristics of proton exchange membrane fuel cell systems, a multi-model strategy based on fuzzy membership weighting is employed. Locally linearized state-space models are established at different load conditions (e.g., different current ranges). By detecting external current signals, the membership degrees of each local model are calculated in real time and then weighted and fused to construct a global prediction model adaptable to a wide range of operating conditions. ; ; In the formula, The number of local models. For membership function, To switch variables, , , For the first The system matrix of each local model, C is the output matrix. Let k be the system state vector at time k. Let k be the system output vector at time k. Let be the system state vector at time k+1.

[0032] 2. Optimized Control: Based on the constructed prediction model and combined with the current state estimate, the lower real-time control layer solves the optimal control problem in the finite time domain at each sampling time, obtaining the optimal control variable sequence for a preset number of future steps. The finite time domain optimization problem is as follows: ; In the formula, The objective function value, Set the optimal output variable value for time k+j. The output variable of the proton exchange membrane fuel cell system at time k+j is predicted at time k. The optimal control variable sequence is calculated for time k+j. and These are the output error weight matrix and the control weight matrix, respectively. To control the increment of the variable.

[0033] The lower-level real-time control layer extracts only the first control variable from the calculated optimal control variable sequence and applies it to the corresponding actuator of the proton exchange membrane fuel cell system. This allows for real-time adjustment of the system's control variables, including air compressor voltage, cooling water mass flow rate, and humidifier power. At the next sampling time, the optimal control variable sequence is repeatedly solved and the control variables of the proton exchange membrane fuel cell system are adjusted to ensure that the actual output variable of the system continuously tracks the optimal output variable setpoint until the deviation between the actual output and the optimal output variable setpoint converges within a preset allowable error range. This achieves synergistic optimization of the system's dynamic response performance and economic performance. The controller design includes hard constraints on the control variables and their rates of change to prevent actuator saturation or drastic fluctuations.

[0034] The following specific examples further illustrate the present invention's method for optimizing economic forecasting and control of a bilayer exchange membrane fuel cell.

[0035] First, a nonlinear mechanism model (controlled object) and a two-layer controller model of the proton exchange membrane fuel cell system are built in the Simulink environment.

[0036] Next, the solution process of the Particle Swarm Optimization (PSO) algorithm for optimizing the net output power of a proton exchange membrane fuel cell system based on PSO is as follows: First, a swarm of particles representing control variables (such as air compressor voltage) is initialized and substituted into the mechanistic model of the proton exchange membrane fuel cell system to simulate to a steady state to obtain output data. Subsequently, the economic indicators of the system are calculated using the steady-state output values, and the individual optimal solution Je of each particle and the global optimal solution Je,best of the swarm are updated by comparing with historical data. Based on this, the particles adjust their positions and velocities according to the update formula, and the above simulation and evaluation process is iterated continuously until the maximum number of iterations M is reached, finally outputting the steady-state operating point that optimizes the net power of the system.

[0037] Then, a prediction model is constructed: In this case, the transfer function matrix of the system is identified by the one-to-one correspondence between input and output variables. This matrix is ​​then directly converted into a state-space form using the Matlab toolbox. The state-space models at different operating points are used as local sub-models. By weighting the membership degrees of these local sub-models based on changes in the external current signal, a new prediction model is obtained, which can more accurately describe the dynamics of the mechanism model. The proposed multi-model prediction model is shown in the following equation: ; In the formula, , and These represent the increment values ​​at the state, input, and disturbance times, respectively. , , ; Next, the optimization problem is solved: the lower-level real-time control layer uses model membership-weighted MPC to track the optimal setpoint of the RTO layer. Kalman filtering is used to estimate the system's state variables in real time, and a nonlinear programming solver is employed to solve the optimization problem of multi-model predictive control in the finite-time domain, as shown in the following equation: ; In the formula, Yr is the optimal output variable setting value obtained based on the upper-level RTO calculation. To control the increment of the variable, The error weight matrix for the output variables. This is the error weight matrix for the control variables.

[0038] The aforementioned optimization problem aims to enable the output variable to quickly track the optimal output variable setpoint calculated by the upper-level RTO, while maintaining the stability of the control variable. The controller's execution steps are as follows: at the current time step, the optimization problem is solved to obtain the optimal control variable sequence. The lower-level real-time control layer extracts only the first control variable from the calculated optimal control variable sequence and applies it to the controlled system. At the next time step, the controller estimates the state variables at the current time step using Kalman filtering, continues to solve the optimization problem, calculates the optimal control variable sequence for the next time step, and applies the first optimal control variable to the controlled system again. This process is repeated until the entire simulation duration ends.

[0039] The simulation environment for this case is as follows: 1. Sampling time: set to 0.1s; 2. Predicted horizon and control horizon: The prediction time domain P is set to 10, and the control time domain M is set to 2; 3. Weighting matrix: Error weighting matrix of controlled variables Control variable weight matrix ; 4. Constraints: Air compressor voltage constraint: [50 300]; Cooling water mass flow rate constraint: [0.001 0.8]; Humidifier power constraint: [0 1200].

[0040] External current simulation experiments such as Figure 6 As shown: The purpose of the simulation experiment is to test the control performance in tracking the optimal operating point under varying loads. The system initially operates stably under the economically optimal steady-state condition with an external current of 130 A. At the 35th second, the external current experiences its first significant step disturbance, suddenly increasing to 150 A and maintaining this high load. Subsequently, at the 235th second, the external current undergoes a second step change, instantaneously decreasing by 10 A (i.e., dropping to 140 A), and continues to operate under this load until the simulation ends after 400 seconds.

[0041] The control performance parameters of RTO-MPC are shown in Table 1: Table 1

[0042] The simulation results of this case are as follows: Figure 5 , Figure 7 As shown in Figure 8, Figure 5The figure illustrates the trends in the stack output voltage, cathode-side relative humidity, and operating temperature under different external currents to achieve optimal net power. The dynamic performance of the operating parameters shows that as the external current increases (from 80 A to 180 A), the optimal output voltage of the stack generally exhibits a significant decreasing trend due to the influence of the polarization characteristic curve. Simultaneously, the optimal cathode-side relative humidity remains remarkably stable at 95%, near the upper limit of the optimization range. This further verifies that maintaining sufficient humidification within the stack is beneficial for maximizing the system's net power output. Figure 7 The simulation curves fully demonstrate that the proposed RTO-MPC control strategy not only has accurate and fast setpoint tracking capabilities, but also effectively overcomes the strong nonlinearity and coupling interference between multiple variables in the proton exchange membrane fuel cell system, ensuring that the system can still safely and quickly transition to a new economically optimal operating point when faced with drastic changes in external load. Figure 8 The graph shows the RTO-MPC control variables of the proton exchange membrane fuel cell system under varying external current. The input air compressor voltage, humidifier power, and cooling water mass flow rate are all operating reasonably within the specified limits and meet the constraints.

[0043] Corresponding to the above method embodiments, this invention also provides a dual-layer optimized economic prediction and control device for exchange membrane fuel cells. The device employs a dual-layer control structure comprising an upper real-time optimization layer and a lower real-time control layer. The device includes: The system model building module is used to build a mechanistic model of the proton exchange membrane fuel cell system and determine the control variables, output variables and external current disturbance signals of the proton exchange membrane fuel cell system. The upper-level real-time optimization module is used to construct a steady-state economic index function based on the principle of maximizing the net output power of the proton exchange membrane fuel cell system according to the current external current disturbance signal, and to use the particle swarm optimization algorithm to solve for the optimal steady-state operating point under a given load, thereby obtaining the optimal output variable setpoints, wherein the optimal output variable setpoints include the fuel cell stack output voltage, cathode-side relative humidity, and fuel cell stack operating temperature. The lower-level real-time control module receives the optimal output variable setpoint calculated by the upper-level real-time optimization layer, solves the finite-time domain optimal control variable sequence based on the constructed proton exchange membrane fuel cell prediction model, and adjusts the control variables of the proton exchange membrane fuel cell system to make the actual output variable of the proton exchange membrane fuel cell system track the optimal output variable setpoint.

[0044] For ease of description, the above apparatus is described in terms of its functions as various system modules. Of course, in implementing this disclosure, the functions of each system module can be implemented in one or more software and / or hardware.

[0045] The apparatus of the above embodiments is used to implement the bilayer optimized economic forecasting control method for exchange membrane fuel cells in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0046] Corresponding to the above method embodiments, this embodiment of the invention also provides a computer device, including: Memory, which is used to store computer programs; The processor is used to execute computer programs to implement the steps of the above-described optimized economic forecasting control method for bilayer exchange membrane fuel cells.

[0047] In this embodiment of the invention, the processor may be a central processing unit (CPU), an application-specific integrated circuit, a digital signal processor, a field-programmable gate array, or other programmable logic devices.

[0048] The processor can call programs stored in the memory. Specifically, the processor can execute the operations in the embodiments of the above-described bilayer optimized economic forecasting control method for exchange membrane fuel cells.

[0049] The memory is used to store one or more programs, which may include program code, including computer operation instructions.

[0050] In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device or other volatile solid-state storage device.

[0051] Corresponding to the above method embodiments, this embodiment of the invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described exchange membrane fuel cell bilayer optimized economic forecasting and control method.

[0052] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0053] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0054] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0055] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0056] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for optimizing economic forecasting and control of a bilayer exchange membrane fuel cell, characterized in that, The method employs a two-layer control structure comprising an upper real-time optimization layer and a lower real-time control layer, and includes the following steps: Step S1: Establish a mechanism model of the proton exchange membrane fuel cell system and determine the control variables, output variables and external current disturbance signals of the proton exchange membrane fuel cell system. Step S2: The upper real-time optimization layer constructs a steady-state economic index function based on the current external current disturbance signal and the principle of maximizing the net output power of the proton exchange membrane fuel cell system. It then uses the particle swarm optimization algorithm to solve for the optimal steady-state operating point under a given load and obtains the optimal output variable setpoints. The optimal output variable setpoints include the fuel cell stack output voltage, cathode-side relative humidity, and fuel cell stack operating temperature. Step S3: The lower real-time control layer receives the optimal output variable setpoint calculated by the upper real-time optimization layer, solves the finite-time domain optimal control variable sequence based on the constructed proton exchange membrane fuel cell prediction model, and adjusts the control variables of the proton exchange membrane fuel cell system to make the actual output variable of the proton exchange membrane fuel cell system track the optimal output variable setpoint.

2. The method for optimized economic forecasting and control of a bilayer exchange membrane fuel cell according to claim 1, characterized in that, In step S2, the formula for maximizing the net output power of the proton exchange membrane fuel cell system includes: ； In the formula, For the output power of the proton exchange membrane fuel cell system, This refers to the operating power of the air compressor. The operating power of the circulating water pump. The operating power of the humidifier. This refers to the operating power of the cooling fan.

3. The method for optimized economic forecasting and control of a bilayer exchange membrane fuel cell according to claim 2, characterized in that, In step S2, the method for constructing a steady-state economic index function based on the principle of maximizing the net output power of the proton exchange membrane fuel cell system includes: The steady-state economic performance index function J integrates economic objectives and safety constraint objectives, and its expression is as follows: ； In the formula, The optimal net output power of the proton exchange membrane fuel cell system, i.e., the economic objective, is to achieve this. This represents the absolute value of the error between the steady-state relative humidity on the cathode side and the optimized range. This represents the absolute value of the error between the steady-state operating temperature of the battery stack and the optimized operating temperature range. and These represent the weighting coefficients of the corresponding performance indicators.

4. The method for optimized economic forecasting and control of a bilayer exchange membrane fuel cell according to claim 1, characterized in that, In step S2, the method for solving the optimal steady-state operating point under a given load using the particle swarm optimization algorithm includes: The optimization algorithm employs particle swarm optimization, and the specific solution steps include: Step S21: Given an external current, initialize a particle swarm that satisfies the control variable constraints, including particle position and velocity. Step S22: Substitute the N groups of particles under the current iteration into the mechanism model of the proton exchange membrane fuel cell system and simulate to the steady state point to obtain the steady state output value of the system under the current parameters; Step S23: Substitute the steady-state output value obtained in step S22 into the preset steady-state economic index function for calculation to obtain the optimization index corresponding to the current particle and its corresponding steady-state state. Step S24: Compare the current particle's optimization index with the global optimum, and sequentially determine and update the individual optimum and the global optimum for the current iteration; after the current iteration is completed, the algorithm updates the particle position and velocity vector for the next iteration based on the latest extreme value information; Step S25: Determine whether the maximum number of iterations has been reached. If so, output the steady-state output value corresponding to the global optimal solution as the optimal output variable setting value of the lower real-time control layer.

5. The method for optimized economic forecasting and control of a bilayer exchange membrane fuel cell according to claim 4, characterized in that, In step S3, the method for constructing the prediction model includes: Locally linearized state-space models are established at different operating points. The membership degree of each local model is calculated by the change of external current signal. The local models are then weighted and fused to obtain the current prediction model. ； ； In the formula, The number of local models. For membership function, To switch variables, , , For the first The system matrix of each local model, C is the output matrix. Let k be the system state vector at time k. Let k be the system output vector at time k. Let be the system state vector at time k+1.

6. The method for optimized economic forecasting and control of a bilayer exchange membrane fuel cell according to claim 5, characterized in that, In step S3, the method for solving the finite-time domain optimal control variable sequence based on the constructed proton exchange membrane fuel cell prediction model includes: The lower-level real-time control layer, based on the constructed prediction model and combined with the current state estimate, solves the optimal control problem in the finite time domain at each sampling time, obtaining the optimal control variable sequence for a preset number of future steps. The finite time domain optimization problem is as follows: ； In the formula, The objective function value, Set the optimal output variable value for time k+j. The output variable of the proton exchange membrane fuel cell system at time k+j is predicted at time k. The optimal control variable sequence is calculated for time k+j. and These are the output error weight matrix and the control weight matrix, respectively. To control the increment of the variable.

7. The method for optimized economic forecasting and control of a bilayer exchange membrane fuel cell according to claim 6, characterized in that, In step S3, the method of adjusting the control variables of the proton exchange membrane fuel cell system to make the actual output variable of the proton exchange membrane fuel cell system track the optimal output variable setpoint includes: The lower real-time control layer extracts only the first control variable from the calculated optimal control variable sequence and applies the control variable to the corresponding actuator of the proton exchange membrane fuel cell system to adjust the control variables of the proton exchange membrane fuel cell system in real time. The control variables include air compressor voltage, cooling water mass flow rate and humidifier power. When entering the next sampling time, the optimal control variable sequence is solved repeatedly and the control variables of the proton exchange membrane fuel cell system are adjusted so that the actual output variable of the proton exchange membrane fuel cell system continuously tracks the optimal output variable set value until the deviation between the actual output of the proton exchange membrane fuel cell system and the optimal output variable set value converges to the preset allowable error range.

8. A bilayer optimized economic forecasting and control device for an exchange membrane fuel cell, characterized in that, The device employs a dual-layer control structure comprising an upper real-time optimization layer and a lower real-time control layer. The device includes: The system model building module is used to build a mechanistic model of the proton exchange membrane fuel cell system and determine the control variables, output variables and external current disturbance signals of the proton exchange membrane fuel cell system. The upper-level real-time optimization module is used to construct a steady-state economic index function based on the principle of maximizing the net output power of the proton exchange membrane fuel cell system according to the current external current disturbance signal, and to use the particle swarm optimization algorithm to solve for the optimal steady-state operating point under a given load, thereby obtaining the optimal output variable setpoints, wherein the optimal output variable setpoints include the fuel cell stack output voltage, cathode-side relative humidity, and fuel cell stack operating temperature. The lower-level real-time control module receives the optimal output variable setpoint calculated by the upper-level real-time optimization layer, solves the finite-time domain optimal control variable sequence based on the constructed proton exchange membrane fuel cell prediction model, and adjusts the control variables of the proton exchange membrane fuel cell system to make the actual output variable of the proton exchange membrane fuel cell system track the optimal output variable setpoint.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method as described in any one of claims 1-7.

Images