Fuel cell air supply system control method based on fusion model prediction

Abstract

The application relates to a fuel cell air supply system control method based on a fusion model prediction, which comprises the following steps: step S1: constructing a high-order nonlinear model of a fuel cell stack and an air supply system, and simplifying the high-order nonlinear model into a six-order model; step S2: linearizing the six-order model by using a Taylor expansion mode to obtain a linearized model; step S3: bringing state parameter data of A and B power points into the linearized model, and converting the linearized model into a form of a state space equation; step S4: designing a controller, and giving constraints and reference outputs to the controller; and step S5: realizing output coupling of two controllers, and designing a compensation amount for the coupled control amount to compensate for the control amount when a current step drops. The application can realize effective control of cathode air supply of a fuel cell.

Description

Technical Field

[0001] This invention relates to the field of fuel cell system control, and more specifically to a control method for a fuel cell air supply system based on fusion model prediction. Background Technology

[0002] The utilization of hydrogen energy plays a crucial role in achieving carbon neutrality and alleviating energy shortage pressures. Hydrogen, with its high energy density and clean reaction products, is considered an excellent alternative to fossil fuels. Among the many automotive fuel cell technologies, proton exchange membrane fuel cells (PEMFCs) are one of the most promising options.

[0003] PEMFC has advantages such as high energy conversion efficiency, good technical controllability, zero emissions, low temperature, and good start-up performance. It can catalyze the reaction of hydrogen and oxygen to produce water.

[0004] To ensure the proper functioning of the internal reactions in a proton exchange membrane fuel cell (PEMFC), several auxiliary systems are required to guarantee sufficient reactant supply, a suitable temperature range for the fuel stack and reactants, and appropriate humidity levels for the reactants. Among these, the air system is crucial, providing the fuel stack with precise oxygen supply. This not only prevents oxygen deficiency but also reduces the power consumption of auxiliary systems. PEMFC oxygen deficiency occurs due to rapid changes in vehicle operating conditions, leading to rapid variations in the required airflow rate, which can potentially cause fuel stack degradation and damage. If the air compressor provides excessive airflow, the fuel cell's net power will suffer a significant loss. Therefore, the air supply system requires precise control to provide the appropriate airflow rate to the fuel stack. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a control method for a fuel cell air supply system based on fusion model prediction, which can achieve effective control of the cathode air supply of the fuel cell.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A control method for a fuel cell air supply system based on fusion model prediction includes the following steps:

[0008] Step S1: Based on the characteristics of the proton exchange membrane fuel cell stack and air supply system, construct a high-order nonlinear model of the fuel cell stack and air supply system, and assume that the anode hydrogen supply meets the requirements and the humidifier is in an ideal state, simplifying the high-order nonlinear model into a sixth-order model.

[0009] Step S2: Obtain the state parameter data of the fuel cell at power points A and B based on the high-order nonlinear model, and linearize the sixth-order model using Taylor expansion to obtain the linearized model;

[0010] Step S3: Substitute the state parameter data of power points A and B into the linearized model, and convert the linearized model into the form of state-space equations;

[0011] Step S4: Using the model prediction algorithm, design a controller for the linearized model at power points A and B, and give the controller constraints and reference output;

[0012] Step S5: Give variable weights to the two model predictive controllers constructed in step S4 to realize the output coupling of the two controllers, and design a compensation quantity for the coupled control quantity to supplement the control quantity when the current steps down, so as to realize the compensation of the control quantity when the current steps down.

[0013] Furthermore, the established high-order nonlinear model of the fuel cell stack and air supply system is as follows:

[0014]

[0015] In the formula For state variables, specifically It is the compressor angular velocity. It's the air pressure in the air supply manifold. It refers to the gas quality in the gas supply manifold. It refers to the oxygen quality inside the fuel cell. It refers to the mass of hydrogen inside the fuel cell. It refers to the mass of nitrogen gas inside the fuel cell. It is the air pressure in the return manifold. It is the mass of water vapor at the anode of the fuel cell. It refers to the mass of water vapor at the cathode of the fuel cell. , and These are the hydrogen input, output, and reaction consumption flow rates at the fuel cell anode. , , and These are the fuel cell anode input, output, water vapor passing through the proton exchange membrane, and output liquid water flow rates. and The nitrogen flow rate at the anode input and output of the fuel cell. , , , and These are the input, output, generation, and liquid water flow rates of the fuel cell cathode, which are the water vapor passing through the proton exchange membrane and the output. , and These are the oxygen input, output, and reaction consumption flow rates at the fuel cell cathode. The temperature of the gas in the exhaust manifold. The gas constant of air. For exhaust manifold volume, This represents the mass flow rate of the gas output from the cathode. This refers to the mass flow rate output from the exhaust manifold. It is the output mass flow rate of the air compressor. This refers to the mass flow rate output from the intake manifold. For constant thermal pressure ratio, For the intake manifold volume, The output gas temperature of the air compressor. This refers to the intake manifold gas temperature. and These are the air compressor torque and load, respectively. It is the equivalent inertia of the air compressor.

[0016] Furthermore, the control input of the fuel cell air supply system is the control voltage of the air compressor, the measurable input disturbance is the current, and the output is the oxygen excess coefficient of the fuel cell stack. The input-output expression is as follows:

[0017]

[0018]

[0019] In the formula, This indicates the current in the fuel cell stack. This indicates the compressor control voltage. This indicates the oxygen intake flow rate of the fuel cell stack. This indicates the oxygen flow rate required for the fuel cell reaction.

[0020] Furthermore, the sixth-order model is specifically as follows:

[0021]

[0022] In the formula , , , , , For state variables, specifically It is the compressor angular velocity. It's the air pressure in the air supply manifold. It refers to the gas quality in the gas supply manifold. It refers to the oxygen quality inside the fuel cell. It refers to the mass of nitrogen gas inside the fuel cell. It is the air pressure in the return manifold. and The nitrogen flow rate at the anode input and output of the fuel cell. , and These are the oxygen input, output, and reaction consumption flow rates at the fuel cell cathode. The temperature of the gas in the exhaust manifold. The gas constant of air. For exhaust manifold volume, This represents the mass flow rate of the gas output from the cathode. This refers to the mass flow rate output from the exhaust manifold. It is the output mass flow rate of the air compressor. This refers to the mass flow rate output from the intake manifold. For constant thermal pressure ratio, For the intake manifold volume, The output gas temperature of the air compressor. This refers to the intake manifold gas temperature. and These are the air compressor torque and load, respectively. It is the equivalent inertia of the air compressor.

[0023] Furthermore, the specific method for obtaining the state parameter data of the fuel cell at power points A and B based on a high-order nonlinear model is as follows:

[0024] ① Establish a mathematical model of a high-order nonlinear model in the Matlab / simulink simulation environment, and the parameter states inside the model can be observed in real time.

[0025] ② Input constant load currents corresponding to power points A and B to simulate the input load of the fuel cell system under steady-state conditions.

[0026] ③ Observe and record the model's internal parameter data under the load current at power points A and B respectively.

[0027] Furthermore, the sixth-order model is linearized using Taylor expansion to obtain a linearized model, specifically as follows:

[0028] The simplified model of the fuel cell air system is linearized using Taylor expansion, and the values ​​of the state variables under operating states A and B are substituted into the linearized model to obtain the model state equations under operating states A and B at power points.

[0029] The linearization method is as follows:

[0030]

[0031] The linearized state equations are shown below:

[0032]

[0033] in .

[0034] In the formula, For state vectors, For control vectors, For measurable interference, Unmeasurable interference The state matrix, For the input matrix, For measurable interference matrix, The interference matrix is ​​unmeasurable. For the output matrix, For direct transition matrix, This is for outputting the interference matrix.

[0035] Furthermore, step S4 specifically includes:

[0036] Discretize the state equations.

[0037]

[0038] Based on the above model, the system prediction equation is derived as follows:

[0039]

[0040] In the formula Let p be the predicted step control output at time k, where p is the prediction time domain. Let M be the M-step control input column vector, where M is the control time domain; , , , It is the prediction matrix:

[0041]

[0042]

[0043] , ,

[0044]

[0045] Given system constraints:

[0046]

[0047]

[0048]

[0049] In the formula, This refers to the control quantity, specifically the compressor voltage. This refers to the change in the control quantity. This refers to the output, i.e., the oxygen excess coefficient;

[0050] Solve for the objective function:

[0051]

[0052] In the formula, and It is the weight matrix of the optimization problem. This is the reference output matrix:

[0053]

[0054]

[0055]

[0056] The model predictive controller for the fuel cell system under power operating conditions A and B is derived.

[0057] Furthermore, step S5 specifically includes:

[0058] ① Under low power current conditions, the weight ratio of the controller designed with power point A data is 1, while the weight ratio of the controller designed with power point B data is 0.

[0059] ② Under medium power current conditions, the weighting ratio of the controller designed using power point A data is a function that decreases as the current increases, while the weighting ratio of the controller designed using power point B data is a function that increases as the current increases;

[0060] Specifically, the functional expressions for the weight ratios A and B are:

[0061]

[0062]

[0063] ③ Under high power current conditions, the weight ratio of the controller designed with power point A data is 0, while the weight ratio of the controller designed with power point B data is 1.

[0064] The final result of the fusion control of the two controllers is the sum of the control quantities of the two controllers multiplied by their corresponding weights.

[0065] Furthermore, the compensation amount is specifically as follows:

[0066]

[0067] In the formula This represents the current difference between the current at the current moment and the previous moment. This indicates the current value at the current moment. This represents the compressor control voltage at the previous moment. and It is a hyperparameter related to the current difference and the current value.

[0068] Compared with the prior art, the present invention has the following advantages:

[0069] 1. This invention simplifies the fuel cell gas supply system and stack model into a sixth-order model and uses it as a prediction model, which effectively ensures the accuracy of the model while reducing the amount of computation.

[0070] 2. This invention comprehensively considers the system constraints and the control performance of the oxygen excess coefficient. In the event of cathode oxygen starvation, it uses a model prediction algorithm to control the air system, which can make the control results reach the reference value quickly and accurately.

[0071] 3. This invention linearizes the models for two current operating points and designs model prediction algorithms based on the linearized models respectively. Combined with a weighting strategy that varies with current, the control results of the model prediction control algorithms designed for the two different operating points are coupled, enabling the control algorithm to achieve better control performance over a wider operating range of the fuel cell.

[0072] 4. This invention designs a compensation strategy to supplement the control quantity of the model predictive control algorithm for the case of large current step, which can effectively suppress the overshoot phenomenon of the control result under the case of large current step. Attached Figure Description

[0073] Figure 1 This is a schematic diagram of the method of the present invention;

[0074] Figure 2 This is a schematic diagram of the compensation method in one embodiment of the present invention;

[0075] Figure 3 This is a comparison chart of oxygen excess ratio control under step current conditions for a 75kW fuel cell in one embodiment of the present invention. Detailed Implementation

[0076] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0077] Please refer to Figure 1 This invention provides a control method for a fuel cell air supply system based on fusion model prediction, comprising the following steps:

[0078] Step S1: Based on the characteristics of the proton exchange membrane fuel cell stack and air supply system, construct a high-order nonlinear model of the fuel cell stack and air supply system, and assume that the anode hydrogen supply meets the requirements and the humidifier is in an ideal state, simplifying the high-order nonlinear model into a sixth-order model.

[0079] Step S2: Obtain the state parameter data of the fuel cell at power points A and B based on the high-order nonlinear model, and linearize the sixth-order model using Taylor expansion to obtain the linearized model;

[0080] Step S3: Substitute the state parameter data of power points A and B into the linearized model, and convert the linearized model into the form of state-space equations;

[0081] Step S4: Using the model prediction algorithm, design a controller for the linearized model at power points A and B, and give the controller constraints and reference output;

[0082] Step S5: Give variable weights to the two model predictive controllers constructed in step S4 to realize the output coupling of the two controllers, and design a compensation quantity for the coupled control quantity to supplement the control quantity when the current steps down, so as to realize the compensation of the control quantity when the current steps down.

[0083] In this embodiment, the established high-order nonlinear model of the fuel cell stack and air supply system is as follows:

[0084]

[0085] In the formula For state variables, specifically It is the compressor angular velocity. It's the air pressure in the air supply manifold. It refers to the gas quality in the gas supply manifold. It refers to the oxygen quality inside the fuel cell. It refers to the mass of hydrogen inside the fuel cell. It refers to the mass of nitrogen gas inside the fuel cell. It is the air pressure in the return manifold. It is the mass of water vapor at the anode of the fuel cell. It refers to the mass of water vapor at the cathode of the fuel cell. , and These are the hydrogen input, output, and reaction consumption flow rates at the fuel cell anode. , , and These are the fuel cell anode input, output, water vapor passing through the proton exchange membrane, and output liquid water flow rates. and The nitrogen flow rate at the anode input and output of the fuel cell. , , , and These are the input, output, generation, and liquid water flow rates of the fuel cell cathode, which are the water vapor passing through the proton exchange membrane and the output. , and These are the oxygen input, output, and reaction consumption flow rates at the fuel cell cathode. The temperature of the gas in the exhaust manifold. The gas constant of air. For exhaust manifold volume, This represents the mass flow rate of the gas output from the cathode. This refers to the mass flow rate output from the exhaust manifold. It is the output mass flow rate of the air compressor. This refers to the mass flow rate output from the intake manifold. For constant thermal pressure ratio, For the intake manifold volume, The output gas temperature of the air compressor. This refers to the intake manifold gas temperature. and These are the air compressor torque and load, respectively. It is the equivalent inertia of the air compressor.

[0086] In this embodiment, the control input of the fuel cell air supply system is the control voltage of the air compressor, the measurable input disturbance is the current, and the output is the oxygen excess coefficient of the fuel cell stack. The input-output expression is as follows:

[0087]

[0088]

[0089] In the formula, This indicates the current in the fuel cell stack. This indicates the compressor control voltage. This indicates the oxygen intake flow rate of the fuel cell stack. This indicates the oxygen flow rate required for the fuel cell reaction.

[0090] In this embodiment, the sixth-order model is specifically as follows:

[0091]

[0092] In the formula , , , , , For state variables, specifically It is the compressor angular velocity. It's the air pressure in the air supply manifold. It refers to the gas quality in the gas supply manifold. It refers to the oxygen quality inside the fuel cell. It refers to the mass of nitrogen gas inside the fuel cell. It is the air pressure in the return manifold. and The nitrogen flow rate at the anode input and output of the fuel cell. , and These are the oxygen input, output, and reaction consumption flow rates at the fuel cell cathode. The temperature of the gas in the exhaust manifold. The gas constant of air. For exhaust manifold volume, This represents the mass flow rate of the gas output from the cathode. This refers to the mass flow rate output from the exhaust manifold. It is the output mass flow rate of the air compressor. This refers to the mass flow rate output from the intake manifold. For constant thermal pressure ratio, For the intake manifold volume, The output gas temperature of the air compressor. This refers to the intake manifold gas temperature. and These are the air compressor torque and load, respectively. It is the equivalent inertia of the air compressor.

[0093] In this embodiment, the specific method for obtaining the state parameter data of the fuel cell at power points A and B based on a high-order nonlinear model is as follows:

[0094] ① Establish a mathematical model of a high-order nonlinear model in the Matlab / simulink simulation environment, and the parameter states inside the model can be observed in real time.

[0095] ② Input constant load currents corresponding to power points A and B to simulate the input load of the fuel cell system under steady-state conditions.

[0096] ③ Observe and record the model's internal parameter data under the load current at power points A and B respectively.

[0097] In this embodiment, the sixth-order model is linearized using Taylor expansion to obtain a linearized model, specifically:

[0098] Using Taylor expansion, the sixth-order model is linearized based on the state parameter data of the fuel cell at power points A and B, and the linearized model is then written in state-space form:

[0099]

[0100] .

[0101] In the formula, For state vectors, For control vectors, For measurable interference, Unmeasurable interference The state matrix, For the input matrix, For measurable interference matrix, The interference matrix is ​​unmeasurable. For the output matrix, For direct transition matrix, This is for outputting the interference matrix.

[0102] In this embodiment, the state-space model is discretized, constraints are added to the system, and a controller is designed for the state-space model at two different current operating points using a model prediction algorithm:

[0103] ;

[0104] In this embodiment, step S5 specifically includes:

[0105] ① Under low power current conditions, the weight ratio of the controller designed with power point A data is 1, while the weight ratio of the controller designed with power point B data is 0.

[0106] ② Under medium power current conditions, the weighting ratio of the controller designed using power point A data is a function that decreases as the current increases, while the weighting ratio of the controller designed using power point B data is a function that increases as the current increases;

[0107] Specifically, the functional expressions for the weight ratios A and B are:

[0108]

[0109]

[0110] ③ Under high power current conditions, the weight ratio of the controller designed with power point A data is 0, while the weight ratio of the controller designed with power point B data is 1.

[0111] The final result of the fusion control of the two controllers is the sum of the control quantities of the two controllers multiplied by their corresponding weights.

[0112] In this embodiment, the compensation amount is specifically:

[0113]

[0114] In the formula This represents the current difference between the current at the current moment and the previous moment. This indicates the current value at the current moment. This represents the compressor control voltage at the previous moment. and It is a hyperparameter related to the current difference and the current value.

[0115] Example 1:

[0116] This embodiment takes a 75kW proton exchange membrane fuel cell as an example for further analysis.

[0117] In this embodiment, the control input of the fuel cell air system of the 75kW proton exchange membrane fuel cell is the control voltage of the air compressor, the measurable input disturbance is the current, and the output is the oxygen excess coefficient of the fuel cell stack. The input-output expression is:

[0118]

[0119]

[0120] By combining the input and output quantities with a nonlinear model and simplifying it, the resulting model is:

[0121]

[0122] The values ​​of the state variables of the fuel cell air system at power points A and B were obtained through nonlinear model simulation.

[0123] The simplified model of the fuel cell air system is linearized using Taylor expansion, and the values ​​of the state variables under operating states A and B are substituted into the linearized model to obtain the model state equations under operating states A and B at power points.

[0124] The linearization method is as follows:

[0125]

[0126] The linearized state equations are shown below:

[0127]

[0128] in,

[0129]

[0130] Discretize the state equations.

[0131]

[0132] Based on the above model, the system prediction equation is derived as follows:

[0133]

[0134] In the formula Let p be the predicted step control output at time k, where p is the prediction time domain. Let M be the M-step control input column vector, where M is the control time domain.

[0135] , , , It is the prediction matrix:

[0136]

[0137]

[0138] , ,

[0139]

[0140] Given system constraints:

[0141]

[0142]

[0143]

[0144] In the formula, This refers to the control quantity, specifically the compressor voltage. This refers to the change in the control quantity. This refers to the output, or the oxygen excess coefficient.

[0145] Solve for the objective function:

[0146]

[0147] In the formula, and It is the weight matrix of the optimization problem. This is the reference output matrix:

[0148]

[0149]

[0150]

[0151] Using this method, a model predictive controller for the fuel cell system operating at power states A and B is derived.

[0152] Combine the control values ​​from the two controllers, such as Figure 1 As shown;

[0153] Specifically, the control quantities of the controller in power state A and the controller in power state B are multiplied by their respective weights, and then the results are added together to obtain the control quantity for the fusion control of the two controllers. The corresponding weights of the two controllers are shown in the figure.

[0154] Based on the fused control quantity, a supplementary control quantity is added for the operating condition of current step drop. The compensation control quantity is calculated using the following method:

[0155]

[0156] In the formula, in the formula This represents the current difference between the current at the current moment and the previous moment. This indicates the current value at the current moment. This represents the compressor control voltage at the previous moment. and It is a hyperparameter related to the current difference and the current value.

[0157] Specifically, the selection strategy for control quantities is as follows: Figure 2 As shown;

[0158] The model predictive control strategy described above enables the control of the compressor in the air supply system, thereby enabling the control of the air supply system for the proton exchange membrane fuel cell.

[0159] like Figure 3 As shown in the figure, this embodiment presents a comparison chart of oxygen excess ratio control for a 75kW fuel cell under step current conditions (the controller being compared is a traditional model predictive controller).

[0160] The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made within the scope of the claims of the present invention should be included in the scope of the present invention.

Claims

1. A control method for a fuel cell air supply system based on fusion model prediction, characterized in that, Includes the following steps: Step S1: Based on the characteristics of the proton exchange membrane fuel cell stack and air supply system, construct a high-order nonlinear model of the fuel cell stack and air supply system, and assume that the anode hydrogen supply meets the requirements and the humidifier is in an ideal state, simplifying the high-order nonlinear model into a sixth-order model. Step S2: Obtain the state parameter data of the fuel cell at power points A and B based on the high-order nonlinear model, and linearize the sixth-order model using Taylor expansion to obtain the linearized model; Step S3: Substitute the state parameter data of power points A and B into the linearized model, and convert the linearized model into the form of state-space equations; Step S4: Using the model prediction algorithm, design a controller for the linearized model at power points A and B, and give the controller constraints and reference output; Step S5: Give variable weights to the two model predictive controllers constructed in step S4 to realize the output coupling of the two controllers, and design a compensation quantity for the coupled control quantity to supplement the control quantity when the current steps down, so as to realize the compensation of the control quantity when the current steps down. Step S5 specifically involves: ① Under low power current conditions, the weight ratio of the controller designed with power point A data is 1, while the weight ratio of the controller designed with power point B data is 0. ② Under medium power current conditions, the weighting ratio of the controller designed using power point A data is a function that decreases as the current increases, while the weighting ratio of the controller designed using power point B data is a function that increases as the current increases; Specifically, the functional expression for the weight ratio of power points A and B is: ③ Under high power current conditions, the weight ratio of the controller designed based on power point A data is 0, while the weight ratio of the controller designed based on power point B data is 1. The final fusion control result of the two controllers is the sum of the control quantities of the two controllers multiplied by their corresponding weights; The compensation amount is specifically as follows: In the formula This represents the current difference between the current at the current moment and the previous moment. This indicates the current value at the current moment. This represents the compressor control voltage at the previous moment. It is a hyperparameter related to the current difference. It is a hyperparameter related to the current value.

2. The control method for a fuel cell air supply system based on fusion model prediction according to claim 1, characterized in that, The established high-order nonlinear model of the fuel cell stack and air supply system is as follows: state variables , specific It is the compressor angular velocity. It's the air pressure in the air supply manifold. It refers to the gas quality in the gas supply manifold. It refers to the oxygen quality inside the fuel cell. It refers to the mass of hydrogen inside the fuel cell. It refers to the mass of nitrogen gas inside the fuel cell. It is the air pressure in the return manifold. It is the mass of water vapor at the anode of the fuel cell. It is the mass of water vapor at the cathode of the fuel cell; , and These are the hydrogen input, output, and reaction consumption flow rates at the fuel cell anode; , , and These are the fuel cell anode input, output, water vapor passing through the proton exchange membrane, and output liquid water flow rates; , These are the nitrogen flow rates at the fuel cell anode input and output, respectively. , , , and These are the flow rates of water vapor entering, exiting, generating, and passing through the proton exchange membrane of the fuel cell cathode, as well as the output liquid water flow rate. , and These are the oxygen input, output, and reaction consumption flow rates at the fuel cell cathode; The temperature of the gas in the exhaust manifold. The gas constant of air. For exhaust manifold volume, This represents the mass flow rate of the gas output from the cathode. The mass flow rate output from the exhaust manifold; It is the output mass flow rate of the air compressor. This refers to the mass flow rate output from the intake manifold. For constant thermal pressure ratio, For the intake manifold volume, The output gas temperature of the air compressor. This refers to the intake manifold gas temperature. and These are the air compressor torque and load, respectively. It is the equivalent inertia of the air compressor.

3. The control method for a fuel cell air supply system based on fusion model prediction according to claim 2, characterized in that, The specific method for obtaining the state parameter data of a fuel cell at power points A and B based on a high-order nonlinear model is as follows: ① Establish a mathematical model of a high-order nonlinear model in the Matlab / simulink simulation environment, and the parameter state inside the model can be observed in real time; ② Input constant load currents corresponding to power points A and B to simulate the input load of the fuel cell system under steady-state conditions; ③ Observe and record the model's internal parameter data under the load current at power points A and B respectively.

4. The control method for a fuel cell air supply system based on fusion model prediction according to claim 1, characterized in that, The control input of the fuel cell air supply system is the control voltage of the air compressor, the measurable input disturbance is the current, and the output is the oxygen excess coefficient of the fuel cell stack. The input and output expressions are as follows: In the formula, This indicates the current in the fuel cell stack. This indicates the compressor control voltage. This indicates the oxygen intake flow rate of the fuel cell stack. This indicates the oxygen flow rate required for the fuel cell reaction.

5. The control method for a fuel cell air supply system based on fusion model prediction according to claim 2, characterized in that, The sixth-order model is specifically as follows: In the formula , , , , , For state variables, specifically It is the compressor angular velocity. It's the air pressure in the air supply manifold. It refers to the gas quality in the gas supply manifold. It refers to the oxygen quality inside the fuel cell. It refers to the mass of nitrogen gas inside the fuel cell. It is the air pressure in the return manifold; and These are the nitrogen flow rates at the fuel cell anode input and output, respectively. , and These are the oxygen input, output, and reaction consumption flow rates at the fuel cell cathode; The temperature of the gas in the exhaust manifold. The gas constant of air. For exhaust manifold volume, This represents the mass flow rate of the gas output from the cathode. The mass flow rate output from the exhaust manifold; It is the output mass flow rate of the air compressor. This refers to the mass flow rate output from the intake manifold. For constant thermal pressure ratio, For the intake manifold volume, The output gas temperature of the air compressor. This refers to the intake manifold gas temperature. , These are the air compressor torque and load. It is the equivalent inertia of the air compressor.

6. The control method for a fuel cell air supply system based on fusion model prediction according to claim 5, characterized in that, The method of using Taylor expansion to linearize the sixth-order model to obtain a linearized model is as follows: the simplified model of the fuel cell air system is linearized using Taylor expansion, and the values ​​of the state variables at power points A and B are substituted into the linearized model to obtain the model state equations at power points A and B. The linearization method is as follows: The linearized state equations are shown below: in In the formula, For state vectors, For control vectors, For measurable interference, Unmeasurable interference The state matrix, For the input matrix, For measurable interference matrix, The interference matrix is ​​unmeasurable. For the output matrix, For direct transition matrix, To output the interference matrix; This indicates the current in the fuel cell stack.

7. The control method for a fuel cell air supply system based on fusion model prediction according to claim 6, characterized in that, Step S4 specifically involves: Discretize the state equations. Based on the above model, the system prediction equation is derived as follows: In the formula Let p be the predicted step control output at time k, where p is the prediction time domain. Let M be the M-step control input column vector, where M is the control time domain; , , , These are all prediction matrices: , , Given system constraints: In the formula, This refers to the control quantity, specifically the compressor voltage. This refers to the change in the control quantity. This refers to the output, i.e., the oxygen excess coefficient; Solve for the objective function J: In the formula, and It is the weight matrix of the optimization problem. This is the reference output matrix: The model predictive controller of the fuel cell system is derived under the operating conditions of power points A and B.

Images