PEMFC-CHP system temperature control method and system

Abstract

The application discloses a kind of PEMFC-CHP system temperature control method and system, it is related to fuel cell combined heat and power technology field, comprising the following steps: the energy conservation model of each component is obtained based on the transmission process of energy in dynamic model;The nonlinear state space expression of PEMFC-CHP system is established based on the energy conservation model of each component, linearization and discretization are carried out to nonlinear state space expression, and discrete state space expression is obtained;MPC controller is designed based on discrete state space expression, and temperature control is carried out to PEMFC-CHP system by MPC controller.The controller of the application combines all information of system, can consider the mutual influence between different state variables, so its temperature control effect is far superior to PID widely used in actual engineering.

Description

Technical Field

[0001] This invention relates to the field of fuel cell combined heat and power technology, and in particular to a temperature control method and system for a PEMFC-CHP system. Background Technology

[0002] Proton exchange membrane fuel cell combined heat and power (PEMFC-CHP) systems can store waste heat generated during fuel cell stack operation in a hot water tank via a heat exchanger, while simultaneously providing heat and electricity to the outside world. Therefore, the overall system efficiency is over 90%, making it a very promising energy technology. The operating temperature of the PEMFC fuel cell stack is crucial to the system's efficient and stable operation, necessitating the introduction of appropriate temperature control algorithms.

[0003] For complex, nonlinear, strongly coupled, multi-input multi-output systems like PEMFC-CHP, the PID controllers widely used in practical engineering can only independently control a single input and its corresponding single output. However, the multi-input multi-output systems of PEMFC-CHP interact with each other, so existing PID control methods cannot fully utilize the various information of the system and are prone to problems such as long settling time and large overshoot, thus failing to achieve good temperature control results. Summary of the Invention

[0004] Based on the shortcomings of the existing technology, the present invention provides a temperature control method and system for PEMFC-CHP system, which solves the problems that the existing PID control method cannot fully utilize all the information of the system, and is prone to problems such as long adjustment time and large overshoot, and cannot achieve good temperature control effect.

[0005] The present invention adopts the following technical solution: In a first aspect, the present invention provides a temperature control method for a PEMFC-CHP system, comprising the following steps: A dynamic model of the PEMFC-CHP system is constructed, and an energy conservation model for each component is obtained based on the energy transfer process in the dynamic model. A nonlinear state-space expression for the PEMFC-CHP system is established based on the energy conservation model of each component. The multiple state variables of the nonlinear state-space expression include multiple output variables of the PEMFC-CHP system and other state variables that interact with it. The control variables are multiple input variables of the PEMFC-CHP system. The nonlinear state-space expression is linearized and discretized to obtain a discrete state-space expression; an MPC controller is designed based on the discrete state-space expression, and the temperature of the PEMFC-CHP system is controlled by the MPC controller.

[0006] Preferably, the PEMFC-CHP system includes a PEMFC battery stack, a water storage tank, a heat exchanger, and a hot water storage tank; multiple input variables include coolant flow rate and cooling water flow rate, multiple output variables include battery stack inlet coolant temperature and battery stack outlet coolant temperature; other state variables include water storage tank outlet coolant temperature, cooling water leaving the heat exchanger temperature, and cooling water entering the heat exchanger temperature.

[0007] Preferably, the nonlinear state-space expression is as follows: ; In the formula, For a set of state variables, This refers to the outlet coolant temperature of the battery stack. The outlet temperature of the coolant in the water tank. This refers to the temperature of the coolant at the battery stack inlet. The temperature at which the cooling water leaves the heat exchanger. The temperature at which cooling water enters the heat exchanger. , , , , and The derivative of the state variable. for, For battery stack quality, For the specific heat capacity of the battery stack, The specific heat capacity of the coolant. This refers to the coolant flow rate. The mass of coolant in the water tank. The mass of the coolant in the heat exchanger. The heat transfer coefficient of the heat exchanger surface. For heat exchange area, For the quality of cooling water in the heat exchanger, For cooling water flow rate, The specific heat capacity of cooling water. The quality of cooling water in the hot water storage tank.

[0008] Preferably, the discrete state-space expression is as follows: ; In the formula, The value of the state variable deviating from the equilibrium point. A and B The discrete state matrix and input matrix are given. k For a moment, To control the deviation of the action quantity from the equilibrium point.

[0009] Preferably, the temperature control of the PEMFC-CHP system via the MPC controller specifically includes the following steps: Obtain the current state of the PEMFC-CHP system, input the current state into the discrete state-space expression, and obtain the future state. N The system state trajectory of the step; Constructing an idea about the future based on system state trajectories N Step the objective function of the control action, solve the objective function, and obtain the optimal control sequence; Temperature control of the PEMFC-CHP system is achieved using the first control action in the optimal control sequence.

[0010] Preferably, the objective function is as follows: In the formula, J Let be the objective function. Q , R , F It is a symmetric positive definite weight matrix. From k Time backward i The predicted state, From k Time backward i The amount of control applied in the step T This is the transpose operator.

[0011] Preferably, an integral state is introduced into the MPC controller to construct a nonlinear state-space expression containing the integral state; the integral state includes the integral of the coolant temperature at the battery stack outlet and the integral of the coolant temperature at the battery stack inlet. The nonlinear state-space expression containing integral states is linearized and discretized. A robust MPC controller is designed based on the discretization results, and the system temperature is controlled by the robust MPC controller.

[0012] Secondly, the present invention provides a temperature control system for a PEMFC-CHP system, comprising: The building module is used to construct a dynamic model of the PEMFC-CHP system and obtain the energy conservation model of each component based on the energy transfer process in the dynamic model. A module is established to establish a nonlinear state-space expression for the PEMFC-CHP system based on the energy conservation model of each component. The multiple state variables of the nonlinear state-space expression include multiple output variables of the PEMFC-CHP system and other state variables that interact with it. The control variables are multiple input variables of the PEMFC-CHP system. The control module is used to linearize and discretize the nonlinear state-space expression to obtain the discrete state-space expression; based on the discrete state-space expression, an MPC controller is designed to control the temperature of the PEMFC-CHP system.

[0013] Compared with the prior art, the above-mentioned at least one technical solution adopted by the present invention can achieve the following beneficial effects: This invention first establishes a nonlinear state-space expression based on the energy conservation model of each component in the PEMFC-CHP system. Then, at each equilibrium operating point of the PEMFC-CHP system, the nonlinear state-space expression is linearized and discretized to obtain a discrete state-space expression, from which a linear MPC controller is designed. Because this controller incorporates all information from the system and can consider the interactions between different state variables, its temperature control performance is far superior to the PID controllers widely used in practical engineering. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the 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.

[0015] Figure 1 This is a structural diagram of the PEMFC-CHP system of the present invention; Figure 2 This is a diagram of the applied current disturbance in an embodiment of the present invention; Figure 3 This is a diagram showing the effect of outlet coolant temperature control in an embodiment of the present invention (PID vs. MPC). Figure 4 This is a diagram showing the effect of inlet coolant temperature control in an embodiment of the present invention (PID vs. MPC). Figure 5 This is a diagram illustrating the effect of outlet coolant temperature control in an embodiment of the present invention (MPC vs. Robust MPC). Figure 6 This is a diagram illustrating the effect of imported coolant temperature control in an embodiment of the present invention (MPC vs. Robust MPC). Figure 7 for Figure 5 Enlarged view of a section in the middle (1200-1800s); Figure 8 for Figure 6 Enlarged view of a section in the middle (1200-1800s); Figure 9This is a diagram showing the effect of outlet coolant temperature control under model mismatch conditions in an embodiment of the present invention (MPC vs. Robust MPC). Figure 10 This is a diagram showing the effect of inlet coolant temperature control under model mismatch conditions in this embodiment of the invention (MPC vs. Robust MPC). Figure 11 This is a flowchart of a temperature control method for a PEMFC-CHP system according to the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] This invention provides a temperature control method for a PEMFC-CHP system, specifically a robust MPC-based temperature control method for a PEMFC-CHP system. This invention employs an advanced model-based control method—Model Predictive Control (MPC)—to achieve high-precision temperature control. The core idea of ​​MPC can be summarized as "model prediction, rolling optimization, and feedback correction," that is, based on model prediction of future dynamics, the optimal control in the finite-time domain is solved online through rolling optimization, and feedback correction is used to update the prediction with the latest measured values ​​in the next cycle. Simultaneously, to address the problem of traditional MPC's over-reliance on model accuracy, this invention improves the robustness of traditional MPC by introducing integral states, enabling it to maintain good control performance even when the model is inaccurate. (Refer to...) Figure 11 This includes the following steps:

[0018] S1: Based on the physical characteristics of the battery stack, heat exchanger, and water tank in the PEMFC-CHP system, a dynamic model of the PEMFC combined heat and power system is established using the Matlab / Simulink simulation platform.

[0019] The PEMFC-CHP system architecture diagram is as follows: Figure 1 As shown, its thermal management system is mainly divided into an internal coolant circuit and an external cooling water circuit. In the internal coolant circuit, the coolant absorbs the waste heat generated by the PEMFC battery stack during operation through bipolar plates, then flows through a heat exchanger to release heat to the external cooling water. After cooling down, it flows back into the battery stack, thus achieving cyclic cooling. In the external cooling water circuit, the cooling water absorbs heat from the internal coolant through a heat exchanger, then flows into the hot water storage tank, thus transferring the heat generated by the battery stack to the hot water storage tank.

[0020] The PEMFC-CHP system mainly consists of a battery stack, a heat exchanger, a water storage tank, and a hot water storage tank. The PEMFC battery stack model is often built using semi-empirical equations.

[0021] ; ; ; ; ; ; ; In the formula, Represents the output voltage. Represents Nernst voltage. Represents activation loss, Represents ohmic loss, This represents concentration loss. There are 9 coefficients, which can be calibrated through experimental data, among which The activation loss coefficient, This is the no-load current. For external circuit resistance, The water content of the exchange membrane, This represents the limiting current density.

[0022] To reflect the dynamic temperature changes of various components in the PEMFC-CHP system, this invention establishes energy conservation models for each component by analyzing energy transfer, transmission, and dissipation. The energy conservation model for the battery stack is as follows:

[0023] ; In the formula, For battery stack quality, For the specific heat capacity of the battery stack, This is the outlet coolant temperature of the battery stack (assuming it is equal to the battery stack temperature). To generate heat for the battery stack, To remove heat from the coolant.

[0024] ; In the formula, Total energy of fuel Energy is carried away by the gas entering and exiting the battery stack. This refers to the radiative heat dissipation of the battery stack. This refers to the power of the battery stack.

[0025] ; In the formula, This refers to the coolant flow rate. The specific heat capacity of the coolant. This refers to the temperature of the coolant inlet to the battery stack.

[0026] The energy conservation model for the water storage tank is constructed using the following equations: ; In the formula, The mass of coolant in the water tank. This refers to the coolant temperature at the water tank outlet.

[0027] The energy conservation model of the heat exchanger is constructed through the following equations, which divide it into a hot water side and a cold water side. The liquid on the hot water side is the coolant after cooling the battery stack, and the liquid on the cold water side is the cooling water flowing in from the hot water storage tank: Hot water side: ; In the formula, The mass of the coolant in the heat exchanger. The heat exchanged between hot and cold water.

[0028] Cold water side: ; In the formula, For the quality of cooling water in the heat exchanger, The specific heat capacity of cooling water. For cooling water flow rate, The temperature at which cooling water enters the heat exchanger. This refers to the temperature at which the cooling water leaves the heat exchanger.

[0029] ; In the formula, The heat transfer coefficient of the heat exchanger surface. This refers to the heat exchange area.

[0030] Energy conservation model for hot water storage tanks: ; In the formula, For the quality of cooling water in the hot water storage tank, For the user's heat load.

[0031] S2: Based on the PEMFC-CHP system model established in S1, design a temperature control strategy and use PID to initially implement temperature control.

[0032] The temperature control system of the PEMFC-CHP system is a 2-input, 2-output system. The input variables are coolant flow rate and cooling water flow rate, and the output variables are inlet coolant temperature and outlet coolant temperature. Although there is interaction between the two inputs and two outputs, the inlet coolant temperature is mainly controlled by the cooling water flow rate, and the outlet coolant temperature is mainly controlled by the coolant flow rate. Therefore, this invention first uses two independent PID controllers to adjust the flow rates of coolant and cooling water to control the coolant outlet temperature and inlet temperature respectively.

[0033] PID controllers are the most widely used controllers in practical engineering. They have advantages such as good robustness, simple structure, and ease of implementation. Essentially, they are controllers that calculate the control action based on the error. ; In the formula, This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients. To control the amount of action, This represents the error between the given value and the actual value. In a PEMFC combined heat and power system, for a PID controller regulating the coolant flow rate, This refers to the coolant flow rate. The difference between the setpoint and actual outlet coolant temperature is used by a PID controller to regulate the coolant flow rate. For cooling water flow rate, This is the difference between the given value and the actual value of the inlet coolant temperature.

[0034] S3: Design an MPC controller based on model information to achieve temperature control of the PEMFC-CHP system and compare the temperature control effect with that of the PID controller in S2.

[0035] Because the PID controller in S2 generates control action only based on error and does not fully utilize model information, and because the two PID loops in the PEMFC-CHP system interact with each other, it is difficult to achieve good control results, often resulting in problems such as large overshoot and long settling time. To overcome the inherent defects of the PID controller, this invention first uses an MPC controller designed based on model information to achieve temperature control in the PEMFC-CHP system.

[0036] MPC controllers are multiple-input multiple-output (MIMO) controllers, capable of simultaneously controlling multiple controlled variables through multiple control actions. Therefore, they are more suitable for MIMO systems than PID controllers. In the PEMFC-CHP system, the MPC controller can simultaneously adjust the coolant flow rate and the cooling water flow rate to simultaneously control the coolant outlet temperature and its inlet temperature.

[0037] First, based on the equations of each component of the PEMFC-CHP system built in S2, its nonlinear state-space expression is derived as follows: ; In the formula, , representing state variables, The outlet coolant temperature of the battery stack , The outlet coolant temperature of the water tank , Temperature of the coolant at the battery stack inlet , Temperature of cooling water leaving the heat exchanger , Temperature of cooling water entering the heat exchanger , Represents the control action quantity. Coolant flow rate , Cooling water flow rate It can be seen that the PEMFC combined heat and power system has 5 state variables, which are coupled to each other. , , , , and This is the derivative corresponding to the state variable. If only two independent PID controllers are used to control... and It is impossible to take into account the interaction between various state variables, but the design of the MPC controller is based on state-space equations, so these model information can be utilized.

[0038] Since this invention uses linear MPC, it can only handle linear state-space equations. Therefore, it is necessary to linearize the above nonlinear state-space expression at the equilibrium operating point (the operating point where the time derivative of the state variables is 0), as shown below: ; In the formula, The state matrix, For the input matrix, The value of the state variable deviating from the equilibrium point. To control the deviation of the action from the equilibrium point, The state at the equilibrium point, The control factor at the equilibrium point; Linearized matrix and as follows: ; ; Since MPC can only handle discrete systems, it is necessary to discretize the locally linearized system. Equation (18) is then transformed into: ; In the formula, matrices A and B can be obtained through... Matrix and Transformation to obtain (zero-order hold): ; ; In the formula, T The sampling interval is... It is the integral variable.

[0039] The core idea of ​​MPC can be summarized as "model prediction, rolling optimization and feedback correction". That is, based on the model to predict future dynamics, the optimal control in the finite time domain is solved online through rolling optimization, and the prediction is updated with the latest measurement value in the next cycle using feedback correction. Model prediction: at each sampling time k Based on the current state and system model Predicting the future through recursion N The system state trajectory of step is obtained according to equation (19) as follows: ; ; ; The matrix can be written in a compact format as follows: ; in: ; ; ; ; Rolling optimization: Online solution to a finite-time optimal control problem to determine the control sequence, its core being to minimize the following objective function: ; In the formula, Q , R , F It is a symmetric positive definite weight matrix. From k Time backward i The predicted state, Fromk Time backward i The amount of control action applied step by step. Substituting into equation (25) and simplifying the formula, we get:

[0040] ; in: ; ; Solving the above optimization problem yields the optimal control sequence: , This is the optimal control sequence. This is the optimal control quantity.

[0041] But only the first control variable in the sequence It acts on the system. The cost function can be seen. J It is about The quadratic programming problem, MPC, is essentially an optimization problem, that is, finding the solution that makes the cost function... J Minimum control sequence This invention uses the interior point method for solving;

[0042] Feedback correction: to the next moment k +1, utilizing new actual measurements The initial state of the problem is updated and optimized, and the above prediction and optimization process is repeated to form a closed-loop control.

[0043] Because the controlled quantity is and Therefore, it is given a large weight. , , Since it does not require special attention, it is assigned a relatively small weight. The weight matrix selected in this invention is as follows: ; ; ; Apply to the system Figure 2 The current disturbance and temperature control effect shown are as follows: Figure 3 and Figure 4 As shown, the linear MPC outperforms the PID controller in both coolant inlet and outlet temperature control, with significantly shorter settling time and almost no overshoot. Table 1 shows that the integral absolute error (IAE) of the linear MPC is much smaller than that of the PID controller, almost an order of magnitude smaller, demonstrating the superiority of the advanced controller designed based on model information.

[0044] Table 1 Comparison of IAE S4: Introduce integral state correction to the MPC controller designed in S3 to improve its robustness.

[0045] To overcome the problem of traditional MPC in S3 relying too heavily on model accuracy, an integral state is introduced to form a robust MPC controller. When faced with model mismatch or external disturbances, the output of a system using traditional MPC may exhibit a steady-state error compared to the desired value. Robust MPC constructs a new state-space model by introducing an integral state, thus embedding integral action within the predictive control framework. This integral state can automatically accumulate historical errors, fundamentally forcing the system's steady-state error to zero, significantly improving the system's anti-interference capability and control accuracy.

[0046] After introducing an integral state in this invention, the state variables are: ,in ~ remain unchanged. For state The points, For state The integral, that is: ; ; After such processing, the matrix in equation (18) and The following corrections have been made: ; Obtain the corrected matrix and Then, the MPC controller designed according to step S3 is the robust MPC. To eliminate steady-state error, the integral state needs to be forced to approach zero; therefore, a state parameter needs to be assigned. and For larger weights, the present invention selects the following weight matrix:

[0047] ; ; ; Apply to the system Figure 2 The current disturbance and temperature control effect shown are as follows: Figure 5 and Figure 6As shown in Table 2, robust MPC exhibits slightly increased overshoot and settling time compared to traditional MPC. In terms of IAE (Inlet Coolant Equivalent) metrics, robust MPC and traditional MPC perform similarly, with robust MPC having a larger outlet coolant temperature IAE but a smaller inlet coolant temperature IAE. In conclusion, assuming accurate model performance, robust MPC slightly reduces control dynamics (overshoot and settling time) compared to traditional MPC.

[0048] Table 2 Comparison of IAE To compare the control steady-state performance (steady-state error) of robust MPC with that of traditional MPC, Figure 5 and Figure 6 A local magnified study was conducted. This invention uses the temperature control effect after the third current disturbance as an example (corresponding to the temperature control process from 1200-1800s), and its local magnified diagram is shown below. Figure 7 and Figure 8 As shown, robust MPC can completely eliminate steady-state error, while traditional MPC still has a certain degree of steady-state error. The control steady-state performance of robust MPC is far superior to that of traditional MPC. In summary, robust MPC sacrifices some control dynamic performance compared to traditional MPC, but greatly improves its control steady-state performance, and can even completely eliminate steady-state error.

[0049] Because the hot water storage tank in this invention cannot reach a balanced state, the average value during the control process can only be used to replace the parameter values ​​at the equilibrium point for controller design. Therefore, a slight model mismatch problem occurs during equilibrium point linearization, resulting in a small steady-state error in traditional MPC. Robust MPC, by introducing an integral state, can automatically accumulate historical errors and incorporate the integral action into the cost function. J In this process, by minimizing the cost function J This fundamentally forces the system's steady-state error to be zero.

[0050] During the actual operation of the PEMFC-CHP system, the user's heat load ( Random external disturbances can cause the temperature of the hot water storage tank to ( ) Fluctuations. If Drastic changes will affect the state A significant deviation from the equilibrium point leads to a severe model mismatch problem. To simulate this phenomenon, this invention directly modifies the state during the control process. Disturbance is applied, and the temperature control effect is as follows: Figure 9 and Figure 10As shown in Table 3, traditional MPC exhibits significant steady-state temperature error, while robust MPC can still achieve error-free control while maintaining good control performance. Table 3 shows that in cases of severe model mismatch, the IAE of robust MPC is much smaller than that of traditional MPC, and robust MPC can still maintain complete steady-state error even under these conditions. In summary, this invention introduces integral state correction to transform traditional MPC into robust MPC, eliminating steady-state error and enabling the PEMFC-CHP system to better cope with random external disturbances such as user heat load, thus better meeting practical engineering requirements.

[0051] Table 3 Comparison of IAE This invention provides a method for controlling a nonlinear system using a linear MPC. Linear state-space equations are obtained by linearizing the PEMFC-CHP system at each equilibrium operating point, and a linear MPC is designed based on this model information. Because this controller incorporates model information, it can take into account the mutual influence between different state variables, thus its temperature control performance is far superior to the PID controller widely used in practical engineering.

[0052] This invention provides a robust MPC control method. Traditional MPC relies too heavily on model accuracy and may exhibit steady-state errors when faced with model mismatch or external disturbances. Therefore, this invention introduces integral state correction to transform traditional MPC into robust MPC, eliminating steady-state errors and enabling the PEMFC-CHP system to better cope with random external disturbances such as user thermal loads.

[0053] The robust MPC proposed in this invention constructs a new state-space model by introducing integral states, thereby embedding integral action within the predictive control framework. These integral states can automatically accumulate historical errors, fundamentally forcing the system's steady-state error to zero. This significantly improves the system's anti-interference capability and control accuracy, enabling the PEMFC-CHP system to better cope with random external disturbances such as user thermal loads.

[0054] Based on the same concept, the present invention also provides a temperature control system for a PEMFC-CHP system, including a construction module, an establishment module, and a control module.

[0055] The building module is used to construct a dynamic model of the PEMFC-CHP system and obtain the energy conservation model of each component based on the energy transfer process in the dynamic model.

[0056] The module is used to establish the nonlinear state-space expression of the PEMFC-CHP system based on the energy conservation model of each component. The multiple state variables in the nonlinear state-space expression include multiple output variables of the PEMFC-CHP system and other state variables that interact with it. The control variables are multiple input variables of the PEMFC-CHP system.

[0057] The control module is used to linearize and discretize the nonlinear state-space expression to obtain the discrete state-space expression; based on the discrete state-space expression, an MPC controller is designed to control the temperature of the PEMFC-CHP system.

[0058] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0059] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A temperature control method for a PEMFC-CHP system, characterized in that, Includes the following steps: A dynamic model of the PEMFC-CHP system is constructed, and an energy conservation model for each component is obtained based on the energy transfer process in the dynamic model. A nonlinear state-space expression for the PEMFC-CHP system is established based on the energy conservation model of each component. The multiple state variables of the nonlinear state-space expression include multiple output variables of the PEMFC-CHP system and other state variables that interact with it. The control variables are multiple input variables of the PEMFC-CHP system. The nonlinear state-space expression is linearized and discretized to obtain a discrete state-space expression; an MPC controller is designed based on the discrete state-space expression, and the temperature of the PEMFC-CHP system is controlled by the MPC controller.

2. The temperature control method for a PEMFC-CHP system as described in claim 1, characterized in that, The PEMFC-CHP system includes a PEMFC battery stack, a water storage tank, a heat exchanger, and a hot water storage tank; multiple input variables include coolant flow rate and cooling water flow rate, multiple output variables include coolant temperature at the battery stack inlet and coolant temperature at the battery stack outlet; other state variables include coolant temperature at the water storage tank outlet, cooling water temperature at the heat exchanger exit, and cooling water temperature at the heat exchanger inlet.

3. The temperature control method for a PEMFC-CHP system as described in claim 2, characterized in that, The specific nonlinear state-space expression is as follows: ； In the formula, For a set of state variables, This refers to the outlet coolant temperature of the battery stack. The outlet temperature of the coolant in the water tank. This refers to the temperature of the coolant at the battery stack inlet. The temperature at which the cooling water leaves the heat exchanger. The temperature at which cooling water enters the heat exchanger. , , , , and The derivative of the state variable. for, For battery stack quality, For the specific heat capacity of the battery stack, The specific heat capacity of the coolant. This refers to the coolant flow rate. The mass of coolant in the water tank. The mass of the coolant in the heat exchanger. The heat transfer coefficient of the heat exchanger surface. For heat exchange area, For the quality of cooling water in the heat exchanger, For cooling water flow rate, The specific heat capacity of cooling water. The quality of cooling water in the hot water storage tank.

4. The temperature control method for a PEMFC-CHP system as described in claim 3, characterized in that, The discrete state-space expression is as follows: ； In the formula, The value of the state variable deviating from the equilibrium point. A and B The discrete state matrix and input matrix are given. k For a moment, To control the deviation of the action quantity from the equilibrium point.

5. The temperature control method for a PEMFC-CHP system as described in claim 1, characterized in that, The temperature control of the PEMFC-CHP system via the MPC controller specifically includes the following steps: Obtain the current state of the PEMFC-CHP system, input the current state into the discrete state-space expression, and obtain the future state. N The system state trajectory of the step; Constructing an idea about the future based on system state trajectories N Step the objective function of the control action, solve the objective function, and obtain the optimal control sequence; Temperature control of the PEMFC-CHP system is achieved using the first control action in the optimal control sequence.

6. The temperature control method for a PEMFC-CHP system as described in claim 5, characterized in that, The specific objective function is as follows: In the formula, J Let be the objective function. Q , R , F It is a symmetric positive definite weight matrix. From k Time backward i The predicted state, From k Time backward i The amount of control applied in the step T This is the transpose operator.

7. The temperature control method for a PEMFC-CHP system as described in claim 2, characterized in that, Also includes: Integral states are introduced into the MPC controller to construct a nonlinear state-space expression that includes integral states; The integral state includes the integral of the coolant temperature at the battery stack outlet and the integral of the coolant temperature at the battery stack inlet. The nonlinear state-space expression containing integral states is linearized and discretized. A robust MPC controller is designed based on the discretization results, and the system temperature is controlled by the robust MPC controller.

8. A temperature control system for a PEMFC-CHP system, characterized in that, include: The building module is used to construct a dynamic model of the PEMFC-CHP system and obtain the energy conservation model of each component based on the energy transfer process in the dynamic model. A module is established to establish a nonlinear state-space expression for the PEMFC-CHP system based on the energy conservation model of each component. The multiple state variables of the nonlinear state-space expression include multiple output variables of the PEMFC-CHP system and other state variables that interact with it. The control variables are multiple input variables of the PEMFC-CHP system. The control module is used to linearize and discretize the nonlinear state-space expression to obtain the discrete state-space expression; based on the discrete state-space expression, an MPC controller is designed to control the temperature of the PEMFC-CHP system.

Images