Modeling method and system of proton exchange membrane methanol reforming hydrogen production fuel cell

Abstract

The invention discloses a modeling method and system for a proton exchange membrane methanol reforming hydrogen production fuel cell. The method comprises the following steps: based on a front-end reformer, inputting a methanol-water solution, calculating a methanol conversion rate and a reactor outlet component composition ratio, and constructing a dynamic sub-model of a front-end methanol reforming device; inputting the high-purity reformed gas into the PEM fuel cell for power generation, and constructing a static polarization model and a dynamic response model of the rear-end proton exchange membrane fuel cell; thermal parameters are obtained through a particle swarm optimization algorithm, and a methanol reforming hydrogen production fuel cell model is constructed by combining a dynamic sub-model of a front methanol reforming device and a static polarization model and a dynamic response model of a rear-end proton exchange membrane fuel cell. According to the method, the modeling precision of the methanol reforming hydrogen production fuel cell model is improved. The modeling method and system for the proton exchange membrane methanol reforming hydrogen production fuel cell can be widely applied to the technical field of fuel cell modeling.

Description

Technical Field

[0001] This invention relates to the field of fuel cell modeling technology, and in particular to a modeling method and system for proton exchange membrane methanol reforming hydrogen fuel cells. Background Technology

[0002] Driven by global warming, green, low-carbon, and environmentally friendly practices have become the development direction for the energy industry. Renewable energy is continuously replacing fossil fuels, and its share in the industry is constantly increasing. Methanol, as a basic raw material for traditional industries and a recycled product of renewable energy (such as biomass energy), is an excellent secondary energy carrier. Hydrogen energy, as a clean and efficient zero-carbon energy source, can effectively promote industrial decarbonization when combined with renewable energy. Reforming reactions combine methanol and hydrogen energy, linking traditional energy with new energy, and are an effective way to optimize the energy industry structure.

[0003] In hydrogen utilization, PEM fuel cells offer advantages such as high efficiency, high power density, and flexible scalability. Combining methanol reformers with PEM fuel cells can address the challenges of hydrogen storage and transportation while utilizing inexpensive methanol production, thereby reducing hydrogen production costs. MRFCs can leverage existing oil and gas storage and transportation networks to efficiently construct distributed energy systems, playing a crucial role in microgrids and even energy islands.

[0004] Modeling of MRFC (Methanol Reforming Fuel Cell) systems is mainly divided into a front-end methanol reforming reaction model, a back-end proton exchange membrane fuel cell model, and a system-level model integrating coupled components. In the MRFC system, the front-end reforming, back-end electrochemical reactions and the thermo-electro-chemical effects of coupled components are interconnected and dynamically influence each other. Therefore, in order to establish a system model that can support simulation analysis, control design and commercial applications, it is necessary to achieve coupling optimization and integrated construction of each sub-model.

[0005] However, in terms of front-end methanol reforming reaction models, traditional research has developed various basic modeling schemes. Some studies have established classical kinetic models for the yield of major reactants under a wide range of operating conditions by analyzing the catalyst surface mechanism through fixed-bed differential reactors. Other studies have explored the generation pathways and suppression strategies of CO in methanol reforming for hydrogen production, or conducted simulation studies using two fuel supply structures, focusing on vehicle-mounted applications. However, existing models generally lack consideration for the heat transfer characteristics of the reformer and actual operating conditions, while reforming reactor models that rely on solar heating limit the integration of MRFC systems and are difficult to adapt to the needs of complex application scenarios.

[0006] Modeling research on back-end proton exchange membrane fuel cells (MRFCs) started relatively early. Traditional approaches have established classical mathematical models and multi-parameter sensitivity analysis methods to quantify the relative importance of each parameter. Subsequent studies have introduced bilayer charging effects, thermodynamic characteristics, and electrochemical impedance spectroscopy to construct equivalent circuit models adaptable to different current ranges, which have been validated on kilowatt-level fuel cell stacks. Some models have even described the static and dynamic behavior under high-frequency pulsed loads. However, existing models still have significant limitations: early models neglected the verification of dynamic response, some MRFC-oriented models did not correct for fuel cell polarization characteristics, and most models failed to adequately model and verify dynamic behavior, making it difficult to meet the accurate simulation requirements under cold start and load fluctuation conditions.

[0007] Regarding system-level models integrating coupled components, traditional research has preliminarily established basic thermal models with oil-based thermal loops, focusing on the thermal energy flow of each independent component. Some zero-dimensional models have confirmed the feasibility of thermal self-sufficiency under catalytic combustion conditions, and some studies have improved system efficiency and shortened start-up time through waste heat recovery design. However, existing system-level models have significant shortcomings: they lack comprehensive consideration of reformer conversion efficiency, detailed thermal analysis, fuel cell VI characteristics and dynamic behavior, and do not adequately characterize system-level thermal coupling effects and cold start processes. Furthermore, they rely heavily on pure mechanism models, making it difficult to capture the nonlinear characteristics of the system and support accurate simulation and control design under all operating conditions. Summary of the Invention

[0008] To address the aforementioned technical problems, the present invention aims to provide a modeling method and system for proton exchange membrane methanol reforming hydrogen fuel cells. This method and system can analyze the front-end reformer reaction model and the back-end static polarization characteristics, as well as the thermal coupling effect between the front and back ends, thereby improving the modeling accuracy of the methanol reforming hydrogen fuel cell model.

[0009] The first technical solution adopted in this invention is: a modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell, comprising the following steps:

[0010] Based on the front-end reformer, inputting a methanol-water solution, calculating the methanol conversion rate and reactor outlet component composition ratio, and constructing a dynamic sub-model of the pre-methanol reforming unit;

[0011] High-purity reformed gas is input into a PEM fuel cell to generate electricity, and a static polarization model and a dynamic response model of the back-end proton exchange membrane fuel cell are constructed.

[0012] Thermodynamic parameters were obtained through particle swarm optimization algorithm, and a methanol reforming hydrogen fuel cell model was constructed by combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell.

[0013] Furthermore, the step of constructing a dynamic sub-model of the pre-processing methanol reforming unit based on the input methanol-water solution from the front-end reformer, calculating the methanol conversion rate and the reactor outlet component composition ratio, specifically includes:

[0014] The methanol-water solution is heated and vaporized to serve as the input gas for the front-end reformer.

[0015] Based on the front-end reformer, the methanol reforming reaction rate, methanol cracking reaction rate and reverse water-gas shift reaction rate are obtained, and a kinetic model of the front-end reformer is constructed.

[0016] The dynamic model of the front-end reformer is described by differential equations, and the initial conditions are set for iterative solution to obtain the reformed gas and the proportion of each component at the reactor outlet.

[0017] The reformed gas is purified by a palladium membrane purifier to obtain high-purity reformed gas.

[0018] The methanol conversion rate is calculated based on the amount of methanol at the inlet and outlet. Combined with the proportion of each component at the reactor outlet, high-purity reformed gas is input into the PEM fuel cell to generate electricity, and a dynamic sub-model of the pre-methanol reforming unit is constructed.

[0019] Furthermore, the specific expression for calculating the methanol conversion rate is as follows: ;

[0020] In the above formula, Indicates methanol conversion rate. This indicates the amount of methanol at the input of the front-end reformer. This indicates the amount of methanol at the output of the front-end reformer.

[0021] Furthermore, the specific formula for calculating the proportions of each component at the reactor outlet is as follows: ; In the above formula, This indicates the proportion of each component at the reactor outlet. Indicates the components in the reformed gas. This indicates the amount of methanol at the output of the front-end reformer. This indicates the amount of water vapor at the output of the front-end reformer. This indicates the amount of hydrogen at the output of the front-end reformer. This indicates the amount of carbon dioxide at the output of the front-end reformer. This indicates the amount of carbon monoxide at the output of the front-end reformer. This indicates the flow rate of various substances at the output end.

[0022] Furthermore, the step of inputting high-purity reformed gas into the PEM fuel cell to generate electricity and constructing the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell specifically includes:

[0023] High-purity reformed gas is input into a PEM fuel cell to generate electricity. The total voltage of the downstream proton exchange membrane fuel cell is obtained, and a static polarization model of the downstream proton exchange membrane fuel cell is constructed. The total voltage includes thermodynamic voltage, activation loss, ohmic loss, and concentration loss.

[0024] Using the Voigt structure, the resistance values ​​corresponding to ohmic, activation, and concentration overpotentials were obtained, and the double-layer capacitance was calculated.

[0025] The voltage response curve was obtained by step current testing, and the system time constant was determined by fitting it with a second-order exponential model.

[0026] By combining the system time constant and the double-layer capacitance value, a dynamic response model of the back-end proton exchange membrane fuel cell is constructed.

[0027] Furthermore, the expression for the total voltage of the back-end proton exchange membrane fuel cell is as follows: ; In the above formula, This indicates the total voltage of the downstream proton exchange membrane fuel cell. Represents thermodynamic voltage. Indicates activation loss. Indicates ohmic loss. This indicates concentration loss.

[0028] Furthermore, the step of acquiring thermodynamic parameters through particle swarm optimization and constructing a methanol reforming hydrogen fuel cell model by combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell specifically includes:

[0029] Dynamic thermodynamic models of the upstream methanol reforming unit and the downstream proton exchange membrane fuel cell were constructed using ordinary differential equations.

[0030] The dynamic thermodynamic models of the pre-methanol reforming unit and the post-proton exchange membrane fuel cell are identified using the particle swarm optimization algorithm to obtain the thermal parameters.

[0031] A data-driven thermal model is constructed based on thermodynamic parameters. By combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell, a methanol reforming hydrogen production fuel cell model is constructed.

[0032] Furthermore, the specific expression of the dynamic thermodynamic model of the pre-methanol reforming unit is as follows: ; In the above formula, Indicates the heat capacity of the reforming unit. This represents the total heat generated by the reforming reaction. Indicates surface radiative heat. This represents the heat exchange coefficient between the reforming unit and the environment. This represents the surface area of ​​the reforming unit. This indicates the temperature difference between the reforming unit and the environment. This indicates the measurement of the thermal effect of a fuel cell. This represents the heat transfer coefficient between the reforming unit and the fuel cell. This indicates the temperature difference between the reforming unit and the fuel cell. This indicates the amount of heat generated by the heating element.

[0033] Furthermore, the specific expression of the dynamic thermodynamic model of the back-end proton exchange membrane fuel cell is as follows: ; In the above formula, This indicates the number of cells in the PEM fuel cell stack. Indicates surface radiative heat. This indicates the measurement of the thermal effect of a fuel cell. Indicates activation loss. Indicates ohmic loss. Indicates concentration loss. Indicates surface radiative heat. Indicates the heat capacity of a fuel cell. Indicates the temperature of the fuel cell. This indicates the heat generated by fuel cell losses. This indicates the output current of the fuel cell. This indicates the temperature difference between the fuel cell and the environment. This indicates the temperature difference between the fuel cell and the front-end reformer.

[0034] The second technical solution adopted in this invention is: a modeling system for proton exchange membrane methanol reforming hydrogen fuel cells, comprising:

[0035] The first module is used to construct a dynamic sub-model of the pre-methanol reforming unit based on the input of methanol-water solution from the front-end reformer, calculate the methanol conversion rate and the composition ratio of reactor outlet components, and build a dynamic sub-model of the pre-methanol reforming unit.

[0036] The second module is used to input high-purity reformed gas into the PEM fuel cell to generate electricity, and to construct the static polarization model and dynamic response model of the back-end proton exchange membrane fuel cell.

[0037] The third module is used to obtain thermodynamic parameters through particle swarm optimization algorithm, and to construct a methanol reforming hydrogen fuel cell model by combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell.

[0038] The beneficial effects of the method and system of this invention are as follows: This invention constructs a dynamic sub-model of the front-end methanol reforming unit by inputting a methanol-water solution based on a front-end reformer, calculating the methanol conversion rate and the composition ratio of the reactor outlet components, thereby improving the accuracy of the front-end reformer modeling and accurately predicting the methanol conversion rate and the proportion of product components. Then, high-purity reformed gas is input into a PEM fuel cell to generate electricity, constructing a static polarization model and a dynamic response model for the back-end proton exchange membrane fuel cell, improving the back-end PEM fuel cell modeling, and comprehensively covering static and dynamic electrochemical characteristics and temperature adaptability. Finally, thermodynamic parameters are obtained through particle swarm optimization algorithm, and combined with the dynamic sub-model of the front-end methanol reforming unit and the static polarization model and dynamic response model of the back-end proton exchange membrane fuel cell, a methanol reforming hydrogen production fuel cell model is constructed, clarifying the front-end and back-end thermal coupling effect and achieving an accurate description of the entire process thermal characteristics from cold start to steady-state operation. Attached Figure Description

[0039] Figure 1 This is a flowchart of the steps in the modeling method of a proton exchange membrane methanol reforming hydrogen fuel cell of the present invention;

[0040] Figure 2 This is a structural block diagram of a modeling system for a proton exchange membrane methanol reforming hydrogen fuel cell according to the present invention.

[0041] Figure 3 This is a schematic diagram of the integrated MRFC system provided in a specific embodiment of the present invention;

[0042] Figure 4 This is a schematic diagram of the integrated MRFC system provided in a specific embodiment of the present invention;

[0043] Figure 5 This is a schematic diagram of the modeling workflow provided in a specific embodiment of the present invention;

[0044] Figure 6 This is a schematic diagram of the front-end reforming reaction process provided in a specific embodiment of the present invention;

[0045] Figure 7 This is a schematic diagram of the equivalent circuit model of the PEMFC model provided in a specific embodiment of the present invention;

[0046] Figure 8 This is a schematic diagram of the step response results of the PEMFC model provided in a specific embodiment of the present invention;

[0047] Figure 9 This is a schematic diagram of the experimental setup for a methanol reforming hydrogen fuel cell system provided in a specific embodiment of the present invention. Detailed Implementation

[0048] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. The step numbers in the following embodiments are only for ease of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.

[0049] First, it needs to be explained that, as Figure 3 The diagram shows a structural schematic of an MRFC system, which can be divided into two main parts: a front-end reformer and a back-end fuel cell. Modeling work for MRFC systems is divided into three aspects: the front-end, the back-end, and the integrated system of their coupling. However, current modeling research on MRFCs has many shortcomings: the thermal analysis of the front-end reformer is not accurate enough, the dynamic / static electrochemical characteristics of the back-end fuel cell are not comprehensively considered, and there is a lack of overall modeling of the coupled operating characteristics of the two systems.

[0050] Based on this, the embodiments of the present invention first construct a basic MRFC mechanism model starting from the methanol conversion rate at the reformer inlet, the composition of the product at the outlet, and the static volt-ampere characteristics of the fuel cell; then, a kilowatt-level experimental platform is built to collect operating data under all operating conditions, including start-up and loading step, and a data-driven method is used to model the dynamic response of the fuel cell, and the thermal parameters and thermal coupling effects during the start-up stage are calibrated using the particle swarm optimization algorithm; finally, the mechanism model is combined with data-driven analysis to establish a complete MRFC system dynamic model that includes the mechanism analysis of the reformer ordinary differential equation, the modified static-dynamic electrochemical model of the fuel cell, and the dynamic thermal model considering cold start and cross-system coupling.

[0051] It should also be noted that the modeling data for the hybrid MRFC model in this embodiment of the invention mainly comes from a kilowatt-level MRFC experimental platform. Data from the entire operating condition, including system startup and loading step, is collected, covering key operating parameters such as temperature, solution flow rate, and pressure, providing solid data support for model construction and parameter optimization. Based on a hybrid architecture of "mechanism modeling + data-driven," a comprehensive MRFC model is constructed by fusing the system's physicochemical laws with experimental data. The model is divided into a core modeling part based on mechanisms and a data-driven parameter optimization part. These two parts are coupled through key parameters such as methanol conversion rate, hydrogen supply rate, and equivalent capacitance. The internal structure and coupling relationship of the model are as follows: Figure 4 As shown, the modeling process is as follows: Figure 5 As shown.

[0052] Reference Figure 1This invention provides a modeling method for proton exchange membrane methanol reforming hydrogen fuel cells, the method comprising the following steps:

[0053] S100. Based on the front-end reformer, input methanol-water solution, calculate methanol conversion rate and reactor outlet component composition ratio, and construct dynamic sub-model of the front-end methanol reforming unit.

[0054] Specifically, a methanol-water solution is heated and vaporized to serve as the input gas for the front-end reformer. Based on the front-end reformer, the methanol reforming reaction rate, methanol cracking reaction rate, and reverse water-gas shift reaction rate are obtained to construct a dynamic model of the front-end reformer. The dynamic model of the front-end reformer is described by differential equations, and initial conditions are set for iterative solution to obtain the reformed gas and the proportions of each component at the reactor outlet. The reformed gas is purified by a palladium membrane purifier to obtain high-purity reformed gas. The methanol conversion rate is calculated based on the amount of methanol at the inlet and outlet. Combined with the proportions of each component at the reactor outlet, the high-purity reformed gas is input into the PEM fuel cell to generate electricity, thus constructing a dynamic sub-model of the front-end methanol reforming unit.

[0055] First, it should be noted that the core modeling based on the mechanism focuses on the intrinsic reactions and characteristics of the system: the front-end methanol reforming reaction is mainly methanol steam reforming (MSR), and the corresponding reaction equation is: ; It also includes two types of side reactions: high-temperature methanol decomposition (MD) and water-gas shift (WGS), and their reaction equations are as follows: ; ; Methanol conversion rate serves as a core indicator for measuring the efficiency of the front-end reforming process. High-purity hydrogen is obtained by removing reaction byproducts through a palladium purifier, and the correlation between the hydrogen supply rate and the output current of the back-end PEM fuel cell constitutes a core coupling effect between the front and back ends. The back-end PEM fuel cell modeling focuses on polarization characteristics and incorporates double-layer capacitance parameters to accurately reflect the cell's dynamic response performance.

[0056] The data-driven parameter optimization section is supported by experimental data and specifically addresses the parameter estimation challenges in mechanism modeling: it employs particle swarm optimization to accurately identify thermally coupled parameters, thereby improving the accuracy of system thermal characteristic modeling; and it estimates the equivalent capacitance of the PEM fuel cell using a time constant fitting method, thereby optimizing the accuracy of dynamic performance characterization.

[0057] The coupling mechanism of the model is reflected in the following aspects: the mechanism part clarifies the physical and chemical correlation and core coupling path of each component of the system, and the data-driven part optimizes key parameters through experimental data to make up for the parameter uncertainty defects of the pure mechanism model. The synergistic effect of the two not only ensures the physical rationality of the model, but also improves the dynamic response and steady-state accuracy, providing reliable support for the control design and operating condition simulation of the MRFC system.

[0058] In this embodiment, the methanol-water solution is heated and vaporized before entering the reformer as input gas. The reformed gas is then purified by a palladium membrane purifier to remove toxic CO that would otherwise be harmful to the downstream fuel cell. The resulting high-purity hydrogen is used as a reactant and fed into the PEM fuel cell to generate electricity. The kinetic model is based on the calculated MSR. MD WGS The reaction rate is constructed, and its expression is as follows: ; ; ; In the above formula, This is the constant reaction rate of MSR. It is the equilibrium constant MSR. It is the total surface concentration of center 1. It is specific surface area. R is the catalyst density, and R is the gas constant. It is the adsorption equilibrium constant.

[0059] in: ; In the above formula, and These are the partial pressure and molar flow rate of component I, while and These are the total pressure and total flow rate. The reaction rate can be obtained from a kinetic model based on partial pressures and reference parameters. From the reactor inlet to the outlet, the partial pressures of each component change continuously as the reforming reaction proceeds. Methanol is gradually converted to hydrogen, while producing byproducts such as carbon monoxide. The kinetic model of the reaction rate shows that temperature has a significant impact on the conversion rate. The reforming process can be described by differential equations: ; in It is the length of the feed inlet of the reforming unit. It is the cross-sectional area.

[0060] Based on the initial conditions, an iterative solution can be used to obtain the proportions of each component at the reactor outlet. For ease of calculation, the initial conditions... The following settings are available: ; like Figure 6 As shown, the concentration of the solution is measured at the input port. and The partial pressure. The kinetic model was validated mainly from two aspects: methanol conversion rate and reactor outlet component composition ratio.

[0061] The methanol conversion rate is calculated based on the amount of methanol at the import and export points. ; The composition ratio of each component in the product can be calculated as follows: ; In the above formula, Represents the components in the reformed gas, including , , , and .

[0062] S200, High-purity reformed gas is input into the PEM fuel cell to generate electricity, and a static polarization model and dynamic response model of the back-end proton exchange membrane fuel cell are constructed.

[0063] Specifically, high-purity reformed gas is input into a PEM fuel cell to generate electricity. The total voltage of the downstream proton exchange membrane fuel cell is obtained, and a static polarization model of the downstream proton exchange membrane fuel cell is constructed. The total voltage includes thermodynamic voltage, activation loss, ohmic loss, and concentration loss. Using a Voigt structure, the resistance values ​​corresponding to ohmic, activation, and concentration overpotentials are obtained, and the double-layer capacitance value is calculated. The voltage response curve is obtained through step current testing, and a second-order exponential model is used for fitting to determine the system time constant. Combining the system time constant and the double-layer capacitance value, a dynamic response model of the downstream proton exchange membrane fuel cell is constructed.

[0064] In this embodiment, the methanol-water solution is converted into reformed gas after reaction in the front-end reformer. To prevent toxic CO from affecting the fuel cell, a palladium membrane purifier is used for purification. The obtained high-purity hydrogen is fed into the PEM fuel cell as a reactant. A 2.5 kW proton exchange membrane fuel cell is integrated on the experimental platform. The fuel cell is modeled from two aspects: static polarization characteristics and dynamic response during operation. The fuel cell voltage decreases with increasing current density. The total voltage can be calculated using the following formula: ; in , , and These represent thermodynamic voltage, activation loss, ohmic loss, and concentration loss, respectively.

[0065] The theoretical maximum voltage can be given by the following formula: ; In the formula For standard state reversible voltage, Let be the ideal gas constant. For battery temperature, It is Faraday's constant. , and These are the partial pressures of hydrogen, oxygen, and water, respectively.

[0066] The formation of this substance is due to the energy barrier related to electron transfer in the electrochemical reaction at the electrode surface. Its formation process is as follows: ; in It is the operating current density. It is the exchange current density. It is the charge of the ion.

[0067] This is caused by the resistance within the fuel cell assembly (including the membrane, electrodes, and current collector). It can be calculated as: ; in The resistivity of the membrane is given by the empirical formula shown below: ; When the reactant concentration decreases or mass transfer limitation reduces the reaction rate at the electrode surface, it will produce Based on experimental data, the battery voltage drops rapidly before the current density reaches its upper limit. To improve the accuracy of the static polarization model, this paper employs the current density limit method. Therefore, It can be represented in the following form: ; in This is the maximum current density. For different operating conditions, The current density is set when the battery voltage drops to zero.

[0068] Due to the charging effect of the double layer at the electrode-electrolyte interface, the voltage change in a PEM fuel cell lags behind the current change. In other words, when faced with a step current, the PEM fuel cell behaves similarly to a capacitor. Therefore, the dynamic response of a proton exchange membrane fuel cell can be simulated by estimating the double-layer capacitance. The dynamic equivalent circuit is as follows: Figure 7 As shown. This embodiment uses the Voigt structure. The ohmic overpotential, activation overpotential, and concentration overpotential correspond to R1, R2, and R3, respectively. The double-layer capacitance value was determined using the time constant method.

[0069] For a steady-state PEM fuel cell, a step current is applied to test its voltage response. The voltage response can be expressed as: ; Since the initial state and the final state can be calculated and Therefore, only the dynamic response of the dual RC circuit needs to be considered. Thus, the dynamic response can be viewed as a second-order inertial element. Then, by fitting the experimental data using a second-order exponential model, the time constant can be generated.

[0070] according to The capacitance can be obtained by dividing a constant by the equivalent resistance. This capacitance is then applied to the dynamic equivalent circuit. The fitted data and simulation results are as follows: Figure 8 As shown.

[0071] S300: Thermodynamic parameters are obtained through particle swarm optimization algorithm, and a methanol reforming hydrogen fuel cell model is constructed by combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell.

[0072] Specifically, a dynamic thermodynamic model of the upstream methanol reforming unit and the downstream proton exchange membrane fuel cell are constructed using ordinary differential equations. Thermal parameters are identified in both models using particle swarm optimization. A data-driven thermal model is then constructed based on these parameters. Finally, a methanol reforming hydrogen fuel cell model is built by combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization and dynamic response models of the downstream proton exchange membrane fuel cell.

[0073] In this embodiment, temperature plays a crucial role in the MRFC system. However, thermal parameters, including heat capacity and heat transfer coefficient, are difficult to identify. Furthermore, the thermal coupling effect between the front-end reformer and the back-end proton exchange membrane fuel cell is difficult to estimate. Particle swarm optimization (PSO) is widely used for parameter estimation due to its simplicity, efficiency, and low computational cost. Therefore, a data-driven approach was adopted, utilizing PSO.

[0074] Methanol conversion rate and the composition of multiple components in reformed gas are significantly affected by operating temperature. Since ordinary differential equations are the mainstream method for thermal modeling, a dynamic thermodynamic model can be established: ; in, Indicates the heat capacity of the reforming unit. This represents the total heat generated by the reforming reaction. Indicates surface radiative heat. This represents the heat exchange coefficient between the reforming unit and the environment. This represents the surface area of ​​the reforming unit. This indicates the temperature difference between the reforming unit and the environment. Used to measure the thermal effect of fuel cells This represents the heat transfer coefficient between the reforming unit and the fuel cell. This indicates the temperature difference between the reforming unit and the fuel cell. This indicates the amount of heat generated by the heating element.

[0075] The performance of PEM fuel cells is also affected by operating temperature. The dynamic thermal model of a proton exchange membrane fuel cell can be expressed as: ; In cases where the parameter definitions in this model are similar, This indicates the number of cells in the PEM fuel cell stack. Since the front-end reformer and the back-end fuel cell are connected via a hydrogen supply line, the thermal coupling effect can be treated as a separate heat source.

[0076] To address the difficulty in measuring thermal parameters, a data-driven approach was used to establish a more realistic thermal model. The Particle Swarm Optimization (PSO) algorithm, with its simplicity, efficiency, and avoidance of local minima, is widely used in solving complex optimization problems. Therefore, PSO is employed for thermal parameter estimation, as shown in the pseudocode. The velocity update equations and position update equations are as follows: ; in It is a particle In the steps speed, This is its current location. It is its optimal position. It is the globally optimal position. It is inertial weight. , It is a learning factor. , It is a random factor.

[0077] In this work, each particle represents a candidate set of parameters: Used for front-end MR, similarly for proton exchange membrane fuel cells. Fitness function evaluation of experimental temperature profiles. The simulated temperature obtained by solving for concentration loss and voltage response ODE The root mean square error between them is shown in Table 1.

[0078] Table 1 Estimation of thermal parameters ; For methanol reformers and proton exchange membrane fuel cells, heat capacity and heat transfer coefficient are parameters that need to be identified. During the experiment, the temperatures at the front and rear ends were measured separately. Therefore, particle swarm optimization was performed separately. To ensure the accuracy and reliability of the simulation, the initial values ​​of the thermal parameters were set based on preliminary experimental measurements. The optimization results of each iteration were fed back to the MATLAB-Simulink simulation platform for verification. After particle swarm optimization, the thermodynamic parameters that best approximate the experimental values ​​were finally obtained.

[0079] In summary, this invention constructed a 2.5kW MRFC integrated platform, collected data, and then built a data-driven-mechanism hybrid model. Mechanistically, the front-end reformer reaction model and back-end static polarization characteristics were analyzed. Data-drivenly, front-end thermal parameters were identified based on PSO, and back-end dynamic response modeling was performed, analyzing the thermal coupling effect between the front and back ends. Finally, the modeling results were compared with the operating data of the experimental platform, and the results verified the effectiveness of the model.

[0080] To perform general validation of the integrated MRFC model, a full-scale comprehensive experiment was conducted. For example... Figure 9 As shown, the 2.5kW MRFC unit consists of three parts: a reforming unit, a PEM fuel cell, and a control unit. Figure 9 This is a schematic diagram of an MRFC system, which can be divided into two main parts: the front-end reformer and the back-end fuel cell. The modeling work for the MRFC system is divided into three aspects: the front-end, the back-end, and the integrated system where the two are coupled.

[0081] Finally, the embodiments of the present invention have the following advantages compared with the prior art:

[0082] 1) Improve the accuracy of front-end reformer modeling and accurately predict methanol conversion rate and product component ratio.

[0083] 2) Improve the back-end PEM fuel cell modeling to fully cover static and dynamic electrochemical characteristics and temperature adaptability.

[0084] 3) Clearly define the front-end and back-end thermal coupling effects to achieve an accurate description of the thermal characteristics of the entire process from cold start to steady-state operation.

[0085] 4) Reduces modeling errors and achieves better accuracy than existing methods in key indicators such as conversion rate, polarization characteristics, thermal characteristics and output power.

[0086] Reference Figure 2A modeling system for a proton exchange membrane methanol reforming hydrogen fuel cell includes:

[0087] The first module 201 is used to construct a dynamic sub-model of the pre-methanol reforming unit based on the input of methanol-water solution from the front-end reformer, calculate the methanol conversion rate and the composition ratio of reactor outlet components, and build a dynamic sub-model of the pre-methanol reforming unit.

[0088] The second module 202 is used to input high-purity reformed gas into the PEM fuel cell to generate electricity and to construct the static polarization model and dynamic response model of the back-end proton exchange membrane fuel cell.

[0089] The third module 203 is used to obtain thermodynamic parameters through particle swarm optimization algorithm, and to construct a methanol reforming hydrogen fuel cell model by combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell.

[0090] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0091] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell, characterized in that, Includes the following steps: Based on the front-end reformer, inputting a methanol-water solution, calculating the methanol conversion rate and reactor outlet component composition ratio, and constructing a dynamic sub-model of the pre-methanol reforming unit; High-purity reformed gas is input into a PEM fuel cell to generate electricity, and a static polarization model and a dynamic response model of the back-end proton exchange membrane fuel cell are constructed. Thermodynamic parameters were obtained through particle swarm optimization algorithm, and a methanol reforming hydrogen fuel cell model was constructed by combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell.

2. The modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell according to claim 1, characterized in that, The step of constructing a dynamic sub-model of the pre-processing methanol reforming unit by inputting a methanol-water solution from the front-end reformer, calculating the methanol conversion rate and the reactor outlet component composition ratio, specifically includes: The methanol-water solution is heated and vaporized to serve as the input gas for the front-end reformer. Based on the front-end reformer, the methanol reforming reaction rate, methanol cracking reaction rate and reverse water-gas shift reaction rate are obtained, and a kinetic model of the front-end reformer is constructed. The dynamic model of the front-end reformer is described by differential equations, and the initial conditions are set for iterative solution to obtain the reformed gas and the proportion of each component at the reactor outlet. The reformed gas is purified by a palladium membrane purifier to obtain high-purity reformed gas. The methanol conversion rate is calculated based on the amount of methanol at the inlet and outlet. Combined with the proportion of each component at the reactor outlet, high-purity reformed gas is input into the PEM fuel cell to generate electricity, and a dynamic sub-model of the pre-methanol reforming unit is constructed.

3. The modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell according to claim 2, characterized in that, The specific formula for calculating the methanol conversion rate is as follows: ； In the above formula, Indicates methanol conversion rate. This indicates the amount of methanol at the input of the front-end reformer. This indicates the amount of methanol at the output of the front-end reformer.

4. The modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell according to claim 3, characterized in that, The specific formula for calculating the proportions of each component at the reactor outlet is as follows: ； In the above formula, This indicates the proportion of each component at the reactor outlet. Indicates the components in the reformed gas. This indicates the amount of methanol at the output of the front-end reformer. This indicates the amount of water vapor at the output of the front-end reformer. This indicates the amount of hydrogen at the output of the front-end reformer. This indicates the amount of carbon dioxide at the output of the front-end reformer. This indicates the amount of carbon monoxide at the output of the front-end reformer. This indicates the flow rate of various substances at the output end.

5. The modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell according to claim 4, characterized in that, The step of inputting high-purity reformed gas into a PEM fuel cell to generate electricity and constructing a static polarization model and a dynamic response model for the downstream proton exchange membrane fuel cell specifically includes: High-purity reformed gas is input into a PEM fuel cell to generate electricity. The total voltage of the downstream proton exchange membrane fuel cell is obtained, and a static polarization model of the downstream proton exchange membrane fuel cell is constructed. The total voltage includes thermodynamic voltage, activation loss, ohmic loss, and concentration loss. Using the Voigt structure, the resistance values ​​corresponding to ohmic, activation, and concentration overpotentials were obtained, and the double-layer capacitance was calculated. The voltage response curve was obtained by step current testing, and the system time constant was determined by fitting it with a second-order exponential model. By combining the system time constant and the double-layer capacitance value, a dynamic response model of the back-end proton exchange membrane fuel cell is constructed.

6. The modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell according to claim 5, characterized in that, The expression for the total voltage of the back-end proton exchange membrane fuel cell is as follows: ； In the above formula, This indicates the total voltage of the downstream proton exchange membrane fuel cell. Represents thermodynamic voltage. Indicates activation loss. Indicates ohmic loss. This indicates concentration loss.

7. The modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell according to claim 6, characterized in that, The step of acquiring thermodynamic parameters through particle swarm optimization and constructing a methanol reforming hydrogen fuel cell model by combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell specifically includes: Dynamic thermodynamic models of the upstream methanol reforming unit and the downstream proton exchange membrane fuel cell were constructed using ordinary differential equations. The dynamic thermodynamic models of the pre-methanol reforming unit and the post-proton exchange membrane fuel cell are identified using the particle swarm optimization algorithm to obtain the thermal parameters. A data-driven thermal model is constructed based on thermodynamic parameters. By combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell, a methanol reforming hydrogen production fuel cell model is constructed.

8. The modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell according to claim 7, characterized in that, The specific expression of the dynamic thermodynamic model of the pre-methanol reforming unit is as follows: ； In the above formula, Indicates the heat capacity of the reforming unit. This represents the total heat generated by the reforming reaction. Indicates surface radiative heat. This represents the heat exchange coefficient between the reforming unit and the environment. This represents the surface area of ​​the reforming unit. This indicates the temperature difference between the reforming unit and the environment. This indicates the measurement of the thermal effect of a fuel cell. This represents the heat transfer coefficient between the reforming unit and the fuel cell. This indicates the temperature difference between the reforming unit and the fuel cell. This indicates the amount of heat generated by the heating element.

9. The modeling method for a proton exchange membrane methanol reforming hydrogen fuel cell according to claim 8, characterized in that, The specific expression of the dynamic thermodynamic model of the back-end proton exchange membrane fuel cell is as follows: ； In the above formula, This indicates the number of cells in the PEM fuel cell stack. Indicates surface radiative heat. This indicates the measurement of the thermal effect of a fuel cell. Indicates activation loss. Indicates ohmic loss. Indicates concentration loss. Indicates surface radiative heat. Indicates the heat capacity of a fuel cell. Indicates the temperature of the fuel cell. This indicates the heat generated by fuel cell losses. This indicates the output current of the fuel cell. This indicates the temperature difference between the fuel cell and the environment. This indicates the temperature difference between the fuel cell and the front-end reformer.

10. A modeling system for a proton exchange membrane methanol reforming hydrogen fuel cell, characterized in that, Includes the following modules: The first module is used to construct a dynamic sub-model of the pre-methanol reforming unit based on the input of methanol-water solution from the front-end reformer, calculate the methanol conversion rate and the composition ratio of reactor outlet components, and build a dynamic sub-model of the pre-methanol reforming unit. The second module is used to input high-purity reformed gas into the PEM fuel cell to generate electricity, and to construct the static polarization model and dynamic response model of the back-end proton exchange membrane fuel cell. The third module is used to obtain thermodynamic parameters through particle swarm optimization algorithm, and to construct a methanol reforming hydrogen fuel cell model by combining the dynamic sub-model of the upstream methanol reforming unit with the static polarization model and dynamic response model of the downstream proton exchange membrane fuel cell.

Images