High-power gold double-bipolar-plate fuel cell thermal mass strengthening method based on dissipation minimization

Abstract

The invention provides a dissipation minimization-based thermal mass strengthening method for a high-power gold double-bipolar-plate fuel cell, and relates to the field of metal bipolar-plate fuel cells, and the method comprises the following steps: S1, determining demand analysis: determining an application scene, a performance demand and a dissipation minimization target of the metal bipolar-plate fuel cell, specific requirements and application scene characteristics of the metal bipolar plate fuel cell on thermal mass enhancement are determined; s2, establishing a dissipation minimization monitoring system, acquiring temperature, humidity, pressure, flow, voltage and current parameters of the metal bipolar plate fuel cell through a sensor, analyzing the parameters, performing real-time evaluation and prediction by establishing a dissipation minimization evaluation model, and optimizing predicted data by using a gradient descent method. By monitoring the parameters of the fuel cell in real time and combining the dissipation minimization evaluation model and the intelligent control algorithm, it can be ensured that the fuel cell can be kept in the optimal operation state under various working conditions.

Description

Technical Field

[0001] The invention relates to the field of metal bipolar plate fuel cells, and in particular to a thermal mass enhancement method for a high-power gold bipolar plate fuel cell based on dissipation minimization. Background Art

[0002] As an efficient and environmentally friendly energy conversion device, high-power fuel cells have shown broad application prospects in many fields. However, in the actual operation process, due to energy conversion and heat loss, a certain amount of dissipation will be generated, which not only affects the performance of the fuel cell, but also limits its service life and application scope. Therefore, studying how to minimize the dissipation of fuel cells and improve their heat and mass transfer efficiency has become an important direction for the development of current fuel cell technology.

[0003] Traditional high-power fuel cell thermal management methods often focus on single parameter control or simple feedback mechanisms, lacking systematicity and comprehensiveness. With the development of sensor technology, data processing technology and intelligent control algorithms, new ideas and methods have been provided for the thermal management of fuel cells. By real-time monitoring of the operating parameters of the fuel cell, combined with evaluation models and intelligent control algorithms, precise control and optimization of the thermal state of the fuel cell can be achieved, thereby further improving its performance and stability.

[0004] For example, the Chinese invention patent (application number: CN202410051112.3) discloses a method, device and readable medium for simulating thermal management of fuel cells based on automobiles. Its specification discloses that fuel cell vehicles have attracted much attention due to their advantages of being pollution-free and highly efficient. The fuel cell carried by the vehicle mainly includes a thermal management system, a hydrogen supply system, an air supply system and an electronic control system. With the application of fuel cell vehicles to high-power and heavy-duty applications, the thermal management problem of fuel cells is becoming more and more serious. Meeting the heat dissipation requirements of high-power vehicles is a key research work for the commercialization of fuel cells. The fuel cell thermal management system directly affects the operating temperature of the stack. Establishing a high-precision and high-performance thermal management system simulation platform and optimizing and improving the fuel cell engine in combination with the simulation results of the system model can effectively improve the performance of the fuel cell and meet the heat exchange requirements of fuel cell vehicles. The current fuel cell thermal management system modeling mostly adopts theoretical modeling methods. A large number of assumptions and theoretical calculations are made during the system construction process, ignoring the change process of most physical quantities in the system, resulting in poor accuracy of the system model. As a coupled system of multiple physical quantities, it is of great significance to establish a high-precision fuel cell thermal management system simulation platform, conduct system structure design for thermal management system simulation, and conduct real-time monitoring of various physical parameters of the thermal management system to guide the development of fuel cell system vehicles. The above patents can prove the defects of the existing technology.

[0005] Therefore, we made improvements to this and proposed a thermal mass enhancement method for high-power gold double bipolar plate fuel cells based on minimizing dissipation. Summary of the invention

[0006] The purpose of the present invention is to address the fact that traditional high-power fuel cell thermal mass management methods often focus on single parameter control or simple feedback mechanism, lacking systematicity and comprehensiveness.

[0007] In order to achieve the above-mentioned purpose of the invention, the present invention provides a thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization to improve the above-mentioned problems.

[0008] The specific application is as follows:

[0009] The steps include:

[0010] S1: Determine the demand analysis, determine the application scenarios, performance requirements and dissipation minimization goals of metal bipolar plate fuel cells, and determine the specific requirements and application scenario characteristics of metal bipolar plate fuel cells for thermal mass enhancement;

[0011] S2: Establish a dissipation minimization monitoring system, obtain the temperature, humidity, pressure, flow, voltage and current parameters of the metal bipolar plate fuel cell through sensors, analyze the parameters, conduct real-time evaluation and prediction by establishing a dissipation minimization evaluation model, and use the gradient descent method to optimize the predicted data;

[0012] S3: Selection of the dissipation minimization function, by presetting the operating mode according to different application scenarios and requirements of the fuel cell system, and the operator manually selects the corresponding operating mode according to the actual application scenario;

[0013] S4: The adaptive selection function of the metal bipolar plate fuel usage scenario is realized, combining the application scenario determination of the metal bipolar plate fuel cell in step S1 and the real-time parameter evaluation in step S2, and adjusting the parameters corresponding to the metal bipolar plate fuel cell according to the application scenario;

[0014] S5: Thermal enhancement of the metal bipolar plate fuel cell, combined with the dissipation minimization monitoring system in step S2, real-time detection of the metal bipolar plate fuel cell parameters, and corresponding adjustments to the parameters.

[0015] As a preferred technical solution of the present application, the establishment of the dissipation minimization evaluation model in step S2 specifically includes the following steps:

[0016] SA: Data collection and preprocessing, through experimental measurement of the parameters of the metal bipolar plate fuel cell under different working conditions, including temperature, humidity, pressure, flow, voltage and current parameters, using COMSOL Multiphysics fuel cell simulation software to conduct simulation experiments, obtain the parameters under the working conditions, and clean, denoise and calibrate the collected parameters;

[0017] SB: Build a model. According to the loss and heat loss in the energy conversion process of the metal bipolar plate fuel cell, analyze the source and mechanism of dissipation, simplify the dissipation evaluation model, establish a mathematical model, and calibrate the model in combination with the experimental data and simulation data in step SA. The energy conversion loss model uses the voltage, current and efficiency of the fuel cell to calculate the energy conversion loss. The specific formula is:

[0018] Energy Conversion Loss=V stack I stack ·(1-η stack )

[0019] Where: Energy Conversion Loss represents energy conversion loss, V stack Indicates the voltage of the battery stack, I stack Indicates the current η of the battery stack stack Indicates the efficiency of the battery stack;

[0020] The heat loss model formula is:

[0021] Thermal Loss = m coolant ·c p ·(T out -T in )+Radiative Loss+Convective Loss

[0022] Where m coolant is the mass flow rate of cooling water, c p is the specific heat capacity of water, T out -T in are the outlet and inlet temperatures of cooling water, Radiative Loss is the radiation heat loss, and Convective Loss is the convection heat loss;

[0023] The total dissipation model adds the energy conversion loss and the heat loss to obtain the total dissipation. The specific formula is:

[0024] Total Dissipation=Energy Conversion Loss+Thermal Loss

[0025] SC: Compare and verify the model prediction results with experimental data and simulation data, and optimize the model according to the verification results, including modifying the model structure and adjusting parameter values;

[0026] SD: Collect the actual operation data of the system during use, compare and analyze it with the model prediction results, and provide timely feedback to the model developers for iterative optimization.

[0027] As the preferred technical solution of the present application, the specific algorithm formulas of the cleaning, denoising and calibration processing in step SA are:

[0028] Cleaning: c k =1 / n k ∑ xi∈Ck x i

[0029] where n k is cluster C k The number of points in x i is a point in the cluster;

[0030] Denoising: x denoised =W -1 (τ(Wx))

[0031] Among them, W -1 is the inverse transform, τ(Wx) represents the threshold processing applied to the coefficients after wavelet transform;

[0032] Calibration; X calibrated =T calibrated +S calibrated +R

[0033] Among them, Tcalibrated and Scalibrated are the trend and seasonal components after being processed by a certain calibration method, and R is the residual component.

[0034] As a preferred technical solution of the present application, the mathematical model established in step SB is established by establishing a partial differential equation, and a partial differential equation describing the system behavior is established according to Fourier's heat conduction law. The specific partial differential equation can be expressed as:

[0035]

[0036] Wherein, T is temperature, x and y are spatial coordinates, and the establishment of the mathematical model also includes constructing a dissipation function, and the specific expression of the dissipation function is:

[0037]

[0038] Where Φ is a physical quantity related to the temperature gradient, k is a constant, T is the temperature, and V is the integrated volume.

[0039] As a preferred technical solution of the present application, the implementation of the metal bipolar plate fuel usage scenario adaptive selection function in step S4 specifically includes the following steps:

[0040] SI: Data processing, using filtering technology to pre-process the data and compress the processed data. The filtering formula used in the filtering technology is:

[0041] y[n]=α·x[n]+(1-α)·y[n-1]

[0042] Where: y[n] is the current smoothed output value, x[n] is the current input value, y[n-1] is the last smoothed output value, and α is a smoothing factor between 0 and 1;

[0043] SII: Pattern recognition uses a computer learning algorithm to take the sensor and the data processed in step SI as input, identify the usage scenario through feature extraction, and use a decision tree with a classification algorithm to recursively divide the data set to build a classification model. The specific formula of the classification algorithm is:

[0044] {Min / w，b}1 / 2|w| 2

[0045]

[0046] Among them, w is the weight vector, b is the bias term, and y i is the label of the ith sample, xi is the feature vector of the ith sample, and the objective function is 1 / 2|w| 2 It is half the square of the magnitude of the weight vector.

[0047] As the preferred technical solution of this application, the following steps are also included:

[0048] SIII: Design a feedback mechanism to monitor the system response in real time and adjust the decision algorithm. The feedback control algorithm formula is:

[0049] u(t)=K p ·e(t)+K i ∫0 t e(τ)dτ+K d ·{de(t) / dt}

[0050] Where: u(t) is the controller output, which is used to control the controlled object, e(t) is the error signal, that is, the difference between the expected output and the actual output, K p is the proportional coefficient, which determines the strength of the proportional effect, K i is the integral coefficient, which determines the strength of the integral effect, Kd is the differential coefficient, which determines the strength of the differential effect, ∫0 t e(τ)dτ is the integral of the error signal, which is used to eliminate static errors, and de(t) / dt is the differential of the error signal, which is used to predict future error changes.

[0051] As a preferred technical solution of the present application, the computer learning algorithm in step SII specifically includes:

[0052] Trend component: T t ≈Loess(X t )

[0053] Among them, X t is the original time series data, and Loess is a nonparametric regression method;

[0054] Seasonal ingredients: S t ≈Smooth(X t -T t )

[0055] Among them, Smooth represents smoothing operation;

[0056] Residual component: R t =X t -T t -S t

[0057] Trend calibration: T calibrated,t =β0+β1t

[0058] Among them, β0 and β1 are the linear regression coefficients estimated by minimizing the residual sum of squares.

[0059] As a preferred technical solution of the present application, the thermal enhancement of the metal bipolar plate fuel cell in step S5 specifically includes the following steps:

[0060] Sa: Use CFD simulation software to perform three-dimensional modeling and simulation analysis on the cooling channel. By changing the channel shape, width, and depth parameters, and integrating the temperature sensor into the metal bipolar plate fuel cell, combined with step SIII, real-time data detection is performed.

[0061] As a preferred technical solution of the present application, the step S1 of determining the demand analysis specifically includes the following steps:

[0062] Step 1: Use data cleaning, screening, clustering, and association rule mining techniques to identify market trends, user preferences, and competitive situations, and analyze user reviews;

[0063] Step 2: Use finite element analysis software to analyze stress distribution and heat conduction during the manufacturing process; use computational fluid dynamics software to analyze gas flow and cooling water circulation fluid behavior;

[0064] Step 3: Establish an automated market intelligence collection system, use RSS subscription and API interface technology to capture industry trends and technological progress information in real time; use Tableau and Power BI data visualization tools to convert market and technical data into intuitive charts and reports.

[0065] As the preferred technical solution of this application, the data clustering and association rule mining technology in step 1 specifically adopts the following algorithm:

[0066] Clustering: J(C,μ)=∑ i＝1 k ∑ x∈Ci d(x,μ i )

[0067] Where J(C,μ) represents the clustering quality index, C represents clustering, μ represents the cluster center, d(x,μ i) Represents the distance between the data point x and the cluster center μi;

[0068] Association rule mining: s = count(X∪Y) / count(D)

[0069] Where s represents support, count(X∪Y) represents the number of transactions containing the item set X∪Y, and count(D) represents the total number of transactions.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] In the scheme of this application:

[0072] 1. In order to solve the problem of poor performance and stability of fuel cells in the prior art, the present application monitors the temperature, humidity, pressure, flow, voltage and current parameters of the fuel cell in real time, and combines the dissipation minimization evaluation model and the intelligent control algorithm. This method can ensure that the fuel cell can maintain the optimal operating state under various working conditions; this not only improves the energy conversion efficiency of the fuel cell, but also significantly enhances its operating stability, so that the fuel cell can more reliably meet the needs of various application scenarios;

[0073] 2. In order to solve the problem of short service life of fuel cells in the prior art, the present application significantly reduces the thermal stress and mechanical stress of the fuel cell during operation by minimizing the dissipation of the fuel cell, thereby reducing its failure rate and maintenance cost; at the same time, since the fuel cell can operate in a more optimal state, its service life is also effectively extended;

[0074] 3. Through adaptive selection function and intelligent decision-making mechanism, fuel cells can better adapt to different application scenarios and needs. Fuel cells can maintain the optimal operating state through intelligent adjustment, thereby improving their adaptability and reliability, and solving the problem of low adaptability and reliability of fuel cells in existing technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 A flow chart of a thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization provided in this application. DETAILED DESCRIPTION

[0076] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0077] As described in the background art, traditional high-power fuel cell thermal management methods often focus on single parameter control or simple feedback mechanisms, lacking systematicity and comprehensiveness.

[0078] In order to solve this technical problem, the present invention provides a high-power gold double bipolar plate fuel cell thermal mass enhancement method based on dissipation minimization, which is applied to adaptive metal bipolar plate fuel cell thermal mass enhancement.

[0079] Specifically, please refer to Figure 1 The thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization specifically comprises the following steps:

[0080] S1: Determine the demand analysis, determine the application scenarios, performance requirements and dissipation minimization goals of metal bipolar plate fuel cells, and determine the specific requirements and application scenario characteristics of metal bipolar plate fuel cells for thermal mass enhancement;

[0081] S2: Establish a dissipation minimization monitoring system, obtain the temperature, humidity, pressure, flow, voltage and current parameters of the metal bipolar plate fuel cell through sensors, and analyze the parameters. Establish a dissipation minimization evaluation model for real-time evaluation and prediction, and use the gradient descent method to optimize the predicted data. The prediction model uses linear regression technology, and its specific formula is:

[0082] y pred =β0+β1·x1+β2·x2+…+β n ·x n

[0083] Where: y pred is the predicted value of the target variable, β0 is the intercept term, which means the predicted value of the target variable when all independent variables are 0, β1,β2,…,β n is the slope coefficient, which indicates the influence of each variable on the predicted value of the target variable, x1, x2,…, x n is the independent variable, i.e., the input feature;

[0084] The specific function used by the gradient descent method is:

[0085]

[0086] Among them, θ t+1 is the updated parameter, θ t is the current parameter, η is the learning rate, which determines the step size of parameter update, It is the gradient of the loss function J(θ) with respect to the parameter θ, indicating the direction of parameter update;

[0087] S3: Selection of the dissipation minimization function, by presetting the operating mode according to different application scenarios and requirements of the fuel cell system, and the operator manually selects the corresponding operating mode according to the actual application scenario;

[0088] S4: The adaptive selection function of the metal bipolar plate fuel usage scenario is realized, combining the application scenario determination of the metal bipolar plate fuel cell in step S1 and the real-time parameter evaluation in step S2, and adjusting the parameters corresponding to the metal bipolar plate fuel cell according to the application scenario;

[0089] S5: Thermal enhancement of the metal bipolar plate fuel cell, combined with the dissipation minimization monitoring system in step S2, real-time detection of the metal bipolar plate fuel cell parameters, and corresponding adjustments to the parameters.

[0090] The thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization provided by the present invention can ensure that the fuel cell can maintain the optimal operating state under various operating conditions by real-time monitoring of the temperature, humidity, pressure, flow, voltage and current parameters of the fuel cell, and combining a dissipation minimization evaluation model and an intelligent control algorithm; this not only improves the energy conversion efficiency of the fuel cell, but also significantly enhances its operating stability, so that the fuel cell can more reliably meet the needs of various application scenarios.

[0091] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings.

[0092] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions in the embodiments may be combined with each other.

[0093] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0094] Example 1, please refer to Figure 1 The thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization, wherein the establishment of the dissipation minimization evaluation model in step S2 specifically includes the following steps:

[0095] SA: Data collection and preprocessing, through experimental measurement of the parameters of the metal bipolar plate fuel cell under different working conditions, including temperature, humidity, pressure, flow, voltage and current parameters, using COMSOL Multiphysics fuel cell simulation software to conduct simulation experiments, obtain the parameters under the working conditions, and clean, denoise and calibrate the collected parameters;

[0096] SB: Build a model. According to the loss and heat loss in the energy conversion process of the metal bipolar plate fuel cell, analyze the source and mechanism of dissipation, simplify the dissipation evaluation model, establish a mathematical model, and calibrate the model in combination with the experimental data and simulation data in step SA. The energy conversion loss model uses the voltage, current and efficiency of the fuel cell to calculate the energy conversion loss. The specific formula is:

[0097] Energy Conversion Loss=V stack I stack ·(1-η stack )

[0098] Where: Energy Conversion Loss represents energy conversion loss, V stack Indicates the voltage of the battery stack, I stack Indicates the current η of the battery stack stack Indicates the efficiency of the battery stack;

[0099] The heat loss model formula is:

[0100] Thermal Loss = m coolant ·c p ·(T out -T in )+Radiative Loss+Convective Loss

[0101] Where m coolant is the mass flow rate of cooling water, cp is the specific heat capacity of water, T out -T in are the outlet and inlet temperatures of cooling water, Radiative Loss is the radiation heat loss, and Convective Loss is the convection heat loss;

[0102] The total dissipation model adds the energy conversion loss and the heat loss to obtain the total dissipation. The specific formula is:

[0103] Total Dissipation=Energy Conversion Loss+Thermal Loss

[0104] SC: Compare and verify the model prediction results with experimental data and simulation data, and optimize the model according to the verification results, including modifying the model structure and adjusting parameter values;

[0105] SD: Collect the actual operation data of the system during use, compare and analyze it with the model prediction results, and provide timely feedback to the model developers for iterative optimization.

[0106] Through data collection and preprocessing, a reliable foundation is provided for establishing an accurate dissipation assessment model. The established dissipation assessment model can comprehensively reflect the energy conversion loss and heat loss of the fuel cell, and provides strong theoretical support for minimizing dissipation. Through comparative verification and iterative optimization, the accuracy and practicality of the model are ensured.

[0107] Further, such as Figure 1 As shown, the specific algorithm formulas of the cleaning, denoising and calibration processing in step SA are:

[0108] Cleaning: c k =1 / n k ∑ xi∈Ck x i

[0109] where n k is cluster C k The number of points in x i is a point in the cluster;

[0110] Denoising: x denoised =W -1 (τ(Wx))

[0111] Among them, W -1 is the inverse transform, τ(Wx) represents the threshold processing applied to the coefficients after wavelet transform;

[0112] Calibration; X calibrated =T calibrated +S calibrated +R

[0113] Among them, Tcalibrated and Scalibrated are the trend and seasonal components after being processed by a certain calibration method, and R is the residual component.

[0114] It provides specific data cleaning, denoising and calibration algorithms, and provides clear methodological guidance for data preprocessing. These algorithms can effectively improve the accuracy and reliability of data, and provide strong support for subsequent analysis and decision-making.

[0115] Further, such as Figure 1 As shown, the mathematical model established in step SB is established by establishing a partial differential equation, and a partial differential equation describing the system behavior is established according to Fourier's heat conduction law. The specific partial differential equation can be expressed as:

[0116]

[0117] Wherein, T is temperature, x and y are spatial coordinates, and the establishment of the mathematical model also includes constructing a dissipation function, and the specific expression of the dissipation function is:

[0118]

[0119] Where Φ is a physical quantity related to the temperature gradient, k is a constant, T is the temperature, V is the integrated volume, and each term in the integrand is the square of the partial derivative of the temperature T with respect to the spatial coordinates x, y, and z, which represent the rate of change of temperature in the x, y, and z directions, respectively.

[0120] The use of partial differential equations to describe the system behavior provides a mathematical tool for a deeper understanding of the heat and mass transfer process of fuel cells. The constructed dissipation function provides an effective means to quantify and analyze the dissipation of fuel cells.

[0121] Example 2 further optimizes the thermal mass enhancement method for a high-power gold dual bipolar plate fuel cell based on dissipation minimization provided in Example 1. Specifically, Figure 1 As shown, the metal bipolar plate fuel usage scenario adaptive selection function in step S4 specifically includes the following steps:

[0122] SI: Data processing, using filtering technology to pre-process the data and compress the processed data. The filtering formula used in the filtering technology is:

[0123] y[n]=α·x[n]+(1-α)·y[n-1]

[0124] Where: y[n] is the current smoothed output value, x[n] is the current input value, y[n-1] is the output value after the previous smoothing, and α is a smoothing factor between 0 and 1, which is used to control the influence of the input value on the current output value. The larger the value of α, the greater the influence of the input value on the current output value, and the smaller the smoothing effect; conversely, the smaller the value of α, the greater the smoothing effect;

[0125] SII: Pattern recognition uses a computer learning algorithm to take the sensor and the data processed in step SI as input, identify the usage scenario through feature extraction, and use a decision tree with a classification algorithm to recursively divide the data set to build a classification model. The specific formula of the classification algorithm is:

[0126] {Min / w，b}1 / 2|w| 2

[0127]

[0128] Among them, w is the weight vector, b is the bias term, and y i is the label of the ith sample (usually -1 or 1), xi is the feature vector of the ith sample, and the objective function is 1 / 2|w| 2 is half the square of the modulus of the weight vector, with the constraint y i (w T x i +b)≥1 ensures that all samples are correctly classified and the confidence of the classification is at least 1.

[0129] Through data processing and pattern recognition, accurate identification of fuel cell usage scenarios is achieved. The constructed classification model provides decision support for the adaptive selection function and improves the intelligence level of the system.

[0130] Further, such as Figure 1 As shown, the following steps are also included:

[0131] SIII: Design a feedback mechanism to monitor the system response in real time and adjust the decision algorithm. The feedback control algorithm formula is:

[0132] u(t)=K p ·e(t)+K i ∫0 t e(τ)dτ+K d ·{de(t) / dt}

[0133] Where: u(t) is the controller output, which is used to control the controlled object, e(t) is the error signal, that is, the difference between the expected output and the actual output, K p is the proportional coefficient, which determines the strength of the proportional effect, Ki is the integral coefficient, which determines the strength of the integral effect, K d is the differential coefficient, which determines the strength of the differential effect, ∫0 t e(τ)dτ is the integral of the error signal, which is used to eliminate static errors, and de(t) / dt is the differential of the error signal, which is used to predict future error changes;

[0134] SIV: Decision-making and execution, making decisions based on real-time data and prediction results, determining the optimal operating parameters of the fuel cell system, executing the decision results through the control system, and adjusting the operating status of the fuel cell system in real time, including fuel supply, oxidant flow and cooling water temperature.

[0135] The designed feedback mechanism can monitor the system response in real time and adjust the decision algorithm as needed, which improves the stability and adaptability of the system. Through the decision-making and execution steps, real-time optimization and control of the fuel cell system is achieved, further improving its operating efficiency.

[0136] Further, such as Figure 1 As shown, the computer learning algorithm in step SII specifically includes:

[0137] Trend component: T t ≈Loess(X t )

[0138] Among them, X t is the original time series data, Loess is a nonparametric regression method used to fit a local weighted polynomial;

[0139] Seasonal ingredients: S t ≈Smooth(X t -T t )

[0140] Among them, Smooth represents smoothing operation;

[0141] Residual component: R t =X t -T t -S t

[0142] Trend calibration: T calibrated,t =β0+β1t

[0143] Among them, β0 and β1 are the linear regression coefficients estimated by minimizing the residual sum of squares.

[0144] Example 3 further optimizes the thermal mass enhancement method for a high-power gold dual bipolar plate fuel cell based on dissipation minimization provided in Example 1 or 2. Specifically, Figure 1As shown, the thermal enhancement of the metal bipolar plate fuel cell in step S5 specifically includes the following steps:

[0145] Sa: Use CFD simulation software to perform three-dimensional modeling and simulation analysis on the cooling channel. By changing the channel shape, width and depth parameters, the flow distribution of the coolant in the channel is optimized, the pressure drop is reduced, and the heat exchange efficiency is improved. The temperature sensor is integrated into the metal bipolar plate fuel cell. Combined with step SIII, the data is detected in real time;

[0146] Sb: Collect the real-time data in step Sa, and remotely monitor and maintain the operating status of the metal bipolar plate fuel cell in combination with step SIV, while integrating the sensor network, data processing unit and control actuator into one to achieve real-time monitoring and intelligent regulation of the fuel cell system.

[0147] Example 4: The thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization provided in the above embodiments is further optimized, such as Figure 1 As shown, the step S1 of determining the demand analysis specifically includes the following steps:

[0148] Step 1: Use data cleaning, screening, clustering, and association rule mining techniques to identify market trends, user preferences, and competitive situations, and analyze user reviews;

[0149] Step 2: Use finite element analysis software to analyze stress distribution and heat conduction during the manufacturing process; use computational fluid dynamics software to analyze gas flow and cooling water circulation fluid behavior;

[0150] Step 3: Establish an automated market intelligence collection system, use RSS subscription and API interface technology to capture industry trends and technological progress information in real time; use Tableau and Power BI data visualization tools to convert market and technical data into intuitive charts and reports.

[0151] Further, such as Figure 1 As shown, the data clustering and association rule mining technology in step 1 specifically adopts the following algorithm:

[0152] Clustering: J(C,μ)=∑ i＝1 k ∑ x∈Ci d(x,μ i )

[0153] Where J(C,μ) represents the clustering quality index, C represents clustering, μ represents the cluster center, d(x,μ i) Represents the distance between the data point x and the cluster center μi;

[0154] Association rule mining: s = count(X∪Y) / count(D)

[0155] Where s represents support, count(X∪Y) represents the number of transactions containing the item set X∪Y, and count(D) represents the total number of transactions.

[0156] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral one; it can be a mechanical connection, an electrical connection, or communication with each other; it can be a direct connection, or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0157] Obviously, the embodiments described above are only some embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific embodiments, or to perform equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of the present invention, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. A thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization, characterized in that: The steps include: S1: Determine the demand analysis, determine the application scenarios, performance requirements and dissipation minimization goals of metal bipolar plate fuel cells, and determine the specific requirements and application scenario characteristics of metal bipolar plate fuel cells for thermal mass enhancement; S2: Establish a dissipation minimization monitoring system, obtain the temperature, humidity, pressure, flow, voltage and current parameters of the metal bipolar plate fuel cell through sensors, analyze the parameters, conduct real-time evaluation and prediction by establishing a dissipation minimization evaluation model, and use the gradient descent method to optimize the predicted data; S3: Selection of the dissipation minimization function, by presetting the operating mode according to different application scenarios and requirements of the fuel cell system, and the operator manually selects the corresponding operating mode according to the actual application scenario; S4: The adaptive selection function of the metal bipolar plate fuel usage scenario is realized, combining the application scenario determination of the metal bipolar plate fuel cell in step S1 and the real-time parameter evaluation in step S2, and adjusting the parameters corresponding to the metal bipolar plate fuel cell according to the application scenario; S5: Thermal enhancement of the metal bipolar plate fuel cell, combined with the dissipation minimization monitoring system in step S2, real-time detection of the metal bipolar plate fuel cell parameters, and corresponding adjustments to the parameters.

2. The thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization according to claim 1 is characterized in that: The establishment of the dissipation minimization evaluation model in step S2 specifically includes the following steps: SA: Data collection and preprocessing, through experimental measurement of the parameters of the metal bipolar plate fuel cell under different working conditions, including temperature, humidity, pressure, flow, voltage and current parameters, using COMSOL Multiphysics fuel cell simulation software to conduct simulation experiments, obtain the parameters under the working conditions, and clean, denoise and calibrate the collected parameters; SB: Construct a model. According to the loss and heat loss in the energy conversion process of the metal bipolar plate fuel cell, analyze the source and mechanism of dissipation, simplify the dissipation evaluation model, establish a mathematical model, and calibrate the model in combination with the experimental data and simulation data in step SA. The energy conversion loss model uses the voltage, current and efficiency of the fuel cell to calculate the energy conversion loss. The specific formula is: Energy Conversion Loss=V stack ·I stack ·(1-η stack ) Where: Energy Conversion Loss represents energy conversion loss, V stack Indicates the voltage of the battery stack, I stack Indicates the current η of the battery stack stack Indicates the efficiency of the battery stack; The heat loss model formula is: Thermal Loss = m coolant ·c p ·(T out -T in )+Radiative Loss+Convective Loss where m coolant is the mass flow rate of cooling water, c p is the specific heat capacity of water, T out -T in are the outlet and inlet temperatures of cooling water, Radiative Loss is the radiation heat loss, and Convective Loss is the convection heat loss; The total dissipation model adds the energy conversion loss and the heat loss to obtain the total dissipation. The specific formula is: Total Dissipation=Energy Conversion Loss+Thermal Loss SC: Compare and verify the model prediction results with experimental data and simulation data, and optimize the model according to the verification results, including modifying the model structure and adjusting parameter values; SD: During use, the actual operation data of the system is collected, compared and analyzed with the model prediction results, and timely feedback is given to the model developers for iterative optimization.

3. The thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization according to claim 2 is characterized in that: The specific algorithm formulas for the cleaning, denoising and calibration processing in step SA are: Cleaning: c k =1 / n k ∑ xi∈Ck x i where n k is cluster C k The number of points in x i is a point in the cluster; Denoising: x denoised =W -1 (τ(Wx)) Among them, W -1 is the inverse transform, τ(Wx) represents the threshold processing applied to the coefficients after wavelet transform; Calibration; X calibrated =T calibrated +S calibrated +R Among them, Tcalibrated and Scalibrated are the trend and seasonal components after being processed by a certain calibration method, and R is the residual component.

4. The thermal mass enhancement method for a high-power gold double bipolar plate fuel cell based on dissipation minimization according to claim 3 is characterized in that: The mathematical model established in step SB is established by establishing a partial differential equation, and a partial differential equation describing the system behavior is established according to Fourier's heat conduction law. The specific partial differential equation can be expressed as: Wherein, T is temperature, x and y are spatial coordinates, and the establishment of the mathematical model also includes constructing a dissipation function, and the specific expression of the dissipation function is: Where Φ is a physical quantity related to the temperature gradient, k is a constant, T is the temperature, and V is the integrated volume.

5. The method for thermal mass enhancement of a high-power gold double bipolar plate fuel cell based on dissipation minimization according to claim 4 is characterized in that: The implementation of the metal bipolar plate fuel usage scenario adaptive selection function in step S4 specifically includes the following steps: SI: Data processing, using filtering technology to pre-process the data and compress the processed data. The filtering formula used in the filtering technology is: y[n]=α·x[n]+(1-α)·y[n-1] Where: y[n] is the current smoothed output value, x[n] is the current input value, y[n-1] is the last smoothed output value, and α is a smoothing factor between 0 and 1; SII: Pattern recognition uses a computer learning algorithm to take the sensor and the data processed in step SI as input, identify the usage scenario through feature extraction, and use a decision tree with a classification algorithm to recursively divide the data set to build a classification model. The specific formula of the classification algorithm is: {Min / w，b }1 / 2∣w∣ 2 Among them, w is the weight vector, b is the bias term, and y i is the label of the i-th sample.

6. The method for thermal mass enhancement of a high-power gold double bipolar plate fuel cell based on dissipation minimization according to claim 5, characterized in that: The following steps are also included: SIII: Design a feedback mechanism to monitor the system response in real time and adjust the decision algorithm. The feedback control algorithm formula is: u(t)=K p ·e(t)+K i ·∫0 t e(τ)dτ+K d ·{de(t) / dt} Where: u(t) is the controller output, which is used to control the controlled object, e(t) is the error signal, that is, the difference between the expected output and the actual output, K p is the proportional coefficient, which determines the strength of the proportional effect, K i is the integral coefficient, which determines the strength of the integral effect, K d is the differential coefficient, which determines the strength of the differential effect, ∫0 t e(τ)dτ is the integral of the error signal, which is used to eliminate static errors, and de(t) / dt is the differential of the error signal, which is used to predict future error changes.

7. The method for thermal mass enhancement of a high-power gold double bipolar plate fuel cell based on dissipation minimization according to claim 6, characterized in that: The computer learning algorithm in step SII specifically includes: Trend component: T t ≈Loess(X t ) Among them, X t is the original time series data, and Loess is a nonparametric regression method; Seasonal ingredients: S t ≈Smooth(X t -T t ) Among them, Smooth represents smoothing operation; Residual component: R t =X t -T t -S t Trend calibration: T calibrated,t =β0+β1t Among them, β0 and β1 are the linear regression coefficients estimated by minimizing the residual sum of squares.

8. The method for thermal mass enhancement of a high-power gold double bipolar plate fuel cell based on dissipation minimization according to claim 7 is characterized in that: The thermal enhancement of the metal bipolar plate fuel cell in step S5 specifically includes the following steps: Sa: Use CFD simulation software to perform three-dimensional modeling and simulation analysis on the cooling channel. By changing the channel shape, width, and depth parameters, and integrating the temperature sensor into the metal bipolar plate fuel cell, combined with step SIII, real-time data detection is performed.

9. The method for thermal mass enhancement of a high-power gold double bipolar plate fuel cell based on dissipation minimization according to claim 8, characterized in that: The step S1 of determining the demand analysis specifically includes the following steps: Step 1: Use data cleaning, screening, clustering, and association rule mining techniques to identify market trends, user preferences, and competitive situations, and analyze user reviews; Step 2: Use finite element analysis software to analyze stress distribution and heat conduction during the manufacturing process; Use computational fluid dynamics software to analyze gas flow and cooling water circulation fluid behavior; Step 3: Establish an automated market intelligence collection system, and use RSS subscription and API interface technology to capture industry trends and technological progress information in real time; use Tableau and Power BI data visualization tools to convert market and technical data into intuitive charts and reports.

10. The method for thermal mass enhancement of a high-power gold double bipolar plate fuel cell based on dissipation minimization according to claim 9, characterized in that: The data clustering and association rule mining technology in step 1 specifically adopts the following algorithm: Clustering: J(C,μ)=∑ i＝1 k ∑ x∈Ci d(x,μ i ) Where J(C,μ) represents the clustering quality index, C represents clustering, μ represents the cluster center, d(x,μ i) Represents the distance between the data point x and the cluster center μi; Association rule mining: s = count(X∪Y) / count(D) Where s represents support, count(X∪Y) represents the number of transactions containing the item set X∪Y, and count(D) represents the total number of transactions.

Images