Construction of Hydrogen Fuel Cell Performance Prediction Model, Performance Analysis Method and Device

Abstract

The present application relates to a method and device for constructing a performance prediction model and analyzing the performance of a hydrogen fuel cell, which are applied to the technical field of hydrogen fuel cells. The method for constructing the battery performance prediction model includes: obtaining the operation data of the hydrogen fuel cell at multiple moments, and calculating the total voltage loss at each moment according to the operation data at a single moment; performing independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses of three components at each moment; using the voltage losses of the components at N consecutive moments as input data and the voltage losses of the components at the next moment of N consecutive moments as label data to construct the training data corresponding to the components; constructing a transformer prediction model corresponding to each component, and training the transformer prediction model according to the training data corresponding to the component; when the termination condition is met, the training ends, and a hydrogen fuel cell performance prediction model is obtained. The present application can improve the accuracy of model training.

Description

Technical Field

[0001] The present application relates to the technical field of hydrogen fuel cells, and particularly to a method and device for constructing a hydrogen fuel cell performance prediction model, a performance analysis method and device, an electronic device, and a storage medium. Background Art

[0002] A hydrogen fuel cell is a clean and environmentally friendly energy conversion device, which is widely used in automobiles, portable energy storage devices, and other fields with high energy efficiency requirements. The performance of a hydrogen fuel cell is affected by various factors. How to effectively evaluate and analyze the performance of a hydrogen fuel cell has become an important research topic in this field.

[0003] The performance analysis methods of hydrogen fuel cells mainly include a mechanism model based on physical principles and a machine learning model based on data-driven. The mechanism model predicts the performance of the battery under different operating conditions by describing phenomena such as reaction processes, heat transfer, and current flow. The machine learning model uses a large amount of battery operation data for training through data-driven and deep learning methods to calculate the performance of the hydrogen fuel cell. However, the existing machine learning models have low prediction accuracy for the performance of hydrogen fuel cells, and the interpretability of the prediction results of machine learning algorithms is poor. Summary of the Invention

[0004] To solve the above technical problems, the present application provides a method and device for constructing a hydrogen fuel cell performance prediction model, a hydrogen fuel cell performance analysis method and device based on an AI large model, an electronic device, and a storage medium.

[0005] According to a first aspect of the present application, there is provided a method for constructing a hydrogen fuel cell performance prediction model, including:

[0006] Obtain the operation data of the hydrogen fuel cell at multiple moments, and calculate the total voltage loss at the moment according to the operation data at a single moment;

[0007] Based on the physical mechanism of the output voltage of the hydrogen fuel cell, perform independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses of three components at each moment; the three components include polarization, ohmic, and concentration;

[0008] For each component, use the voltage losses of the component at consecutive N moments as input data, and the voltage losses of the component at the next moment of the consecutive N moments as label data to construct the training data corresponding to the component; N is a positive integer;

[0009] Construct a transformer prediction model corresponding to each component, and train the transformer prediction model according to the training data corresponding to the component;

[0010] When the termination condition is met, the training ends, and a hydrogen fuel cell performance prediction model is obtained.

[0011] Optionally, based on the physical mechanism of the output voltage of the hydrogen fuel cell, independent component analysis is performed on the total voltage losses at multiple moments to obtain the voltage losses of three components at each moment, including:

[0012] Divide the multiple moments into M cycles, and based on the physical mechanism of the output voltage of the hydrogen fuel cell and the actual output voltages of the hydrogen fuel cell at multiple moments in each cycle, determine multiple constant parameters corresponding to the cycle for calculating the voltage losses of the three components; M is a positive integer;

[0013] Select any moment from each cycle, and calculate the reference voltage losses of the three components at this moment according to the operation data and multiple constant parameters at this moment;

[0014] Based on the reference voltage losses of the three components at M moments, perform independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses of the three components at each moment.

[0015] Optionally, the performing independent component analysis on the total voltage losses at multiple moments based on the reference voltage losses of the three components at M moments to obtain the voltage losses of the three components at each moment includes:

[0016] With the objective function: min||X - A·S|| 2 +λ||S - S T || as the constraint, perform independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses S of the three components;

[0017] where X represents the total voltage loss at a single moment, A represents the signal fusion matrix calculated by independent component analysis, S T represents the reference voltage losses of the three components, and λ represents the weight coefficient.

[0018] Optionally, the method further includes:

[0019] During the process of training the transformer prediction model according to the training data corresponding to the components, use an optimization algorithm to optimize the hyperparameters of the transformer prediction model.

[0020] According to the second aspect of the present application, there is provided a hydrogen fuel cell performance analysis method based on an AI large model, including:

[0021] Obtain a user instruction, and use a pre-constructed hydrogen fuel cell professional large language model to identify the user instruction and determine whether it belongs to a hydrogen fuel cell performance analysis instruction;

[0022] If the user instruction does not belong to the hydrogen fuel cell performance analysis instruction, output the corresponding query result;

[0023] If the user instruction belongs to the hydrogen fuel cell performance analysis instruction, obtain the hydrogen fuel cell operation data corresponding to the user instruction, and call the model corresponding to the user instruction from the pre-constructed hydrogen fuel cell performance analysis model to process the hydrogen fuel cell operation data to obtain a performance analysis result;

[0024] Among them, the hydrogen fuel cell performance analysis model includes a hydrogen fuel cell performance prediction model, and the hydrogen fuel cell performance prediction model is constructed based on the method described in the first aspect.

[0025] Optionally, the construction method of the hydrogen fuel cell professional large language model includes:

[0026] Obtain the hydrogen fuel cell knowledge base corpus and convert the hydrogen fuel cell knowledge base corpus into question-and-answer pair data;

[0027] Obtain a general large language model and fine-tune the general large language model using the question-and-answer pair data to obtain the hydrogen fuel cell professional large language model.

[0028] According to the third aspect of the present application, a hydrogen fuel cell performance prediction model construction device is provided, including:

[0029] A total voltage loss calculation module, configured to obtain the operation data of the hydrogen fuel cell at multiple moments, and calculate the total voltage loss at the moment according to the operation data at a single moment;

[0030] An independent component analysis module, configured to perform independent component analysis on the total voltage loss at multiple moments based on the physical mechanism of the hydrogen fuel cell output voltage to obtain the voltage loss of three components at each moment; the three components include polarization, ohmic, and concentration difference;

[0031] A training data construction module, configured to, for each component, use the voltage loss of the component at N consecutive moments as input data and the voltage loss of the next moment of the N consecutive moments as label data to construct the training data corresponding to the component; N is a positive integer;

[0032] A transformer prediction model training module, configured to construct a transformer prediction model corresponding to each component, and train the transformer prediction model according to the training data corresponding to the component;

[0033] A hydrogen fuel cell performance prediction model generation module, configured to end the training when the termination condition is met to obtain a hydrogen fuel cell performance prediction model.

[0034] Optionally, the independent component analysis module is specifically configured to divide the multiple moments into M cycles, determine multiple constant parameters corresponding to each cycle for calculating the voltage losses of the three components based on the physical mechanism of the output voltage of the hydrogen fuel cell and the actual output voltage of the hydrogen fuel cell at multiple moments in each cycle; M is a positive integer; select any moment from each cycle, calculate the reference voltage losses of the three components at the moment according to the operation data and multiple constant parameters at the moment; perform independent component analysis on the total voltage losses at multiple moments based on the reference voltage losses of the three components at M moments, and obtain the voltage losses of the three components at each moment.

[0035] Optionally, the independent component analysis module is specifically configured to perform independent component analysis on the total voltage losses at multiple moments based on the reference voltage losses of the three components at M moments through the following steps to obtain the voltage losses of the three components at each moment:

[0036] With the objective function: min||X - A·S|| 2 +λ||S - S T || as a constraint, perform independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses S of the three components;

[0037] wherein, X represents the total voltage loss at a single moment, A represents the signal fusion matrix calculated through independent component analysis, and S T represents the reference voltage losses of the three components, and λ represents the weight coefficient.

[0038] Optionally, the transformer prediction model training module is further configured to optimize the hyperparameters of the transformer prediction model by using an optimization algorithm during the process of training the transformer prediction model according to the training data corresponding to the components.

[0039] According to the fourth aspect of the present application, there is provided a hydrogen fuel cell performance analysis device based on an AI large model, including:

[0040] A user instruction acquisition module for acquiring user instructions;

[0041] A user instruction recognition module for using a pre-constructed hydrogen fuel cell professional large language model to recognize the user instructions and determine whether they belong to hydrogen fuel cell performance analysis instructions;

[0042] A query result output module for outputting a corresponding query result if the user instruction recognition module determines that the user instruction does not belong to hydrogen fuel cell performance analysis instructions;

[0043] A performance analysis structure output module, configured to, if the user instruction recognition module determines that the user instruction belongs to a hydrogen fuel cell performance analysis instruction, obtain the hydrogen fuel cell operation data corresponding to the user instruction, and call the model corresponding to the user instruction from a pre-constructed hydrogen fuel cell performance analysis model to process the hydrogen fuel cell operation data to obtain a performance analysis result;

[0044] Wherein, the hydrogen fuel cell performance analysis model includes a hydrogen fuel cell performance prediction model, and the hydrogen fuel cell performance prediction model is constructed based on the method described in the first aspect.

[0045] Optionally, the hydrogen fuel cell performance analysis device based on the AI large model further includes:

[0046] A hydrogen fuel cell professional large language model construction module, configured to obtain hydrogen fuel cell knowledge base corpus, convert the hydrogen fuel cell knowledge base corpus into question-and-answer pair data; obtain a general large language model, and fine-tune the general large language model using the question-and-answer pair data to obtain the hydrogen fuel cell professional large language model.

[0047] According to the fifth aspect of the present application, there is provided an electronic device, including: a processor, the processor is configured to execute a computer program stored in a memory, and when the computer program is executed by the processor, the method described in the first aspect or the second aspect is implemented.

[0048] According to the sixth aspect of the present application, there is provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method described in the first aspect or the second aspect is implemented.

[0049] According to the seventh aspect of the present application, there is provided a computer program product, when the computer program product runs on a computer, causing the computer to execute the method described in the first aspect or the second aspect.

[0050] The technical solution provided by the embodiments of the present application has the following advantages compared with the prior art:

[0051] Based on the physical mechanism of hydrogen fuel cell voltage loss, that is, the actual output voltage of a hydrogen fuel cell is the difference between the theoretical voltage of the hydrogen fuel cell and the activation voltage loss, ohmic voltage loss, and concentration difference voltage loss. It can be known that the total voltage loss calculated by collecting the actual output voltage of the hydrogen fuel cell includes three components: activation voltage loss, ohmic voltage loss, and concentration difference voltage loss. Therefore, the three voltage losses can be separated from the total voltage loss by independent component analysis. Furthermore, a transformer prediction model for the three components is constructed to analyze the change trend of each component loss, with higher accuracy. At the same time, in the result analysis, the loss changes of each component can be corresponded to enhance the model interpretability. Brief Description of the Drawings

[0052] The drawings herein are incorporated into and form a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0053] To more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0054] Figure 1 It is the architecture diagram of the hydrogen fuel cell performance analysis system based on the AI large model in the embodiments of the present application;

[0055] Figure 2 It is a flowchart of a method for constructing a hydrogen fuel cell performance prediction model in the embodiments of the present application;

[0056] Figure 3 It is a schematic diagram of the independent component analysis method based on physical information constraints in the embodiments of the present application;

[0057] Figure 4 It is a schematic diagram of a method for constructing a hydrogen fuel cell performance prediction model in the embodiments of the present application;

[0058] Figure 5 It is a flowchart of a method for analyzing the performance of a gas turbine based on the AI large model in the embodiments of the present application;

[0059] Figure 6 It is a schematic structural diagram of a device for constructing a hydrogen fuel cell performance prediction model in the embodiments of the present application;

[0060] Figure 7 It is a schematic structural diagram of a device for analyzing the performance of a gas turbine based on the AI large model in the embodiments of the present application;

[0061] Figure 8 It is a schematic structural diagram of an electronic device in the embodiments of the present application. Detailed Description of the Embodiments

[0062] In order to be able to more clearly understand the above objects, features, and advantages of the present application, the following will further describe the solutions of the present application. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0063] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only a part of the embodiments of the present application, rather than all of the embodiments.

[0064] As a natural language processing technology based on deep learning, large language models have powerful text understanding and generation capabilities. With the Transformer algorithm as the core, through pre-training on large-scale text data, they have achieved complex language generation, question answering and other tasks. In recent years, with the improvement of computing power and the accumulation of large amounts of data, related technologies based on large language models have developed rapidly and achieved remarkable application effects in multiple fields. Through natural language interaction, they can effectively improve the user experience and operation efficiency, and provide more intelligent and automated problem-solving solutions.

[0065] However, although general large language models have made significant progress in the field of natural language processing, there are still certain biases in the understanding of hydrogen fuel cell expertise. Their understanding of specific problems and the generation of specific answers often lack pertinence, resulting in insufficient depth and accuracy in the explanation of specific problems.

[0066] Based on this, the embodiments of the present application propose a professional large language model based on a hydrogen fuel cell professional knowledge base, namely, a hydrogen fuel cell professional large language model. The hydrogen fuel cell professional large language model can receive the operating data of hydrogen fuel cells under different working conditions, and through natural language interaction with user instructions, achieve efficient current performance calculation, future performance prediction and result analysis under complex working conditions of hydrogen fuel cells.

[0067] Figure 1 This is the architecture diagram of the hydrogen fuel cell performance analysis system based on the AI large model in the embodiments of the present application. This system takes the hydrogen fuel cell professional large language model as the core, and based on intelligent agent technology, integrates two functional modules: a hydrogen fuel cell knowledge base and a hydrogen fuel cell performance analysis model, to achieve efficient natural language interaction for hydrogen fuel cell performance analysis.

[0068] Among them, the corpus of the hydrogen fuel cell knowledge base includes: basic theoretical knowledge of key components of hydrogen fuel cells such as the composition structure, operating principle, catalyst, proton exchange membrane, gas diffusion layer, bipolar plate, etc.; basic scientific knowledge in related fields such as reaction kinetics, fluid mechanics, thermodynamics, heat transfer, etc.; and related technical materials such as hydrogen fuel cell performance calculation methods, equipment failure types, cause analysis, and control schemes. The corpus sources include authoritative professional books, technical specifications, and high-level paper literatures to ensure the accuracy of the knowledge base content. Quality assessment and data cleaning are performed on the collected corpus, and through OCR (Optical Character Recognition) technology, the text in the corpus is extracted to provide high-quality text data for the input of the large language model.

[0069] Optionally, the method for constructing a hydrogen fuel cell specialized large language model includes: obtaining the corpus of the hydrogen fuel cell knowledge base and converting the corpus of the hydrogen fuel cell knowledge base into question-and-answer pair data. Obtaining a general large language model, using the general large language model as a base, and fine-tuning the general large language model with the question-and-answer pair data to obtain a hydrogen fuel cell specialized large language model.

[0070] The hydrogen fuel cell performance analysis model includes: a hydrogen fuel cell power generation hydrogen consumption model, a hydrogen fuel cell efficiency calculation model, a hydrogen fuel cell rated power average decay rate model, and a hydrogen fuel cell performance prediction model. The hydrogen fuel cell specialized large language model uses intelligent agent technology to select an appropriate performance calculation model for analysis and calculation according to the user instruction requirements, and outputs the analysis results in natural language.

[0071] Among them, the hydrogen fuel cell power generation hydrogen consumption model can be expressed as a formula:

[0072] H is the hydrogen consumption for power generation of the hydrogen fuel cell under the rated power, with the unit of kg / kWh, m H2 is the mass of hydrogen consumed by the hydrogen fuel cell under the rated power, with the unit of kg / s, P is the power generation power of the hydrogen fuel cell, with the unit of kW.

[0073] The hydrogen fuel cell efficiency calculation model can be expressed as a formula:

[0074] η H2 is the hydrogen fuel cell efficiency, m H2,输入 、m H2,输出 are the hydrogen inlet and outlet flow rates respectively, with the unit of kg / s, Pa is the power consumption of the battery auxiliary system, Ps is the fuel cell stack, V cell is the voltage of a single hydrogen fuel cell, E0 is the theoretical voltage of the hydrogen fuel cell, and by default, it is taken as 1.23V.

[0075] The hydrogen fuel cell rated power average decay rate model can be expressed as a formula:

[0076] S is the rated power decay rate of the hydrogen fuel cell, U0 is the voltage at the rated current of a brand-new hydrogen fuel cell, and U1 is the voltage at the rated current after the hydrogen fuel cell operates to the current state according to the power generation working condition.

[0077] The main function of the hydrogen fuel cell performance prediction model is to accurately predict the remaining life of the hydrogen fuel cell. Aiming at the physical mechanism of voltage loss of the hydrogen fuel cell, a supervised decomposition analysis of the voltage loss is carried out based on the independent component analysis method, and the decomposed different independent components are input into the transformer prediction model for training to realize the performance prediction of the hydrogen fuel cell.

[0078] See Figure 2 , Figure 2 which is a flowchart of a method for constructing a hydrogen fuel cell performance prediction model in an embodiment of the present application, and may include the following steps:

[0079] Step S202: Obtain the operating data of the hydrogen fuel cell at multiple moments, and calculate the total voltage loss at each moment according to the operating data at a single moment.

[0080] The operating data of the hydrogen fuel cell during operation includes: hydrogen fuel cell voltage, current, and operating time parameters, etc.

[0081] According to the formula: Calculate the theoretical voltage E0 of the hydrogen fuel cell,

[0082] where ΔG is the change in Gibbs free energy of the chemical reaction of the hydrogen fuel cell, with the unit of J, n is the number of electrons transferred in the chemical reaction, the value of n is 2, and F is the Faraday constant 96485 C / mol.

[0083] According to the formula: E loss = mE0 - E a , calculate the total voltage loss E loss .

[0084] m is the number of single cells in the hydrogen fuel cell stack, and E a is the actual output voltage of the hydrogen fuel cell.

[0085] Step S204: Based on the physical mechanism of the hydrogen fuel cell output voltage, perform independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses of three components at each moment.

[0086] Based on the physical mechanism of the hydrogen fuel cell output voltage: V = E0 - V act - V ohm - V trans ;

[0087] where V, V act , V ohm , V trans are the output voltage, activation voltage loss, ohmic voltage loss, and concentration difference voltage loss of the hydrogen fuel cell, respectively.

[0088] It can be seen that the total voltage loss of the hydrogen fuel cell includes three parts: activation voltage loss, ohmic voltage loss, and concentration difference voltage loss. Therefore, performing independent component analysis on the total voltage loss can decompose it into activation voltage loss, ohmic voltage loss, and concentration difference voltage loss. Subsequently, corresponding transformer prediction models can be constructed for the three components respectively.

[0089] In some embodiments, in order to improve the accuracy of blind source separation of independent component analysis, one or more target moments can be selected from multiple moments, and the activation voltage loss, ohmic voltage loss, and concentration difference voltage loss at the target moments are calculated. Using the activation voltage loss, ohmic voltage loss, and concentration difference voltage loss at the target moments as the supervision constraints for the feature decomposition of the total voltage loss, the feature decomposition result of the total voltage loss of the hydrogen fuel cell is made to better conform to the physical characteristics of the polarization, ohmic, and concentration difference voltage losses.

[0090] For a hydrogen fuel cell, the theoretical calculation formula for the activation voltage loss is:

[0091]

[0092] The theoretical calculation formula for the ohmic voltage loss is: V ohm = ir;

[0093] The theoretical calculation formula for the concentration difference voltage loss is:

[0094] Where, A is the Tafel slope, R is the gas constant 8.314 J / (mol·K), T is the operating temperature of the hydrogen fuel cell, α is the electrochemical reaction transfer coefficient, which is default taken as 0.2 - 0.5, i is the working current density of the hydrogen fuel cell, i loss is the leakage current density of the hydrogen fuel cell, i0 is the exchange current density of the hydrogen fuel cell. r is the internal resistance of the hydrogen fuel cell, B is the internal diffusion constant of the battery, i l is the limiting current density of the hydrogen fuel cell.

[0095] When the relevant parameters can be obtained, the activation voltage loss, ohmic voltage loss, and concentration difference voltage loss can be directly calculated according to the above formulas. However, the parameters such as the electrochemical reaction transfer coefficient α, leakage current density i loss , exchange current density i0, limiting current density i l , internal resistance r of the hydrogen fuel cell, Tafel slope A, and internal diffusion constant B of the battery are all constant parameters related to the battery structure state and operation time. These constant parameters cannot be directly obtained, and the activation voltage loss, ohmic voltage loss, and concentration difference voltage loss cannot be calculated. Therefore, the above constant parameters can be obtained by calculation, and then the activation voltage loss, ohmic voltage loss, and concentration difference voltage loss can be calculated. Using the calculated activation voltage loss, ohmic voltage loss, and concentration difference voltage loss as constraints for independent component analysis.

[0096] The independent component analysis method based on physical information constraints is as Figure 3As shown. In the embodiments of the present application, it can be considered that the values of the constant parameters are fixed within a short period of time. Divide multiple moments into M cycles, and based on the physical mechanism of the output voltage of the hydrogen fuel cell and the actual output voltage of the hydrogen fuel cell at multiple moments in each cycle, determine multiple constant parameters corresponding to the cycle and used to calculate the voltage losses of the three components; M is a positive integer.

[0097] For example, it can be as the objective function, that is, to minimize the error between the actual voltage and the theoretical voltage, and calculate the above constant parameters by the least squares method.

[0098] Select any moment from each cycle, and calculate the reference voltage losses of the three components at the moment according to the operation data at the moment and multiple constant parameters. According to the reference voltage losses of the three components at M moments, perform independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses of the three components at each moment. That is, during the process of independent component analysis, simultaneously refer to the reference voltage losses of the three components at M moments to improve the accuracy of the decomposition results.

[0099] Optionally, it can be with the objective function: min||X - A·S|| 2 +λ||S - S T || as a constraint, perform independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses S of the three components. Among them, X represents the total voltage loss at a single moment, A represents the signal fusion matrix calculated by independent component analysis, S T represents the reference voltage losses of the three components, λ represents the weight coefficient, and the value can be 1.

[0100] Step S206, for each component, use the voltage losses of the component at consecutive N moments as input data and the voltage losses of the component at the next moment of the consecutive N moments as label data to construct the training data corresponding to the component.

[0101] For each component, N is the number of voltage losses of the component input to the transformer prediction model, and N is a positive integer. According to the voltage losses of the component at multiple moments obtained by decomposition, the voltage losses of the component at consecutive N moments can be used as a set of input data, and the voltage losses of the component at the next moment as label data to obtain multiple sets of training data.

[0102] Step S208, construct a transformer prediction model corresponding to each component, and train the transformer prediction model according to the training data corresponding to the component.

[0103] The setting of hyperparameters (such as the number of transformer layers, the dimension of the hidden layer, and the prediction step) of the transformer prediction model is generally based on the experience of the designer. In some embodiments, in order to obtain the optimal hyperparameters, during the training of the transformer prediction model according to the training data corresponding to each component, an optimization algorithm (such as the particle swarm optimization algorithm, the GA genetic algorithm, the ABC bee colony algorithm, etc.) can be used to optimize the hyperparameters of the transformer prediction model.

[0104] Figure 4 This is a schematic diagram of a method for constructing a hydrogen fuel cell performance prediction model in an embodiment of the present application. After decomposing the voltage losses of three components through the physical information-constrained independent component analysis algorithm, the voltage losses of each component are input into the corresponding transformer prediction model. The hyperparameters of the transformer prediction model (the number of transformer layers is 2-6, the dimension of the hidden layer is 2-8, and the prediction step is 1-10) are cyclically optimized through the particle swarm optimization algorithm. The population size of the particle swarm can be set to 18, the maximum number of iterations can be set to 80, and the fitness function of the particle swarm is set to the MSE (mean square error) of voltage loss prediction. When the particle swarm algorithm reaches the maximum number of iterations, the optimal hyperparameter combination of the transformer prediction model is output.

[0105] Step S210, when the termination condition is satisfied, the training ends, and a hydrogen fuel cell performance prediction model is obtained.

[0106] The method for constructing a hydrogen fuel cell performance prediction model in the embodiment of the present application is based on the physical mechanism of the voltage loss of the hydrogen fuel cell, that is, the actual output voltage of the hydrogen fuel cell is the difference between the theoretical voltage of the hydrogen fuel cell and the activation voltage loss, the ohmic voltage loss, and the concentration difference voltage loss. It can be seen that the total voltage loss calculated by collecting the actual output voltage of the hydrogen fuel cell includes three components: the activation voltage loss, the ohmic voltage loss, and the concentration difference voltage loss. The three voltage losses can be more accurately separated from the total voltage loss through the independent component analysis method based on physical information. Furthermore, a transformer prediction model for three components is constructed to analyze the change trend of each component loss, making the accuracy of each trained transformer prediction model higher. At the same time, in the result analysis, the loss changes of each component can be corresponding to enhance the model interpretability. The optimization algorithm is used to realize the automatic search and optimization of the hyperparameters of the prediction model, so as to obtain the optimal hydrogen fuel cell performance prediction model, avoiding the uncertainty of artificially setting hyperparameters and further improving the model prediction accuracy. Generally speaking, through this architecture design, the hydrogen fuel cell performance prediction model not only improves in terms of performance prediction accuracy, but also makes progress in terms of model interpretability and optimization efficiency.

[0107] See Figure 5 , Figure 5 which is a flowchart of a gas turbine performance analysis method based on an AI large model in an embodiment of the present application, including:

[0108] Step S502, obtain a user instruction.

[0109] Step S504, use a pre - constructed large language model for hydrogen fuel cells to identify the user instruction and determine whether it belongs to a hydrogen fuel cell performance analysis instruction.

[0110] The large language model for hydrogen fuel cells can, according to its own understanding ability, identify the user instruction and determine whether the user instruction belongs to a hydrogen fuel cell performance analysis instruction. If the user instruction does not belong to a hydrogen fuel cell performance analysis instruction, execute step S506; if the user instruction belongs to a hydrogen fuel cell performance analysis instruction, execute step S508.

[0111] For example, if the user instruction is "What is the activation of a hydrogen fuel cell?", the large language model for hydrogen fuel cells can identify that this user instruction is a conventional knowledge answering instruction and directly output the query result corresponding to this user instruction. If the user instruction is "How long is the remaining life of a hydrogen fuel cell?", it can be identified that this user instruction belongs to a hydrogen fuel cell performance analysis instruction.

[0112] Step S506, output the corresponding query result.

[0113] Step S508, obtain the hydrogen fuel cell operation data corresponding to the user instruction, and call the model corresponding to the user instruction from a pre - constructed hydrogen fuel cell performance analysis model to process the hydrogen fuel cell operation data and obtain a performance analysis result.

[0114] Among them, the hydrogen fuel cell performance analysis model includes a hydrogen fuel cell performance prediction model, and the hydrogen fuel cell performance prediction model is constructed based on the above - mentioned hydrogen fuel cell performance prediction model construction method.

[0115] For example, for the user instruction "How long is the remaining life of a hydrogen fuel cell?", the corresponding operation data can be obtained, such as the current operating voltage, operating current, and operating time of the hydrogen fuel cell. Then calculate the total voltage loss of the hydrogen fuel cell, and obtain the voltage losses of three components through independent component analysis. Input the voltage losses of the three components into the corresponding transformer prediction models respectively to obtain the predicted voltage losses of the three components. Calculate the predicted voltage of the hydrogen fuel cell based on the predicted voltage losses of the three components, and calculate the remaining life of the hydrogen fuel cell with the standard of regarding the voltage lower than 90% of the standard condition voltage as battery failure.

[0116] The gas turbine performance analysis method based on the AI large model in the embodiment of the present application improves the ability of the professional large language model for hydrogen fuel cells to understand problems, solve problems, and analyze results by pre - constructing a professional knowledge base for hydrogen fuel cells. By integrating the hydrogen fuel cell power generation hydrogen consumption model, the hydrogen fuel cell efficiency calculation model, the hydrogen fuel cell rated power average attenuation rate model, and the hydrogen fuel cell performance prediction model, based on intelligent agent technology, it can automatically select the most suitable calculation model according to the user input instruction and provide a detailed natural language explanation, thus solving the difficulties in selection and the complexity of analysis existing in the current performance calculation models.

[0117] Corresponding to the above - mentioned method embodiment, the embodiment of the present application also provides a device for constructing a hydrogen fuel cell performance prediction model. Refer to Figure 6 , the device 600 for constructing a hydrogen fuel cell performance prediction model includes:

[0118] The total voltage loss calculation module 602 is configured to obtain the operation data of the hydrogen fuel cell at multiple moments and calculate the total voltage loss at a moment according to the operation data at a single moment;

[0119] The independent component analysis module 604 is configured to perform independent component analysis on the total voltage losses at multiple moments based on the physical mechanism of the output voltage of the hydrogen fuel cell to obtain the voltage losses of three components at each moment; the three components include polarization, ohmic, and concentration difference;

[0120] The training data construction module 606 is configured to, for each component, construct the training data corresponding to the component by using the voltage losses of the component at consecutive N moments as input data and the voltage losses of the next moment of the component at consecutive N moments as label data; N is a positive integer;

[0121] The transformer prediction model training module 608 is configured to construct a transformer prediction model corresponding to each component and train the transformer prediction model according to the training data corresponding to the component;

[0122] The hydrogen fuel cell performance prediction model generation module 610 is configured to end the training when the termination condition is met to obtain the hydrogen fuel cell performance prediction model.

[0123] Optionally, the independent component analysis module 604 is specifically configured to divide multiple moments into M cycles, determine multiple constant parameters corresponding to each cycle for calculating the voltage losses of three components based on the physical mechanism of the output voltage of the hydrogen fuel cell and the actual output voltages of the hydrogen fuel cell at multiple moments in each cycle; M is a positive integer; select any moment from each cycle, calculate the reference voltage losses of the three components at the moment according to the operation data at the moment and the multiple constant parameters; perform independent component analysis on the total voltage losses at multiple moments based on the reference voltage losses of the three components at M moments to obtain the voltage losses of the three components at each moment.

[0124] Optionally, the independent component analysis module 604 is specifically configured to implement independent component analysis on the total voltage losses at multiple moments based on the reference voltage losses of the three components at M moments through the following steps to obtain the voltage losses of the three components at each moment:

[0125] With the objective function: min||X - A·S|| 2 +λ||S - S T || as the constraint, perform independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses S of the three components;

[0126] where X represents the total voltage loss at a single moment, A represents the signal fusion matrix calculated through independent component analysis, S T represents the reference voltage losses of the three components, and λ represents the weight coefficient.

[0127] Optionally, the transformer prediction model training module 608 is further configured to optimize the hyperparameters of the transformer prediction model by using an optimization algorithm during the process of training the transformer prediction model according to the training data corresponding to the components.

[0128] The embodiment of the present application also provides a hydrogen fuel cell performance analysis device based on an AI large model. Refer to Figure 7 , the hydrogen fuel cell performance analysis device 700 based on the AI large model includes:

[0129] A user instruction acquisition module 702, configured to acquire user instructions;

[0130] A user instruction recognition module 704, configured to use a pre-constructed professional large language model for hydrogen fuel cells to recognize user instructions and determine whether they belong to hydrogen fuel cell performance analysis instructions;

[0131] A query result output module 706, configured to output the corresponding query result if the user instruction recognition module 704 determines that the user instruction does not belong to hydrogen fuel cell performance analysis instructions;

[0132] A performance analysis structure output module 708, which is configured to, if a user instruction recognition module 704 determines that a user instruction belongs to a hydrogen fuel cell performance analysis instruction, obtain hydrogen fuel cell operation data corresponding to the user instruction, and call a model corresponding to the user instruction from a pre-constructed hydrogen fuel cell performance analysis model to process the hydrogen fuel cell operation data to obtain a performance analysis result;

[0133] Among them, the hydrogen fuel cell performance analysis model includes a hydrogen fuel cell performance prediction model, and the hydrogen fuel cell performance prediction model is constructed based on a hydrogen fuel cell performance prediction model construction method.

[0134] Optionally, the hydrogen fuel cell performance analysis device 700 based on an AI large model further includes:

[0135] A hydrogen fuel cell professional large language model construction module, which is configured to obtain hydrogen fuel cell knowledge base corpus, convert the hydrogen fuel cell knowledge base corpus into question-and-answer pair data; obtain a general large language model, and use the question-and-answer pair data to fine-tune the general large language model to obtain a hydrogen fuel cell professional large language model.

[0136] The specific details of each module or unit in the above device have been described in detail in the corresponding method, so they will not be repeated here.

[0137] It should be noted that although several modules or units of a device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0138] An embodiment of the present application also provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein, the processor is configured to execute the above hydrogen fuel cell performance prediction model construction or the gas turbine performance analysis method based on an AI large model in the present exemplary embodiment.

[0139] Refer to Figure 8 , Figure 8 which is a schematic structural diagram of an electronic device in an embodiment of the present application, and the specific implementation of the electronic device is not limited in the specific embodiments of the present application.

[0140] As Figure 8 shown, the electronic device may include: a processor 802, a communication interface 804, a memory 806, and a communication bus 808.

[0141] Among them, the processor 802, the communication interface 804, and the memory 806 communicate with each other through the communication bus 808.

[0142] The communication interface 804 is used to communicate with other electronic devices or servers.

[0143] The processor 802 is used to execute the program 810, and specifically can execute the relevant steps in the above method embodiments.

[0144] Specifically, the program 810 may include program code, and the program code includes computer operation instructions.

[0145] The processor 802 may be a central processing unit, or a specific integrated circuit, or one or more integrated circuits configured to implement the embodiments of the present application. One or more processors included in the intelligent device may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0146] The memory 806 is used to store the program 810. The memory 806 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

[0147] The program 810 is specifically used to enable the processor 802 to execute the steps in the above embodiments of the gas turbine performance analysis method based on the AI large model.

[0148] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the above-described devices and modules can refer to the corresponding process descriptions in the foregoing method embodiments, and will not be elaborated herein.

[0149] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the above method for constructing a hydrogen fuel cell performance prediction model or the gas turbine performance analysis method based on an AI large model.

[0150] It should be noted that the computer-readable storage medium shown in this application can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memories, read-only memories, erasable programmable read-only memories (EPROMs or flash memories), optical fibers, portable compact disk read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this application, a computer-readable storage medium can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable storage medium can be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, radio frequency, etc., or any suitable combination of the above.

[0151] In an embodiment of this application, a computer program product is further provided. When the computer program product runs on a computer, it causes the computer to execute the above-mentioned hydrogen fuel cell performance prediction model construction method or the gas turbine performance analysis method based on the AI large model.

[0152] It should be noted that in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including", or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or device that includes a series of elements includes not only those elements, but also other elements that are not explicitly listed, or elements that are inherent to such process, method, article, or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article, or device that includes the element.

[0153] The above are only specific embodiments of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application will not be limited to these embodiments described herein, but rather will conform to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing a performance prediction model of a hydrogen fuel cell, characterized in that, Comprising: Obtain the operation data of the hydrogen fuel cell at multiple moments, and calculate the total voltage loss at the moment according to the operation data at a single moment; Based on the physical mechanism of the output voltage of the hydrogen fuel cell, perform independent component analysis on the total voltage losses at multiple moments to obtain the voltage losses of three components at each moment; the three components include polarization, ohmic, and concentration difference; For each component, use the voltage losses of the component at N consecutive moments as input data and the voltage losses of the component at the next moment of the N consecutive moments as label data to construct the training data corresponding to the component; N is a positive integer; Construct a transformer prediction model corresponding to each component, and train the transformer prediction model according to the training data corresponding to the component; When the termination condition is met, the training ends to obtain a hydrogen fuel cell performance prediction model; Wherein, the performing independent component analysis on the total voltage losses at multiple moments based on the physical mechanism of the output voltage of the hydrogen fuel cell to obtain the voltage losses of three components at each moment includes: Divide the multiple moments into M cycles, and based on the physical mechanism of the output voltage of the hydrogen fuel cell and the actual output voltages of the hydrogen fuel cell at multiple moments in each cycle, determine multiple constant parameters corresponding to the cycle for calculating the voltage losses of the three components; M is a positive integer; Select any moment from each cycle, and calculate the reference voltage losses of the three components at the moment according to the operation data at the moment and the multiple constant parameters; Perform independent component analysis on the total voltage losses at multiple moments according to the reference voltage losses of the three components at M moments to obtain the voltage losses of the three components at each moment.

2. The method according to claim 1, wherein The performing independent component analysis on the total voltage losses at multiple moments according to the reference voltage losses of the three components at M moments to obtain the voltage losses of the three components at each moment includes: With the objective function: min||X - AS|| 2 + λ||S - S T || as the constraint, perform independent component analysis on the total voltage loss at multiple moments to obtain the voltage losses S of the three components; Among them, X represents the total voltage loss at a single moment, A represents the signal fusion matrix calculated by independent component analysis, and S T represents the reference voltage losses of three components, and λ represents the weight coefficient.

3. The method according to claim 1, wherein The method further includes: During the process of training the transformer prediction model according to the training data corresponding to the component, use an optimization algorithm to optimize the hyperparameters of the transformer prediction model.

4. A method for analyzing the performance of a hydrogen fuel cell based on a large AI model, characterized in that, Comprising: Obtain a user instruction, use a pre-constructed hydrogen fuel cell professional large language model to identify the user instruction, and determine whether it belongs to a hydrogen fuel cell performance analysis instruction; If the user instruction does not belong to a hydrogen fuel cell performance analysis instruction, output the corresponding query result; If the user instruction belongs to a hydrogen fuel cell performance analysis instruction, obtain the hydrogen fuel cell operation data corresponding to the user instruction, and call the model corresponding to the user instruction from the pre-constructed hydrogen fuel cell performance analysis model to process the hydrogen fuel cell operation data to obtain a performance analysis result; Wherein, the hydrogen fuel cell performance analysis model includes a hydrogen fuel cell performance prediction model, and the hydrogen fuel cell performance prediction model is constructed based on the method according to any one of claims 1 to 3.

5. The method according to claim 4, wherein The construction method of the hydrogen fuel cell professional large language model includes: Obtain hydrogen fuel cell knowledge base corpus, and convert the hydrogen fuel cell knowledge base corpus into question-and-answer pair data; Obtain a general large language model and fine-tune the general large language model using the Q&A pair data to obtain the hydrogen fuel cell professional large language model.

6. An apparatus for constructing a performance prediction model of a hydrogen fuel cell, characterized in that, It includes: A total voltage loss calculation module, configured to obtain the operation data of the hydrogen fuel cell at multiple moments and calculate the total voltage loss at the moment according to the operation data at a single moment; An independent component analysis module, configured to perform independent component analysis on the total voltage losses at multiple moments based on the physical mechanism of the output voltage of the hydrogen fuel cell to obtain the voltage losses of three components at each moment; the three components include polarization, ohmic, and concentration difference; A training data construction module, configured to, for each component, use the voltage losses of the component at N consecutive moments as input data and the voltage losses of the component at the next moment of the N consecutive moments as label data to construct the training data corresponding to the component; N is a positive integer; A transformer prediction model training module, configured to construct a transformer prediction model corresponding to each component and train the transformer prediction model according to the training data corresponding to the component; A hydrogen fuel cell performance prediction model generation module, configured to end the training when the termination condition is met to obtain a hydrogen fuel cell performance prediction model; Wherein, the independent component analysis module is specifically configured to divide the multiple moments into M cycles, and based on the physical mechanism of the output voltage of the hydrogen fuel cell and the actual output voltage of the hydrogen fuel cell at multiple moments in each cycle, determine multiple constant parameters corresponding to the cycle for calculating the voltage losses of the three components; M is a positive integer; Select any moment from each cycle, and calculate the reference voltage losses of the three components at the moment according to the operation data at the moment and the multiple constant parameters; perform independent component analysis on the total voltage losses at multiple moments according to the reference voltage losses of the three components at M moments to obtain the voltage losses of the three components at each moment.

7. The device according to claim 6, wherein The independent component analysis module is specifically configured to implement independent component analysis on the total voltage losses at multiple moments according to the reference voltage losses of the three components at M moments to obtain the voltage losses of the three components at each moment through the following steps: With the objective function: min||X - A·S|| 2 + λ||S - S T || as the constraint, perform independent component analysis on the total voltage loss at multiple moments to obtain the voltage losses S of the three components; Among them, X represents the total voltage loss at a single moment, A represents the signal fusion matrix calculated by independent component analysis, and S T represents the reference voltage losses of three components, and λ represents the weight coefficient.

8. A hydrogen fuel cell performance analysis device based on an AI large model, characterized in that, It includes: A user instruction acquisition module, configured to acquire user instructions; A user instruction recognition module, configured to use the pre-constructed hydrogen fuel cell professional large language model to recognize the user instructions and determine whether they belong to hydrogen fuel cell performance analysis instructions; A query result output module, configured to output the corresponding query result if the user instruction recognition module determines that the user instruction does not belong to hydrogen fuel cell performance analysis instructions; A performance analysis result output module, configured to, if the user instruction recognition module determines that the user instruction belongs to hydrogen fuel cell performance analysis instructions, acquire the hydrogen fuel cell operation data corresponding to the user instruction and call the model corresponding to the user instruction from the pre-constructed hydrogen fuel cell performance analysis model to process the hydrogen fuel cell operation data to obtain a performance analysis result; Among them, the hydrogen fuel cell performance analysis model includes a hydrogen fuel cell performance prediction model, and the hydrogen fuel cell performance prediction model is constructed based on the method described in any one of claims 1 to 3.

Images