Fuel cell system operation scheme prediction method and device and computer equipment

Abstract

The invention relates to a fuel cell system operation scheme prediction method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a preset operation parameter item of a to-be-predicted fuel cell system, and acquiring a preset value range and preset value precision of the preset operation parameter item; randomly generating a plurality of groups of operation parameter values of the preset operation parameter item according to a preset value range and preset value precision; respectively converting each operation parameter value into gene information of individuals in a genetic algorithm to obtain an initial population; and obtaining preset constraint condition information and a plurality of preset objective functions, and obtaining a predicted operation scheme of the to-be-predicted fuel cell system according to a preset constraint condition, each preset objective function and the initial population. By adopting the method, the running scheme prediction efficiency of the fuel cell system can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, apparatus, computer device, computer-readable storage medium and computer program product for predicting an operation plan of a fuel cell system. Background Art

[0002] In the research and development and application of fuel cell systems, the performance parameters of fuel cell systems, such as current density, hydrogen utilization, system efficiency and durability, directly determine their technological advancement and market competitiveness.

[0003] However, traditional optimization methods usually use simulation tests as an alternative. Simulation tests usually simply divide each parameter evenly and combine the calculations. This fails to fully consider the differences in the impact of different parameters on system performance and economy, adds unnecessary computational burden, and results in low optimization efficiency. The efficiency of predicting the operation plan of the fuel cell system is poor. Summary of the invention

[0004] Based on this, it is necessary to provide a fuel cell system operation plan prediction method, device, computer equipment, computer readable storage medium and computer program product that can improve the efficiency of fuel cell system operation plan prediction in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for predicting an operation scheme of a fuel cell system, comprising:

[0006] Obtaining a preset operating parameter item of the fuel cell system to be predicted, and obtaining a preset value range and a preset value accuracy of the preset operating parameter item;

[0007] Randomly generate multiple groups of operating parameter values ​​for the preset operating parameter items according to the preset value range and the preset value accuracy;

[0008] Converting each of the operating parameter values ​​into genetic information of individuals in the genetic algorithm to obtain an initial population;

[0009] Preset constraint information and a plurality of preset objective functions are obtained, and a predicted operation plan of the fuel cell system to be predicted is obtained according to the preset constraint information, each of the preset objective functions and the initial population.

[0010] In one embodiment, the step of converting each of the operating parameter values ​​into gene information of an individual in a genetic algorithm comprises:

[0011] Determine the encoding length corresponding to the preset operating parameter item according to the preset value range and the preset value precision;

[0012] Determining the encoding relationship of the preset operating parameter items according to the encoding length;

[0013] According to the coding relationship, each of the operating parameter values ​​is converted into coding information as the gene information of the individual in the genetic algorithm.

[0014] In one embodiment, obtaining the predicted operation scheme of the fuel cell system to be predicted according to the preset constraint conditions, each of the preset objective functions and the initial population includes:

[0015] According to the preset constraint conditions and each of the preset objective functions, obtaining fitness functions of individuals in multiple genetic algorithms;

[0016] Taking the initial population as the current population, calculating the fitness of each individual in the current population;

[0017] Generate a descendant population according to the fitness and the current population, use the descendant population as a new current population, and return to execute the step of calculating the fitness of each individual in the current population until a descendant population that meets the preset conditions is obtained;

[0018] A predicted operation scheme of the fuel cell system to be predicted is determined according to the offspring population that meets the preset conditions.

[0019] In one embodiment, generating a descendant population according to the fitness and the current population includes:

[0020] According to the fitness, the individuals in the current population are non-dominated and sorted to determine the non-dominated level of each individual in the current population; each non-dominated level contains at least one individual;

[0021] From the current population, remove the individuals with the lowest non-dominant level to obtain a parent population;

[0022] A child population is generated according to the parent population.

[0023] In one embodiment, generating a child population according to the parent population includes:

[0024] According to the parent population, a first child population is obtained through a crossover operation and a mutation operation;

[0025] Calculate the crowding distance of the individual with the highest non-dominated level in the parent population;

[0026] According to the crowding distance, a preset number of individuals are selected from the individuals with the highest non-dominated level in the parent population, and the individuals are integrated into a second offspring population;

[0027] The first progeny population and the second progeny population are collectively referred to as the progeny population.

[0028] In one embodiment, calculating the fitness of each individual in the current population includes:

[0029] In a physical simulation tool, constructing a simulation model of the fuel cell system to be predicted;

[0030] In the physical simulation tool, the fitness of each individual in the current population is calculated according to the simulation model;

[0031] The step of generating a descendant population according to the fitness and the current population includes:

[0032] An optimization prediction tool is used to generate a progeny population according to the fitness and the current population.

[0033] In a second aspect, the present application also provides a fuel cell system operation scheme prediction device, comprising:

[0034] An information acquisition module, used to acquire a preset operating parameter item of the fuel cell system to be predicted, and to acquire a preset value range and a preset value accuracy of the preset operating parameter item;

[0035] A parameter generation module, used for randomly generating a plurality of groups of operating parameter values ​​of the preset operating parameter items according to the preset value range and the preset value precision;

[0036] A gene conversion module, used to convert each of the operating parameter values ​​into gene information of individuals in the genetic algorithm to obtain an initial population;

[0037] The scheme prediction module is used to obtain preset constraint information and multiple preset objective functions, and obtain the predicted operation scheme of the fuel cell system to be predicted based on the preset constraint information, each preset objective function and the initial population.

[0038] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0039] Obtaining a preset operating parameter item of the fuel cell system to be predicted, and obtaining a preset value range and a preset value accuracy of the preset operating parameter item;

[0040] Randomly generate multiple groups of operating parameter values ​​for the preset operating parameter items according to the preset value range and the preset value accuracy;

[0041] Converting each of the operating parameter values ​​into genetic information of individuals in the genetic algorithm to obtain an initial population;

[0042] Preset constraint information and a plurality of preset objective functions are obtained, and a predicted operation plan of the fuel cell system to be predicted is obtained according to the preset constraint information, each of the preset objective functions and the initial population.

[0043] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0044] Obtaining a preset operating parameter item of the fuel cell system to be predicted, and obtaining a preset value range and a preset value accuracy of the preset operating parameter item;

[0045] Randomly generate multiple groups of operating parameter values ​​for the preset operating parameter items according to the preset value range and the preset value accuracy;

[0046] Converting each of the operating parameter values ​​into genetic information of individuals in the genetic algorithm to obtain an initial population;

[0047] Preset constraint information and a plurality of preset objective functions are obtained, and a predicted operation plan of the fuel cell system to be predicted is obtained according to the preset constraint information, each of the preset objective functions and the initial population.

[0048] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0049] Obtaining a preset operating parameter item of the fuel cell system to be predicted, and obtaining a preset value range and a preset value accuracy of the preset operating parameter item;

[0050] Randomly generate multiple groups of operating parameter values ​​for the preset operating parameter items according to the preset value range and the preset value accuracy;

[0051] Converting each of the operating parameter values ​​into genetic information of individuals in the genetic algorithm to obtain an initial population;

[0052] Preset constraint information and a plurality of preset objective functions are obtained, and a predicted operation plan of the fuel cell system to be predicted is obtained according to the preset constraint information, each of the preset objective functions and the initial population.

[0053] In the above-mentioned fuel cell system operation plan prediction method, first, the preset operating parameter items of the fuel cell system to be predicted are obtained, as well as the preset value range and preset value accuracy of the preset operating parameter items. By obtaining the preset operating parameter items and their value range and value accuracy, it is ensured that the optimization process can be carried out based on the actual system operation data and boundary conditions, thereby ensuring the effectiveness and accuracy of the entire optimization process, and being able to reflect the actual operating status of the system through precise values. This precision setting avoids the accuracy of the optimization results being affected by parameter values ​​that are too coarse or insufficiently refined, and ensures the feasibility and stability of the generated operation plan in practice; then, according to the preset value range and preset value precision, multiple groups of operation parameter values ​​for the preset operation parameter items are randomly generated. By randomly generating multiple groups of operation parameter values, the diversity of system parameter combinations is increased, and the parameter limitations caused by simple uniform division in traditional simulation methods are avoided. It can cover a wider solution space, help explore the system performance under different parameter combinations, and improve the comprehensiveness of the optimization process and the diversity of the final solution. At the same time, through the precision setting, the invalid calculation burden is effectively avoided, and the suboptimal solutions caused by too low precision and the excessive calculation caused by too high precision are reduced; then, each operation parameter value is converted into the genetic information of the individual in the genetic algorithm to obtain the initial population, and the operation parameter value is converted into the genetic algorithm. The genetic information in the gene is used to generate the initial population, which provides an effective decoding method, so that the genetic algorithm can use genetic operations such as crossover and mutation to generate new parameter combinations, and gradually optimize the solution through the evolution process; finally, the preset constraint information and multiple preset objective functions are obtained, and the predicted operation plan of the fuel cell system to be predicted is obtained according to the preset constraints, each preset objective function and the initial population. The genetic algorithm combines the preset constraints and multiple objective functions in the optimization process, so that it can adaptively adjust the evolution direction and concentrate resources on the key parameters that have a greater impact on performance and economy. Then, in the iterative process, the changes in those parameters that have a smaller impact on system performance will gradually decrease, thereby reducing the calculation requirements for these minor parameters and improving the efficiency of the algorithm. Through fitness evaluation, suboptimal solutions are gradually eliminated, and resources are concentrated on more potential solutions, avoiding the inefficient process of equal calculation of all parameters. In the above method, through the selective optimization mechanism of the genetic algorithm, solutions that do not meet the target can be gradually eliminated, avoiding indiscriminate calculation of each parameter combination. Compared with the uniform division in the traditional method, combined with the value accuracy of each preset operating parameter item, the genetic algorithm gradually reduces the focus on minor parameters through selection, crossover, and mutation in the evolutionary process, focusing on the optimization of key parameters, significantly reducing unnecessary computational burden. Ultimately, the prediction efficiency of the fuel cell system operation plan and the utilization rate of computing resources are improved, making the entire optimization process more computationally efficient while maintaining accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0055] FIG1 is a schematic flow chart of a method for predicting an operation scheme of a fuel cell system in one embodiment;

[0056] FIG2 is a schematic flow chart of the steps of obtaining a prediction operation plan in one embodiment;

[0057] FIG3 is a schematic flow chart of a method for predicting an operation plan of a fuel cell system in another embodiment;

[0058] FIG4 is a schematic diagram of a fuel cell system to be predicted in one embodiment;

[0059] FIG5 is a schematic flow chart of a method for predicting an operation plan of a fuel cell system in one embodiment;

[0060] FIG6 is a structural block diagram of a fuel cell system operation scheme prediction device in one embodiment;

[0061] FIG. 7 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0063] In one embodiment, as shown in FIG1 , a method for predicting the operation scheme of a fuel cell system is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptops, smart phones, and tablet computers. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. In this embodiment, the method includes the following steps:

[0064] Step S101 : obtaining a preset operating parameter item of a fuel cell system to be predicted, and obtaining a preset value range and a preset value accuracy of the preset operating parameter item.

[0065] Among them, the preset operating parameter items refer to the parameter items that affect the key factors such as the performance and economy of the fuel cell system during operation, usually including current density, operating temperature, inlet temperature, water-carbon ratio, etc., as well as the basic parameters of the fuel cell: number of cells, flow channel size, etc. These parameter items play an important role in the design and optimization of the system.

[0066] The preset value range refers to the possible range of values ​​that each operating parameter item may take in actual operation, which is used to limit the exploration interval of each parameter in the optimization process of the algorithm to ensure that the calculation results are within a reasonable physical and engineering range. The preset value accuracy refers to the minimum unit change or allowed numerical accuracy of the operating parameter item in the system, which affects the precision of the parameter representation in the optimization process and determines the calculation accuracy of the algorithm and the degree of refinement of the solution.

[0067] Exemplarily, when the terminal obtains the preset operating parameter items of the fuel cell system, it extracts these parameter items from the system design documents, experimental data or the existing system operation history. For each parameter item, the terminal also obtains its preset value range and value accuracy. For example, for the current density parameter, the value range may be 0 to 100A / cm², and the value accuracy may be 0.01A / cm². The parameter range of hydrogen utilization may be 50% to 90%, and its value accuracy is 0.1%. These value ranges and accuracy ensure that in the subsequent optimization calculations, the generated parameter values ​​are consistent with the actual operation of the physical system and can achieve the required calculation accuracy, so as to avoid the problem of too coarse or too fine numerical representation during the optimization process.

[0068] Step S102: randomly generate multiple groups of operating parameter values ​​for preset operating parameter items according to a preset value range and a preset value precision.

[0069] The operating parameter value refers to the specific value of each set of parameter items generated according to the preset value range and preset value accuracy, which is used for individual gene expression in the subsequent genetic algorithm.

[0070] Exemplarily, the terminal uses a random number generation algorithm or a Monte Carlo method to randomly generate multiple sets of operating parameter values ​​according to the preset value range and precision of each parameter. The generation process takes into account the upper and lower limits of each parameter, and discretizes the parameters according to the precision requirements. For example, if the current density ranges from 0 to 100A / cm² and the precision is 0.01A / cm², the terminal will generate a random number with two decimal places between 0 and 100 as the value of the parameter. Similarly, the hydrogen utilization rate ranges from 50% to 90% with a precision of 0.1%, and the parameter value generated by the terminal will be taken within this range in steps of 0.1%. This generation method ensures that each parameter value varies within a reasonable range and meets specific precision requirements.

[0071] Step S103, converting each operating parameter value into the gene information of individuals in the genetic algorithm to obtain an initial population.

[0072] Among them, gene information is the coding form used to represent individuals in genetic algorithms. Through the coding operation, each operating parameter value is converted into a gene representation that can be processed by the genetic algorithm. The initial population is the first generation solution set of the genetic algorithm, which contains multiple individuals. Each individual is represented by a set of gene information and serves as the starting point of the optimization process.

[0073] Exemplarily, after obtaining multiple sets of operating parameter values, the terminal converts these parameter values ​​into genetic information in the genetic algorithm through specific coding rules (such as binary coding, real number coding or other suitable problem coding forms). The terminal first determines the coding method of each operating parameter item to ensure that the value of each parameter item can be processed in a unified genetic expression method. For example, for continuous parameters such as current density, the terminal can convert it into a real number coding form, or use binary coding for discrete parameters. Each set of converted genetic information is combined to form an individual representing the operating plan of the fuel cell system, which in turn constitutes the entire initial population. The diversity of the initial population provides a rich solution space basis for subsequent genetic operations (such as selection, crossover, and mutation), helping the genetic algorithm to effectively explore the optimal solution.

[0074] Step S104, obtaining preset constraint information and a plurality of preset objective functions, and obtaining a predicted operation plan of the fuel cell system to be predicted according to the preset constraint information, the preset objective functions and the initial population.

[0075] The preset constraint information refers to the physical, engineering or economic restrictions that the fuel cell system must meet during operation, such as boundary conditions such as temperature, pressure, fuel utilization, etc. The preset objective function is a number of performance indicators used to measure the optimization results, usually involving multiple aspects such as system efficiency, economy, and durability.

[0076] Exemplarily, the terminal first obtains the constraint information of the fuel cell system to ensure that each parameter combination during operation meets the basic operating restrictions. For example, the constraints may include the temperature difference of each heat exchanger, etc. Then, the terminal obtains multiple preset objective functions, which are used to evaluate the pros and cons of each individual, which may include multi-objective functions such as maximizing system efficiency, minimizing fuel consumption, or minimizing costs. The terminal compares each individual in the initial population with these objective functions and constraints, calculates the fitness of each individual, and uses a genetic algorithm to iteratively optimize the population. The terminal selects individuals with better performance according to the fitness value for genetic operations (such as selection, crossover, and mutation), generates a new generation of individuals, and continuously updates the population. After several generations of iterative optimization, the terminal finally selects individuals that meet all constraints and perform better on multiple preset objective functions, and outputs them as predicted operation plans.

[0077] In the above-mentioned fuel cell system operation plan prediction method, first, the preset operating parameter items of the fuel cell system to be predicted are obtained, as well as the preset value range and preset value accuracy of the preset operating parameter items. By obtaining the preset operating parameter items and their value range and value accuracy, it is ensured that the optimization process can be carried out based on the actual system operation data and boundary conditions, thereby ensuring the effectiveness and accuracy of the entire optimization process, and being able to reflect the actual operating status of the system through precise values. This precision setting avoids the accuracy of the optimization results being affected by parameter values ​​that are too coarse or insufficiently refined, and ensures the feasibility and stability of the generated operation plan in practice; then, according to the preset value range and preset value precision, multiple groups of operation parameter values ​​for the preset operation parameter items are randomly generated. By randomly generating multiple groups of operation parameter values, the diversity of system parameter combinations is increased, and the parameter limitations caused by simple uniform division in traditional simulation methods are avoided. It can cover a wider solution space, help explore the system performance under different parameter combinations, and improve the comprehensiveness of the optimization process and the diversity of the final solution. At the same time, through the precision setting, the invalid calculation burden is effectively avoided, and the suboptimal solutions caused by too low precision and the excessive calculation caused by too high precision are reduced; then, each operation parameter value is converted into the genetic information of the individual in the genetic algorithm to obtain the initial population, and the operation parameter value is converted into the genetic algorithm. The genetic information in the gene is used to generate the initial population, which provides an effective decoding method, so that the genetic algorithm can use genetic operations such as crossover and mutation to generate new parameter combinations, and gradually optimize the solution through the evolution process; finally, the preset constraint information and multiple preset objective functions are obtained, and the predicted operation plan of the fuel cell system to be predicted is obtained according to the preset constraints, each preset objective function and the initial population. The genetic algorithm combines the preset constraints and multiple objective functions in the optimization process, so that it can adaptively adjust the evolution direction and concentrate resources on the key parameters that have a greater impact on performance and economy. Then, in the iterative process, the changes in those parameters that have a smaller impact on system performance will gradually decrease, thereby reducing the calculation requirements for these minor parameters and improving the efficiency of the algorithm. Through fitness evaluation, suboptimal solutions are gradually eliminated, and resources are concentrated on more potential solutions, avoiding the inefficient process of equal calculation of all parameters. In the above method, through the selective optimization mechanism of the genetic algorithm, solutions that do not meet the target can be gradually eliminated, avoiding indiscriminate calculation of each parameter combination. Compared with the uniform division in the traditional method, combined with the value accuracy of each preset operating parameter item, the genetic algorithm gradually reduces the focus on minor parameters through selection, crossover, and mutation in the evolutionary process, focusing on the optimization of key parameters, significantly reducing unnecessary computational burden. Ultimately, the prediction efficiency of the fuel cell system operation plan and the utilization rate of computing resources are improved, making the entire optimization process more computationally efficient while maintaining accuracy.

[0078] In an exemplary embodiment, the above-mentioned step S103 converts each operating parameter value into genetic information of an individual in a genetic algorithm, and also includes: determining the coding length corresponding to a preset operating parameter item based on a preset value range and a preset value accuracy; determining the coding relationship of the preset operating parameter items based on the coding length; and converting each operating parameter value into coding information based on the coding relationship as the genetic information of an individual in the genetic algorithm.

[0079] Exemplarily, the terminal first determines the corresponding coding length according to the value range and value precision of each preset operation parameter item. For example, if the value range of a certain operation parameter is 0 to 100 and the value precision is 0.01, the terminal will need enough bits to represent each possible value between 0 and 100. This means that for this parameter, the coding length may need to be longer to ensure that all precise values ​​can be represented. Next, the terminal determines the coding relationship of the parameter item based on the coding length. For some parameters, binary coding can be used to convert the numerical value into a corresponding binary bit string; for continuous parameters with higher precision requirements, real number coding may be required for representation. The terminal converts each operation parameter value into corresponding coding information according to the coding relationship, thereby generating individual genetic information, which will be used as the individual's genetic representation in the genetic algorithm and participate in the subsequent optimization process.

[0080] In this embodiment, by determining the encoding length according to the value range and precision of each parameter, the accurate expression of different parameter items is ensured, so that the optimization step size of the parameter in the genetic algorithm can be flexibly adjusted according to the importance of its impact on the system. This not only avoids the same degree of treatment of all parameters in the traditional uniform parameter adjustment method, but also enables more effective optimization according to the different characteristics of the parameter items. For key parameters, a smaller encoding step size can ensure fine adjustment, while for minor parameters with less impact, a larger step size speeds up the optimization process and reduces unnecessary computational burden. Ultimately, this optimization method improves the efficiency and accuracy of the prediction of the fuel cell system operation plan, allowing the system to converge to the optimal solution faster and more accurately.

[0081] In an exemplary embodiment, the above step S101 obtains the preset value range and preset value precision of the preset operating parameter item, and also includes: performing sensitivity analysis on each preset operating parameter item to obtain analysis results; and determining the preset value precision corresponding to each preset operating parameter item based on the analysis results.

[0082] Exemplarily, after obtaining the value range of each preset operating parameter item, the terminal first performs a sensitivity analysis on each parameter. The terminal adjusts the value of each parameter and observes its impact on the performance of the fuel cell system (such as efficiency, life or cost), so as to determine which parameters have a significant impact on the results and which parameters have a smaller impact. For example, if the change in current density has a greater impact on the system efficiency, the parameter requires a higher value accuracy to ensure that it can be fine-tuned during the optimization process. For parameters with less impact, such as certain environmental condition variables, a lower value accuracy can be used to avoid excessive calculation burden. According to the results of the sensitivity analysis, the terminal determines the preset value accuracy for each parameter item. For example, the value accuracy of current density can be set to 0.01A / cm² to ensure that high-precision adjustments can be made during the optimization process, while the accuracy of hydrogen utilization may be set to 0.1% to reduce calculation overhead. In this way, parameters with high sensitivity obtain higher accuracy, while parameters with low sensitivity can be optimized with lower accuracy, thereby improving the overall calculation efficiency of the system.

[0083] In this embodiment, by performing sensitivity analysis on each preset operating parameter item and adjusting the value accuracy according to the analysis results, the computing resources can be reasonably allocated according to the importance of each parameter. Sensitivity analysis ensures that key parameters are accurately adjusted during the optimization process to improve system performance and the accuracy of optimization results. At the same time, for minor parameters that have a smaller impact on the system, by reducing their value accuracy, unnecessary computing burden is reduced and the efficiency of system resource utilization is optimized. This not only speeds up the optimization speed of the fuel cell system operation plan, but also improves the accuracy and reliability of the overall prediction plan.

[0084] In an exemplary embodiment, as shown in FIG. 2 , the above step S104 obtains a predicted operation scheme of the fuel cell system to be predicted according to preset constraints, various preset objective functions and the initial population, and can also be implemented by the following steps:

[0085] Step S201, obtaining fitness functions of individuals in multiple genetic algorithms according to preset constraints and preset objective functions;

[0086] Step S202, taking the initial population as the current population, and calculating the fitness of each individual in the current population;

[0087] Step S203, generating a descendant population according to the fitness and the current population, taking the descendant population as the new current population, and returning to the step of calculating the fitness of each individual in the current population until a descendant population that meets the preset conditions is obtained;

[0088] Step S204: determining a predicted operation plan of the fuel cell system to be predicted based on the offspring population that meets the preset conditions.

[0089] Exemplarily, the terminal first defines a fitness function according to each preset objective function and constraint. Each fitness function reflects the performance of an individual on the objective function and ensures that all solutions meet the basic constraints. For example, if the system requires maximization of efficiency and minimization of cost, the fitness function can include these two goals at the same time, and punish individuals that violate the constraints, so as to ensure that the individual always meets the basic operating restrictions, that is, the constraints are taken as a separate goal. If the constraints are not met, the fitness corresponding to this goal will be extremely low, and if they are met, the fitness corresponding to this goal will be extremely high. Then, the terminal brings the individuals of the initial population into the fitness function one by one for calculation, and evaluates the performance of each individual on the objective function. The result of the calculation is the fitness value of each individual, which is used to measure the performance of the individual in multi-objective optimization. For example, an individual with a higher fitness value may represent a higher system efficiency and lower cost, which meets the preset goals. Then, the terminal selects the current population according to the fitness value of each individual, and generates a descendant population through crossover and mutation operations in the genetic algorithm. The selection operation tends to select individuals with higher fitness, so that they have a greater probability of participating in the generation of the next generation, thereby gradually eliminating unfit individuals. After the offspring population is generated, the terminal uses it as the new current population and repeats the steps of calculating the fitness of each individual until the optimal solution that meets the preset conditions is found. This process will continue until the set stop condition (such as convergence of fitness value or maximum number of iterations) is reached. After the terminal obtains the optimal offspring population that meets the preset objective function and constraints through multiple generations of iterations, it will be used as the final solution set, and the final operation plan will be selected from it as the operation plan of the fuel cell system to be predicted. This plan maximizes performance or economy while satisfying the system constraints, and ultimately provides a valuable reference for the actual operation of the fuel cell system.

[0090] Among them, the final operation plan is selected from the final solution set. Specifically, it can be: due to the existence of multiple objective functions, different solutions may perform differently on different objectives, so there will not be a single "optimal solution", but a Pareto frontier will be formed, that is, a set of solutions, each of which is not dominated by other solutions on one or more objectives. Then, the weighted sum of each solution can be calculated by assigning different weights to each objective function. Then, from the Pareto optimal solution set, the solution with the smallest (or largest) weighted sum is selected as the final plan. It is also possible to select solutions based on the priority of a certain goal according to user needs. It is also possible to define the ideal value of multiple objective functions (i.e., the optimal value of the objective function) as an ideal point, calculate the distance of each solution to the ideal point, and select the solution with the shortest distance as the final plan.

[0091] In this embodiment, through the above steps, the iterative optimization mechanism of the genetic algorithm can be effectively used to gradually screen out the optimal fuel cell system operation plan. By generating the offspring population through multiple iterations and calculating its fitness, the system can gradually optimize the solution set under multi-objective constraints, and the final solution can not only meet the constraints, but also perform well in multiple performance goals. This method not only improves the optimization efficiency of the system, but also ensures the accuracy and reliability of the final solution.

[0092] In an exemplary embodiment, the above step S203 generates a child population based on the fitness and the current population, and also includes: according to the fitness, sorting the individuals in the current population by non-dominance to determine the non-dominated level of each individual in the current population; each non-dominated level contains at least one individual; from the current population, eliminating the individual with the lowest non-dominated level to obtain the parent population; and generating a child population based on the parent population.

[0093] Among them, non-dominated sorting refers to dividing individuals in a population into different levels according to their performance on multiple objective functions.

[0094] Among them, the non-dominated level represents the ranking level of each individual in the population. Individuals in higher levels perform better and are not dominated by other individuals on all goals.

[0095] Exemplarily, the terminal first performs non-dominated sorting on all individuals in the current population. The principle of non-dominated sorting is that the first layer (i.e., individuals with the highest non-dominated level) includes all individuals that are not completely dominated by other individuals on any target, that is, these individuals perform well on one or more targets. The second layer includes individuals that are not dominated by the remaining individuals after the first layer individuals are removed, and so on. In this way, the terminal can divide the individuals in the population into multiple levels according to the non-dominated sorting, and the individuals in each level have similar advantages and disadvantages. The individuals in each non-dominated level determine their ranking according to their multi-objective fitness. The terminal assigns these individuals to different levels, and the individuals in non-dominated level 1 perform best, the individuals in non-dominated level 2 perform second, and so on. These levels are used to guide the selection and elimination of individuals to ensure that individuals with excellent performance are retained first during the evolution process. According to the fitness and non-dominated level sorting, the terminal removes the individuals with the lowest non-dominated level in the current population. These individuals perform poorly on all targets, so they are no longer retained in the next generation population. By removing individuals at the lowest level, the overall quality of the population can be effectively improved, and individuals with better performance on multiple objectives can be retained. After removing individuals with poor performance, the terminal uses the retained parent population to generate a new offspring population. The process of generating offspring includes the following steps: Selection: Select according to the parent individuals with higher non-dominated levels, and give priority to individuals with better fitness. Crossover: Perform crossover operations between parent individuals to generate new individuals. The crossover operation combines the genetic information of the parent individuals, so that the offspring individuals may show better characteristics. Mutation: Perform mutation operations on some offspring individuals, introduce new gene combinations, ensure the diversity of the population, and avoid falling into local optimality. The generated offspring population will serve as the new current population and continue to participate in the next round of iterative optimization.

[0096] In this embodiment, by performing non-dominated sorting on the current population, the pros and cons of individuals can be effectively distinguished, and a suitable parent population can be selected according to its non-dominated level. Eliminating individuals with poor performance not only improves the overall quality of the population, but also generates a more diverse and potential offspring population through crossover and mutation operations, ultimately helping the genetic algorithm to find the optimal solution faster in multi-objective optimization problems. This ensures that the genetic algorithm can balance multiple objectives during the optimization process and gradually screen out the optimal fuel cell system operation plan.

[0097] In an exemplary embodiment, the above-mentioned generation of a child population based on a parent population also includes: obtaining a first child population based on the parent population through a crossover operation and a mutation operation; calculating the crowding distance of the individual with the highest non-dominated level in the parent population; based on the crowding distance, screening a preset number of individuals from the individuals with the highest non-dominated level in the parent population, and integrating them into a second child population; and combining the first child population and the second child population as a whole into the child population.

[0098] Among them, the crowding distance is used to measure the sparse distribution of individuals in the population in the solution space.

[0099] Exemplarily, the terminal first performs a crossover operation on the individuals in the parent population, and generates a new individual by combining the genetic information of the two parent individuals. The new individual inherits part of the genetic information of the parent, but has a new gene combination, which improves the diversity of the population. Next, the terminal performs a mutation operation, and performs random gene mutations on a few individuals to introduce a new solution space and prevent the population from falling into a local optimal solution. After the crossover and mutation operations, the first child population is generated, which inherits the dominant characteristics of the parent while retaining the diversity of the solution. In order to ensure the diversity of the population and the uniformity of the distribution of solutions, the terminal calculates the crowding distance of the individual with the highest non-dominated level in the parent population. The crowding distance measures the sparseness of the distribution between individuals on the Pareto front. Individuals with a large crowding distance indicate that the solutions around them are relatively sparse, which means that these individuals represent more diverse solutions and can help the genetic algorithm search in a wider solution space. The terminal selects a preset number of individuals from the individuals with the highest non-dominated level in the parent population based on the calculated crowding distance. During the screening process, individuals with larger crowding distances are preferred because the solution space where these individuals are located is more sparse, which can increase the diversity of the population and prevent the population from converging to the local optimum too early. The screened individuals will be used as the second offspring population. Finally, the terminal integrates the first offspring population generated by crossover and mutation operations with the second offspring population screened according to the crowding distance to form the final offspring population. The integrated offspring population includes both new individuals generated by genetic operations and parent individuals with better diversity, thereby balancing the exploration and convergence of the solution.

[0100] In this embodiment, a more diverse and representative offspring population is generated by combining crossover, mutation operations and crowding distance screening. Crossover and mutation operations ensure the evolution of the population and the generation of new solutions, while crowding distance screening retains high-quality individuals with uniform distribution and sparse solution space. This not only improves the diversity of the population, but also avoids the algorithm from falling into the local optimal solution, ultimately making the prediction of the fuel cell system operation plan more accurate and efficient.

[0101] In an exemplary embodiment, the above step S202 calculates the fitness of each individual in the current population, and also includes: constructing a simulation model of the fuel cell system to be predicted in a physical simulation tool; and calculating the fitness of each individual in the current population in the physical simulation tool based on the simulation model.

[0102] Furthermore, the above step S203 generates a progeny population according to the fitness and the current population, and further includes: using an optimization prediction tool to generate a progeny population according to the fitness and the current population.

[0103] Among them, physical simulation tools are used to build a simulation model of the fuel cell system and calculate the performance of each individual based on the model to evaluate its fitness. Common physical simulation tools include EES (Engineering Equation Solver), Aspen Plus, COMSOL Multiphysics and GT-SUITE.

[0104] Among them, the optimization prediction tool is a tool used to generate the offspring population according to the selection, crossover and mutation operations of the genetic algorithm after the fitness calculation. It can be MATLAB (especially the genetic algorithm toolbox of MATLAB), Python with DEAP, GAMS (General Algebraic Modeling System) and Genetic Algorithm for Python (GAFT) (a Python library dedicated to genetic algorithms that supports the solution of complex optimization problems).

[0105] Exemplarily, the terminal builds a simulation model of the fuel cell system by calling a physical simulation tool (such as EES), which includes key elements such as the system's operating parameters, thermodynamic equilibrium, and chemical reactions. By inputting each individual (i.e., parameter combination) in the current population into the simulation model, the physical simulation tool can calculate the performance of each individual and convert it into a fitness value. For example, the simulation tool can evaluate the pros and cons of each individual based on objective functions such as system efficiency and hydrogen utilization. The terminal performs genetic algorithm operations on the current population by calling an optimization prediction tool (such as MATLAB). According to the fitness value, the terminal preferentially selects individuals with higher fitness for genetic operations to generate new offspring populations. The crossover operation is used to combine the genes of the parent individuals to generate offspring individuals with new parameter combinations; the mutation operation introduces randomness to ensure the diversity of the population and prevent the population from falling into a local optimum.

[0106] In this embodiment, the physical simulation tool accurately calculates the fitness of each individual, and combined with the genetic algorithm operation of the optimization prediction tool, the terminal can efficiently generate a new offspring population. The combined use of the tools not only ensures the accuracy of the system simulation calculation, but also enables rapid iterative optimization, helping the fuel cell system solution prediction to achieve the best performance and economy.

[0107] In another exemplary embodiment, as shown in FIG3 , the present application provides a method for predicting an operation scheme of a fuel cell system, the method comprising the following steps:

[0108] Step S301, obtaining a preset operating parameter item of a fuel cell system to be predicted, and obtaining a preset value range of the preset operating parameter item.

[0109] Step S302: Perform sensitivity analysis on each preset operating parameter item to obtain analysis results, and determine the preset value accuracy corresponding to each preset operating parameter item based on the analysis results.

[0110] Step S303: randomly generate multiple groups of operating parameter values ​​for preset operating parameter items according to a preset value range and a preset value precision.

[0111] Step S304: Determine the encoding length corresponding to the preset operating parameter item according to the preset value range and the preset value precision.

[0112] Step S305, determining the coding relationship of the preset operating parameter items according to the coding length, and converting each operating parameter value into coding information according to the coding relationship, as the gene information of the individual in the genetic algorithm, to obtain the initial population.

[0113] Step S306, obtaining preset constraint information and multiple preset objective functions, and obtaining fitness functions of individuals in multiple genetic algorithms according to the preset constraint information and the preset objective functions.

[0114] Step S307: taking the initial population as the current population and calculating the fitness of each individual in the current population.

[0115] Step S308: sort the individuals in the current population by non-dominated order according to the fitness, and determine the non-dominated level of each individual in the current population.

[0116] Each non-dominated level contains at least one individual.

[0117] Step S309: Eliminate the individuals with the lowest non-dominant level from the current population to obtain the parent population, and obtain the first child population through crossover and mutation operations based on the parent population.

[0118] Step S310, calculating the crowding distance of the individual with the highest non-dominated level in the parent population, and selecting a preset number of individuals from the individuals with the highest non-dominated level in the parent population according to the crowding distance, and integrating them into the second child population.

[0119] Step S311, the first offspring population and the second offspring population are all offspring populations, the offspring population is used as a new current population, and the process returns to step S307 until an offspring population that meets the preset conditions is obtained.

[0120] Step S312: determining a predicted operation plan of the fuel cell system to be predicted based on the offspring population that meets the preset conditions.

[0121] In this embodiment, through the selective optimization mechanism of the genetic algorithm, solutions that do not meet the target can be gradually eliminated to avoid indiscriminate calculation of each parameter combination. Compared with the uniform division in the traditional method, combined with the value accuracy of each preset operating parameter item, the genetic algorithm gradually reduces the focus on minor parameters through selection, crossover, and mutation in the evolutionary process, focusing on the optimization of key parameters, significantly reducing unnecessary computational burden. Ultimately, the prediction efficiency of the fuel cell system operation plan and the utilization rate of computing resources are improved, so that the entire optimization process has higher computational efficiency while maintaining accuracy. It not only effectively overcomes the limitations of traditional optimization methods in dealing with complex, nonlinear and multi-objective problems, but also significantly improves the design and optimization efficiency of fuel cell systems through intelligent means, providing strong support for technological progress in the field of new energy.

[0122] In an exemplary embodiment, as shown in FIG. 4 , it is a schematic diagram of a fuel cell system to be predicted for executing a fuel cell system operation scenario prediction method.

[0123] Specifically, the fuel cell system consists of three parts: solid oxide fuel cell system, transcritical carbon dioxide cycle and LNG (Liquefied Natural Gas) cold energy utilization system. The SOFC (Solid Oxide FuelCell) system includes two-stage compression intercooling air compressors (AC1 and AC2), condensers (HE1), fuel compressors (FC), water pumps (WP), two air preheaters (PH1 and PH2 respectively), fuel preheaters (PH3), water preheaters (PH4), mixers (M), turbines (T3), planar direct internal reforming (DIR) SOFC stacks, AC / DC inverters, afterburners (AB), and gas turbines (GT). The transcritical carbon dioxide cycle includes a heat recovery steam generator (HRVG), a carbon dioxide turbine (T2), a regenerator (R), a carbon dioxide pump (P1), and a condenser (Con). The LNG cold energy utilization system consists of an LNG storage tank, an LNG pump (P2), a cold storage (HE2), and a heat exchanger (HE3).

[0124] In a fuel cell system, the net output power of the fuel cell system is one of the important indicators often used to evaluate the working capacity of the fuel cell. In this system, it can be expressed as:

[0125]

[0126] Among them, W GT is the output power of the gas turbine in the SOFC system, kW; W c,fuel is the power consumption of the fuel compressor in the SOFC system, kW; W c,airis the power consumption of the air compressor / kW; W pump is the pump power consumption / kW.

[0127] In the bottom cycle of this system, the transcritical carbon dioxide cycle driven by the waste heat provided by the waste heat boiler and the LNG refrigeration exergy utilization system are connected in series. The output work of the series cycle system represents the amount of waste heat recovered from the main system, which is expressed in this system as:

[0128]

[0129] Among them, W SOFC,net is the net output power of the SOFC system, kW; W CO2,net is the net output power of the transcritical carbon dioxide cycle, kW; W LNG,net is the net output power of the LNG system, kW.

[0130] Thermal efficiency of TRCC (Transcritical CO2 Cycle) and LNG combined bottoming cycle system:

[0131]

[0132] Among them, h is the enthalpy value, which is a physical quantity; m is the mass flow rate; h 19 It represents the enthalpy value of node 19 in Figure 4, h 24 It represents the enthalpy value of node 24 in Figure 4; m CO2 , represents the mass flow rate of carbon dioxide.

[0133] The thermal efficiency of the SOFC system is expressed as:

[0134]

[0135] Among them, Q SOFC is the input energy of the fuel in the fuel cell system.

[0136] The total thermal efficiency of the system is:

[0137]

[0138] In this system, the economic feasibility evaluation criteria are divided into the following three points: product operation and maintenance costs, internal rate of return and discounted payback period. The input parameters of economic analysis are shown in Table 1, Table 2 and Table 3, where the total cost rate unit is: US dollars per second.

[0139] Table 1 Input parameters for economic analysis

[0140]

[0141] Table 2 Economic analysis input parameters

[0142]

[0143] Table 3 Economic analysis input parameters

[0144]

[0145] In the above economic analysis input parameter table, the definitions of the parameters are shown in Table 4 below.

[0146] Table 4 Parameter interpretation table

[0147]

[0148] Based on the parameters listed in the table above, the capital costs of SOFC / GT, TRCC and LNG subsystems are evaluated. The unit time cost rate of each investment project is calculated as follows:

[0149]

[0150] Among them, Z k , CRF, and N represent capital cost, capital recovery factor, maintenance factor, and annual operating time, respectively. CRF is the ratio of the fixed annuity to the present value of the annuity received during the investment period, calculated as follows:

[0151]

[0152] Where i and n are the interest rate and system lifetime.

[0153] In addition to the fixed costs of the above components, the system cost also needs to calculate the fuel cost of the system. The fuel cost rate is evaluated based on the unit cost of natural gas and is defined as follows:

[0154]

[0155] Among them, m f is the fuel mass.

[0156] The optimization goal of the above system is to maximize the total thermal efficiency / exergy efficiency of the fuel cell cogeneration system while minimizing the total system cost rate. The more the number of cells in the fuel cell stack is, the higher the battery input temperature is, and the lower the circuit density is, the higher the system efficiency is, and the higher the corresponding system cost is, showing a contradictory relationship between the two. Moreover, the two are independent of each other and have a small correlation, which meets the selection conditions of the objective function in multi-objective optimization.

[0157] The constraint condition of the above system is mainly the temperature difference of each heat exchanger, which requires that the temperature difference of each port of each heat exchanger is correct and meets the temperature conditions of the hot and cold ends.

[0158] Exemplarily, a comprehensive evaluation model of fuel cell system performance parameters and economic indicators is first constructed to clarify the specific quantitative indicators and weight distribution of each target. Subsequently, the sensitivity analysis technology is used to identify key parameters that have a significant impact on system performance and economy, such as fuel utilization, current density, catalyst activity, etc., and the sensitivity matrix of these parameters is constructed to provide a scientific basis for the subsequent optimization process. On this basis, the present invention innovatively applies genetic algorithms to multi-objective optimization of fuel cell systems. Through the encoding mechanism, the value range of each key parameter is converted into a chromosome structure in the genetic algorithm, and each chromosome represents a set of potential optimization schemes. The algorithm uses a fitness function to evaluate the comprehensive performance and economy of different chromosomes (i.e., different parameter combinations). The function comprehensively considers the improvement of system performance parameters and the economic cost required to achieve these improvements. In the iterative process of the genetic algorithm, excellent individuals are retained by selection operations, the advantages of different excellent individuals are integrated by crossover operations, and new gene combinations are introduced by mutation operations, and better chromosomes (parameter combinations) are continuously generated and screened. In this process, combined with the results of sensitivity analysis, more sophisticated coding and search strategies are used for highly sensitive parameters to ensure the accuracy of their optimization; while for low-sensitivity parameters, the search range is appropriately relaxed to improve optimization efficiency. As the number of iterations increases, the algorithm gradually approaches the global optimal solution, and eventually obtains one or more groups of optimization solutions that can significantly improve the performance parameters of the fuel cell system and effectively control the economic cost. The optimization method of the present invention not only effectively overcomes the limitations of traditional optimization methods in dealing with complex, nonlinear and multi-objective problems, but also significantly improves the design and optimization efficiency of the fuel cell system through intelligent means, providing strong support for technological progress in the field of new energy.

[0159] In practical applications, the simulation process of the system is run in the commercial software EES, relying on the sound gas thermal physical property parameter database of EES. However, in the subsequent system multi-objective optimization process, EES cannot meet the optimization requirements of the system. Here, a multi-software collaborative simulation idea is proposed: that is, the MATLAB genetic algorithm toolbox is combined with EES. The state parameters of each point in the system are calculated through the EES physical property parameter database, and then a matrix document is generated. The optimization toolbox is called with the MATLAB program to read the state parameters of each point. The genetic algorithm calculates the input parameters of the next generation system and then feeds them back to EES to continue to calculate the state of each point and continue to output, forming a collaborative simulation closed loop. As shown in Figure 5, the final optimization data is obtained by repeatedly calling parameters and iterative calculations.

[0160] In this embodiment, we not only focus on the optimization of the performance parameters of the fuel cell system, but also use the genetic algorithm, an efficient global search technology, to deeply explore and balance the various performance indicators of the system, such as efficiency, power density and durability, and also include economic analysis in the core considerations, such as cost-effectiveness, operation and maintenance costs and investment payback period. By building a multi-objective optimization model, we strive to effectively control and optimize the economic cost while improving the performance of the system. At the same time, a multi-software joint call calculation mechanism is established to form a unified optimization method, optimize the calculation process, and improve the calculation efficiency.

[0161] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0162] Based on the same inventive concept, the embodiment of the present application also provides a fuel cell system operation scheme prediction device for implementing the above-mentioned fuel cell system operation scheme prediction method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more fuel cell system operation scheme prediction device embodiments provided below can refer to the limitations of the fuel cell system operation scheme prediction method above, and will not be repeated here.

[0163] In an exemplary embodiment, as shown in FIG6 , a fuel cell system operation scheme prediction device is provided, comprising: an information acquisition module 601, a parameter generation module 602, a gene conversion module 603 and a scheme prediction module 604, wherein:

[0164] The information acquisition module 601 is used to acquire the preset operating parameter items of the fuel cell system to be predicted, and to acquire the preset value range and preset value accuracy of the preset operating parameter items;

[0165] A parameter generation module 602, used to randomly generate multiple groups of operation parameter values ​​of preset operation parameter items according to a preset value range and a preset value precision;

[0166] Gene conversion module 603, used to convert each operating parameter value into gene information of individuals in the genetic algorithm to obtain an initial population;

[0167] The scheme prediction module 604 is used to obtain preset constraint information and multiple preset objective functions, and obtain a predicted operation scheme of the fuel cell system to be predicted based on the preset constraint information, the preset objective functions and the initial population.

[0168] In one embodiment, the above-mentioned gene conversion module 603 is also used to determine the coding length corresponding to the preset operating parameter item according to the preset value range and the preset value accuracy; determine the coding relationship of the preset operating parameter item according to the coding length; and convert each operating parameter value into coding information according to the coding relationship as the genetic information of the individual in the genetic algorithm.

[0169] In one of the embodiments, the above-mentioned scheme prediction module 604 is also used to obtain the fitness function of individuals in multiple genetic algorithms according to preset constraints and various preset objective functions; take the initial population as the current population, and calculate the fitness of each individual in the current population; generate a child population according to the fitness and the current population, take the child population as the new current population, and return to execute the step of calculating the fitness of each individual in the current population until a child population that meets the preset conditions is obtained; determine the predicted operation scheme of the fuel cell system to be predicted according to the child population that meets the preset conditions.

[0170] In one embodiment, the above-mentioned scheme prediction module 604 is also used to sort the individuals in the current population according to fitness, determine the non-dominated level of each individual in the current population; each non-dominated level contains at least one individual; from the current population, eliminate the individual with the lowest non-dominated level to obtain the parent population; generate the child population based on the parent population.

[0171] In one embodiment, the above-mentioned scheme prediction module 604 is also used to obtain a first child population based on the parent population through crossover operations and mutation operations; calculate the crowding distance of the individual with the highest non-dominated level in the parent population; based on the crowding distance, select a preset number of individuals from the individuals with the highest non-dominated level in the parent population and integrate them into a second child population; and the first child population and the second child population are combined into a child population.

[0172] In one of the embodiments, the above-mentioned scenario prediction module 604 is also used to construct a simulation model of the fuel cell system to be predicted in a physical simulation tool; in the physical simulation tool, the fitness of each individual in the current population is calculated according to the simulation model; and the optimization prediction tool is used to generate a descendant population according to the fitness and the current population.

[0173] Each module in the above-mentioned fuel cell system operation scheme prediction device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0174] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG7. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for predicting an operation scheme of a fuel cell system is implemented. The display unit of the computer device is used to form a visually visible picture, which may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0175] Those skilled in the art will understand that the structure shown in FIG. 7 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0176] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0177] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0178] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0179] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0180] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0181] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0182] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for predicting a fuel cell system operation scheme, characterized in that: The method comprises: Obtaining a preset operating parameter item of the fuel cell system to be predicted, and obtaining a preset value range and a preset value accuracy of the preset operating parameter item; Randomly generate multiple groups of operating parameter values ​​for the preset operating parameter items according to the preset value range and the preset value accuracy; Converting each of the operating parameter values ​​into genetic information of individuals in the genetic algorithm to obtain an initial population; Preset constraint information and a plurality of preset objective functions are obtained, and a predicted operation plan of the fuel cell system to be predicted is obtained according to the preset constraint information, each of the preset objective functions and the initial population.

2. The method according to claim 1, characterized in that The step of converting each of the operating parameter values ​​into gene information of individuals in the genetic algorithm comprises: Determine the encoding length corresponding to the preset operating parameter item according to the preset value range and the preset value precision; Determining the encoding relationship of the preset operating parameter items according to the encoding length; According to the coding relationship, each of the operating parameter values ​​is converted into coding information as the gene information of the individual in the genetic algorithm.

3. The method according to claim 1, characterized in that The step of obtaining a predicted operation scheme of the fuel cell system to be predicted according to the preset constraint conditions, each preset objective function and the initial population includes: According to the preset constraint conditions and each of the preset objective functions, obtaining fitness functions of individuals in multiple genetic algorithms; Taking the initial population as the current population, calculating the fitness of each individual in the current population; Generate a descendant population according to the fitness and the current population, use the descendant population as a new current population, and return to execute the step of calculating the fitness of each individual in the current population until a descendant population that meets the preset conditions is obtained; A predicted operation scheme of the fuel cell system to be predicted is determined according to the offspring population that meets the preset conditions.

4. The method according to claim 3, characterized in that The step of generating a descendant population according to the fitness and the current population includes: According to the fitness, the individuals in the current population are non-dominated and sorted to determine the non-dominated level of each individual in the current population; each non-dominated level contains at least one individual; From the current population, remove the individuals with the lowest non-dominant level to obtain a parent population; A child population is generated according to the parent population.

5. The method according to claim 4, characterized in that The step of generating a child population according to the parent population includes: According to the parent population, a first child population is obtained through a crossover operation and a mutation operation; Calculate the crowding distance of the individual with the highest non-dominated level in the parent population; According to the crowding distance, a preset number of individuals are selected from the individuals with the highest non-dominated level in the parent population, and the individuals are integrated into a second offspring population; The first progeny population and the second progeny population are collectively referred to as the progeny population.

6. The method according to claim 3, characterized in that The calculating the fitness of each individual in the current population comprises: In a physical simulation tool, constructing a simulation model of the fuel cell system to be predicted; In the physical simulation tool, the fitness of each individual in the current population is calculated according to the simulation model; The step of generating a descendant population according to the fitness and the current population includes: An optimization prediction tool is used to generate a progeny population according to the fitness and the current population.

7. A fuel cell system operation plan prediction device, characterized in that: The device comprises: An information acquisition module, used to acquire a preset operating parameter item of the fuel cell system to be predicted, and to acquire a preset value range and a preset value accuracy of the preset operating parameter item; A parameter generation module, used for randomly generating a plurality of groups of operating parameter values ​​of the preset operating parameter items according to the preset value range and the preset value precision; A gene conversion module, used to convert each of the operating parameter values ​​into gene information of individuals in the genetic algorithm to obtain an initial population; The scheme prediction module is used to obtain preset constraint information and multiple preset objective functions, and obtain the predicted operation scheme of the fuel cell system to be predicted based on the preset constraint information, each preset objective function and the initial population.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images