A method and system for optimizing the operation of an all-vanadium redox flow battery system

Abstract

The present invention discloses a method and system for optimizing the operation of a vanadium redox flow battery system, which relates to the technical field of battery energy storage, and includes the following steps: obtaining a plurality of operating parameters and a plurality of evaluation indexes of the vanadium redox flow battery, and dividing different evaluation indexes based on different application scenarios; respectively obtaining a plurality of evaluation index values corresponding to different values of a plurality of operating parameters through prediction models under different application scenarios; iterating the values of the plurality of operating parameters during the optimization process of the optimal solution of each objective function to obtain an optimized operating parameter solution set corresponding to the application scenario; scoring multiple groups of operating parameters in the optimized operating parameter solution sets under different application scenarios, and taking the multiple groups of operating parameters with scores greater than a set threshold as the optimized operating parameters. The present invention can effectively avoid the problem of local optimal solutions existing in the existing methods, effectively reduce the overall computational load, and improve the accuracy of the optimization results.

Description

Technical Field

[0001] The present invention relates to the technical field of battery energy storage, and particularly to a method and system for optimizing the operation of a vanadium redox flow battery system. Background Art

[0002] With the continuous growth of the demand for renewable energy, energy storage technology has become increasingly important. Against this background, the vanadium redox flow battery (VRFB), as an innovative energy storage solution, has received extensive attention due to its significant potential in high-proportion intermittent renewable energy power generation systems. Technological advancements have continuously promoted the optimization of VRFB performance, covering multiple aspects such as battery design, fluid management, charge and discharge strategies, and system integration. Currently, the indicators for evaluating VRFB performance include battery efficiency, discharge capacity, energy density, and power density, etc. However, in the existing technology, mostly 1 - 2 indicators are used to evaluate the battery performance. In practical applications, it is not comprehensive enough to rely solely on 1 - 2 indicators to evaluate the battery performance. Therefore, developing a comprehensive and applicable evaluation system is crucial for promoting the development and application of VRFB technology.

[0003] The multi-objective operation condition optimization technology of VRFB is a complex task, which involves multiple key aspects, including the selection of optimization parameters, the acquisition method of sample data, the formulation of optimization objectives, the applicability of optimization algorithms, and the screening of optimization solution sets. In the optimization research of VRFB systems, methods such as multi-physics field coupling modeling and simulation, system-level simulation, and experimental analysis have been widely used to obtain sample data. Optimization algorithms such as parametric scanning method, genetic algorithm, particle swarm optimization, and NSGA-II have also been applied to battery performance prediction or operation condition optimization. However, for the optimization parameters of VRFB, such as flow channel size, operation conditions (electrolyte flow rate and charge-discharge current density), and electrode thickness, etc., the current optimization objectives usually focus on a single target. At the same time, the coupling effect of the operating parameters of VRFB, such as electrolyte flow rate, charge-discharge current density, and electrode compression ratio, on battery performance is a complex non-linear relationship. Traditional optimization methods, such as particle swarm optimization and parametric scanning, etc., which rely on simple mathematical models and heuristic rules, are prone to local optimum problems during the optimization process. Therefore, their prediction accuracy and search ability are limited, and it is difficult to handle this highly complex non-linear optimization problem, and the calculation cost and implementation complexity are relatively high. In addition, traditional methods usually do not have the ability to handle high-dimensional data and are also lacking in dynamic change or real-time adjustment.

[0004] In summary, the existing methods for optimizing the operation of battery systems only rely on one or two indicators to evaluate battery performance, and cannot adapt to diverse application scenarios. At the same time, due to the complex non-linear relationship of the coupling effect of the operating parameters of VRFB on battery performance, the existing optimization methods are prone to local optimum problems during the optimization of the optimization parameters, and the calculation cost is relatively high, resulting in inaccurate optimization results of the operating parameters, thereby reducing the performance and efficiency of VRFB. Summary of the Invention

[0005] The present invention provides a method and system for optimizing the operation of a vanadium redox flow battery system, develops a comprehensive performance evaluation system applicable to different application scenarios and requirements, and solves the problems that existing optimization methods are difficult to handle some highly complex non-linear optimization problems, which are prone to local optimum problems in the model, resulting in relatively high calculation costs and implementation complexities.

[0006] In the first aspect, the present invention provides a method for optimizing the operation of a vanadium redox flow battery system, including the following steps:

[0007] Obtain multiple operating parameters of the vanadium redox flow battery system and multiple evaluation indicators for evaluating the performance of the vanadium redox flow battery, and divide different evaluation indicators based on different application scenarios;

[0008] Construct prediction models for different application scenarios through multiple operating parameters and multiple evaluation indicators for different application scenarios; respectively obtain multiple evaluation indicator values corresponding to different values of multiple operating parameters through the prediction models for different application scenarios;

[0009] Establish corresponding objective functions through multiple evaluation indicator values for different application scenarios, and iterate the values of multiple operating parameters during the optimization process of the optimal solution of each objective function to obtain an optimized operating parameter solution set for the corresponding application scenario;

[0010] Score multiple groups of operating parameters in the optimized operating parameter solution sets for different application scenarios, and use multiple groups of operating parameters with scores greater than the set threshold as the optimized operating parameters.

[0011] Preferably, the multiple operating parameters include electrode compression ratio, electrolyte flow rate, and current density, the multiple evaluation indicators include energy efficiency, system efficiency, power density, energy density, and discharge capacity, and the application scenarios include economic application scenarios, continuous application scenarios, load capacity application scenarios, and environmental protection application scenarios.

[0012] Preferably, the dividing different evaluation indicators based on different application scenarios includes:

[0013] For economic application scenarios, the indicators include energy efficiency, system efficiency, power density, and energy density;

[0014] For continuous application scenarios, its indicators include energy efficiency, system efficiency, power density, energy density, and discharge capacity;

[0015] For load capacity application scenarios, its indicators include power density and discharge capacity;

[0016] For environmental protection application scenarios, its indicators include energy efficiency, system efficiency, energy density, and discharge capacity.

[0017] Preferably, constructing a prediction model for different application scenarios through multiple operating parameters and multiple evaluation indicators of different application scenarios includes the following steps:

[0018] Obtain the battery charge and discharge curves under multiple operating parameters, and obtain multiple evaluation indicator values under different application scenarios through the battery charge and discharge curves to obtain a data set;

[0019] Taking multiple operating parameters as inputs and multiple evaluation indicators under different application scenarios as outputs, establish multiple neural network models;

[0020] Train multiple neural network models through the data set to obtain a prediction model for different application scenarios.

[0021] Preferably, the value range of the electrode compression ratio is 0% - 80%, the value range of the electrolyte flow rate is 10 mL / min - 310 mL / min, and the value range of the current density is 100 A / m 2 ~3600 A / m 2 .

[0022] Preferably, in the optimization process of each objective function's optimal solution, iterating the values of multiple operating parameters to obtain an optimized operating parameter solution set for the corresponding application scenario includes the following steps:

[0023] Set the population size, iteration conditions, objective function, and boundary values of multiple operating parameters of the genetic algorithm;

[0024] Randomly generate multiple individuals as the initial population, where the individual is a set of operating parameters;

[0025] Use the trained prediction models for different application scenarios to obtain multiple objective functions corresponding to the initial population and screen out the initial optimal solution of the objective function;

[0026] Perform crossover and mutation operations on the initial population to obtain a new population, obtain multiple objective functions corresponding to the new population, and screen out the new optimal solution of the objective function;

[0027] Perform the above-mentioned crossover, mutation, and comparison iterations on the new population, and output the optimal solution of the objective function and the corresponding set of optimized operating parameters when the iteration condition is met.

[0028] Preferably, scoring the multiple sets of operating parameters in the optimized operating parameter solutions for different application scenarios includes the following steps:

[0029] Construct a normalized decision matrix with the evaluation index values corresponding to different application scenarios;

[0030] Perform positive and standardization processing on the normalized decision matrix, and extract the ideal optimal solution and the ideal worst solution;

[0031] Obtain the distances between each evaluation index and the ideal optimal solution and the ideal worst solution to obtain the positive ideal solution and the negative ideal solution;

[0032] Fit the positive ideal solution and the negative ideal solution to obtain the scores of multiple sets of operating parameters.

[0033] In a second aspect, the present invention provides an optimized operation system for a vanadium redox flow battery system, including:

[0034] An acquisition module for acquiring multiple operating parameters of the vanadium redox flow battery system and multiple evaluation indexes for evaluating the performance of the vanadium redox flow battery, and dividing different evaluation indexes based on different application scenarios;

[0035] A prediction module for constructing prediction models for different application scenarios through multiple operating parameters and multiple evaluation indexes for different application scenarios; respectively obtaining multiple evaluation index values corresponding to different values of multiple operating parameters through the prediction models for different application scenarios;

[0036] An optimization module for establishing corresponding objective functions through multiple evaluation index values for different application scenarios, and iterating the values of multiple operating parameters during the optimization process of the optimal solution of each objective function to obtain a set of optimized operating parameters for the corresponding application scenario;

[0037] A scoring module for scoring multiple sets of operating parameters in the optimized operating parameter solutions for different application scenarios, and taking the multiple sets of operating parameters with scores greater than the set threshold as the optimized operating parameters.

[0038] Compared with the prior art, the beneficial effects of the present invention are:

[0039] The present invention first divides different evaluation indicators based on different application scenarios, and clarifies the performance evaluation indicators for different application scenarios. Then, for each application scenario, a prediction model is constructed through multiple operating parameters and multiple evaluation indicators of different application scenarios to quickly capture the complex non-linear relationship between the input and output. Finally, the values of multiple evaluation indicators in the current application scenario are used as the objective function, and the values of multiple operating parameters are iteratively optimized during the optimization process of the optimal solution of the objective function. By combining the prediction model, the optimization process, and the optimal solution screening and scoring process, the present invention can comprehensively search for the values of all operating parameters, effectively avoid the local optimal solution problem existing in the existing methods, effectively reduce the overall computational load, improve the accuracy of the optimization results, and enhance the performance and efficiency of VRFB. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0041] Figure 1 It is a flowchart of an operation optimization method for a vanadium redox flow battery system of the present invention;

[0042] Figure 2 It is a schematic diagram of the optimization result of the economy of the present invention;

[0043] Figure 3 It is a schematic diagram of the optimization result of the sustainability of the present invention;

[0044] Figure 4 It is a schematic diagram of the optimization result of the load capacity of the present invention;

[0045] Figure 5 It is a schematic diagram of the optimization result of the environmental protection of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0047] An operating optimization method for a vanadium redox flow battery system, specifically an optimal operating condition evaluation and selection method for a vanadium redox flow battery system based on a neural network combined with a genetic algorithm, which addresses the complex problem of multi-objective optimization to find the optimal parameter configuration under different application scenarios, referring to Figure 1 , and specifically includes the following steps:

[0048] Step 1: Obtain multiple operating parameters and multiple evaluation indicators of the vanadium redox flow battery.

[0049] The multiple operating parameters of the vanadium redox flow battery system include the electrode compression ratio CR , the electrolyte flow rate Q in , and the current density I app . The multiple evaluation indicators include the energy efficiency EE , the system efficiency SE , the power density P power , the energy density E power , and the discharge capacity C capacity . These indicators not only reflect the key performance of the battery in terms of energy conversion and overall system operating efficiency, but also are closely related to the battery's ability to store and release energy efficiently.

[0050] Step 2: Configure the operating parameters.

[0051] Determine the parameter configuration of the input variables. In order to find the optimal parameter configuration suitable for different scenarios and avoid local optima, the present invention expands the range of parameter configuration. Among them, the value range of the electrode compression ratio CR is 0% - 80%, with a step size of 10%. The value range of the electrolyte flow rate Q in is 10 mL / min - 310 mL / min, with a step size of 30 mL / min. The value range of the current density I app is 100 A / m 2 - 3600 A / m 2 , with a step size of 300 A / m 2 .

[0052] Step 3: Divide different evaluation indicators according to different application scenarios.

[0053] To comprehensively evaluate the performance of the battery, the present invention emphasizes the integration of optimization criteria in multiple aspects such as efficiency, utilization rate, and safety. These criteria are systematically divided into the following four categories to adapt to different application scenarios of the vanadium redox flow battery, specifically including load capacity, environmental friendliness, sustainability, and economy. Among them, the evaluation indicators of the load capacity include Ppower and C capacity The environmental protection evaluation indicators include EE , SE , E power and C capacity , continuous evaluation indicators include EE , SE , P power , E power and C capacity The economic evaluation indicators include EE , SE , P power and E power .

[0054] Economics is a key criterion for evaluating all-vanadium flow batteries, which includes comprehensive cost-effectiveness from initial investment to operation and maintenance. EE and SE Reduce operating costs by improving energy conversion and reducing transmission losses, P power and E power Improving power density reduces startup investment and operating expenses, and improves market competitiveness. EE , SE , P power and E power Divided into application scenarios of battery economy. Optimizing these indicators can reduce costs, increase market share and return on investment, thereby enhancing the company's market competitiveness and development prospects. The key to sustainability is to improve energy density, extend cycle life, and ensure environmentally friendly material recycling and processing. Systems with strong continuous operation capabilities can provide frequency regulation and load balancing in a timely manner to stabilize load fluctuations during operation. Improve the reliability of power supply, reduce maintenance costs, and improve overall economic benefits. Therefore, key indicators include EE , SE , P power , E power and C capacity Load capacity is a key indicator for evaluating energy output capacity and load bearing capacity in practical applications. P power and C capacity. High power density can meet applications with higher instantaneous energy demands, while large capacity represents the battery's service life and the ability to sustain energy supply to ensure continuous power supply, which is suitable for long discharge cycles. The optimization of power density and capacity is critical to battery energy storage systems and renewable energy applications. In terms of environmental protection, the consumption of natural resources and environmental pollution can be reduced by optimizing VRFB performance. Key indicators include EE , SE and E power These three indicators are of great significance for promoting sustainable development and ensuring ecological health.

[0055] Step 4: Obtain multiple evaluation index values corresponding to multiple operating parameters in different application scenarios.

[0056] The present invention uses the VRFB numerical model established by COMSOL to perform fixed-step parametric scanning for calculation, obtains the battery charge and discharge curves under different operating parameters, and calculates the corresponding evaluation index value through a formula.

[0057] The collected data was screened for outliers. The screening steps were as follows: (1) At low compression rates, the internal resistance of the battery will also be high. When the current density is too high, a large polarization will occur, which will affect the normal calculation and solution of the model. Therefore, these extreme conditions were deleted. (2) Data with performance indicators equal to 0 and less than 0.01 were deleted to prevent large errors in neural network fitting.

[0058] In order to avoid the adverse effects of data span on subsequent neural network training and make neural network training more effective, the data after removing outliers is normalized and the index range is processed to [0, -1]. The processing formula is as follows:

[0059] ;

[0060] In the formula, is the normalized evaluation index, x is the evaluation index before normalization, X The dataset is used for evaluation indicators.

[0061] Step 5: Take multiple operating parameters as input and multiple evaluation index values corresponding to different scenarios as output, and train multiple prediction models for different application scenarios.

[0062] Load volume divided by application scenario ( P power and C capacity )、Environmental protection( EE , SE , Epower and C capacity ), persistence ( EE , SE , P power , E power and C capacity ) and economy ( EE , SE , P power and E power ), taking 4 optimization objectives as output variables, Q in , I app and CR as input variables. Using the neural network toolbox in MATLAB, 4 neural network models are trained. By adjusting the size of the network layers and the proportion of the training set, the training data percentage is controlled between 60% and 80%. The Bayesian regularization technique is used to train the network models, and four independent neural network prediction models are established respectively to adapt to the non-linear relationship between the input variables and the output variables and improve the generalization ability of the training process.

[0063] Obtain multiple evaluation index values corresponding to different scenarios when multiple operating parameters take different values through multiple prediction models respectively.

[0064] Step 6: Multi-objective optimization through genetic algorithm.

[0065] Establish corresponding objective functions through multiple evaluation index values under different application scenarios, and iterate the values of multiple operating parameters through the optimization process of the optimal value of each objective function to obtain the solution set of the optimized operating parameters corresponding to the application scenarios. Specifically, it includes:

[0066] (1) Set the population size, iteration conditions (maximum number of iterations, objective fitness or no significant improvement for multiple consecutive generations) and boundary values of multiple operating parameters through the genetic algorithm.

[0067] (2) Randomly generate multiple individuals as the initial population, where an individual is a set of operating parameters, and use the objective function as the fitness value.

[0068] (3) Use the trained prediction models under different application scenarios to obtain multiple objective functions corresponding to the initial population and screen out the initial optimal solutions of the objective functions.

[0069] (4) Perform crossover and mutation operations on the initial population to obtain a new population, obtain multiple objective functions corresponding to the new population and screen out the new optimal solutions of the objective functions.

[0070] (5) Perform the above-mentioned crossover, mutation, and comparison iterations on the new population. When the iteration condition is met, output the optimal solution of the objective function and the corresponding solution set of the optimized operating parameters.

[0071] The genetic algorithm sets the population size as the total number of samples, the maximum number of iterations as 300, the limit of stagnant iterations as 200, and the evaluation indicators under different application scenarios as the objective function, with its tolerance being 1e -10 , and at the same time specifies the plotting function for real-time display of the Pareto front. The lower and upper bounds of the optimized operating parameters (compression ratio, electrolyte flow rate, and current density) are the minimum boundary [0 100 10] and the maximum boundary [0.8 3600 310] of the operating parameters respectively. The gamultiobj function is used to find the optimal solution of the operating parameters and return the solution and the corresponding objective function value, obtaining the solution set of the Pareto front as the data set of the optimized operating parameters under this application scenario.

[0072] Utilize the global search ability of the genetic algorithm to explore potential optimal parameter configurations, prevent the results from falling into local optima, and obtain the solution set of the optimized operating parameters.

[0073] Step 7: TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) multi-objective decision-making for comprehensive scoring and ranking.

[0074] First, construct a normalized decision matrix from the evaluation indicator values corresponding to each application scenario. Since all the evaluation indicators are maximization indicators, direct positive normalization is performed:

[0075] (1);

[0076] Among them, z ij is the value after positive normalization, x ij is the eigenvalue in the matrix, i is the row, j is the column, n is the total number of rows. Through the processes of positive normalization and standardization, the present invention can extract the vectors representing the ideal optimal solution z + and the ideal worst solution z - :

[0077] (2).

[0078] Use the positive ideal solution ( d + ) and the negative ideal solution ( d- )Calculate the distances of each evaluation index from the ideal optimal solution and the ideal worst solution:

[0079] (3);

[0080] (4);

[0081] Among them, m is the total number of columns, z j + is the optimal solution of the column, z j - is the worst solution of the column.

[0082] The scores of each set of operating parameters calculated according to the following formula:

[0083] (5).

[0084] Sort multiple scores, and take multiple sets of operating parameters with scores greater than the set threshold as the optimized operating parameters.

[0085] According to the sorting results, screen out the best operating conditions of the four evaluation systems (economy, sustainability, load capacity, and environmental friendliness) under different application scenarios.

[0086] Obtain the final optimization result, and through sorting analysis, obtain the optimized parameter configuration under different application scenarios, as Figures 2 - 4 shown, and obtain the optimization result diagrams under 4 application scenarios.

[0087] The present invention takes the evaluation indexes included in four different scenarios as four groups of output variables respectively, CR , Q in and I app as input variables, and uses the optimization technology combining neural network and genetic algorithm to find the optimal parameter configuration. By applying the TOPSIS algorithm to comprehensively evaluate and sort the optimal solution sets generated by the genetic algorithm, the operating condition configurations most suitable for each specific scenario are accurately screened out. This process not only enhances the performance of VRFB, but also ensures its adaptability and practicability in different environments.

[0088] Compared with traditional optimization methods such as parametric scanning, the present invention demonstrates significant superiority in the face of multi-objective optimization challenges. Generally speaking, this method combines the dual advantages of neural networks and genetic algorithms, specifically targeting complex and multi-dimensional optimization problems. Through comprehensive search and adaptive mechanisms, this method can effectively locate the Pareto front solution set of the optimal parameter configuration, thereby improving the efficiency of optimization and the quality of the solution. At the same time, traditional methods are often limited by being trapped in local optima and high computational complexity.

[0089] The present invention uses a method that combines a neural network with a genetic algorithm and TOPSIS decision-making to evaluate and select the optimal operating parameter configuration of a vanadium redox flow battery, providing an accurate and efficient solution to the complex problem of finding the optimal operating parameter configuration for multi-objective evaluation index optimization under different application scenarios.

[0090] Based on the same concept, the present invention also provides a vanadium redox flow battery system operation optimization system, including an acquisition module, a prediction module, an optimization module, and a scoring module.

[0091] The acquisition module is used to acquire multiple operating parameters of the vanadium redox flow battery system and multiple evaluation indexes for evaluating the performance of the vanadium redox flow battery, and divide different evaluation indexes based on different application scenarios.

[0092] The prediction module is used to construct prediction models for different application scenarios through multiple operating parameters and multiple evaluation indexes for different application scenarios; respectively obtain multiple evaluation index values corresponding to different values of multiple operating parameters through the prediction models for different application scenarios.

[0093] The optimization module is used to establish corresponding objective functions through multiple evaluation index values for different application scenarios, and iterate the values of multiple operating parameters during the optimization process of the optimal solution of each objective function to obtain the optimized operating parameter solution set for the corresponding application scenario.

[0094] The scoring module is used to score multiple groups of operating parameters in the optimized operating parameter solution sets for different application scenarios, and use the multiple groups of operating parameters with scores greater than the set threshold as the optimized operating parameters.

[0095] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0096] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A method for optimizing the operation of a vanadium redox flow battery system, characterized in that, It includes the following steps: Obtain multiple operating parameters of the all-vanadium redox flow battery system and multiple evaluation indicators for evaluating the performance of the all-vanadium redox flow battery, and divide different evaluation indicators based on different application scenarios; Construct prediction models for different application scenarios through multiple operating parameters and multiple evaluation indicators in different application scenarios; respectively obtain multiple evaluation indicator values corresponding to different values of multiple operating parameters through the prediction models in different application scenarios; Establish corresponding objective functions through multiple evaluation indicator values in different application scenarios, and iterate the values of multiple operating parameters during the optimization process of the optimal solution of each objective function to obtain the optimized operating parameter solution sets corresponding to the application scenarios; Score multiple groups of operating parameters in the optimized operating parameter solution sets for different application scenarios, and use the multiple groups of operating parameters with scores greater than the set threshold as the optimized operating parameters; The multiple operating parameters include electrode compression ratio, electrolyte flow rate, and current density, the multiple evaluation indicators include energy efficiency, system efficiency, power density, energy density, and discharge capacity, and the application scenarios include economic application scenarios, continuous application scenarios, load capacity application scenarios, and environmental protection application scenarios; The dividing of different evaluation indicators based on different application scenarios includes: For economic application scenarios, the indicators include energy efficiency, system efficiency, power density, and energy density; For continuous application scenarios, the indicators include energy efficiency, system efficiency, power density, energy density, and discharge capacity; For load capacity application scenarios, the indicators include power density and discharge capacity; For environmental protection application scenarios, the indicators include energy efficiency, system efficiency, energy density, and discharge capacity; The value range of the electrode compression ratio is 0% to 80%, the value range of the electrolyte flow rate is 10 mL / min to 310 mL / min, and the value range of the current density is 100 A / m 2 ~3600 A / m 2 ; The constructing of prediction models for different application scenarios through multiple operating parameters and multiple evaluation indicators in different application scenarios includes the following steps: Obtain the charge-discharge curves of the battery under multiple operating parameters, and obtain multiple evaluation indicator values in different application scenarios through the charge-discharge curves of the battery to obtain a data set; Establish multiple neural network models with multiple operating parameters as inputs and multiple evaluation indicators in different application scenarios as outputs; Train the multiple neural network models through the data set to obtain prediction models for different application scenarios; The iterating of the values of multiple operating parameters during the optimization process of the optimal solution of each objective function to obtain the optimized operating parameter solution sets corresponding to the application scenarios includes the following steps: Set the population size, iteration conditions, objective function, and boundary values of multiple operating parameters of the genetic algorithm; Randomly generate multiple individuals as the initial population, where the individual is a group of operating parameters; Use the trained prediction models for different application scenarios to obtain multiple objective functions corresponding to the initial population, and screen out the initial optimal solution of the objective function; Perform crossover and mutation operations on the initial population to obtain a new population, obtain multiple objective functions corresponding to the new population, and screen out the new optimal solution of the objective function; Perform the above-mentioned crossover, mutation, and comparison iterations on the new population, and output the optimal solution of the objective function and the corresponding optimized operating parameter solution set when the iteration conditions are met; Scoring multiple sets of operating parameters in the optimized operating parameter solution sets for different application scenarios includes the following steps: Construct a normalized decision matrix with the evaluation index values corresponding to different application scenarios; Perform positive and standardization processing on the normalized decision matrix, and extract the ideal optimal solution and the ideal worst solution; Obtain the distances between each evaluation index and the ideal optimal solution and the ideal worst solution to obtain the positive ideal solution and the negative ideal solution; Fit the positive ideal solution and the negative ideal solution to obtain the scores of multiple sets of operating parameters.

2. An operating optimization system for a vanadium redox flow battery system, characterized in that, Including: An acquisition module for acquiring multiple operating parameters of the all-vanadium redox flow battery system and multiple evaluation indexes for evaluating the performance of the all-vanadium redox flow battery, and dividing different evaluation indexes based on different application scenarios; A prediction module for constructing prediction models for different application scenarios through multiple operating parameters and multiple evaluation indexes for different application scenarios; respectively obtaining multiple evaluation index values corresponding to multiple operating parameters at different values through the prediction models for different application scenarios; An optimization module for establishing corresponding objective functions through multiple evaluation index values for different application scenarios, and iterating the values of multiple operating parameters during the optimization process of the optimal solution of each objective function to obtain the optimized operating parameter solution sets for the corresponding application scenarios; A scoring module for scoring multiple sets of operating parameters in the optimized operating parameter solution sets for different application scenarios, and taking the multiple sets of operating parameters with scores greater than the set threshold as the optimized operating parameters; The multiple operating parameters include electrode compression ratio, electrolyte flow rate, and current density, the multiple evaluation indexes include energy efficiency, system efficiency, power density, energy density, and discharge capacity, and the application scenarios include economic application scenarios, sustainability application scenarios, load capacity application scenarios, and environmental protection application scenarios; The dividing different evaluation indexes based on different application scenarios includes: For economic application scenarios, the indexes include energy efficiency, system efficiency, power density, and energy density; For sustainability application scenarios, the indexes include energy efficiency, system efficiency, power density, energy density, and discharge capacity; For load capacity application scenarios, the indexes include power density and discharge capacity; For environmental protection application scenarios, the indexes include energy efficiency, system efficiency, energy density, and discharge capacity; The value range of the electrode compression ratio is 0% to 80%, the value range of the electrolyte flow rate is 10 mL / min to 310 mL / min, and the value range of the current density is 100 A / m 2 ~3600 A / m 2 ; The constructing prediction models for different application scenarios through multiple operating parameters and multiple evaluation indexes for different application scenarios includes the following steps: Obtain the charge-discharge curves of the battery under multiple operating parameters, and obtain multiple evaluation index values for different application scenarios through the charge-discharge curves of the battery to obtain a data set; Establish multiple neural network models with multiple operating parameters as inputs and multiple evaluation indexes for different application scenarios as outputs; Train the multiple neural network models through the data set to obtain prediction models for different application scenarios; The iterating the values of multiple operating parameters during the optimization process of the optimal solution of each objective function to obtain the optimized operating parameter solution sets for the corresponding application scenarios includes the following steps: Set the population size, iteration conditions, objective function, and boundary values of multiple operating parameters of the genetic algorithm; Randomly generate multiple individuals as the initial population, where the individuals are a set of operating parameters; Use the trained prediction models under different application scenarios to obtain multiple objective functions corresponding to the initial population, and screen out the initial optimal solutions of the objective functions; Perform crossover and mutation operations on the initial population to obtain a new population, obtain multiple objective functions corresponding to the new population, and screen out the new optimal solutions of the objective functions; Perform the above crossover, mutation, and comparison iterations on the new population. When the iteration conditions are met, output the optimal solution of the objective function and the corresponding set of optimized operating parameters; The scoring of multiple sets of operating parameters in the optimized operating parameter solution sets under different application scenarios includes the following steps: Construct a normalized decision matrix with the evaluation index values corresponding to different application scenarios; Perform positive and standardization processing on the normalized decision matrix, and extract the ideal optimal solution and the ideal worst solution; Obtain the distances between each evaluation index and the ideal optimal solution and the ideal worst solution to obtain the positive ideal solution and the negative ideal solution; Fit the positive ideal solution and the negative ideal solution to obtain the scores of multiple sets of operating parameters.

Images