Formula proportion optimization method and system of electrolyte, electronic equipment and storage medium

Abstract

The invention provides an electrolyte formula proportion optimization method and system, electronic equipment and a storage medium, and relates to the technical field of batteries, the method comprises the following steps: obtaining electrolyte experiment data, the electrolyte experiment data at least comprising an electrolyte formula proportion and performance index data of a corresponding battery; constructing a prediction model of each performance index according to the electrolyte experiment data, and obtaining a plurality of performance index prediction models; performing multi-objective optimization by adopting a non-dominated sorting genetic algorithm in combination with each performance index prediction model to obtain an electrolyte formula proportion set; and selecting a target electrolyte formula proportion scheme from the electrolyte formula proportion set according to the target performance index demand. According to the method, the performance index prediction models are constructed, meanwhile, the non-dominated sorting genetic algorithm is combined with the performance index prediction models for multi-objective optimization, and the formula proportion set meeting the multiple performance indexes can be quickly screened out from mass data, so that the research and development efficiency is greatly improved, and the cost is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of batteries, in particular to a formula proportion optimization method and system of electrolyte, an electronic device and a storage medium. BACKGROUND

[0002] As an important component of batteries, the formula proportion of electrolyte directly affects the performance of batteries. Traditional electrolyte formula development mainly relies on the experience of experimenters and a large number of experiments for verification. This way not only has a long development cycle, but also has high costs, which seriously restricts the rapid iteration and development of battery technology.

[0003] At the same time, existing electrolyte formula optimization techniques mainly focus on improving a single performance, which makes it difficult to achieve the ideal state of the comprehensive performance of batteries. The existing methods lack research on the interaction mechanism between electrolyte components, which leads to a certain blindness in improving the performance of batteries. Therefore, the traditional electrolyte formula development has the problems of low efficiency and inability to achieve balanced optimization of the comprehensive performance of batteries. SUMMARY

[0004] In view of the above problems of the prior art, the present application provides a formula proportion optimization method and system of electrolyte, an electronic device and a storage medium, which effectively solve the problems of low efficiency and inability to achieve balanced optimization of the comprehensive performance of batteries in traditional electrolyte formula development.

[0005] In a first aspect, the present application provides a formula proportion optimization method of electrolyte, which comprises: obtaining electrolyte experimental data, the electrolyte experimental data at least including electrolyte formula proportion and performance index data of corresponding batteries; constructing a prediction model of each performance index according to the electrolyte experimental data, to obtain a plurality of performance index prediction models; performing multi-objective optimization by using a non-dominated sorting genetic algorithm in combination with each performance index prediction model, to obtain a set of electrolyte formula proportions; selecting a target electrolyte formula proportion scheme from the set of electrolyte formula proportions according to target performance index requirements.

[0006] In an optional embodiment, constructing a prediction model of each performance index data according to the electrolyte experimental data, to obtain a plurality of performance index prediction models, comprises: respectively constructing a prediction model of each performance index data; dividing the electrolyte experimental data into a training set and a validation set; respectively performing model training on each prediction model by using the training set, to obtain a plurality of initial performance index prediction models; Adjust parameters of each of the initial performance index prediction models respectively by using the verification set to obtain a plurality of performance index prediction models.

[0007] In an optional embodiment, the non-dominated sorting genetic algorithm is combined with each of the performance index prediction models to perform multi-objective optimization to obtain an electrolyte formula proportion set, including: Setting working parameters of the non-dominated sorting genetic algorithm; Generating an initial population of electrolyte formulas according to the working parameters, the initial population including a plurality of individuals, each of the individuals representing an electrolyte formula proportion scheme; Performing performance index data prediction on the individuals in the initial population according to each of the performance index prediction models to obtain performance index prediction data; Performing non-dominated sorting on the individuals in the initial population according to the performance index prediction data to obtain a front sequence; Calculating a crowding distance of each of the individuals in the front sequence; Selecting a parent population according to the front sequence and the crowding distance, performing simulated binary crossover and polynomial mutation on the selected parent population to generate a child population; Merging the parent population and the child population, performing non-dominated sorting and crowding distance calculation, and iteratively generating a next generation population until the number of iterations reaches a preset number, and taking non-dominated individuals in the current population as the electrolyte formula proportion set.

[0008] In an optional embodiment, calculating a crowding distance of each of the individuals in the front sequence includes: Calculating an Euclidean distance of each of the individuals in the front sequence from other individuals; Selecting a target number of neighboring individuals with the smallest Euclidean distance, calculating an average value of the Euclidean distances corresponding to the neighboring individuals to obtain a neighborhood crowding distance; Calculating a local density of each of the individuals in the front sequence in a target space by using a Gaussian kernel density, and obtaining a Gaussian crowding distance according to the local density; Weightedly fusing the neighborhood crowding distance and the Gaussian crowding distance to obtain the crowding distance.

[0009] In an optional embodiment, the performance index data at least includes a battery cycle capacity retention rate, a battery voltage drop value, and a battery internal resistance.

[0010] In an optional embodiment, the prediction model includes at least one or more of a support vector machine model, a gradient boosting decision tree model, and a deep neural network model.

[0011] In an optional embodiment, the working parameters at least include population size, iteration number, crossover probability, mutation probability and proportion range of each electrolyte material.

[0012] In a second aspect, the present application provides an electrolyte formula proportion optimization system, the system comprising: a data acquisition module configured to acquire electrolyte experimental data, the electrolyte experimental data at least including electrolyte formula proportion and performance index data of a corresponding battery; a model construction module configured to construct a prediction model of each performance index according to the electrolyte experimental data, to obtain a plurality of performance index prediction models; a target optimization module configured to perform multi-objective optimization by using a non-dominated sorting genetic algorithm in combination with each performance index prediction model, to obtain a set of electrolyte formula proportions; a scheme selection module configured to select a target electrolyte formula proportion scheme from the set of electrolyte formula proportions according to a target performance index requirement.

[0013] In a third aspect, the present application provides an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor executing the computer program to implement the electrolyte formula proportion optimization method according to any one of the preceding embodiments.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, the computer program being executable by a processor to implement the electrolyte formula proportion optimization method according to any one of the preceding embodiments.

[0015] The electrolyte formula proportion optimization method, system, electronic device and storage medium provided by the present application clearly present the correlation between formula proportion and performance by constructing performance index prediction models, providing accurate guidance for research and development. Meanwhile, the non-dominated sorting genetic algorithm is used in combination with the performance index prediction models to perform multi-objective optimization, which can quickly filter out a set of formula proportions meeting multiple performance indexes from massive data, reduce the number of experiments, and reduce material consumption and equipment use cost, thereby greatly improving research and development efficiency and reducing cost. Through multi-objective optimization, the multiple key performance indexes of the battery are taken into account, and the filtered electrolyte formula proportion can make the battery exhibit good comprehensive performance in different use scenarios, improving the reliability and market competitiveness of the product. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as limiting the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0017] Figure 1 is the first schematic diagram of the electrolyte formula proportion optimization method provided by the embodiments of the present application; Figure 2 is the second schematic diagram of the electrolyte formula proportion optimization method provided by the embodiments of the present application; Figure 3 is the third schematic diagram of the electrolyte formula proportion optimization method provided by the embodiments of the present application; Figure 4 is the schematic diagram of the structure of the electrolyte formula proportion optimization system provided by the embodiments of the present application; Figure 5 is the schematic diagram of the structure of the electronic device provided by the embodiments of the present application.

[0018] Main element symbol explanation: 200, electrolyte formula proportion optimization system; 210, data acquisition module; 220, model construction module; 230, target optimization module; 240, scheme selection module; 300, electronic device; 310, processor; 320, communication interface; 330, memory; 340, communication bus. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the present application more clear, the following will combine the drawings in the embodiments of the present application to make the technical solutions of the present application further clear and complete. It should be noted that the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0020] In addition, the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more than two, unless otherwise specifically limited.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0022] The traditional electrolyte formula development mode has many drawbacks. On the one hand, it mainly relies on the experience of experimenters and combines a large number of experimental verifications. This way not only has a long development cycle, but also has high costs, which seriously restricts the rapid iteration and development of battery technology. On the other hand, the existing electrolyte formula optimization technology mainly focuses on the improvement of a single performance, and it is difficult to balance and optimize the overall performance of the battery, resulting in that the comprehensive performance of the battery is difficult to reach the ideal state. In addition, the traditional method does not deeply study the interaction mechanism between each component of the electrolyte, so that there is a lack of clear theoretical guidance when improving the performance of the battery, and the optimization process has a certain blindness, which cannot accurately and efficiently develop a high-performance electrolyte formula, hindering the battery industry to move towards higher performance and lower cost.

[0023] Embodiment 1 The embodiments of the present application provide a formula proportion optimization method of an electrolyte, which effectively solves the problems of low efficiency and inability to balance and optimize the overall performance of the battery in the traditional electrolyte formula development. Figure 1 is the first schematic diagram of the formula proportion optimization method of the electrolyte provided by the embodiments of the present application, as shown in Figure 1 The method specifically comprises the following steps: S100, obtaining electrolyte experimental data, the electrolyte experimental data at least including electrolyte formula proportion and corresponding battery performance index data.

[0024] In the embodiments of the present application, when obtaining the electrolyte experimental data, the setting of the electrolyte formula proportion is the key starting point, which can fully consider the chemical properties, physical characteristics of different electrolyte materials and the possible interactions therebetween. The gradient setting method is used to plan the material ratio in the electrolyte, from low concentration to high concentration, gradually adjusting the proportion of each component of the electrolyte, and recording the electrolyte formula proportion.

[0025] According to the set electrolyte formula proportion, various electrolyte materials are accurately weighed, and are mixed and prepared in a strictly controlled experimental environment. In the production process, the temperature, stirring speed and time and other parameters are accurately controlled to ensure the quality stability and consistency of the electrolyte. After completing the electrolyte production, the battery is produced by using a strictly set process to ensure that the performance of the battery is mainly affected by the electrolyte formula, rather than the difference of the production process.

[0026] After the battery is completed, the performance indicators of the prepared battery are measured using professional related instruments. Optionally, the performance indicator data includes but is not limited to battery cycle capacity retention rate, battery voltage drop value and battery internal resistance. The battery cycle capacity retention rate is calculated by simulating the charge and discharge cycle process of the battery in actual use, recording the capacity change after each cycle, and then calculating the cycle capacity retention rate, so as to evaluate the capacity attenuation of the battery in the long-term use. The battery voltage drop value can be measured when the battery is subjected to specific charge and discharge operation, and the change of the battery voltage can be accurately monitored, the amplitude and speed of the voltage drop can be analyzed, and the voltage stability of the battery can be reflected. The battery internal resistance measurement can use the alternating current internal resistance test method to accurately obtain the internal resistance value of the battery. The internal resistance directly affects the charge and discharge efficiency and power output capacity of the battery. For example, the 100th cycle capacity retention rate of the battery, the 100th cycle voltage drop value and the direct current internal resistance of the battery can be selected as the performance indicator data.

[0027] Optionally, due to the influence of external factors such as temperature fluctuation and humidity change during the experiment, some performance indicator data may deviate. In order to ensure the accuracy and reliability of the experimental data, the data obtained from 3-5 batteries prepared with the same electrolyte formula can be preprocessed. For example, by data cleaning, those data obviously deviating from the normal range, which may be caused by external factors, are excluded, and the remaining data are standardized by calculating the average value and standard deviation, etc. to provide a solid data foundation for subsequent construction of performance indicator prediction model.

[0028] S200, constructing a prediction model for each performance indicator according to the electrolyte experimental data, and obtaining a plurality of performance indicator prediction models.

[0029] In the embodiment of the present application, a prediction model can be constructed for each performance indicator based on the electrolyte experimental data. The input of the prediction model is the electrolyte formula ratio, and the output is the specified performance indicator. The prediction model includes but is not limited to support vector machine model, gradient boosting decision tree model and deep neural network model, etc. which can be selected according to actual situation. Figure 2 is the second schematic diagram of the electrolyte formula ratio optimization method provided by the embodiment of the present application, as shown in Figure 2 The construction of the performance indicator prediction model specifically includes the following steps: S210, respectively constructing a prediction model for each performance indicator data.

[0030] Optionally, in the embodiment of the present application, a gradient boosting decision tree model is used as the prediction model, and a gradient boosting decision tree model corresponding to the battery cycle capacity retention rate, the battery voltage drop value and the battery internal resistance is constructed respectively to predict the battery cycle capacity retention rate at the 100th cycle, the battery voltage drop value at the 100th cycle and the direct current internal resistance of the battery. The gradient boosting decision tree model fits the data through the iteration combination of multiple decision trees. A simple decision tree is used to fit the initial prediction value of the data initially, and each tree fits the residual of the prediction of the previous round of model subsequently. The fitting effect is improved by continuously correcting the error.

[0031] S220, divide the electrolyte experimental data into a training set and a validation set.

[0032] In the embodiment of the present application, the training set and the validation set are divided according to the proportions of 70% and 30% respectively. In this way, it is ensured that the training set has sufficient data quantity to enable the model to learn the internal law between the electrolyte formula and the performance index, and a suitable amount of data is left for the validation set to evaluate the performance of the model on unseen data, thereby preventing overfitting of the model.

[0033] S230, model training is performed on each prediction model respectively using the training set to obtain multiple initial performance index prediction models.

[0034] Firstly, the gradient boosting decision tree models corresponding to the battery cycle capacity retention rate, the battery voltage drop value and the battery internal resistance are initialized. The mean value of the performance index data in the training set can be used as the initial prediction value of the corresponding gradient boosting decision tree model. Then, the negative gradient is calculated. In each iteration, the residual between the current model prediction value and the true value of the training set is calculated, and this residual is used as the fitting target of the subsequent decision tree. Then, the decision tree is constructed. The features that have a great influence on the residual prediction are selected from the electrolyte formula features, and the node splitting and other operations are performed until the stop condition is met, so that a decision tree that can preliminarily fit the residual is formed. Next, the contribution of the decision tree output is calculated. The prediction value of the sample residual by the decision tree is multiplied by the learning rate to obtain its contribution to the final prediction. Then, the contribution is added to the prediction value of the previous round to update the model prediction value. Finally, the steps of calculating the residual, constructing the decision tree and updating the prediction value are repeated for multiple iterations until the maximum number of iterations is reached or the model performance no longer improves significantly. At this time, the model training is completed, and multiple initial performance index prediction models are obtained, which can accurately predict the performance index according to the input electrolyte formula proportion.

[0035] S240, the parameters of each initial performance index prediction model are adjusted respectively using the validation set to obtain multiple performance index prediction models.

[0036] In the embodiments of the present application, the parameters of the initial performance index prediction model are dynamically adjusted by observing the performance of the validation set. Optionally, the mean square error can be used as the loss function, which can intuitively measure the difference between the predicted value and the true value of the model. When the mean square error on the validation set is large, it indicates that the model is not accurate enough in predicting the performance index of the electrolyte under the current parameter setting. At this time, the model can be optimized by adjusting the key parameters such as the depth of the decision tree, the learning rate, and the subsampling ratio. For example, increasing the depth of the decision tree can allow the model to learn more complex feature combinations, and reducing the learning rate can make the model converge more stably during iteration. By continuously adjusting these parameters and observing the changes in the mean square error on the validation set, a performance index prediction model that performs well on both the training set and the validation set is obtained, that is, the model can give more accurate performance index prediction values on both the training data and the validation data.

[0037] S300, using a non-dominated sorting genetic algorithm to combine each performance index prediction model for multi-objective optimization to obtain an electrolyte formula proportion set.

[0038] Generally, there is a conflict between performance indicators, and it is difficult to ensure that one performance indicator is good while another performance indicator is also good, so it is difficult to find the best set solution, and only a relatively better set solution can be given. The embodiments of the present application use a non-dominated sorting genetic algorithm to perform multi-objective optimization on the output data of each performance index prediction model to obtain a relatively better electrolyte formula proportion set. Figure 3 is a third schematic diagram of the electrolyte formula proportion optimization method provided by the embodiments of the present application, as shown in Figure 3 The non-dominated sorting genetic algorithm for multi-objective optimization specifically includes the following steps: S310, setting the working parameters of the non-dominated sorting genetic algorithm.

[0039] In the embodiments of the present application, the working parameters include but are not limited to population size, iteration number, crossover probability, mutation probability, and proportion range of each electrolyte material, etc. Among them, the population size determines the search breadth and diversity of the algorithm, the larger the population size, the wider the solution space covered, and the more likely to find the global optimal solution, but the higher the calculation cost, and the population size can be adjusted according to the complexity of the electrolyte material combination. The iteration number controls the convergence degree of the algorithm, the more the iteration number, the more likely the algorithm approaches the optimal solution, but the calculation time is prolonged, and the iteration number can be set in combination with the number or complexity of the performance indicators. The crossover probability determines the frequency of gene exchange between parent individuals, affecting the generation speed of new solutions, if the proportion range of the electrolyte material is narrow, the crossover probability can be appropriately reduced to retain effective combinations, and if the proportion range of the electrolyte material is wide, a higher crossover probability can be set to explore the combination. The mutation probability introduces random variation, increases population diversity, and avoids premature convergence, and the mutation probability can be set according to the number of material types and the proportion range of the electrolyte. The proportion range of the electrolyte material refers to the proportion range of each material in the electrolyte, which is used to define the boundary of the search space and directly affects the feasibility and quality of the solution.

[0040] In the embodiments of the present application, the population size can be set to 10000, the iteration number can be set to 1000, the crossover probability can be set to 0.7, the mutation probability can be set to 0.01, and the proportion range of the electrolyte material can be set according to the actual situation and specific requirements of different electrolytes.

[0041] S320, generating an initial population of electrolyte formulations according to the working parameters, the initial population including a plurality of individuals, each individual representing a proportion scheme of an electrolyte formulation.

[0042] 10000 initial populations are randomly generated in the given electrolyte material proportion range, the initial population including a plurality of individuals, each individual representing a proportion scheme of an electrolyte formulation. In the embodiments of the present application, the total sum of the proportion of each electrolyte material in each proportion scheme of the electrolyte formulation needs to be controlled to be 100%, and the proportion of each material in the electrolyte can be generated one by one, and needs to meet the set range and the remaining range. For example, the set range of the Nth material is 30%-40%, the total sum of the generated proportions of the first N-1 materials is 68%, and the generated proportion of the Nth material needs to meet the set range 30%-40% and the remaining 0%-34% at the same time, that is, the generated proportion of the Nth material is within 30%-34%. The proportion of the last material can be 100% minus the total sum of the proportions of the previous materials, but needs to meet the set range requirement of the material.

[0043] S330, respectively performing performance indicator data prediction on the individuals in the initial population according to each performance indicator prediction model, to obtain performance indicator prediction data.

[0044] In the embodiment of the present application, the three performance index prediction models corresponding to the battery cycle capacity retention rate, the battery voltage drop value and the battery internal resistance are used to predict the individuals in the initial population, and each electrolyte formula proportion scheme is input to output the corresponding performance index prediction data.

[0045] In the embodiment of the present application, if the individual A is better than the individual B in all performance indexes, the individual A is considered to dominate the individual B, but if all individuals in the initial population do not dominate the individual A, the individual A is considered to be a non-dominated individual. According to the fast non-dominated sorting algorithm, all individuals not dominated by other individuals are found, thereby obtaining the front sequence, which is the optimal non-dominated solution set.

[0046] In the embodiment of the present application, if the individual A is better than the individual B in all performance indexes, the individual A is considered to dominate the individual B, but if all individuals in the initial population do not dominate the individual A, the individual A is considered to be a non-dominated individual. According to the fast non-dominated sorting algorithm, all individuals not dominated by other individuals are found, thereby obtaining the front sequence, which is the optimal non-dominated solution set.

[0047] S350, calculate the crowding distance of each individual in the front sequence.

[0048] In the embodiment of the present application, the crowding distance of the individual is calculated by combining the K nearest neighbor distance and the spatial density estimation. First, the Euclidean distance between each individual in the front sequence and other individuals is calculated, and then the target number of nearest neighbors with the smallest Euclidean distance is selected, and the average value of the Euclidean distance corresponding to these nearest neighbors is calculated to obtain the nearest neighbor crowding degree. The larger the value of the nearest neighbor crowding degree, the sparser the solution. The target number can be 5-10, which can be adjusted according to the size of the initial population. Then, the local density of each individual in the front sequence in the target space is calculated using the Gaussian kernel density, and the Gaussian crowding degree is obtained according to the local density. The Gaussian crowding degree is defined as the reciprocal of the local density, and the larger the value of the Gaussian crowding degree, the lower the density of the region where the solution is located. Finally, the nearest neighbor crowding degree and the Gaussian crowding degree are weighted and fused to obtain the crowding distance, wherein the weight coefficients of the nearest neighbor crowding degree and the Gaussian crowding degree can be adjusted according to the actual situation.

[0049] S360, select the parent population according to the front sequence and the crowding distance, and generate the child population by simulating binary crossover and polynomial mutation on the selected parent population.

[0050] In the embodiment of the present application, the tournament algorithm is used to select the parent, and the binary crossover algorithm and the polynomial mutation algorithm are used to generate the child population. In the tournament algorithm, the individual with high ranking in the front sequence is selected, and if the levels of the front sequence are the same, the individual with large crowding distance is selected as the parent. The binary crossover operation is simulated to generate the child individual, and the polynomial probability mutation is performed on the child individual to introduce diversity. The generation of the child individual needs to control the proportion of each material of the electrolyte to meet the variable range, while ensuring that the material proportion of the electrolyte is 100%. Optionally, the variable range is sorted, and the variable with small variable range is processed first, and then the variable with large variable range is processed. The previous variable is crossed or mutated normally, and is controlled within the variable range. If it exceeds the range, the boundary value is taken. For the last variable, 100% is subtracted from the proportion of the previous variable to obtain the value of the last variable. It is judged whether the variable is within the corresponding variable range. If not, the child individual is regenerated.

[0051] S370, the parent population and the child population are merged, non-dominated sorting and crowding distance calculation are performed, the next generation population is iteratively generated, and the iteration is performed until the iteration number reaches the preset number. The non-dominated individuals in the current population are used as the electrolyte formula proportion set.

[0052] In the embodiment of the present application, after the parent population and the child population are merged, the non-dominated sorting and the crowding distance calculation are performed again. According to the size of the front sequence and the crowding distance, 10000 individuals are selected to form the next generation population. The non-dominated sorting and the crowding distance calculation are iteratively performed to generate the next generation population, and the iteration is performed until the iteration number reaches 1000 times. The non-dominated individuals in the current population are used as the electrolyte formula proportion set.

[0053] The non-dominated sorting genetic algorithm combined with the performance index prediction model is used for multi-objective optimization in the embodiment of the present application. The formula proportion set meeting the multi-performance index can be quickly screened from the massive data, the number of experiments is reduced, the material consumption and the equipment use cost are reduced, and thus the research and development efficiency is greatly improved and the cost is reduced.

[0054] S400, a target electrolyte formula proportion scheme is selected from the electrolyte formula proportion set according to the target performance index requirement.

[0055] In the embodiment of the present application, a suitable electrolyte formula proportion scheme can be selected according to different performance index requirements. For example, for the performance index of cycle capacity retention rate, the electrolyte formula proportion scheme with the largest cycle capacity retention rate in the set is selected as the target electrolyte formula proportion scheme. If the scheme with the maximum comprehensive value is selected according to the proportion of the performance index, the performance index comprehensive value of all the electrolyte formula proportion schemes is counted, and the electrolyte formula proportion scheme with the maximum performance index comprehensive value is selected as the target electrolyte formula proportion scheme.

[0056] Embodiment 2 Based on the same technical concept as Embodiment 1, the present embodiment provides a formula proportion optimization system of electrolyte. Figure 4 is a structural schematic diagram of the formula proportion optimization system of electrolyte provided by the present embodiment, as shown in Figure 4 The formula proportion optimization system of electrolyte 200 includes: A data acquisition module 210 is configured to acquire electrolyte experimental data, the electrolyte experimental data at least including electrolyte formula proportion and performance index data of corresponding battery.

[0057] A model construction module 220 is configured to construct a prediction model of each performance index according to the electrolyte experimental data, and obtain a plurality of performance index prediction models.

[0058] A target optimization module 230 is configured to perform multi-objective optimization by using a non-dominated sorting genetic algorithm in combination with the prediction model of each performance index, and obtain a set of electrolyte formula proportions. A scheme selection module 240 is configured to select a target electrolyte formula proportion scheme from the set of electrolyte formula proportions according to target performance index requirements.

[0059] The formula proportion optimization system of electrolyte provided by the present embodiment takes into account multiple key performance indexes of the battery through multi-objective optimization, and the selected electrolyte formula proportion scheme can make the battery exhibit good comprehensive performance in different use scenarios, thereby improving the reliability and market competitiveness of the product.

[0060] It can be understood that the implementation of the formula proportion optimization method of electrolyte described in Embodiment 1 is also applicable to the present embodiment and can achieve the same technical effects, and therefore will not be described again.

[0061] Embodiment 3 Based on the same concept, the present embodiment further provides an electronic device, Figure 5 is a structural schematic diagram of an electronic device provided by the present embodiment, as shown in Figure 5 The electronic device 300 can include a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 complete communication with each other through the communication bus 340. The processor 310 can invoke logical instructions in the memory 330 to execute steps of the formula proportion optimization method of electrolyte in each of the above embodiments. For example, including: S100, acquiring electrolyte experimental data, the electrolyte experimental data at least including electrolyte formula proportion and performance index data of corresponding battery; S200, constructing a prediction model of each performance index according to electrolyte experimental data, and obtaining a plurality of performance index prediction models; S300, performing multi-objective optimization by using a non-dominated sorting genetic algorithm in combination with the prediction models of the performance indexes, and obtaining an electrolyte formula proportion set; S400, selecting a target electrolyte formula proportion scheme from the electrolyte formula proportion set according to a target performance index requirement.

[0062] The processor 310 can be a central processing unit (CPU). The processor can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination thereof.

[0063] In addition, the logical instructions in the memory 330 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the technical solutions that essentially contribute to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0064] The memory 330 can include a program storage area and a data storage area. The program storage area can store an operating system, at least one application required by the at least one function, and the like. The data storage area can store data created by the processor and the like. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory can optionally include a memory disposed remotely from the processor, which can be connected to the processor through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0065] Embodiment 4 Based on the same concept, the embodiments of the present application also provide a computer readable storage medium, which stores a computer program containing at least one code, the at least one code being executable by a host device to control the host device to implement the steps of the electrolyte formula proportion optimization method in the above embodiments. For example, it includes: S100, obtaining electrolyte experimental data, the electrolyte experimental data at least including electrolyte formula proportion and corresponding battery performance index data; S200, constructing a performance index prediction model according to the electrolyte experimental data, to obtain a plurality of performance index prediction models; S300, performing multi-objective optimization by using a non-dominated sorting genetic algorithm in combination with the performance index prediction models, to obtain an electrolyte formula proportion set; S400, selecting a target electrolyte formula proportion scheme from the electrolyte formula proportion set according to a target performance index requirement.

[0066] Based on the same technical concept, the embodiments of the present application also provide a computer program, which, when executed by a host device, is used to implement the above method embodiments.

[0067] The computer program can be stored in whole or in part on a computer readable storage medium packaged together with the processor, or can be stored in part or in whole on a memory not packaged together with the processor.

[0068] Based on the same technical concept, the embodiments of the present application also provide a processor, which is used to implement the above method embodiments. The processor can be a chip.

[0069] To sum up, the application provides the electrolyte formula proportion optimization method, system, electronic equipment and storage medium, the performance index prediction model is constructed, the association between the formula proportion and the performance is clearly presented, and accurate guidance is provided for research and development. Meanwhile, the non-dominated sorting genetic algorithm is combined with the performance index prediction model for multi-objective optimization, a formula proportion set meeting multiple performance indexes can be quickly screened from massive data, the number of experiments is reduced, material consumption and equipment use cost are reduced, research and development efficiency is greatly improved, and cost is reduced. Through multi-objective optimization, multiple key performance indexes of the battery are considered, and the screened electrolyte formula proportion can make the battery exhibit good comprehensive performance in different use scenarios, and improve product reliability and market competitiveness.

[0070] Reference herein to“an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. It is expressly understood that the embodiments described herein are merely possible examples of the application and are thus not limiting to the scope of the application.

[0071] The above-described embodiments only express several implementation manners of the application, which are described in detail and specifically, but should not be understood as limitation to the patent scope of the application. It should be pointed out that, for ordinary skilled in the art, several modifications and improvements can be made without departing from the concept of the application, which are all within the protection scope of the application.

[0072] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the application, but not limit the application; although the application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that: the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application.

Claims

1. A method of optimizing the formulation ratio of an electrolyte, characterized in that, The method comprises: obtaining electrolyte experimental data, the electrolyte experimental data at least including electrolyte formula proportion and corresponding battery performance index data; constructing a prediction model of each of the performance index data according to the electrolyte experimental data, to obtain a plurality of performance index prediction models; performing multi-objective optimization by using a non-dominated sorting genetic algorithm in combination with each of the performance index prediction models, to obtain an electrolyte formula proportion set; selecting a target electrolyte formula proportion scheme from the electrolyte formula proportion set according to a target performance index requirement.

2. The method for optimizing the formulation ratio of an electrolyte according to claim 1, characterized in that, The method comprises: constructing a prediction model of each of the performance index data; dividing the electrolyte experimental data into a training set and a validation set; performing model training on each of the prediction models by using the training set, to obtain a plurality of initial performance index prediction models; adjusting parameters of each of the initial performance index prediction models by using the validation set, to obtain a plurality of the performance index prediction models.

3. The method for optimizing the formulation ratio of an electrolyte according to claim 1, characterized in that, The method comprises: setting working parameters of the non-dominated sorting genetic algorithm; generating an initial population of electrolyte formulae according to the working parameters, the initial population including a plurality of individuals, each of the individuals representing an electrolyte formula proportion scheme; performing performance index data prediction on the individuals in the initial population according to each of the performance index prediction models, to obtain performance index prediction data; performing non-dominated sorting on the individuals in the initial population according to the performance index prediction data, to obtain a front sequence; calculating a crowding distance of each of the individuals in the front sequence; selecting a parent population according to the front sequence and the crowding distance, performing simulated binary crossover and polynomial mutation on the selected parent population, to generate a child population; merging the parent population and the child population, performing non-dominated sorting and crowding distance calculation, and iteratively generating a next generation population, until the number of iterations reaches a preset number, and non-dominated individuals in the current population are taken as the electrolyte formula proportion set.

4. The method for optimizing the formulation ratio of an electrolyte according to claim 3, characterized in that, The method comprises: calculating a Euclidean distance between each of the individuals in the front sequence and other individuals; selecting a target number of nearest neighbors with the smallest Euclidean distance, calculating an average value of the Euclidean distances corresponding to the nearest neighbors, to obtain a nearest neighbor crowding distance; calculating a local density of each of the individuals in the front sequence in a target space by using a Gaussian kernel density, to obtain a Gaussian crowding distance according to the local density; performing weighted fusion of the nearest neighbor crowding distance and the Gaussian crowding distance, to obtain the crowding distance.

5. The method for optimizing the formulation ratio of an electrolyte according to claim 1, characterized in that, The performance index data at least includes battery cycle capacity retention rate, battery voltage drop value and battery internal resistance.

6. The method for optimizing the formulation ratio of an electrolyte according to claim 1, characterized in that, The prediction model includes at least one or more of a support vector machine model, a gradient boosting decision tree model and a deep neural network model.

7. The method for optimizing the formulation ratio of an electrolyte according to claim 3, characterized in that, The working parameters at least include population size, iteration number, crossover probability, mutation probability and proportion range of each electrolyte material.

8. A system for optimizing the formulation ratio of an electrolyte, characterized in that, The system comprises: a data acquisition module configured to acquire electrolyte experiment data, the electrolyte experiment data at least including electrolyte formula proportion and performance index data of corresponding batteries; a model construction module configured to construct a prediction model of each of the performance index data according to the electrolyte experiment data, and obtain a plurality of performance index prediction models; a target optimization module configured to perform multi-objective optimization by using a non-dominated sorting genetic algorithm in combination with each of the performance index prediction models, and obtain a set of electrolyte formula proportions; a scheme selection module configured to select a target electrolyte formula proportion scheme from the set of electrolyte formula proportions according to a target performance index requirement.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the electrolyte formula proportion optimization method according to any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the electrolyte formula proportion optimization method according to any one of claims 1-7.

Images