Gel solid electrolyte prediction method and device, electronic equipment and storage medium

Abstract

The application discloses a gel solid electrolyte prediction method and device, electronic equipment and a storage medium, and relates to the technical field of data processing. The method comprises the following steps: obtaining a plurality of characteristics of a gel solid electrolyte; using the characteristics to determine a target hyperparameter combination of an initial regression analysis model with the conductivity and the capacitance of the gel solid electrolyte as regression targets; using the characteristics as training samples and the conductivity and the capacitance of the gel solid electrolyte as training labels to train a target regression analysis model; using the initial regression analysis model with the target hyperparameter combination as the target regression analysis model; using the trained target regression analysis model to determine the numerical value corresponding to each characteristic and a first target value as a second target value; and the first target value is the target numerical value of the conductivity and the capacitance of the gel solid electrolyte. The application can quickly and accurately predict the numerical value of the conductivity and the capacitance of the gel solid electrolyte without physical testing, and can obtain the numerical value of each characteristic through reverse design.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, electronic device and storage medium for predicting gel solid electrolytes. Background Technology

[0002] Due to their unique physicochemical properties, gel solid electrolytes have wide applications in biomedical engineering, drug delivery, flexible electronics, and other fields. Conductivity and capacitance, as important performance indicators, directly affect their application effectiveness.

[0003] The related technologies have the following problems: 1) The measurement process of conductivity and capacitance of gel solid electrolytes is complex, time-consuming and costly; 2) It is difficult to manufacture gel solid electrolytes with target values ​​of conductivity and capacitance. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, electronic device, and storage medium for predicting gel solid electrolytes, so as to improve the measurement efficiency of the conductivity and capacitance of gel solid electrolytes and improve the efficiency of manufacturing gel solid electrolytes with target conductivity and capacitance values.

[0005] To achieve the above objectives, one aspect of this application proposes a method for predicting gel solid electrolytes, the method comprising the following steps:

[0006] To obtain multiple characteristics of gel solid electrolytes;

[0007] Using the aforementioned characteristics, the conductivity and capacitance of the gel solid electrolyte are used as regression targets to determine the target hyperparameter combination of the initial regression analysis model;

[0008] Using the aforementioned features as training samples and the conductivity and capacitance of the gel solid electrolyte as training labels, a target regression analysis model is trained; wherein, the initial regression analysis model using the target hyperparameter combination is used as the target regression analysis model;

[0009] The trained target regression analysis model is used to determine the numerical values ​​corresponding to each feature and the first target value as the second target value; wherein, the first target value is the target value of the conductivity and capacitance of the gel solid electrolyte.

[0010] In some embodiments, obtaining multiple characteristics of the gel solid electrolyte includes the following steps:

[0011] Obtain the characteristic data of the gel solid electrolyte;

[0012] The characteristics of the gel solid electrolyte are constructed based on the aforementioned characteristic data.

[0013] In some embodiments, determining the target hyperparameter combination of the initial regression analysis model using the conductivity and capacitance of the gel solid electrolyte as regression targets based on each of the aforementioned features includes the following steps:

[0014] The initial hyperparameter combinations of the regression analysis model are determined using a grid search.

[0015] Based on the regression objective, k-fold cross-validation is performed on each of the features to select a second hyperparameter combination from each of the first hyperparameter combinations; where k is a positive integer;

[0016] The importance score of each feature is output based on the initial regression analysis model using the second hyperparameter;

[0017] The features whose importance scores are lower than a preset score threshold are removed to obtain the remaining features.

[0018] Returning to the method of using grid search to determine each of the first hyperparameter combinations of the initial regression analysis model, we obtain the current combinations of the first hyperparameters;

[0019] Based on the remaining features and the current combinations of first hyperparameters, the k-fold cross-validation method is performed on each feature based on the regression target to select a second hyperparameter combination from the combinations of first hyperparameters; the currently obtained second hyperparameter combination is used as the target parameter combination.

[0020] In some embodiments, performing k-fold cross-validation on each of the features based on the regression objective to select a second hyperparameter combination from each of the first hyperparameter combinations includes the following steps:

[0021] Each of the current features is randomly divided into k parts as k data subsets;

[0022] One of the data subsets is used as the validation set, and the remaining data subsets are used as the training set. Based on the regression objective and the initial regression analysis model, the fitting calculation is performed using each of the training sets, and the average coefficient of determination corresponding to the validation set is calculated.

[0023] The subset of data for which the corresponding average coefficient of determination has not been calculated is used as the current validation set, and the remaining subset of data is used as the current training set. The process is repeated, with one subset of data used as the validation set and the remaining subset of data used as the training set. Based on the regression objective and the initial regression analysis model, the model is fitted using each training set, and the average coefficient of determination corresponding to the validation set is calculated, until the average coefficient of determination corresponding to all subsets of data is calculated.

[0024] The first hyperparameter combination in the initial regression analysis model is determined as the second hyperparameter combination when the average coefficient of determination with the maximum value is calculated.

[0025] In some embodiments, determining the numerical values ​​corresponding to each feature and the first target value using the trained target regression analysis model as the second target value includes the following steps:

[0026] Repeatedly adjust the values ​​of each feature and input each feature into the trained target regression analysis model until the conductivity and capacitance values ​​output by the trained target regression analysis model match the first target value, and obtain the current value of each feature as the second target value.

[0027] In some embodiments, the step of repeatedly adjusting the values ​​of each feature and inputting each feature into the trained target regression analysis model until the conductivity and capacitance values ​​output by the trained target regression analysis model match the first target value, and obtaining the current value of each feature as the second target value, includes the following steps:

[0028] A population is created, comprising multiple particles; wherein each particle represents a numerical combination of the various features, and the numerical values ​​of the various features serve as the positions of the corresponding particles.

[0029] Initialize the positions of each of the particles;

[0030] Each particle is input into the trained target regression analysis model, and the sum of squares of the differences between the output value of the trained target regression analysis model and the target value is calculated as the fitness of the corresponding particle.

[0031] If the current fitness of the particle is lower than its historical lowest fitness, then its historical lowest fitness is updated to the current fitness, and its historical best position is updated to the current position.

[0032] If the current fitness of the particle is lower than the historical lowest fitness of the population, then the historical lowest fitness of the population is updated to the current lowest fitness, and the historical best position of the population is updated to the current best position.

[0033] The positions of each particle are updated within the limit values ​​of velocity and position;

[0034] Returning to the step of inputting each particle into the trained target regression analysis model, the sum of squares of the differences between the output value of the trained target regression analysis model and the target value is calculated as the fitness of the corresponding particle, until a preset termination condition is reached, and the current historical best position of the population is determined as the second target value.

[0035] In some embodiments, the method further includes the following steps:

[0036] The actual values ​​of conductivity and capacitance of a gel solid electrolyte sample are measured; wherein the gel solid electrolyte sample is manufactured according to the second target value of each of the aforementioned characteristics.

[0037] By comparing the actual value with the first target value, a comparison result is obtained;

[0038] If the comparison result does not meet the expected conditions, then return to the previous step of repeatedly adjusting the values ​​of each feature and inputting each feature into the trained target regression analysis model until the current comparison result meets the expected conditions.

[0039] To achieve the above objectives, another aspect of this application provides a gel solid electrolyte prediction device, the device comprising:

[0040] The feature acquisition unit is used to acquire multiple features of the gel solid electrolyte;

[0041] The hyperparameter determination unit is used to determine the target hyperparameter combination of the initial regression analysis model by utilizing each of the aforementioned features, with the conductivity and capacitance of the gel solid electrolyte as regression targets.

[0042] The model training unit is used to train a target regression analysis model using the features as training samples and the conductivity and capacitance of the gel solid electrolyte as training labels; wherein the initial regression analysis model using the target hyperparameter combination is used as the target regression analysis model.

[0043] The reverse design unit is used to determine the values ​​corresponding to the first target value for each feature using the trained target regression analysis model as the second target value; wherein the first target value is the target value of the conductivity and capacitance of the gel solid electrolyte.

[0044] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method for predicting gel solid electrolytes.

[0045] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for predicting gel solid electrolytes.

[0046] The embodiments of this application include at least the following beneficial effects:

[0047] This application can acquire multiple features of a gel solid electrolyte; utilize each feature, with the conductivity and capacitance of the gel solid electrolyte as regression targets, to determine the target hyperparameter combination of the initial regression analysis model; use the features as training samples, and the conductivity and capacitance of the gel solid electrolyte as training labels, to train the target regression analysis model; wherein, the initial regression analysis model using the target hyperparameter combination is used as the target regression analysis model; the trained target regression analysis model is used to determine the values ​​corresponding to each feature and the first target value as the second target value; wherein, the first target value is the target value of the conductivity and capacitance of the gel solid electrolyte. Based on the multiple features of the gel solid electrolyte, this application can quickly and accurately predict the specific values ​​of the conductivity and capacitance of the gel solid electrolyte through the trained target regression analysis model without physical testing, improving measurement efficiency and reducing costs; in addition, this application can also reverse design the specific values ​​of each feature in the gel solid electrolyte based on the target values ​​of conductivity and capacitance, and based on the feature values ​​obtained by reverse design, gel solid electrolytes with conductivity and capacitance as target values ​​can be manufactured efficiently. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 A schematic flowchart illustrating a method for predicting gel solid electrolytes provided in an embodiment of this application;

[0050] Figure 2 This is a schematic diagram of the structure of a gel solid electrolyte prediction device provided in an embodiment of this application;

[0051] Figure 3 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0053] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0054] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0055] 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 herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0056] This application provides a method, apparatus, electronic device, and storage medium for predicting gel solid electrolytes. The technical solution includes: acquiring multiple features of the gel solid electrolyte; using each feature and the conductivity and capacitance of the gel solid electrolyte as regression targets to determine a target hyperparameter combination for an initial regression analysis model; using the features as training samples and the conductivity and capacitance of the gel solid electrolyte as training labels to train the target regression analysis model; wherein the initial regression analysis model using the target hyperparameter combination is used as the target regression analysis model; and using the trained target regression analysis model to determine the values ​​corresponding to each feature and a first target value as second target values; wherein the first target value is the target value of the conductivity and capacitance of the gel solid electrolyte. Based on multiple features of the gel solid electrolyte, this application can quickly and accurately predict the specific values ​​of the conductivity and capacitance of the gel solid electrolyte through the trained target regression analysis model without physical testing, improving measurement efficiency and reducing costs. Furthermore, this application can also reverse-engineer the specific values ​​of each feature in the gel solid electrolyte based on the target values ​​of conductivity and capacitance, and based on the reverse-engineered feature values, gel solid electrolytes with target conductivity and capacitance values ​​can be manufactured efficiently.

[0057] This application provides a method for predicting gel solid electrolytes, relating to the field of data processing technology. The gel solid electrolyte prediction method provided in this application can be applied to a terminal, a server, or software running on a terminal or server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the gel solid electrolyte prediction method, but is not limited to the above forms.

[0058] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0059] Reference Figure 1 This application provides a method for predicting gel solid electrolytes, which may include, but is not limited to, steps S100 to S130, as follows:

[0060] S100: Acquire multiple characteristics of gel solid electrolytes.

[0061] Specifically, parameters such as the polymer, crosslinking agent, initiator, solution, solute, and manufacturing environment conditions required for manufacturing gel solid electrolytes can be used as characteristics of gel solid electrolytes.

[0062] Furthermore, S100 may include S101 to S102:

[0063] S101: Obtain the characteristic data of the gel solid electrolyte.

[0064] For example, the characteristic data of the gel solid electrolyte in this embodiment may include:

[0065] Polymer composition: The type of polymer in the gel (e.g., polyacrylamide, polyvinyl alcohol).

[0066] Polymer proportions: the mass percentage of each polymer in the gel, the mass of each polymer, and the total mass of the polymers.

[0067] Crosslinking agent: percentage of crosslinking agent by mass, mass of crosslinking agent.

[0068] Initiator: Initiator mass percentage, initiator mass.

[0069] The types and mass percentages of solutes in an electrolyte solution.

[0070] The type and mass percentage of solvent in the electrolyte solution.

[0071] The molar concentration of the electrolyte solution.

[0072] Reaction time: The duration of the reaction during gel synthesis.

[0073] Reaction conditions: including reaction temperature and whether defoaming treatment is performed.

[0074] Ambient temperature and length of time for drying and settling.

[0075] Conductivity and capacitance of gel solid electrolytes.

[0076] S102: Construct each of the aforementioned features of the gel solid electrolyte based on the aforementioned characteristic data.

[0077] Optionally, before constructing features, this embodiment can first preprocess the feature data, such as data cleaning to remove duplicate records in the feature data; missing value filling, such as filling missing ambient temperature data with 25℃.

[0078] Then, features related to conductivity and capacitance are extracted and constructed based on the characteristic data:

[0079] Create a separate column for features, where a value of 0 represents a single network and a value of 1 represents a dual network.

[0080] For each polymer, a corresponding column is created, with Boolean values ​​indicating whether it exists. A Boolean value of 0 in the column means that the polymer does not exist, and a Boolean value of 1 in the column means that the polymer exists.

[0081] The total mass percentage of polymers in the gel, the mass percentage of each polymer in the gel, and the total mass of the polymers are used as characteristics.

[0082] The mass percentage and mass of crosslinking agent and initiator in the gel are used as characteristics.

[0083] Solutes are characterized by their ionic radius, ionic charge, relative atomic mass of the corresponding atom, molar concentration of the ion, and mass percentage in the gel.

[0084] Solvents are characterized by their dielectric constant, molecular weight, boiling point, and mass percentage of the solution in the gel.

[0085] The molar concentration of the electrolyte solution is used as a characteristic.

[0086] The reaction time and temperature during gel synthesis are used as characteristics, and whether defoaming treatment is performed is indicated by Boolean values.

[0087] Ambient temperature and the length of time for drying and settling are also considered characteristics.

[0088] Standardization is performed by scaling the features of the solvent and solute components to a Min-Max range of 0 to 1. For the solute, scaling is applied to the ionic radius and the relative atomic mass of the corresponding atom. For the solvent, scaling is applied to the dielectric constant, molecular weight, and boiling point.

[0089] S110: Using the characteristics described above, the conductivity and capacitance of the gel solid electrolyte are used as regression targets to determine the target hyperparameter combination of the initial regression analysis model.

[0090] It is understood that in this embodiment, the conductivity and capacitance of the gel solid electrolyte are used as regression targets, and the initial regression analysis model is used to analyze each feature to determine the target hyperparameter combination of the initial regression analysis model.

[0091] Among them, the regression analysis model with undetermined hyperparameters is used as the initial regression analysis model, and the target hyperparameter combination is a number of hyperparameters that make the predictive performance of the initial regression analysis model optimal.

[0092] Furthermore, S110 may include S111 to S116:

[0093] S111: Use grid search to determine the various combinations of first hyperparameters of the initial regression analysis model;

[0094] S112: Based on the regression objective, perform k-fold cross-validation on each of the features to select a second hyperparameter combination from each of the first hyperparameter combinations; where k is a positive integer;

[0095] S113: Output importance scores for each of the features based on the initial regression analysis model using the second hyperparameters;

[0096] S114: Remove the features whose importance scores are lower than a preset score threshold to obtain the remaining features;

[0097] S115: Return to the first hyperparameter combinations determined by grid search for the initial regression analysis model, and obtain the current first hyperparameter combinations;

[0098] S116: Based on the remaining features and the current combinations of first hyperparameters, perform k-fold cross-validation on each feature based on the regression target to select a second hyperparameter combination from the combinations of first hyperparameters; the currently obtained second hyperparameter combination is used as the target parameter combination.

[0099] Furthermore, S112 may include S1121 to S1124:

[0100] S1121: Randomly divide each of the current features into k parts as k data subsets;

[0101] S1122: Using one of the data subsets as the validation set and the remaining data subsets as the training set, the regression target and the initial regression analysis model are used to perform fitting calculations using each of the training sets, and the average coefficient of determination corresponding to the validation set is calculated.

[0102] S1123: The data subset for which the corresponding average coefficient of determination has not been calculated is used as the current validation set, and the remaining data subset is used as the current training set. The process is repeated, using one data subset as the validation set and the remaining data subset as the training set. Based on the regression objective and the initial regression analysis model, the model is fitted using each training set, and the average coefficient of determination corresponding to the validation set is calculated, until the average coefficient of determination corresponding to all data subsets is calculated.

[0103] S1124: The first hyperparameter combination in the initial regression analysis model will be determined as the second hyperparameter combination when the average coefficient of determination with the maximum value is calculated.

[0104] For example, the following description will be given in conjunction with specific embodiments.

[0105] Optionally, this embodiment uses a random forest model as the initial regression analysis model, dividing each feature into an 80% training set and a 20% validation set. The features in the validation set can be used to verify the accuracy of the initial regression analysis model derived from the training set.

[0106] In some embodiments, the k-fold cross-validation method can be a 10-fold cross-validation method, that is, k is 10. k can also be any positive integer less than 10 (e.g., 5, 7, 8, etc.) or greater than 10 (e.g., 11, 12, 13, etc.).

[0107] Multiple combinations of first hyperparameters are obtained using grid search, and second hyperparameter combinations are obtained by filtering from each combination of first hyperparameters using 10-fold cross-validation.

[0108] In the grid search, for each combination of the first hyperparameters, the training set is randomly divided into 10(k) equal subsets, each of which serves as a validation set. The remaining 9 subsets are merged to form the training set. An initial regression model is trained using the current combination of the first hyperparameters and this training set. The trained initial regression model is then used to predict the validation set data, and the determination coefficient at the k-th fold is calculated. The formula is:

[0109]

[0110] Among them, y i It is the label value of the i-th feature in the validation set. It is the label prediction value of the initial regression analysis model for the i-th feature in the validation set. This represents the average of all features in the validation set corresponding to the labels, and n is the number of features in the validation set. Two random forest models are trained for each label, and the methods for selecting the optimal hyperparameters and filtering features are the same.

[0111] Optionally, k can take any positive integer from 1 to 10. When k is 10, the average coefficient of determination R0 2 The formula is:

[0112]

[0113] When k takes other values, the 10 in the formula for calculating the average coefficient of determination is replaced with the value of k.

[0114] The fitting calculation returns the average coefficient of determination for each validation set. The first hyperparameter combination with the highest average coefficient of determination for the 10 validation sets is taken as the second hyperparameter combination.

[0115] In the case of the second hyperparameter combination, the importance score calculated by the random forest model for each feature is output. Features with importance scores lower than the preset score threshold (e.g., one percent) are removed. After removal, grid search and 10-fold cross-validation are performed again. The hyperparameter combination with the highest average determination coefficient of the 10 validation subsets in the current cross-validation is taken as the target hyperparameter combination.

[0116] Retrain the random forest model using the target hyperparameter combination, validate the results on the 20% validation set, and save the trained random forest model if the average determination coefficients of the validation set and the training set are similar.

[0117] S120: Using the features as training samples and the conductivity and capacitance of the gel solid electrolyte as training labels, train the target regression analysis model; wherein, the initial regression analysis model using the target hyperparameter combination is used as the target regression analysis model.

[0118] It is understood that this embodiment can train an initial regression analysis model with a target hyperparameter combination (i.e., a target regression analysis model), wherein each feature labeled with specific values ​​of conductivity and capacitance is used as training data, that is, each feature and its specific value are input into the target regression analysis model, and the target regression analysis model outputs the corresponding specific values ​​of conductivity and capacitance.

[0119] S130: The trained target regression analysis model is used to determine the numerical values ​​corresponding to each feature and the first target value as the second target value; wherein, the first target value is the target value of the conductivity and capacitance of the gel solid electrolyte.

[0120] Specifically, in practical applications, it may be necessary to have gel solid electrolytes with specific conductivity and capacitance values. In this embodiment, the target values ​​of conductivity and capacitance can be defined first, and then the characteristic values ​​required to manufacture the gel solid electrolyte with the target values ​​can be reverse-engineered using the trained target regression analysis model.

[0121] It should be noted that the first target value includes the target values ​​for conductivity and capacitance, while the second target value includes the target values ​​for each characteristic.

[0122] Furthermore, S130 may include:

[0123] S131: Repeatedly adjust the values ​​of each of the features and input each of the features into the trained target regression analysis model until the conductivity and capacitance values ​​output by the trained target regression analysis model match the first target value, and obtain the current values ​​of each feature as the second target value.

[0124] Specifically, if the difference between the value output by the target regression analysis model and the first target value is within a preset range, then the two can be considered to match. At this time, the current value of each feature is the second target value. That is, the conductivity and capacitance values ​​of the gel solid electrolyte manufactured with the current values ​​of each feature match the first target value.

[0125] Furthermore, S131 may include:

[0126] S1311: Create a population, the population comprising multiple particles; wherein, one particle represents a numerical combination of the various features, and the numerical values ​​of the various features serve as the positions of the corresponding particles;

[0127] S1312: Initialize the position of each particle;

[0128] S1313: Input each particle into the trained target regression analysis model, and then calculate the sum of squares of the differences between the output value of the trained target regression analysis model and the target value as the fitness of the corresponding particle;

[0129] S1314: If the current fitness of the particle is lower than its historical lowest fitness, then update its historical lowest fitness to the current fitness and update its historical best position to the current position.

[0130] S1315: If the current fitness of the particle is lower than the historical lowest fitness of the population, then update the historical lowest fitness of the population to the current lowest fitness, and update the historical best position of the population to the current best position.

[0131] S1316: Update the position of each particle within the limit values ​​of velocity and position;

[0132] S1317: Return to the step of inputting each particle into the trained target regression analysis model, and then calculate the sum of squares of the differences between the output value of the trained target regression analysis model and the target value as the fitness of the corresponding particle, until the preset termination condition is reached, and determine the current historical best position of the population as the second target value.

[0133] Specifically, the lowest historical fitness of the population is the lowest historical fitness among all individual particles, and the best historical position of the population is the best historical position among all individual particles.

[0134] The following description will be based on specific embodiments.

[0135] First, target values ​​for conductivity and capacitance are defined as the starting point for reverse engineering.

[0136] Adjust the values ​​of each feature in the target regression analysis model after input training. For example, adjust the mass percentage of polymer in gel, mass and mass percentage of crosslinking agent and initiator, molar concentration of solution and mass percentage of solute solvent, reaction time, reaction temperature, etc., to determine the combination of feature values ​​that can make the conductivity and capacitance output by the target regression analysis model reach the target values.

[0137] Create a population consisting of multiple particles; where each particle represents a numerical combination of features, and the numerical values ​​of each feature serve as the position of the corresponding particle.

[0138] Particle initialization: Uniformly distribute particles to a fixed polymer solution solute combination, and randomly initialize variable characteristics.

[0139] Polymer type: Fixed polymer type, fixed single and double network.

[0140] Solution type: Fixed solution type, such as solute type and solvent type.

[0141] Randomly initialized variable characteristics: polymer mass percentage in gel, polymer mass, mass of crosslinking agent and initiator, molar concentration of solution, mass percentage of solute, mass percentage of solvent, reaction time, reaction temperature, drying and standing time, etc.

[0142] Using defined values ​​of conductivity and capacitance as targets, fitness is the sum of squares of the differences between the conductivity and capacitance predicted by the target regression analysis model based on the particle's input features and the target values. The smaller the sum of squares, the closer the combination of input features corresponding to the particle is to the required values.

[0143] If the current particle's fitness S J Below its individual historical lowest fitness S J,best Then update the individual's historical minimum fitness S. J,best and the individual's historical best position X J,best If the current particle's fitness S J Below the lowest fitness in the population's history S all,best Then update the population's historical lowest fitness S. all,best and the population's historical best position X all,best .

[0144] Update the position of each particle. For particle velocities and positions that exceed the maximum value range, limit them to the maximum value. For particle positions and velocities that exceed the minimum value range, limit them to the minimum value.

[0145] The calculation stops when a preset termination condition is met. This preset termination condition can be that the sum of the squares of the differences between the conductivity and capacitance predicted by the target regression analysis model and the target values ​​is less than a threshold, or that the maximum number of iterations has been reached. The current historical best position of the population is then output as the target value for each feature.

[0146] Furthermore, to further improve the accuracy of the actual values ​​of conductivity and capacitance, embodiments of this application may also include S141 to S143:

[0147] S141: Measure the actual values ​​of conductivity and capacitance of the gel solid electrolyte sample; wherein the gel solid electrolyte sample is manufactured according to the second target value of each of the aforementioned characteristics;

[0148] S142: Compare the actual value with the first target value to obtain a comparison result;

[0149] S143: If the comparison result does not meet the expected conditions, then return to the process of repeatedly adjusting the values ​​of each feature and inputting each feature into the trained target regression analysis model until the current comparison result meets the expected conditions.

[0150] Specifically, in this embodiment, experiments can be designed based on the target values ​​of each feature, and corresponding gel solid electrolyte samples can be manufactured. Then, the actual conductivity and capacitance of the manufactured gel solid electrolyte samples can be measured and compared with the expected results (first target values) to obtain comparison results.

[0151] If the measured actual conductivity and capacitance do not match the first target value, that is, the comparison result does not meet the expected conditions, the values ​​of each feature are adjusted until the comparison result meets the expected conditions. Optionally, in this embodiment, S131 can be repeatedly executed to adjust the values ​​of each feature until the comparison result meets the expected conditions.

[0152] The beneficial effects of the embodiments of this application may include:

[0153] A method is provided to predict the conductivity and capacitance of gel solid electrolytes without physical measurement, thus significantly reducing experimental time and cost. Furthermore, this method can continuously improve prediction accuracy by updating training data, and can inversely design the specific values ​​of various characteristics of the gel solid electrolyte based on the target values ​​of conductivity and capacitance.

[0154] Reference Figure 2 This application also provides a gel solid electrolyte prediction device, which can implement the above-described gel solid electrolyte prediction method. The device includes:

[0155] The feature acquisition unit is used to acquire multiple features of the gel solid electrolyte;

[0156] The hyperparameter determination unit is used to determine the target hyperparameter combination of the initial regression analysis model by utilizing each of the aforementioned features, with the conductivity and capacitance of the gel solid electrolyte as regression targets.

[0157] The model training unit is used to train a target regression analysis model using the features as training samples and the conductivity and capacitance of the gel solid electrolyte as training labels; wherein the initial regression analysis model using the target hyperparameter combination is used as the target regression analysis model.

[0158] The reverse design unit is used to determine the values ​​corresponding to the first target value for each feature using the trained target regression analysis model as the second target value; wherein the first target value is the target value of the conductivity and capacitance of the gel solid electrolyte.

[0159] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0160] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described gel solid electrolyte prediction method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0161] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0162] Please see Figure 3 , Figure 3 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0163] The processor 301 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0164] The memory 302 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 302 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 302 and called and executed by the processor 301 using the gel solid electrolyte prediction method of the embodiments of this application.

[0165] Input / output interface 303 is used to implement information input and output;

[0166] The communication interface 304 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0167] Bus 305 transmits information between various components of the device (e.g., processor 301, memory 302, input / output interface 303, and communication interface 304);

[0168] The processor 301, memory 302, input / output interface 303, and communication interface 304 are connected to each other within the device via bus 305.

[0169] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described gel solid electrolyte prediction method.

[0170] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0171] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0172] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0173] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0174] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0175] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0176] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0177] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0178] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0179] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0180] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0181] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0182] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for predicting gel solid electrolytes, characterized in that, The method includes the following steps: To obtain multiple characteristics of gel solid electrolytes; Using the aforementioned characteristics, the conductivity and capacitance of the gel solid electrolyte are used as regression targets to determine the target hyperparameter combination of the initial regression analysis model; Using the aforementioned features as training samples and the conductivity and capacitance of the gel solid electrolyte as training labels, a target regression analysis model is trained; wherein, the initial regression analysis model using the target hyperparameter combination is used as the target regression analysis model; The trained target regression analysis model is used to determine the numerical values ​​corresponding to each feature and the first target value as the second target value; wherein, the first target value is the target value of the conductivity and capacitance of the gel solid electrolyte; The process involves repeatedly adjusting the values ​​of each feature and inputting each feature into the trained target regression analysis model until the conductivity and capacitance values ​​output by the trained target regression analysis model match the first target value, thereby obtaining the current values ​​of each feature as the second target value. This includes the following steps: A population is created, comprising multiple particles; wherein each particle represents a numerical combination of the various features, and the numerical values ​​of the various features serve as the positions of the corresponding particles. Initialize the positions of each of the particles; Each particle is input into the trained target regression analysis model, and the sum of squares of the differences between the output value of the trained target regression analysis model and the target value is calculated as the fitness of the corresponding particle. If the current fitness of the particle is lower than its historical lowest fitness, then its historical lowest fitness is updated to the current fitness, and its historical best position is updated to the current position. If the current fitness of the particle is lower than the historical lowest fitness of the population, then the historical lowest fitness of the population is updated to the current lowest fitness, and the historical best position of the population is updated to the current best position. The positions of each particle are updated within the limit values ​​of velocity and position; Returning to the step of inputting each particle into the trained target regression analysis model, the sum of squares of the differences between the output value of the trained target regression analysis model and the target value is calculated as the fitness of the corresponding particle, until a preset termination condition is reached, and the current historical best position of the population is determined as the second target value.

2. The method for predicting gel solid electrolytes according to claim 1, characterized in that, The process of obtaining multiple characteristics of the gel solid electrolyte includes the following steps: Obtain the characteristic data of the gel solid electrolyte; The characteristics of the gel solid electrolyte are constructed based on the aforementioned characteristic data.

3. The method for predicting gel solid electrolytes according to claim 1, characterized in that, The process of determining the target hyperparameter combination of the initial regression analysis model using the conductivity and capacitance of the gel solid electrolyte as regression targets, based on each of the aforementioned features, includes the following steps: The initial hyperparameter combinations of the regression analysis model are determined using a grid search. Based on the regression objective, k-fold cross-validation is performed on each of the features to select a second hyperparameter combination from each of the first hyperparameter combinations; where k is a positive integer; The importance score of each feature is output based on the initial regression analysis model using the second hyperparameter; The features whose importance scores are lower than a preset score threshold are removed to obtain the remaining features. Returning to the method of using grid search to determine each of the first hyperparameter combinations of the initial regression analysis model, we obtain the current combinations of the first hyperparameters; Based on the remaining features and the current combinations of first hyperparameters, the k-fold cross-validation method is performed on each feature based on the regression target to select a second hyperparameter combination from the combinations of first hyperparameters; the currently obtained second hyperparameter combination is used as the target hyperparameter combination.

4. The method for predicting gel solid electrolytes according to claim 3, characterized in that, The step of performing k-fold cross-validation on each feature based on the regression objective to select a second hyperparameter combination from each of the first hyperparameter combinations includes the following steps: Each of the current features is randomly divided into k parts as k data subsets; One of the data subsets is used as the validation set, and the remaining data subsets are used as the training set. Based on the regression objective and the initial regression analysis model, the fitting calculation is performed using each of the training sets, and the average coefficient of determination corresponding to the validation set is calculated. The subset of data for which the corresponding average coefficient of determination has not been calculated is used as the current validation set, and the remaining subset of data is used as the current training set. The process is repeated, with one subset of data used as the validation set and the remaining subset of data used as the training set. Based on the regression objective and the initial regression analysis model, the model is fitted using each training set, and the average coefficient of determination corresponding to the validation set is calculated, until the average coefficient of determination corresponding to all subsets of data is calculated. The first hyperparameter combination in the initial regression analysis model is determined as the second hyperparameter combination when the average coefficient of determination with the maximum value is calculated.

5. The method for predicting gel solid electrolytes according to claim 1, characterized in that, The method further includes the following steps: The actual values ​​of conductivity and capacitance of a gel solid electrolyte sample are measured; wherein the gel solid electrolyte sample is manufactured according to the second target value of each of the aforementioned characteristics. By comparing the actual value with the first target value, a comparison result is obtained; If the comparison result does not meet the expected conditions, then return to the previous step of repeatedly adjusting the values ​​of each feature and inputting each feature into the trained target regression analysis model until the current comparison result meets the expected conditions.

6. A gel solid electrolyte prediction device, characterized in that, The device includes: The feature acquisition unit is used to acquire multiple features of the gel solid electrolyte; The hyperparameter determination unit is used to determine the target hyperparameter combination of the initial regression analysis model by utilizing each of the aforementioned features, with the conductivity and capacitance of the gel solid electrolyte as regression targets. The model training unit is used to train a target regression analysis model using the features as training samples and the conductivity and capacitance of the gel solid electrolyte as training labels; wherein the initial regression analysis model using the target hyperparameter combination is used as the target regression analysis model. The reverse design unit is used to determine the numerical values ​​corresponding to the first target value for each feature using the trained target regression analysis model as the second target value; wherein, the first target value is the target value of the conductivity and capacitance of the gel solid electrolyte; The process involves repeatedly adjusting the values ​​of each feature and inputting each feature into the trained target regression analysis model until the conductivity and capacitance values ​​output by the trained target regression analysis model match the first target value, thereby obtaining the current values ​​of each feature as the second target value. This includes the following steps: A population is created, comprising multiple particles; wherein each particle represents a numerical combination of the various features, and the numerical values ​​of the various features serve as the positions of the corresponding particles. Initialize the positions of each of the particles; Each particle is input into the trained target regression analysis model, and the sum of squares of the differences between the output value of the trained target regression analysis model and the target value is calculated as the fitness of the corresponding particle. If the current fitness of the particle is lower than its historical lowest fitness, then its historical lowest fitness is updated to the current fitness, and its historical best position is updated to the current position. If the current fitness of the particle is lower than the historical lowest fitness of the population, then the historical lowest fitness of the population is updated to the current lowest fitness, and the historical best position of the population is updated to the current best position. The positions of each particle are updated within the limit values ​​of velocity and position; Returning to the step of inputting each particle into the trained target regression analysis model, the sum of squares of the differences between the output value of the trained target regression analysis model and the target value is calculated as the fitness of the corresponding particle, until a preset termination condition is reached, and the current historical best position of the population is determined as the second target value.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement a gel solid electrolyte prediction method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a gel solid electrolyte prediction method as described in any one of claims 1 to 5.

Images