Battery material database construction method and system

Abstract

The invention discloses a battery material database construction method and system. The method comprises the following steps: manually inputting raw material and battery material information; automatically acquiring experimental data of the battery material; converting the acquired unstructured warehousing data into structured warehousing data; carrying out model training, testing and evaluation to obtain an optimized model; on the basis of the optimized model, online analysis is carried out on warehousing data, and the potential relation between the performance and the characteristics of the battery material is searched for actual prediction or classification tasks. According to the method and the system, the integrity and the richness of database information are ensured by collecting the relevant information of the whole life cycle of the battery material, a solid data foundation is laid for application, and by performing data mining and analysis on the warehousing data, the safety of the battery material is improved. And a scientific basis is provided for performance evaluation, life prediction and formulation of a personalized research and development scheme of the battery material.

Description

Technical Field

[0001] The present invention relates to the technical field of battery material database construction, and in particular to a battery material database construction method and system. Background Art

[0002] The field of battery materials research faces challenges such as complex data, long performance evaluation cycles, numerous influencing factors, and uneven R&D results. With the continuous advancement of machine learning and artificial intelligence technologies, significant progress has been made in the R&D efficiency and performance optimization of battery materials. However, many unresolved issues remain, such as: What are the key factors affecting the cycle stability and energy density of battery materials? How to build accurate performance prediction models to guide the design of new materials? The R&D process of battery materials generates a massive amount of experimental data, including material composition, structural characteristics, electrochemical properties, preparation process parameters, testing conditions, and environmental factors. These data may have complex correlations with the ultimate performance of the battery materials.

[0003] For example, the invention application with application number 202211091213.0 discloses a battery database processing method, device, electronic device, and storage medium. This application scheme reduces the probability of errors in the first battery information, reduces malicious uploads, illegal uploads, and other upload operations that are not permitted by the second node, and improves the information security of the main database. However, this scheme can only collect information on finished batteries, with the purpose of improving the information security of the main database, and cannot collect and process battery materials.

[0004] Currently, major domestic battery manufacturers and research institutions lack dedicated battery material databases to systematically collect, organize, and analyze this data. Traditional data recording methods, such as paper-based lab records, are not only inefficient but also prone to data loss or incompleteness, hindering their full potential in guiding new material development and optimizing the performance of existing materials. Therefore, there is an urgent need to improve existing technical solutions to overcome one or more of these issues. Summary of the Invention

[0005] In response to the above-mentioned problems, the purpose of the present invention is to provide a method and system for constructing a battery material database, collect information on the entire life cycle of battery materials, analyze and discover the potential relationships and patterns between battery material performance and characteristics, and provide a scientific basis for performance evaluation, life prediction and formulation of personalized R&D plans for battery materials.

[0006] Embodiments of the present invention provide a method and system for constructing a battery material database.

[0007] A first aspect: A method for constructing a battery material database, comprising:

[0008] S1. Manually enter the basic information of raw materials, basic information of battery materials and processing parameter information;

[0009] S2, automatically collect experimental data of battery materials through software programming;

[0010] S3. Use structured processing technology to convert the collected unstructured data into structured data;

[0011] S4. Perform model training, testing, and evaluation based on structured stored data to obtain an optimized model;

[0012] S5. Perform online analysis of incoming data based on the optimized model to find the potential relationship between battery material performance and characteristics for actual prediction or classification tasks.

[0013] By combining manual entry and automated collection, the comprehensiveness and accuracy of battery material data are ensured. Combined with structured processing technology and machine learning models, it can efficiently process and analyze data, discover the potential relationship between battery material performance and characteristics, and provide support for actual prediction and classification tasks.

[0014] Optionally, in S1, the basic information of the raw materials includes: raw material material information, manufacturer information, batch information, and physical and chemical parameters; wherein the raw material material information includes: metals and their compounds, silicon-carbon materials, lithium salts, organic solvents, additives, polymer materials, and ceramic materials;

[0015] Basic information of battery materials includes: product type, product sub-item, project number, experiment number, experiment type, molecular formula, project leader, project participants, and experiment start and end dates; among them, product types include: positive electrode, negative electrode, separator, electrolyte, and auxiliary materials;

[0016] The processing parameter information of battery materials includes: material synthesis formula and process parameters.

[0017] The information scope of raw materials, battery materials and processing parameters is defined in detail to ensure the integrity and traceability of the database, providing a solid foundation for subsequent data analysis and model training.

[0018] Optionally, in S2, the experimental data of the battery material includes: a physical and chemical test characterization report and an electrical performance test report.

[0019] The types of experimental data are clarified, including physical and chemical test and electrical performance test reports, to ensure that the data in the database is scientific and practical, and can better support the performance evaluation of battery materials.

[0020] Optionally, the S3 includes the steps of:

[0021] S31. Collect unstructured data such as characters, pictures, charts and tables;

[0022] S32. Processing unstructured data based on structured processing technology to obtain structured data;

[0023] S33, performing denoising and normalization processing on the structured data to obtain processed structured data;

[0024] S34. Storing the processed structured data in a database to obtain the stored data;

[0025] Among them, structured processing technologies include: data cleaning and preprocessing technology, natural language processing technology, text mining technology, entity recognition and relationship extraction technology, deep learning technology and OCR (optical character recognition) technology.

[0026] Through structured processing technology, unstructured data is converted into structured data, and denoising and normalization are performed to ensure the quality and consistency of the data, providing high-quality data input for subsequent model training.

[0027] Optionally, the S4 includes the following steps:

[0028] S41. Use feature engineering to convert incoming data into feature vectors;

[0029] S42. Select an appropriate machine learning model based on the specific task of the incoming data;

[0030] S43. Use the extracted feature vectors and other relevant features to perform model training, testing, and evaluation;

[0031] S44. Obtain the optimized model and deploy it to the production environment.

[0032] Through feature engineering and machine learning model training, testing, and evaluation, the model performance is optimized and deployed in the production environment to ensure the effectiveness and reliability of the model in practical applications.

[0033] Optionally, the feature engineering includes: data preprocessing, feature construction, feature conversion, feature encoding and feature selection technology;

[0034] Machine learning model types include: logistic regression, support vector machines, random forests, and neural networks.

[0035] The specific techniques of feature engineering and types of machine learning models are listed to ensure that data can be effectively converted into feature vectors, and to select appropriate models for training to improve the model's prediction and classification capabilities.

[0036] Optionally, the online analysis function in S5 includes: linear regression analysis, logistic regression analysis, decision tree analysis, vector machine analysis and Bayesian classification analysis.

[0037] It provides a variety of online analysis functions, including linear regression, logistic regression, decision tree, vector machine and Bayesian classification analysis, ensuring that the performance and characteristic relationship of battery materials can be analyzed from different angles and promoting the research and development of new materials.

[0038] A second aspect: A battery material database construction system, comprising:

[0039] The data entry module is used to manually enter the basic information of raw materials, basic information of battery materials and processing parameter information, as well as software programming to automatically collect experimental data of battery materials;

[0040] The structured processing module is used to convert the collected and sorted unstructured data into structured data for storage;

[0041] The feature extraction and model training module is used to vectorize the incoming data, select appropriate models for training, testing, and evaluation, and deploy the optimized models in the actual production environment for actual prediction or classification tasks;

[0042] The online analysis module, based on the optimized model, is used to analyze and find the potential relationship between battery material performance and characteristics online, promoting the research and development of new battery materials;

[0043] The database global statistics module performs statistical analysis on the battery material data in the incoming data and displays it visually in a variety of chart styles.

[0044] A complete battery material database system has been built, including data entry, structured processing, model training and online analysis modules, to ensure that the system can efficiently process and analyze battery material data and promote the research and development of new battery materials.

[0045] A third aspect: An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method provided in the first aspect are implemented.

[0046] A fourth aspect: A non-transitory computer-readable storage medium having a computer program stored thereon, which implements the steps of the method provided in the first aspect when executed by a processor.

[0047] Beneficial effects of the present invention:

[0048] 1. The method and system of the present invention comprehensively collect all relevant information throughout the entire life cycle of battery materials, ensuring the integrity and richness of database information, laying a solid data foundation for the establishment and application of the database. The dual-channel mode of "manual entry + automated collection" builds a full-dimensional database covering raw material properties (material / batch / physicochemical parameters), processing technology, experimental data and performance indicators, enhancing data traceability and realizing the ability to integrate data throughout the entire life cycle of battery materials.

[0049] 2. The present invention structures the data entering the database so that each data entering the database has a unified format and standard, and is convenient for data mining and analysis in the later stage.

[0050] 3. The present invention uses machine learning technology to extract feature vectors and train models from stored data, providing a scientific basis for performance evaluation, life prediction, and formulation of personalized R&D plans for battery materials.

[0051] 4. The battery material database constructed by the method of this application can significantly improve the standardization of experimental and production data collection. Its construction process is more intelligent and efficient, which can greatly save labor costs. This not only provides valuable experience for the construction of databases in the field of domestic battery material research and development, but also provides a solid "big data" support for optimizing the research and development process and production process of battery materials. These technological innovations have significantly improved the data-driven capabilities of battery material research and development. Actual measurements can shorten the development cycle of new positive electrode materials and reduce research and development costs, which can provide the industry with an efficient digital research and development infrastructure. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flow chart of a method for constructing a battery material database according to an embodiment of the present invention;

[0053] Figure 2 This is a flow chart of using structured processing technology to perform structured processing on the input data in an embodiment of the present invention;

[0054] Figure 3 This is a flowchart of model training for structured data based on feature engineering and machine learning in an embodiment of the present invention;

[0055] Figure 4 A schematic diagram of modules of a battery material database construction system according to an embodiment of the present invention;

[0056] Figure 5 It is a schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0057] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar symbols throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0058] Currently, major domestic battery manufacturers and research institutions lack dedicated battery material databases to systematically collect, organize, and analyze this data. Traditional data recording methods, such as paper-based lab records, are not only inefficient but also prone to data loss or incompleteness, hindering their full potential in guiding new material research and development and optimizing the performance of existing materials.

[0059] In view of the above problems, the present invention provides a method for constructing a battery material database. Figure 1 As shown, the method includes:

[0060] S1. Manually enter the basic information of raw materials, basic information of battery materials and processing parameter information.

[0061] Manually enter the basic information of raw materials, including: raw material information, manufacturer information, batch information, physical and chemical parameters, etc.

[0062] The raw material information includes: metals and their compounds, silicon-carbon materials, lithium salts, organic solvents, additives, polymer materials and ceramic materials, etc.

[0063] Specifically, the physical parameters include: mechanical properties (elastic properties, plastic properties, hardness and impact resistance, etc.), thermal properties (heat capacity, thermal expansion coefficient, thermal conductivity, thermal diffusion coefficient, etc.), optical properties (refractive index, reflectivity, transmittance, etc.), magnetic properties (magnetic permeability, magnetic induction intensity, etc.) and other properties (density, specific gravity, porosity, hydrophilicity, etc.).

[0064] Chemical parameters include: acidity, alkalinity, ion mobility, electronic conductivity, redox reaction activity, chemical stability, lithium ion intercalation and deintercalation ability, etc.

[0065] Manually enter the basic information and processing parameter information of battery materials. The basic information of battery materials includes product type, product sub-item, project number, experiment number, experiment type, molecular formula, project leader, project participants, experiment start and end dates, etc.

[0066] The product types include: positive electrode, negative electrode, diaphragm, electrolyte and auxiliary materials.

[0067] In addition, different product types in the basic information of battery materials correspond to different product sub-items. For example, the product sub-items corresponding to positive electrode products include lithium iron phosphate, lithium manganese iron phosphate, and lithium nickel cobalt manganese oxide, while the product sub-items corresponding to negative electrode products include graphite, silicon carbon, and silicon oxide. The specific product sub-items are determined by the R&D status of the organization that established the database.

[0068] The processing parameter information of battery materials includes: material synthesis formula and process parameters, etc.

[0069] S2. Automatically collect experimental data of battery materials through software programming.

[0070] Software programming automates the collection of experimental data for battery materials. This data includes physical and chemical characterization reports and electrical performance test reports. This data is derived from professional test and characterization equipment and cannot be subsequently modified to prevent data tampering.

[0071] In addition to automatically collecting experimental data on battery materials through software programming, R&D teams and personnel for different materials can also manually enter experimental data into the corresponding sub-database according to the database part they are responsible for to enrich the completeness of the information; to avoid errors in manually entering data, subsequent modifications are allowed to increase the flexibility of data entry.

[0072] S3. Use structured processing technology to convert the collected unstructured data into structured data.

[0073] Structured processing technology is used to perform artificial intelligence extraction on the collected and sorted unstructured incoming data and convert it into structured incoming data.

[0074] Battery material data collected through manual entry and software coding automation exist in various formats, including unstructured data such as characters, pictures, charts and tables. Since the collected and organized data are of different types, various types of data need to be structured before being stored in the database to ensure that the data follows certain rules and formats. At the same time, it is easier to be processed by analysis and mining tools later to extract valuable information.

[0075] Denoising and normalizing structured data in a unified format. In the field of battery materials research, facing huge data sets, which are often mixed with irrelevant data items and data points with significant differences, denoising and normalization techniques can effectively filter out irrelevant data that interferes with analysis while reducing data differences, significantly improving the stability and accuracy of battery materials datasets.

[0076] Among them, structured processing technologies include: data cleaning and preprocessing technology, natural language processing technology, text mining technology, entity recognition and relationship extraction technology, deep learning technology, OCR (optical character recognition) technology, etc.; unstructured stored data includes: characters, pictures, charts and tables, etc.

[0077] For example, the process of converting scanning electron microscope photographs of materials into structured data mainly relies on image analysis and data processing technology. By preprocessing the image and using image processing technologies such as threshold segmentation and edge detection, specific targets or features in the image are identified, and the morphological features of the segmented target are extracted, such as size (diameter, thickness, area), shape (circularity, rectangularity), etc., and the above feature information is organized into a structured data table to meet the requirements of subsequent machine learning.

[0078] Specifically, such as Figure 2 As shown in the figure, the steps for converting unstructured incoming data into structured data in a unified format through structured processing technology include:

[0079] S31. Collect unstructured data such as characters, pictures, charts and tables;

[0080] S32. Processing unstructured data based on structured processing technology to obtain structured data;

[0081] S33, performing denoising and normalization processing on the structured data to obtain processed structured data;

[0082] S34. Storing the processed structured data in a database.

[0083] In the embodiment of the present invention, structuring and denoising normalization of the data to be stored is a key step to improve data quality, enhance data analysis efficiency, promote data mining and machine learning model performance, and enhance system stability.

[0084] S4. Perform model training, testing, and evaluation based on structured stored data to obtain the optimized model.

[0085] Feature engineering techniques are used to transform structured data into vector features that better represent potential issues. Through sophisticated feature engineering, efficient and powerful vector feature sets are constructed, providing a rich information foundation for machine learning models to accurately capture the inherent laws and patterns of the data.

[0086] The data after feature engineering is input into the machine learning model for model training. During the training process, the model algorithm will learn the mapping relationship between the data based on the input vector features and labels, and build a prediction model. It will then be trained, tested, and evaluated based on the training set, test set, and validation set.

[0087] The model is evaluated to check whether its performance meets the requirements. Evaluation indicators usually include accuracy, recall rate, F1 score, etc.

[0088] Specifically, such as Figure 3 As shown, the steps include:

[0089] S41. Use feature engineering to convert incoming data into feature vectors;

[0090] S42. Select an appropriate machine learning model based on the specific task of the incoming data;

[0091] S43. Use the extracted feature vectors and other relevant features to perform model training, testing, and evaluation;

[0092] S44. Obtain the optimized model and deploy it to the production environment.

[0093] In the embodiments of the present invention, feature engineering plays a vital role. It is a bridge between structured data and machine learning models and has a key impact on improving model performance, accuracy, and generalization ability.

[0094] Feature engineering can include techniques such as data preprocessing, feature construction, feature conversion, feature encoding, and feature selection. By converting data into features that better represent the underlying problem, machine learning performance can be improved. Machine learning methods include logistic regression, support vector machines, random forests, and neural networks.

[0095] S5. Perform online analysis of incoming data based on the optimized model to find the potential relationship between battery material performance and characteristics for actual prediction or classification tasks.

[0096] Based on the optimized model, the online analysis function of the machine learning model is used to conduct in-depth analysis of the battery material inventory data, accurately evaluate the factors affecting the performance of battery materials, and provide a basis for the formulation of subsequent material optimization plans.

[0097] Machine learning online analysis functions include: linear regression analysis, logistic regression analysis, decision tree analysis, vector machine analysis and Bayesian classification analysis, etc.

[0098] The present invention also discloses a battery material database construction system based on the construction method.

[0099] like Figure 4 As shown, the system includes a data entry module, a structured processing module, a feature extraction and model training module, an online analysis module, and a database global statistics module, among which:

[0100] The data entry module is used to manually enter the basic information of raw materials, basic information of battery materials and processing parameter information, as well as software programming to automatically collect experimental data of battery materials;

[0101] The structured processing module is used to convert the collected and sorted unstructured data into structured data for storage, where unstructured data includes characters, pictures, charts, and tables;

[0102] The feature extraction and model training module is used to vectorize the incoming data, select appropriate models for training, testing, and evaluation, and deploy the optimized models into actual production environments for actual prediction or classification tasks.

[0103] The online analysis module is based on the optimized model and is used to conduct online analysis to find the potential relationship between battery material performance and characteristics, promote the research and development of new battery materials, accurately evaluate the factors affecting battery material performance, and provide a basis for the formulation of subsequent material optimization plans. Among them, the machine learning online analysis functions include linear regression analysis, logistic regression analysis, decision tree analysis, vector machine analysis and Bayesian classification analysis.

[0104] The database's global statistics module performs statistical analysis on incoming battery material data and presents it visually in a variety of chart styles. For example, it can perform a simple analysis of data such as energy density and cycle efficiency, and then present it more intuitively in the form of a line chart, allowing R&D personnel to directly understand the results of this data and quickly draw conclusions.

[0105] In the above-mentioned invention embodiments, the basic information of raw materials includes: raw material material information, manufacturer information, batch information, and physical and chemical parameters; raw material material information includes: metals and their compounds, silicon-carbon materials, lithium salts, organic solvents, additives, polymer materials, and ceramic materials; basic information of battery materials includes: product type, product sub-item, project number, experiment number, experiment type, molecular formula, project leader, project participants, and experiment start and end dates; battery material product types include: positive electrode, negative electrode, separator, electrolyte, and auxiliary materials. Battery material processing parameter information includes: material synthesis formula and process parameters; battery material experimental data includes: physical and chemical test characterization report and electrical performance test report.

[0106] Based on the system structure of the present invention, it is understandable that the functionality of the database can be further expanded to allow specific data to be filtered out from the database and exported to different file formats, such as Excel, CSV, PDF, etc., to facilitate further analysis, report preparation or data sharing.

[0107] Optionally, the structured processing module includes using structured processing technologies to perform artificial intelligence extraction on the collected and organized information, converting it into structured data for storage. Structured processing technologies include: data cleaning and preprocessing technologies, natural language processing technologies, text mining technologies, entity recognition and relationship extraction technologies, deep learning technologies, and OCR (optical character recognition) technologies.

[0108] Optionally, in some embodiments, the feature extraction and model training module includes a feature vector extraction unit and a machine learning model training unit.

[0109] Among them, the feature vector extraction unit is used to convert data into features that can better represent potential problems through feature engineering technology; the machine learning model training unit is used to use the extracted feature vectors and other related features as model inputs for model training, and deploy the trained model to the production environment for actual prediction or classification tasks.

[0110] Optionally, in some embodiments, the system may further include an approval and release module.

[0111] Among them, the approval module is used to ensure that all data has been properly reviewed and approved to prevent unauthorized or potentially destructive changes from being directly applied to the production environment; the release module is used to safely and controllably deploy approved database changes to the target environment to minimize risks during the change process and ensure system stability and availability.

[0112] Optionally, in some embodiments, the system may further include a visualization module, which is used to display complex data in the database in an intuitive and easy-to-understand graphical manner, thereby helping users better understand the data, analyze the data and make decisions.

[0113] Optionally, in some embodiments, the system may further include a privacy and security module, which is used to ensure that the data stored in the database is not accessed, leaked, tampered with or destroyed without authorization, so as to fully protect the privacy and security of the data.

[0114] It should be clear that the technical architecture and effectiveness of the system of the present invention are closely consistent with the construction method previously described, so the specific details will not be repeated here.

[0115] It is worth noting that the division of modules within the system is only exemplary and is not fixed. According to different implementation scenarios of the present invention, it is entirely possible that the characteristics and functions that originally belonged to two or more modules can be integrated into a single module; conversely, the functions of a single module can also be subdivided into multiple modules to carry them. In addition, as a display component, the physical form and position of the module are not fixed. It can be a centralized physical unit or multiple parts distributed in the network. The implementation of the present invention is flexible and diverse. Users can choose a suitable module combination according to actual needs without the need for creative redesign. For those of ordinary skill in the art, understanding and implementing this solution should be intuitive and feasible.

[0116] The present invention also provides an electronic device, Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 5 As shown, the electronic device may include: a processor, a communications interface, a memory, and a communication bus, wherein the processor, the communications interface, and the memory communicate with each other via the communication bus. The processor may call logic instructions in the memory, for example, to execute the following method:

[0117] S1. Manually enter the basic information of raw materials, basic information of battery materials and processing parameter information;

[0118] S2, automatically collect experimental data of battery materials through software programming;

[0119] S3. Use structured processing technology to convert the collected unstructured data into structured data;

[0120] S4. Perform model training, testing, and evaluation based on structured stored data to obtain an optimized model;

[0121] S5. Perform online analysis of incoming data based on the optimized model to find the potential relationship between battery material performance and characteristics for actual prediction or classification tasks.

[0122] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0123] An embodiment of the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method provided in each of the above embodiments is implemented, for example, including:

[0124] S1. Manually enter the basic information of raw materials, basic information of battery materials and processing parameter information;

[0125] S2, automatically collect experimental data of battery materials through software programming;

[0126] S3. Use structured processing technology to convert the collected unstructured data into structured data;

[0127] S4. Perform model training, testing, and evaluation based on structured stored data to obtain an optimized model;

[0128] S5. Perform online analysis of incoming data based on the optimized model to find the potential relationship between battery material performance and characteristics for actual prediction or classification tasks.

[0129] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0130] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for constructing a battery material database, characterized in that: include: S1. Manually enter the basic information of raw materials, basic information of battery materials and processing parameter information; S2, automatically collect experimental data of battery materials through software programming; S3. Use structured processing technology to convert the collected unstructured data into structured data; S4. Perform model training, testing, and evaluation based on structured stored data to obtain an optimized model; S5. Perform online analysis of incoming data based on the optimized model to find the potential relationship between battery material performance and characteristics for actual prediction or classification tasks.

2. The method for constructing a battery material database according to claim 1, characterized in that: In S1: The basic information of raw materials includes: raw material information, manufacturer information, batch information, and physical and chemical parameters; raw material information includes: metals and their compounds, silicon-carbon materials, lithium salts, organic solvents, additives, polymer materials, and ceramic materials; Basic information of battery materials includes: product type, product sub-item, project number, experiment number, experiment type, molecular formula, project leader, project participants, and experiment start and end dates; among them, product types include: positive electrode, negative electrode, separator, electrolyte, and auxiliary materials; The processing parameter information of battery materials includes: material synthesis formula and process parameters.

3. The method for constructing a battery material database according to claim 1, characterized in that: In said S2, the experimental data of the battery material includes: physical and chemical test characterization report and electrical performance test report.

4. The method for constructing a battery material database according to claim 1, characterized in that: The S3 comprises the steps of: S31. Collect unstructured data such as characters, pictures, charts and tables; S32. Processing unstructured data based on structured processing technology to obtain structured data; S33, performing denoising and normalization processing on the structured data to obtain processed structured data; S34. Storing the processed structured data in a database to obtain the stored data; Among them, structured processing technologies include: data cleaning and preprocessing technology, natural language processing technology, text mining technology, entity recognition and relationship extraction technology, deep learning technology and OCR technology.

5. The method for constructing a battery material database according to claim 4, characterized in that: The S4 comprises the following steps: S41. Use feature engineering to convert incoming data into feature vectors; S42. Select an appropriate machine learning model based on the specific task of the incoming data; S43. Use the extracted feature vectors and other relevant features to perform model training, testing, and evaluation; S44. Obtain the optimized model and deploy it to the production environment.

6. The method for constructing a battery material database according to claim 5, characterized in that: The feature engineering includes: data preprocessing, feature construction, feature conversion, feature encoding and feature selection technology; Machine learning model types include: logistic regression, support vector machines, random forests, and neural networks.

7. The method for constructing a battery material database according to claim 1, characterized in that: The online analysis functions in S5 include: linear regression analysis, logistic regression analysis, decision tree analysis, vector machine analysis and Bayesian classification analysis.

8. A battery material database construction system applied to the method according to any one of claims 1 to 7, characterized in that: The system comprises: The data entry module is used to manually enter the basic information of raw materials, basic information of battery materials and processing parameter information, as well as software programming to automatically collect experimental data of battery materials; The structured processing module is used to convert the collected and sorted unstructured data into structured data for storage; The feature extraction and model training module is used to vectorize the incoming data, select appropriate models for training, testing, and evaluation, and deploy the optimized models in the actual production environment for actual prediction or classification tasks; The online analysis module, based on the optimized model, is used to analyze and find the potential relationship between battery material performance and characteristics online, promoting the research and development of new battery materials; The database global statistics module performs statistical analysis on the battery material data in the incoming data and displays it visually in a variety of chart styles.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of a method for constructing a battery material database according to any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for constructing a battery material database as claimed in any one of claims 1 to 7 are implemented.

Images