CN118275903B - Battery performance test method based on data analysis - Google Patents

Abstract

The application particularly relates to a battery performance test method based on data analysis, which comprises the following steps: constructing a dynamic data set, synchronizing and archiving data in real time, cleaning and checking quality of battery performance data, automatically identifying and eliminating abnormal values of the battery performance data, and filling missing battery performance data; training a support vector machine classification model capable of accurately distinguishing battery performance levels; and (3) automatic parameter adjustment and model self-optimization, and testing the battery performance by using a support vector machine classification model after the automatic parameter adjustment and model self-optimization. The application constructs a highly integrated, automatic and intelligent battery performance test system, and solves the limitations of the traditional test method in the aspects of efficiency, precision, response speed and cost control.

Description

A battery performance testing method based on data analysis

Technical Field

The invention belongs to the field of batteries, and in particular relates to a battery performance testing method based on data analysis.

Background Art

In the context of the rapid development of new energy technology and the electric vehicle industry, accurate testing and analysis of battery performance has become a key factor in ensuring product safety, extending service life, and improving user experience. With the continuous advancement of battery technology, new energy storage solutions including lithium-ion batteries and solid-state batteries have attracted increasing attention, and these technological innovations have put forward higher requirements for battery performance testing. Traditional battery testing methods often rely on static laboratory testing and manual data analysis, which is difficult to meet the needs of rapid and accurate evaluation of battery performance in large-scale production and diversified application scenarios. In recent years, the integrated application of technologies such as big data analysis, the Internet of Things (IoT), cloud computing, and artificial intelligence (AI) has brought revolutionary changes to battery performance testing. In particular, data analysis technology, by integrating massive historical and real-time test data, can reveal subtle trends in battery performance, predict potential failures, and optimize battery design and manufacturing processes. However, how to efficiently build dynamic data sets, synchronize data in real time, perform advanced data processing, and use advanced machine learning algorithms for performance prediction remains a technical challenge facing the industry.

Although the existing technology also discloses solid-state battery performance testing technology, its core is to optimize the detection process by dynamically adjusting the detection ratio and in-depth data analysis, aiming to improve detection efficiency and accuracy. However, the subjectivity and timeliness of the preset standards of this type of technology: the system relies on preset standard curves and thresholds to make performance judgments. The setting of these standards may be highly subjective, and with the development of battery technology and the advancement of materials, the standards need to be updated regularly, otherwise it may affect the accuracy of the test results.

Summary of the invention

The purpose of the present invention is to provide a battery performance testing method based on data analysis to solve the problems raised in the above background technology.

In order to solve the above technical problems, the present invention provides the following technical solutions:

A battery performance testing method based on data analysis, dynamic data set construction, and establishment of a database containing historical and real-time battery performance test data. The battery performance data includes battery cells and physical performance indicators under various working conditions;

Real-time data synchronization and archiving, using the Internet of Things, directly capture data from battery performance test equipment in real time and automatically upload it to the cloud server; at the same time, data archiving is performed to store historical data by time series or batch classification;

Perform battery performance data cleaning and quality inspection, automatically identify and remove abnormal values in battery performance data, and fill in missing battery performance data;

Based on historical battery performance test data, the support vector machine algorithm is used to select the optimal kernel function and parameter settings through cross-validation to train a support vector machine classification model that can accurately distinguish battery performance levels; in the feature selection stage, recursive feature elimination or correlation coefficient analysis-based methods are used to determine the key parameters that have the greatest impact on battery performance. The key parameters include charging rate changes, discharge platform stability, and internal resistance growth rate. The key parameters are used as inputs to the support vector machine classification model;

Automatic parameter adjustment and model self-optimization, configure online learning of the support vector machine classification model, so that the support vector machine classification model can periodically review the newly collected test data without interrupting service, and automatically adjust the support vector machine classification model parameters; use gradient descent or particle swarm optimization algorithms to find better hyperparameter combinations to adapt to subtle changes in battery performance over time and technological evolution;

The battery performance is tested using a support vector machine classification model after automatic parameter adjustment and model self-optimization.

Furthermore, real-time data capture from the battery performance test equipment and automatic uploading to the cloud server specifically includes:

The battery performance test equipment is upgraded to IoT, integrating smart sensors and microcontroller units (MCUs) that are responsible for accurately collecting battery performance data under various working conditions and real-time monitoring of equipment operating status; using the MQTT (Message Queuing Telemetry Transport) low-power wide area network (LPWAN) communication protocol to achieve efficient and secure data transmission between the test equipment and the cloud server; deploying edge computing nodes between the test equipment and the cloud to perform preliminary data preprocessing and screening tasks;

During the data transmission process, TLS/SSL protocol is used to encrypt all communication data to ensure the security and privacy of data during transmission. RESTful API interface is developed on the cloud server side to receive data push from edge computing nodes or directly from devices, automatically parse and store it in a distributed database. The database uses cloud-native databases such as Amazon DynamoDB or Google Cloud Spanner.

A real-time data monitoring platform is built in the cloud to display the latest data and trend analysis of each test point. At the same time, a threshold alarm mechanism is set up. Once abnormal battery performance or test equipment failure is detected, relevant personnel will be notified immediately by email, SMS or APP for rapid response and processing.

Furthermore, data archiving specifically includes:

A time series database management system is used to optimize the storage and query of data on battery performance changes over time. During the data archiving process, a unique identifier is created for each batch of batteries, and an efficient index structure is established in the database.

Furthermore, the battery performance data is cleaned and quality checked to automatically identify and remove abnormal values of the battery performance data and fill in missing battery performance data, including the following steps:

Build an automated data preprocessing pipeline, which is integrated before data is uploaded to the cloud and stored. The data has been cleaned and quality controlled before entering the core database. Use box plot analysis, Z-score method, and clustering algorithm-based anomaly detection technology to automatically identify outliers in battery performance data. For missing values in time series data, linear interpolation, spline interpolation, or time series prediction methods based on ARIMA model are used to fill in missing values to ensure data continuity.

For missing values in non-time series data, the K-nearest neighbor regression model or random forest regression model can be trained with complete data to predict missing values;

Analyze patterns or associations in the data and use data characteristics of similar batches or material groups as references to make reasonable estimates of missing values through nearest neighbor matching or mean/median filling methods.

Furthermore, a support vector machine algorithm is used to select the optimal kernel function and parameter settings through cross-validation, and a support vector machine classification model that can accurately distinguish battery performance levels is trained, including: using multiple kernel functions to determine the optimal kernel function and corresponding parameters through grid search combined with cross-validation; using multiple evaluation indicators to comprehensively evaluate model performance, including accuracy, recall rate, F1 score, and area under the AUC-ROC curve, to ensure that the model can not only accurately classify, but also balance the prediction effects of various categories.

Furthermore, the key parameters that have the greatest impact on battery performance are determined by recursive feature elimination or correlation coefficient analysis:

In the feature selection stage, first, the RFE process is started from the complete set of battery performance features, focusing on the accuracy, recall and F1 score evaluation indicators; using the initially trained support vector machine model, the importance score of each feature is calculated; one or several features with the lowest score are removed from the current feature set, the support vector machine model is retrained, and the model performance is re-evaluated;

Set a threshold to decide when to stop feature elimination, and verify the validity of feature subsets at each step through cross-validation to ensure the robustness of the feature selection process;

Before performing correlation coefficient analysis, the battery performance data was standardized to eliminate the dimension effect, and the Pearson correlation coefficient or Spearman rank correlation coefficient was used to evaluate the correlation between each pair of features;

Identify features related to battery performance while considering multicollinearity between features; eliminate redundant features that have low correlation with the target variable and are highly correlated with other features, and retain the key parameters that have the greatest impact on battery performance and are independent of each other, including charging rate changes, discharge platform stability, and internal resistance growth rate;

The features screened out based on the correlation coefficient analysis were merged with the results of the RFE process, and the features confirmed as important in both methods were given priority to finally determine the feature set for input into the support vector machine classification model.

Furthermore, the battery performance test specifically includes:

After the new battery performance test data is uploaded to the cloud server in real time via the IoT device, it goes through the data preprocessing pipeline to perform format unification, missing value checking and filling, and outlier identification and processing preprocessing steps;

The preprocessed data is then fed into the latest optimized support vector machine classification model;

The support vector machine model evaluates the performance of each batch of batteries based on the selected key features and classifies them into different performance levels.

Beneficial effects: The technical solution proposed in this application has achieved a significant technological leap in the field of battery performance testing:

Significantly improve test accuracy and efficiency: By building a dynamic data set containing historical and real-time data, combined with advanced IoT technology and edge computing, this solution achieves instant collection and efficient transmission of battery performance data, significantly shortening the test cycle. The intelligent application of the support vector machine (SVM) classification model, coupled with automatic parameter adjustment and self-optimization mechanisms, not only ensures high accuracy of test results, but also improves test efficiency and model generalization capabilities through continuous learning to adapt to changes in battery performance as technology develops.

Enhanced decision-making and response speed: The establishment of a real-time data monitoring platform and alarm system, combined with real-time data analysis in the cloud, enables abnormal battery performance or equipment failure to be quickly identified and trigger response measures, effectively improving safety in production and use, reducing downtime caused by failures, and enhancing the company's operational efficiency.

Optimize resource management and reduce costs: Intelligent data archiving and efficient data storage strategies have greatly optimized the data management process and reduced storage and operation and maintenance costs. Data preprocessing and model self-optimization reduce manual intervention and labor costs, while improving data quality and analysis efficiency, providing valuable data resources for long-term performance tracking and research and development.

Improve system flexibility and adaptability: Through online learning and intelligent feature selection strategies, this technical solution can flexibly adapt to the continuous innovation of battery technology and ensure that the test model always maintains the most cutting-edge analysis capabilities, which is crucial for rapidly changing battery materials and designs, and helps battery manufacturers respond quickly to market and technological changes.

To sum up, the technological progress of this application lies in the construction of a highly integrated, automated and intelligent battery performance testing system, which solves the limitations of traditional testing methods in efficiency, accuracy, response speed and cost control by deeply integrating big data, Internet of Things, cloud computing and AI technology.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a flow chart of the method of the present invention.

DETAILED DESCRIPTION

The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

The present application discloses a battery performance testing method based on data analysis, as shown in FIG1 , comprising the following steps:

Dynamic data set construction, establish a database containing historical and real-time battery performance test data. Battery performance data includes battery cells and physical performance indicators under various working conditions. The battery performance database is updated regularly to ensure the timeliness and comprehensiveness of model training data; the battery performance database includes the battery's production batch number, material composition, manufacturing date, basic information on technical specifications, and detailed test data of battery cell performance (such as number of charge and discharge cycles, capacity retention rate, internal resistance change) and physical properties (such as thermal stability, mechanical strength, safety performance test results); each battery performance test data is associated with the battery performance test environment parameters (such as temperature and humidity) to ensure the traceability and comparability of the test results;

Real-time data synchronization and archiving, using the Internet of Things, directly captures data from battery performance test equipment in real time and automatically uploads it to the cloud server to ensure data freshness;

Specific implementation details of capturing data from battery performance test equipment in real time and automatically uploading it to the cloud server:

IoT architecture configuration and implementation

Intelligent transformation of equipment: Upgrade battery performance test equipment to the Internet of Things, integrate smart sensors and microcontroller units (MCUs). These components are responsible for accurately collecting battery performance data under various working conditions, including but not limited to current, voltage, temperature, humidity, and real-time monitoring of equipment operating status to ensure the continuity and accuracy of data collection;

Standardized communication protocol: Adopt MQTT (Message Queuing Telemetry Transport) low-power wide area network (LPWAN) communication protocol to achieve efficient and secure data transmission between test equipment and cloud servers; MQTT protocol is particularly suitable for real-time data transmission requirements in IoT applications due to its lightweight and low bandwidth consumption, and supports publish/subscribe mode, which facilitates the management and data distribution of large-scale devices;

Edge computing optimization: Deploy edge computing nodes between the test equipment and the cloud to perform preliminary data preprocessing and screening tasks; edge computing can effectively reduce data transmission delays and reduce the processing burden of cloud servers, such as data denoising and outlier filtering operations, to ensure that the data uploaded to the cloud is both fresh and high-value;

Security encryption and identity authentication: During the data transmission process, TLS/SSL protocol is used to encrypt all communication data to ensure the security and privacy of data during transmission; at the same time, a unique identity is assigned to each test device, and strict access control and identity authentication mechanisms are implemented to prevent unauthorized access and data tampering;

Cloud platform interface and data storage: Develop a RESTful API interface on the cloud server side to receive data push from edge computing nodes or directly from devices, automatically parse and store it in a distributed database; the database uses cloud-native databases such as Amazon DynamoDB or Google Cloud Spanner to ensure high availability, persistence, and elastic expansion capabilities of data;

Real-time monitoring and alarm system: Build a real-time data monitoring platform in the cloud to display the latest data and trend analysis of each test point, and set a threshold alarm mechanism. Once abnormal battery performance or test equipment failure is detected, relevant personnel will be notified immediately by email, SMS or APP for rapid response and processing;

At the same time, data archiving is performed to store historical data by time series or batch classification to facilitate long-term tracking and analysis;

The specific implementation details of data archiving include the following aspects:

Time series data archiving strategy: Use a time series database management system (such as InfluxDB or TimescaleDB) to optimize the storage and query of data on battery performance changes over time. This type of database can efficiently handle high-frequency data writes and complex time series queries, and supports fast retrieval of historical data by time range and batch number, facilitating long-term performance trend analysis.

Batch classification indexing mechanism: During the data archiving process, a unique identifier is created for each batch of batteries, and an efficient indexing structure is established in the database; the index includes not only the batch number, but also key metadata such as production date, material type, and technical specifications, ensuring that the battery performance records of a specific batch or material group can be quickly located, facilitating performance comparison and trend research between batches;

Data compression and storage optimization: Considering the massive amount of battery performance data, an intelligent data compression algorithm is used to reduce storage space without affecting data integrity and analysis accuracy. At the same time, a data tiered storage strategy is implemented to store frequently accessed recent data on high-speed storage media (such as SSDs), while older historical data is migrated to low-cost cold storage solutions to balance storage costs and access efficiency.

Data lifecycle management: Develop data retention policies to automatically manage the data lifecycle; for example, set rules to automatically clean up historical data that exceeds the specified age, or dynamically adjust storage levels based on data access frequency to ensure that important and frequently used data is always easily accessible while properly releasing storage resources;

Data visualization and report generation: Develop customized data visualization tools or integrate existing BI (business intelligence) platforms, such as Tableau or Power BI, so that researchers and engineers can intuitively view and analyze long-term archived data; support the generation of periodic performance reports, including but not limited to batch performance comparison reports and time series trend analysis reports, to help decision makers quickly grasp the laws of battery performance changes and potential problems;

Data backup and recovery plan: Ensure high availability and disaster recovery capabilities of the data archiving system, and copy data to off-site or cloud storage services through regular full backup and incremental backup strategies; use services such as AWS S3 or Google Cloud Storage to provide high reliability and rapid recovery capabilities, ensuring that historical data can be quickly restored even in the face of unexpected situations, without affecting the continuity of long-term tracking and analysis work;

Through the above detailed implementation strategy, the data archiving process not only ensures the safe storage and efficient organization of historical data, but also promotes in-depth insights into battery performance and long-term trend analysis, performs battery performance data cleaning and quality inspection, automatically identifies and removes abnormal values in battery performance data, fills in missing battery performance data, and ensures the accuracy and completeness of the stored data;

The specific implementation details of battery performance data cleaning and quality inspection include the following steps:

1. Data preprocessing pipeline construction: Build an automated data preprocessing pipeline, which is integrated before the data is uploaded to the cloud and stored. The data has been cleaned and quality controlled before entering the core database. The pipeline includes but is not limited to data analysis, format unification, outlier detection and processing, and missing value filling key links;

2. Identification and processing of outliers:

Combining statistical methods with machine learning: Using box plot analysis, Z-score method and clustering algorithm-based anomaly detection technology to automatically identify outliers in battery performance data; for outliers found, adopt different processing strategies according to their distribution characteristics and business logic, such as direct elimination, marking as abnormal but retaining, or smoothing based on adjacent normal values;

Context-aware adjustment: Considering that battery performance is affected by test conditions (such as temperature and humidity), the outlier identification process will combine the test environment parameters and make context-sensitive adjustments to avoid misjudgment;

3. Missing value filling strategy:

Interpolation based on time series: For missing values in time series data, linear interpolation, spline interpolation or time series prediction methods based on ARIMA model are used to fill them to ensure data continuity;

Model prediction filling: For missing values of non-time series data, the K-nearest neighbor regression model or random forest regression model can be trained with complete data to predict missing values;

Pattern filling: Analyze patterns or associations in the data, use data characteristics of similar batches or material groups as references, and make reasonable estimates of missing values through nearest neighbor matching or mean/median filling methods;

4. Data quality monitoring and feedback loop:

Real-time data quality monitoring: embed a real-time monitoring module in the data cleaning process to continuously track data quality indicators, such as the proportion of missing values and the frequency of abnormal values, and trigger an early warning once the preset threshold is exceeded;

Closed-loop feedback and continuous optimization: Establish a data quality feedback mechanism to record and feed back the problems identified and measures taken during the cleaning process to the front-end data collection and equipment maintenance team to promote quality improvement at the source of data collection; at the same time, continuously adjust the cleaning algorithm and parameters based on feedback information to form a closed loop of continuous optimization;

Through the above detailed implementation details, the data cleaning and quality inspection process is closely integrated with the overall technical solution, ensuring the high quality and reliability of battery performance test data, laying a solid foundation for subsequent model training and performance evaluation;

Based on historical battery performance test data, the support vector machine algorithm is used to select the optimal kernel function and parameter settings through cross-validation to train a support vector machine classification model that can accurately distinguish battery performance levels;

Using the support vector machine algorithm, the optimal kernel function and parameter settings are selected through cross-validation to train a support vector machine classification model that can accurately distinguish the battery performance level, including: using multiple kernel functions (such as linear kernel, RBF kernel, polynomial kernel), and determining the optimal kernel function and corresponding parameters (such as C, γ) through grid search (Grid Search) combined with cross-validation (using k-fold cross-validation, the k value can be selected according to the amount of data and computing resources, such as 5 fold or 10 fold); but when implementing automatic parameter adjustment, you can consider using optimization strategies such as SMO (Sequential Minimal Optimization) and LibSVM built-in, and if necessary, fine-tune the loss function (such as hinge loss function) to adapt to the particularity of battery performance data;

A variety of evaluation indicators are used to comprehensively evaluate the model performance, including accuracy, recall, F1 score, and area under the AUC-ROC curve to ensure that the model can not only accurately classify but also balance the prediction effects of each category;

When there are multiple models with similar performance under different kernel functions and parameter combinations, the prediction results of multiple models can be integrated through voting, bagging, and boosting ensemble learning methods to improve the stability and accuracy of the overall prediction.

In the feature selection stage, recursive feature elimination (RFE) or correlation coefficient analysis-based methods are used to determine the key parameters that have the greatest impact on battery performance, such as charging rate variation, discharge platform stability, and internal resistance growth rate, as inputs to the support vector machine classification model;

In the feature selection stage, in order to ensure that the selected features can characterize the battery performance to the greatest extent and improve the accuracy of model prediction, we adopt the following specific implementation details to perform recursive feature elimination (RFE) and correlation coefficient analysis-based methods:

Recursive Feature Elimination (RFE)

Initialization and goal definition: First, the RFE process is started from a complete set of battery performance characteristics, with a clear goal of optimizing the performance of the support vector machine classification model, especially focusing on the accuracy, recall and F1 score evaluation indicators;

Feature scoring and ranking: Using the initially trained SVM model, calculate the importance score for each feature; this is based on the feature weights, i.e., how much they contribute to the model’s decision boundary; then, rank the features in descending order of their scores;

Stepwise feature elimination: Remove one or several features with the lowest scores from the current feature set, retrain the support vector machine model, and re-evaluate the model performance; this process is repeated, and the model is re-evaluated after each feature removal until the predetermined feature number threshold is reached or the model performance is no longer significantly improved;

Threshold and iteration strategy: Set a threshold to decide when to stop feature elimination, such as when the model performance drops by more than a certain percentage after eliminating the next feature; in addition, the validity of the feature subset can be verified at each step through cross-validation to ensure the robustness of the feature selection process;

Based on correlation coefficient analysis

Data preprocessing: Before performing correlation coefficient analysis, the battery performance data is standardized to eliminate the dimension effect and ensure that the correlation between different features is fair and meaningful;

Calculate the correlation matrix: Use the Pearson correlation coefficient or the Spearman rank correlation coefficient to evaluate the correlation between each pair of features;

Feature selection: Identify features related to battery performance (such as classification labels) while considering multicollinearity between features; remove redundant features that have low correlation with the target variable and are highly correlated with other features, and retain key parameters that have the greatest impact on battery performance and are independent of each other, such as charging rate changes, discharge platform stability, and internal resistance growth rate;

Combine RFE results: Combine the features selected based on correlation coefficient analysis with the results of the RFE process, give priority to the features that are confirmed as important in both methods, and finally determine the feature set that is input into the support vector machine classification model;

Comprehensive Implementation

In the entire feature selection process, a combination of iteration and cross-validation is used to ensure that the selected feature set can not only reflect the key indicators of battery performance, but also maintain stable performance on different data subsets. Through this comprehensive method, we can obtain a streamlined and efficient feature set, further improve the generalization ability and practicality of the support vector machine classification model, and ensure its accuracy and reliability in battery performance testing.

Automatic parameter adjustment and model self-optimization, configure online learning of the support vector machine classification model, so that the support vector machine classification model can periodically review the newly collected test data without interrupting service, and automatically adjust the support vector machine classification model parameters; this process uses gradient descent or particle swarm optimization algorithms to find a better combination of hyperparameters to adapt to the subtle changes in battery performance over time and technological evolution; at the same time, the support vector machine classification model will periodically perform self-verification and ensure the accuracy and reliability of the support vector machine classification model by comparing it with manually annotated test results;

Specific steps of “automatic parameter adjustment and model self-optimization”:

Online learning mechanism configuration

1. Real-time data access and buffering

Data access layer: A data access module is configured and integrated into the API interface of the cloud server, which is responsible for receiving the latest battery performance test data pushed by the edge computing node in real time; the data is temporarily stored in a high-speed buffer to reduce the direct impact on the online learning module and ensure service stability;

2. Timing trigger and data batching

Scheduler: Implement an intelligent scheduler to automatically trigger the model review and parameter adjustment process according to preset time intervals (such as daily, weekly, or on demand). The scheduler is also responsible for sorting the new data in the buffer into batches to form small batch data sets suitable for model training to control the computing resource consumption and response time of training.

3. Parameter optimization algorithm implementation

Gradient descent optimization: For the parameters of the support vector machine (such as C and γ), stochastic gradient descent (SGD) or batch gradient descent (BGD) is used. During each model review, the hyperparameters are gradually adjusted according to the gradient information on the new data set to optimize the model performance.

Particle Swarm Optimization (PSO): As a global optimization algorithm, PSO simulates the foraging behavior of bird flocks and searches for the optimal solution in the solution space through multiple groups of particles (representing different parameter combinations). The PSO algorithm is used to explore the hyperparameter space in parallel to find the parameter configuration that can maximize the accuracy of the model.

4. Model Validation and Self-Assessment

Cross-validation and A/B testing: After each parameter adjustment, apply cross-validation techniques (such as k-fold cross-validation) to verify the new model version to ensure the improvement of model performance; at the same time, implement A/B testing to let the new and old models process a part of the actual test data in parallel, compare the prediction results, and ensure the effectiveness of the improvement;

5. Dynamic adjustment and rollback mechanism

Performance monitoring and dynamic adjustment: Deploy a performance monitoring module to track the model's prediction accuracy, response time, and resource consumption in real time. If the performance of a new model version degrades or the resource consumption is too large, the system automatically rolls back to the previous stable version and records the abnormal log for further analysis.

Adaptive learning rate adjustment: dynamically adjust the learning rate according to the model convergence and changes in data distribution; reduce the learning rate when the model is close to the optimal solution for fine-tuning; and temporarily increase the learning rate when encountering sudden changes in data distribution to accelerate adaptation to new situations;

6. Security and stability guarantee

Model version control: Use Git or a similar version control system to record each version of the model adjustment to ensure that you can quickly trace back to the previous stable version at any time point;

Resource isolation and load balancing: The online learning process runs in an isolated computing environment, separated from the main service. The cloud platform's load balancing technology ensures that the online adjustment process does not affect real-time data processing and service stability.

The implementation of these specific details not only ensures that the support vector machine classification model can continuously optimize itself to adapt to the dynamic changes in battery performance test data, but also ensures the continuity of service and the reliability of the model;

Battery performance test using support vector machine classification model after automatic parameter adjustment and model self-optimization;

The battery performance test specifically includes:

Real-time data access and processing:

After the new battery performance test data is uploaded to the cloud server in real time via the IoT device, it first passes through the data preprocessing pipeline to perform format unification, missing value checking and filling, and outlier identification and processing preprocessing steps to ensure that the data quality meets the model input requirements;

The preprocessed data is then fed into the latest optimized support vector machine classification model; since the model has online learning capabilities, it can process the data instantly without offline retraining, ensuring the timeliness of the test results;

Performance testing and grading:

The support vector machine model evaluates the performance of each batch of batteries based on selected key features (such as charging rate variation, discharge platform stability, and internal resistance growth rate) and divides them into different performance levels. The model can identify which batteries have excellent performance and which require attention or elimination, providing a basis for subsequent quality control and product improvement.

The technical solution proposed in this application has achieved a significant technological leap in the field of battery performance testing. Compared with traditional methods, it has demonstrated multi-dimensional technological progress and advantages, which are specifically reflected in the following technical effects:

Significantly improve test accuracy and efficiency: By building a dynamic data set containing historical and real-time data, combined with advanced IoT technology and edge computing, this solution achieves instant collection and efficient transmission of battery performance data, significantly shortening the test cycle. The intelligent application of the support vector machine (SVM) classification model, coupled with automatic parameter adjustment and self-optimization mechanisms, not only ensures high accuracy of test results, but also improves test efficiency and model generalization capabilities through continuous learning to adapt to changes in battery performance as technology develops.

Enhanced decision-making and response speed: The establishment of a real-time data monitoring platform and alarm system, combined with real-time data analysis in the cloud, enables abnormal battery performance or equipment failure to be quickly identified and trigger response measures, effectively improving safety in production and use, reducing downtime caused by failures, and enhancing the company's operational efficiency.

Optimize resource management and reduce costs: Intelligent data archiving and efficient data storage strategies have greatly optimized the data management process and reduced storage and operation and maintenance costs. Data preprocessing and model self-optimization reduce manual intervention and labor costs, while improving data quality and analysis efficiency, providing valuable data resources for long-term performance tracking and research and development.

Improve system flexibility and adaptability: Through online learning and intelligent feature selection strategies, this technical solution can flexibly adapt to the continuous innovation of battery technology and ensure that the test model always maintains the most cutting-edge analysis capabilities, which is crucial for rapidly changing battery materials and designs, and helps battery manufacturers respond quickly to market and technological changes.

Promoting technological innovation and competitiveness upgrade: This technical solution not only improves the scientificity and efficiency of battery performance testing, but also accelerates the research and development of battery technology by providing a powerful data analysis platform, improving product quality and market competitiveness. It provides strong technical support for the electric vehicle industry and related research institutions, and promotes technological innovation and sustainable development of the entire new energy industry chain.

To sum up, the technological progress of this application lies in the construction of a highly integrated, automated and intelligent battery performance testing system. This system solves the limitations of traditional testing methods in efficiency, accuracy, response speed and cost control by deeply integrating big data, the Internet of Things, cloud computing and AI technologies. It has brought a technological innovation to the battery industry and promoted the industry to develop in a more efficient, intelligent and sustainable direction.

It should be noted that each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments. The device and system embodiments described above are merely schematic, in which the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.

Claims (1)

1. The battery performance test method based on data analysis is characterized by comprising the following steps:

Constructing a dynamic data set, and establishing a database containing historical and real-time battery performance test data, wherein the battery performance data comprises electric cores and physical performance indexes under various working conditions;

The data are synchronized and archived in real time, the Internet of things is adopted, the data are directly captured from the battery performance testing equipment in real time, and the data are automatically uploaded to the cloud server; meanwhile, data archiving is carried out, and historical data is stored according to time sequence or batch classification;

cleaning and quality inspection of battery performance data are carried out, abnormal values of the battery performance data are automatically identified and removed, and missing battery performance data are filled;

Based on historical battery performance test data, using a support vector machine algorithm, selecting an optimal kernel function and parameter setting through cross validation, and training a support vector machine classification model capable of accurately distinguishing battery performance levels; in the characteristic selection stage, determining key parameters with the greatest influence on the battery performance by adopting a recursive characteristic elimination or correlation coefficient analysis-based method, wherein the key parameters comprise charge rate change, discharge platform stability and internal resistance growth rate, and the key parameters are used as support vector machine classification model input;

Automatic parameter adjustment and model self-optimization, on-line learning of a support vector machine classification model is configured, so that the support vector machine classification model can periodically review newly acquired test data without interrupting service, and parameters of the support vector machine classification model are automatically adjusted; searching for a better super-parameter combination by utilizing a gradient descent or particle swarm optimization algorithm so as to adapt to subtle changes of battery performance along with time and technical evolution;

Testing the battery performance by using a support vector machine classification model after automatic parameter adjustment and model self-optimization;

Capturing data from battery performance testing equipment in real time, and automatically uploading the data to a cloud server specifically comprises:

The battery performance test equipment is subjected to Internet of things upgrading, an intelligent sensor and a microcontroller unit are integrated, and the components are responsible for accurately acquiring performance data of the battery under various working conditions; and monitoring the running state of the equipment in real time; adopting an MQTT low-power-consumption wide area network communication protocol to realize efficient and safe data transmission between the test equipment and the cloud server; deploying edge computing nodes, which are positioned between the testing equipment and the cloud end, and executing preliminary data preprocessing and screening tasks;

In the data transmission process, all communication data are encrypted by adopting a TLS/SSL protocol, so that the safety and privacy protection of the data in the transmission process are ensured; the cloud server develops a RESTful API interface, receives data pushing from an edge computing node or directly from equipment, automatically analyzes the data pushing and stores the data pushing into a distributed database; the database adopts Amazon DynamoDB or Google Cloud Spanner cloud native database;

setting up a real-time data monitoring platform at a cloud end, displaying the latest data and trend analysis of each test point, setting a threshold alarm mechanism, and immediately notifying related personnel through an email, a short message or an APP once abnormal battery performance or test equipment failure is detected, so as to quickly respond to the processing;

The data archiving method specifically comprises the following steps:

Adopting a time sequence database management system, and storing and inquiring and optimizing the data of the battery performance which changes along with time; in the process of data archiving, creating a unique identifier for each batch of batteries, and establishing an efficient index structure in a database;

cleaning and quality inspection of battery performance data are carried out, abnormal values of the battery performance data are automatically identified and removed, and missing battery performance data are filled, and the method comprises the following steps:

An automatic data preprocessing pipeline is constructed, the automatic data preprocessing pipeline is integrated before the data is stored after being uploaded to the cloud, and the data is cleaned and controlled in quality before entering a core database; automatically identifying abnormal values in the battery performance data by using a box graph analysis method, a Z-score method and an abnormality detection technology based on a clustering algorithm; filling missing values in the time sequence data by adopting a linear interpolation, spline interpolation or a time sequence prediction method based on an ARIMA model, so as to ensure data continuity;

for the missing value of the non-time series data, the K-nearest neighbor regression model or the random forest regression model can be trained through the complete data to predict the missing value;

Analyzing patterns or correlations in the data, reasonably estimating the missing values by using data characteristics of similar batches or material groups as references through a nearest neighbor matching or mean/median filling method;

Selecting the optimal kernel function and parameter setting through cross validation by using a support vector machine algorithm, and training a support vector machine classification model capable of accurately distinguishing the battery performance level comprises the following steps: determining an optimal kernel function and corresponding parameters by combining grid search with cross-validation by using a plurality of kernel functions; the performance of the model is comprehensively evaluated by adopting various evaluation indexes, including accuracy, recall rate, F1 fraction and area under an AUC-ROC curve, so that the model can be accurately classified and the prediction effect of each category can be balanced;

The method for determining key parameters with the greatest influence on the battery performance by adopting recursive feature elimination or based on correlation coefficient analysis comprises the following steps:

In the feature selection stage, firstly, starting an RFE process from a complete battery performance feature set, and focusing on accuracy, recall and F1 score evaluation indexes; calculating importance scores of each feature by using a support vector machine model of preliminary training; removing one or more features with the lowest scores from the current feature set, retraining the support vector machine model, and reevaluating model performance;

setting a threshold value to determine when to stop feature elimination, and verifying the validity of the feature subset at each step through cross verification to ensure the robustness of the feature selection process;

Before the correlation coefficient analysis, carrying out standardization processing on the battery performance data, eliminating dimension influence, and evaluating the correlation between each pair of characteristics by using a Pierson correlation coefficient or a Speermann correlation coefficient;

identifying features associated with battery performance while taking into account multiple collinearity between the features; eliminating redundant characteristics which have low correlation with a target variable and are highly correlated with other characteristics, and reserving key parameters which have the greatest influence on the battery performance and are mutually independent, wherein the key parameters comprise charge rate change, discharge platform stability and internal resistance increase rate;

Combining the characteristics screened based on the correlation coefficient analysis with the result of the RFE process, and finally determining a characteristic set input into a support vector machine classification model;

The test for battery performance specifically includes:

After the new battery performance test data is uploaded to a cloud server in real time through the Internet of things equipment, the steps of format unification, missing value inspection and filling, abnormal value identification and processing pretreatment are carried out through a data preprocessing pipeline;

the preprocessed data is then sent to the latest optimized support vector machine classification model;

the support vector machine model evaluates the performance of each battery batch based on the selected key features and classifies it into different performance levels.