Battery data collection, analysis and management method and system based on remote AI monitoring

Abstract

The present invention relates to the technical field of data processing, and provides a battery data collection, analysis and management method and system based on remote AI monitoring, which obtains the voltage, current and temperature data of the battery, processes the voltage data to obtain the voltage value, analyzes the voltage value and the current data to obtain the associated data; detects abnormal points of the temperature data through the associated data to obtain the evaluation results, and generates the battery health status based on the evaluation results; generates optimization suggestions based on the battery health status, and sends the optimization suggestions to the user interface of the terminal. Through multi-level data collection and analysis and remote artificial intelligence monitoring, the accuracy and reliability of the battery data are ensured, and a comprehensive evaluation of the battery status is achieved. Then, through systematic evaluation and optimization analysis, the user can obtain intuitive battery health status feedback and optimization suggestions from its user interface, improving the problems of untimely data processing and low accuracy of abnormality detection.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a battery data collection, analysis and management method and system based on remote AI monitoring. Background Art

[0002] With the rapid development of science and technology and the improvement of environmental awareness, batteries play a vital role as energy storage devices in various applications. From smartphones, electric vehicles to renewable energy storage systems, the performance and life of batteries directly affect the efficiency and reliability of these devices. Therefore, effective monitoring and management of batteries has become a key research area in the industry.

[0003] Among the related technical means, the battery data collection, analysis and management method usually involves real-time monitoring of the battery voltage, current and temperature through physical sensors, and then analyzing these data using computer models to evaluate the battery status. This method can monitor the battery health status to a certain extent and help users predict the remaining battery life and maintenance needs.

[0004] Regarding the above technical solution, although the accuracy of battery status assessment can be improved through real-time monitoring and data analysis, there are problems of incomplete data analysis and slow processing response when detecting abnormal battery conditions and predicting battery life, which affects the accurate judgment of battery health status and the formulation of maintenance strategies. Summary of the invention

[0005] In order to improve the problems of untimely data processing and low accuracy of anomaly detection, the present application provides a battery data collection, analysis and management method and system based on remote AI monitoring, which can perform anomaly detection and life prediction more efficiently, thereby providing maintenance suggestions in a timely manner.

[0006] The present invention provides a battery data collection, analysis and management method based on remote AI monitoring, comprising: acquiring voltage data, current data and temperature data of a battery, performing standardization processing on the voltage data to obtain a standardized voltage value, performing correlation analysis on the standardized voltage value and the current data to obtain voltage-current correlation data; performing abnormal point detection on the temperature data through the voltage-current correlation data to obtain abnormal temperature data and normal temperature data, and correcting the standardized voltage value using the abnormal temperature data to obtain a corrected voltage value; inputting the corrected voltage value and the normal temperature data into a preset battery management system to generate a battery status report, performing data mining on the battery status report to obtain key performance indicators and performance prediction data; performing trend analysis on the key performance indicators to obtain a performance trend analysis result, optimizing the performance prediction data using the performance trend analysis result to obtain optimized performance prediction data, and inputting the optimized performance prediction data into a preset remote artificial intelligence monitoring platform for analysis to obtain a battery life prediction; evaluating the battery life prediction to obtain an evaluation result, and generating a battery health state based on the evaluation result; generating optimization suggestions based on the battery health state, and sending the optimization suggestions to a user interface of a terminal.

[0007] As a preferred solution, the steps of acquiring voltage data, current data and temperature data of a battery, performing standardization processing on the voltage data to obtain a standardized voltage value, performing correlation analysis between the standardized voltage value and the current data to obtain voltage-current correlation data include: using a battery monitoring sensor to acquire the voltage data of the battery, using a current measuring device to acquire the current data of the battery, and using a temperature sensor to acquire the temperature data of the battery; performing noise filtering processing on the voltage data by a digital signal processor to obtain purified voltage data and residual noise data, performing linear regression analysis on the purified voltage data by a data analysis algorithm to obtain a linear regression voltage value, and performing error correction on the residual noise data by using the linear regression voltage value to obtain a standardized voltage value; and performing Pearson correlation analysis on the standardized voltage value and the current data by using statistical analysis software to obtain voltage-current correlation data.

[0008] As a preferred solution, the step of performing abnormal point detection on the temperature data through voltage and current correlation data to obtain abnormal temperature data and normal temperature data, and using the abnormal temperature data to correct the standardized voltage value to obtain a corrected voltage value includes: clustering the voltage and current correlation data through a machine learning model to obtain a current high load interval and a current low load interval, performing abnormal detection on the high current load interval through an abnormal detection algorithm to obtain abnormal current data; using dynamic time bending technology to compare and analyze the abnormal current data with historical data to obtain current trend data, matching the current trend data with the low current load interval through a nonlinear optimization model, and calculating a current correction coefficient based on the matching result; performing abnormal point detection on the current correction coefficient and the temperature data through a neural network model to obtain abnormal temperature data and normal temperature data; performing nonlinear regression analysis on the abnormal temperature data using a Gaussian process regression method to obtain a temperature abnormality correction coefficient, and using a dynamic adjustment algorithm to update the temperature abnormality correction coefficient in real time to obtain an updated temperature abnormality correction coefficient; correcting the standardized voltage value using the updated temperature abnormality correction coefficient to obtain a corrected voltage value.

[0009] As a preferred solution, the step of inputting the corrected voltage value and the normal temperature data into a preset battery management system, generating a battery status report, and performing data mining on the battery status report to obtain key performance indicators and performance prediction data includes: using an integrated algorithm to perform data fusion on the corrected voltage value and the temperature data to obtain comprehensive voltage-temperature data, performing feature extraction on the comprehensive voltage-temperature data through a data preprocessing algorithm to obtain a voltage-temperature feature vector, and performing regression analysis on the current data using the voltage-temperature feature vector to obtain current feature data; inputting the current feature data and the voltage-temperature feature vector into a preset battery management system. A management system, in which a pattern recognition algorithm is used to perform pattern matching on the current characteristic data and the voltage-temperature characteristic vector to obtain a battery status report; a data mining tool is used to perform an in-depth analysis on the battery status report to obtain a battery performance evaluation result and a trend prediction result, and a comprehensive analysis is performed on the battery performance evaluation result and the trend prediction result to obtain key performance indicators; a time series analysis is performed on the key performance indicators through a time series prediction algorithm to obtain performance stability data, performance fluctuation data, and performance trend data, and the performance stability data is compared and analyzed with the performance fluctuation data, and performance prediction data is generated based on the analysis results.

[0010] As a preferred solution, the step of performing trend analysis on the key performance indicators to obtain performance trend analysis results, optimizing the performance prediction data using the performance trend analysis results to obtain optimized performance prediction data, and inputting the optimized performance prediction data into a preset remote artificial intelligence monitoring platform for analysis to obtain a battery life prediction comprises: performing multivariate analysis on the key performance indicators using a statistical analysis algorithm to obtain performance correlation data, performance influencing factor data, and performance influencing weight data, and performing regression analysis on the performance correlation data using a preset multivariate regression model to obtain performance correlation results; classifying the performance correlation results using a decision tree algorithm to obtain a comprehensive performance analysis result, and optimizing the comprehensive performance analysis result using an optimization algorithm to obtain optimized performance prediction data; inputting the optimized performance prediction data into a preset remote artificial intelligence monitoring platform, and in the remote artificial intelligence monitoring platform, identifying the optimized performance prediction data using a deep learning algorithm to obtain an identification result, and generating a battery life prediction based on the identification result; wherein the identification result includes the discharge and charging behavior of the battery under different usage conditions.

[0011] As a preferred scheme, the step of evaluating the battery life prediction to obtain an evaluation result, and generating a battery health status based on the evaluation result includes: performing a reliability evaluation on the battery life prediction through a preset risk analysis model to obtain battery life reliability data, identifying influencing factors on the battery life reliability data through a sensitivity analysis method to obtain battery life influencing factor data and battery life influence degree data; performing a comprehensive health status evaluation on the battery life influencing factor data and the battery life influence degree data to obtain a battery health status.

[0012] As a preferred scheme, the step of generating optimization suggestions based on the battery health status and sending the optimization suggestions to the user interface of the terminal includes: performing trend prediction on the battery health status through a preset prediction model to obtain future health trend data and potential risk data of the battery, and using a data analysis platform to conduct in-depth analysis on the future health trend data of the battery to obtain performance improvement strategies and maintenance suggestion data; optimizing the performance improvement strategies and the maintenance suggestion data through a decision optimization algorithm to obtain optimized maintenance suggestions; and sending the optimized maintenance suggestions to the user interface of the terminal through a communication interface.

[0013] The present application also provides a battery data acquisition, analysis and management system based on remote AI monitoring, including: an acquisition unit, used to acquire voltage data, current data and temperature data of the battery, standardize the voltage data to obtain a standardized voltage value, perform correlation analysis on the standardized voltage value and the current data to obtain voltage-current correlation data; a detection unit, used to perform abnormal point detection on the temperature data through the voltage-current correlation data to obtain abnormal temperature data and normal temperature data, and use the abnormal temperature data to correct the standardized voltage value to obtain a corrected voltage value; a mining unit, used to input the corrected voltage value and the normal temperature data into a preset battery management system to generate A battery status report, performing data mining on the battery status report to obtain key performance indicators and performance prediction data; an analysis unit, used to perform trend analysis on the key performance indicators to obtain performance trend analysis results, optimizing the performance prediction data using the performance trend analysis results to obtain optimized performance prediction data, inputting the optimized performance prediction data into a preset remote artificial intelligence monitoring platform for analysis to obtain a battery life prediction; an evaluation unit, used to evaluate the battery life prediction to obtain an evaluation result, and generating a battery health status based on the evaluation result; a generation unit, used to generate optimization suggestions based on the battery health status, and sending the optimization suggestions to the user interface of the terminal.

[0014] The present application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the battery data collection, analysis and management method based on remote AI monitoring described above is implemented.

[0015] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the processor executes the battery data collection, analysis and management method based on remote AI monitoring as described above.

[0016] Compared with the prior art, this application has the following beneficial effects: timely data processing and high detection accuracy. Through multi-level data collection and analysis and remote artificial intelligence monitoring, and using data standardization processing, correlation analysis, anomaly detection and correction technology, the accuracy and reliability of battery data are ensured; combined with status report generation, data mining, trend analysis and life prediction, a comprehensive evaluation of the battery status is achieved. Through a systematic evaluation and optimization model, users can obtain intuitive battery health status feedback and optimization suggestions from their user interface, which not only improves the accuracy and intelligence of battery management, but also improves the user experience through a fast feedback mechanism, and improves the problems of untimely data processing and low anomaly detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0018] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0019] Figure 1 It is a flowchart of a battery data collection, analysis and management method based on remote AI monitoring provided by an embodiment of the present invention; Figure 2 It is a schematic block diagram of the structure of a battery data acquisition, analysis and management system based on remote AI monitoring provided by an embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0020] Description of reference numerals: 10. Battery data acquisition, analysis and management system based on remote AI monitoring; 11. Acquisition unit; 12. Detection unit; 13. Mining unit; 14. Analysis unit; 15. Evaluation unit; 16. Generation unit; 20. Electronic device; 21. Memory; 22. Processor. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in 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 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.

[0022] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0023] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0024] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0025] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0026] Embodiment 1: like Figure 1 As shown, the battery data collection, analysis and management method based on remote AI monitoring provided in an embodiment of the present application includes steps S100 to S600.

[0027] Step S100, obtaining voltage data, current data and temperature data of the battery, standardizing the voltage data to obtain a standardized voltage value, and performing correlation analysis between the standardized voltage value and the current data to obtain voltage-current correlation data.

[0028] In this step, the voltage data, current data and temperature data of the battery are first collected in real time through sensors, and the data is input into the computing system; specifically, the voltage data is standardized by a normalization method, and the voltage data of different ranges are unified into the same dimension for subsequent analysis. Then, the correlation between the standardized voltage value and the current data is calculated using a statistical method to generate voltage-current correlation data. This process is achieved by calculating the correlation coefficient, which can reveal the relationship between voltage and current.

[0029] For example, when a battery is charging, changes in current can cause significant fluctuations in voltage, and correlation analysis can identify the characteristics of this fluctuation.

[0030] Step S200: detecting abnormal points of temperature data using voltage-current correlation data to obtain abnormal temperature data and normal temperature data, and correcting the standardized voltage value using the abnormal temperature data to obtain a corrected voltage value.

[0031] In this step, the voltage-current correlation data obtained in the previous step is used as a reference standard for temperature anomaly detection, and anomaly detection rules are formulated; specifically, anomaly detection algorithms are used, such as the 3σ principle based on statistical distribution, to identify abnormal points in temperature data. After identification, these temperature anomaly data are used to correct the standardized voltage value to offset the impact of temperature anomaly and obtain a more accurate corrected voltage value.

[0032] For example, when an abnormal temperature rise is detected, the correction process reduces the error in the voltage value, making the voltage reading more accurate.

[0033] Step S300: input the corrected voltage value and normal temperature data into a preset battery management system, generate a battery status report, perform data mining on the battery status report, and obtain key performance indicators and performance prediction data.

[0034] In this step, the corrected voltage value and normal temperature data are transmitted to the battery management system, which has data analysis functions. Specifically, data mining algorithms, such as cluster analysis and regression analysis, are used to extract key indicators and prediction models of battery performance from massive data to obtain a complete battery status report. Through such a report, the current health status and future trends of the battery can be displayed in detail.

[0035] For example, the report will indicate important information such as the battery's charge and discharge cycle life and current capacity percentage.

[0036] Step S400: Perform trend analysis on key performance indicators to obtain performance trend analysis results, use the performance trend analysis results to optimize the performance prediction data to obtain optimized performance prediction data, input the optimized performance prediction data into a preset remote artificial intelligence monitoring platform for analysis, and obtain a battery life prediction.

[0037] In this step, time series analysis is applied to the key performance indicators extracted from the battery status report to conduct trend research; specifically, linear regression, moving average and other models are used to analyze the law of indicator changes over time, and this law is used to optimize the previous performance prediction data to improve the accuracy of the prediction.

[0038] For example, if trend analysis reveals that a certain battery model degrades faster under high load operation, the predictive model can be adjusted to reflect this characteristic.

[0039] Step S500: Evaluate the battery life prediction to obtain an evaluation result, and generate a battery health status based on the evaluation result.

[0040] In this step, the accuracy and reliability of the life prediction results are evaluated by comparing them with historical data or experimentally measured values. Specifically, error analysis or other evaluation indicators, such as mean absolute percentage error (MAPE), are used to quantify the prediction quality and make a conclusion on the health status of the battery.

[0041] For example, if the evaluation results show that the expected battery life is consistent with the usage history, the battery health status is evaluated as good.

[0042] Step S600: Generate optimization suggestions based on the battery health status, and send the optimization suggestions to the user interface of the terminal.

[0043] In this step, targeted optimization and maintenance suggestions are made based on the health status of the battery; specifically, the suggestions include adjustments to charging and discharging operations, temperature management strategies, and tips on usage habits. These suggestions are sent to the interface on the user's device through wireless communication technologies such as Wi-Fi or Bluetooth to help users make informed decisions.

[0044] For example, users are advised to reduce heavy-load applications in high-temperature operating environments to extend battery life.

[0045] In this embodiment, the voltage data, current data and temperature data of the battery are obtained, and the voltage data is standardized to obtain a standardized voltage value. Then, the standardized voltage value is correlated with the current data to generate voltage-current correlation data. The data is used to detect abnormal points in the temperature data to distinguish abnormal temperature data from normal temperature data. After that, the standardized voltage value is corrected by the abnormal temperature data to generate a corrected voltage value. The corrected voltage value and normal temperature data are then input into a preset battery management system to generate a battery status report. The report extracts key performance indicators and performance prediction data after data mining. Subsequently, trend analysis is performed on the key performance indicators and the performance prediction data is optimized. The optimized data is then input into the remote artificial intelligence monitoring platform for analysis, and the battery life prediction is finally obtained. Finally, the battery life prediction is evaluated, and the battery health status is generated based on the evaluation results and optimization suggestions are provided, which are communicated to the end user through the user interface.

[0046] The use of remote AI monitoring can significantly improve the efficiency and accuracy of battery data collection and analysis. Standardization and correlation analysis improve the accuracy of data processing. Abnormal point detection can timely identify abnormal battery conditions. Trend analysis and optimization processing significantly improve the scientificity and reliability of performance prediction. Ultimately, through remote monitoring and intelligent evaluation, users can obtain timely battery health status and optimization suggestions, greatly improving the automation of battery management and user experience.

[0047] Embodiment 2: In step S100, a battery monitoring sensor is used to obtain voltage data of the battery, a current measuring device is used to obtain current data of the battery, and a temperature sensor is used to obtain temperature data of the battery.

[0048] By combining different types of sensor systems, it is ensured that the true status information of the battery is collected; specifically, the voltage data is read by a high-precision voltage sensor, the current data is recorded by current measuring devices such as Hall sensors, and the temperature data is monitored by a high-sensitivity temperature sensor. The three types of data are transmitted to the central processing unit in real time through the data collection module for integration.

[0049] For example, the data acquisition system records voltage, current, and temperature data every second and stores it in a central database for subsequent processing.

[0050] The voltage data is subjected to noise filtering processing by a digital signal processor to obtain purified voltage data and residual noise data. A linear regression analysis is performed on the purified voltage data using a data analysis algorithm to obtain a linear regression voltage value. The linear regression voltage value is used to perform error correction on the residual noise data to obtain a standardized voltage value.

[0051] By selecting an appropriate noise filtering algorithm, such as a low-pass filter, the high-frequency noise in the signal caused by hardware such as sensors is removed; specifically, the filtered purified voltage data is compared with a linear model established based on historical data to evaluate and correct the remaining low-frequency errors. In addition, the standardized voltage value is precisely adjusted in the error correction step to improve the accuracy and consistency of the data.

[0052] For example, a Butterworth filter is used to attenuate the 0-50 Hz voltage signal noise, thereby significantly reducing interference in data analysis.

[0053] Statistical analysis software was used to perform Pearson correlation analysis on the standardized voltage and current data to obtain voltage-current correlation data.

[0054] Pearson correlation analysis is performed using statistical software to evaluate the linear relationship between battery voltage and current. Specifically, the correlation coefficient value is calculated to determine the closeness between the two. This type of statistical analysis is used to help determine the load status and operating efficiency of the battery.

[0055] For example, during a specific period of time, the correlation coefficient between current and voltage was found to be 0.85, indicating that there is a strong positive correlation between them.

[0056] In step S200, the voltage-current correlation data is clustered by a machine learning model to obtain a current high-load interval and a current low-load interval, and an abnormality detection algorithm is used to perform abnormality detection on the current high-load interval to obtain abnormal current data.

[0057] Through trained machine learning models, such as the K-Means clustering algorithm, the data is divided into different load intervals according to the voltage and current correlation data; specifically, the patterns and trends in the data are identified, and how these parameters are divided into high load and low load are determined. Through anomaly detection algorithms, such as the Isolation Forest algorithm, the high load intervals are analyzed to obtain abnormal current data.

[0058] For example, when the machine learning model detects that the current high-load interval has irregular fluctuation characteristics, it is marked as a potential abnormal interval.

[0059] Dynamic time bending technology is used to compare and analyze abnormal current data with historical data to obtain current trend data. The current trend data is matched with the current low load interval through a nonlinear optimization model, and the current correction coefficient is calculated based on the matching result.

[0060] Dynamic time warping (DTW) is used to optimally align historical and current abnormal current data on the time axis to identify trend changes. Subsequently, a nonlinear optimization model is used to obtain the best match and calculate accurate current correction coefficients to compensate for the deviation between data.

[0061] For example, DTW can be used to identify abnormal current data that matches historical high-load operating patterns, allowing these data to be corrected quickly.

[0062] The abnormal point detection of the current correction coefficient and temperature data is performed through the neural network model to obtain abnormal temperature data and normal temperature data.

[0063] By building a well-trained neural network model, the current correction coefficient is combined with the temperature data to detect potential anomalies. Specifically, a deep learning framework such as LSTM or CNN is used to automatically adjust the weights to improve the accuracy of anomaly recognition of complex data patterns, and the output is a set of temperature data sets classified as abnormal and normal.

[0064] For example, with better than 99 percent accuracy, the neural network successfully identified anomalies early in the course of a temperature rise, indicating a potential problem.

[0065] The Gaussian process regression method is used to perform nonlinear regression analysis on the temperature anomaly data to obtain the temperature anomaly correction coefficient. The temperature anomaly correction coefficient is updated in real time using a dynamic adjustment algorithm to obtain the updated temperature anomaly correction coefficient.

[0066] By using the Gaussian process regression method, nonlinear modeling is performed with temperature anomaly data as input; specifically, the model is compared with known temperature behavior patterns to generate temperature anomaly correction coefficients. At the same time, the correction coefficients are continuously updated using a dynamic adjustment algorithm to ensure their adaptability.

[0067] For example, the system automatically adjusts the parameters of the Gaussian process model to optimize the response speed of the correction coefficients, thereby ensuring that the model reacts quickly to environmental changes.

[0068] The standardized voltage value is corrected using the updated temperature anomaly correction coefficient to obtain a corrected voltage value.

[0069] The accuracy of the data is improved by gradually correcting it; specifically, the updated temperature anomaly correction coefficient is applied to the correction process of the standardized voltage value to compensate for the impact of temperature fluctuations and ensure that the voltage data reflects a more realistic battery operating condition.

[0070] For example, the data after precise adjustment at each stage simulates the ideal battery response characteristics, reducing the error of the voltage value to 2% of the original.

[0071] In step S300, an integrated algorithm is used to fuse the corrected voltage value and temperature data to obtain comprehensive voltage-temperature data, and a data preprocessing algorithm is used to extract features of the comprehensive voltage-temperature data to obtain a voltage-temperature feature vector. The voltage-temperature feature vector is used to perform regression analysis on the current data to obtain current feature data.

[0072] Through the fusion of different algorithms, various types of data are associated and comprehensively analyzed; specifically, integrated algorithms such as random forest or AdaBoost are used to comprehensively process the corrected voltage value and temperature data to improve data accuracy, and then data preprocessing algorithms are used for cleaning, scaling and encoding. The result is a high-dimensional feature vector for analyzing current characteristics.

[0073] For example, efficient extraction of feature vectors helps identify the energy conversion efficiency of the battery in all operating stages.

[0074] The current characteristic data and the voltage-temperature characteristic vector are input into a preset battery management system, in which a pattern recognition algorithm is used to perform pattern matching on the current characteristic data and the voltage-temperature characteristic vector to obtain a battery status report.

[0075] Pattern recognition algorithms such as SVM are used to analyze and pair feature vectors to focus on and identify specific behavior patterns, and the final output is the overall status of the battery. Through preset logic and artificial intelligence models, the analysis output is automatically generated into an easy-to-understand battery status report.

[0076] For example, the system successfully identified the degradation characteristics of batteries when exposed to direct sunlight or high current loads.

[0077] The battery status report is deeply analyzed through data mining tools to obtain battery performance evaluation results and trend prediction results. The battery performance evaluation results and trend prediction results are comprehensively analyzed to obtain key performance indicators.

[0078] Through advanced data mining tools, the performance information extracted from the report is labeled, classified and quantified; specifically, algorithms such as association rules and classification trees are used to process the performance secondary data.

[0079] For example, by collating and analyzing several performance reports, we can extract the impact of the battery's internal chemical stability on its external service life.

[0080] The time series analysis of key performance indicators is performed through the time series prediction algorithm to obtain performance stability data, performance fluctuation data and performance trend data. The performance stability data and performance fluctuation data are compared and analyzed, and performance prediction data is generated based on the analysis results.

[0081] By applying ARIMA or neural network time series models, we can predict the future change trajectory of performance indicators, form a comprehensive performance analysis, and ensure intelligent system response and adjustment strategy optimization. When comparing performance stability data, we can generate more valuable design solution tips.

[0082] For example, by verifying and comparing with previous forecast results, this sequence analysis can accurately determine the average capacity retention of the battery over the next five years.

[0083] In step S400, a multivariate analysis is performed on the key performance indicators using a statistical analysis algorithm to obtain performance correlation data, performance impact factor data, and performance impact weight data. A preset multivariate regression model is used to perform regression analysis on the performance correlation data to obtain performance correlation results.

[0084] By applying methods such as principal component analysis (PCA), important variation directions are extracted from multiple variables, the dimensions of the data are reduced, and the main performance influencing factors are extracted; specifically, by calculating the correlation matrix between various performance indicators, it is possible to identify which indicators have a significant contribution to the overall performance and assign corresponding weights.

[0085] For example, it was found that the internal resistance and temperature changes of the battery have a strong correlation with the performance, so the performance fluctuations can be predicted more accurately by further analyzing these two variables.

[0086] The performance correlation results are classified using a decision tree algorithm to obtain comprehensive performance analysis results, which are then optimized using an optimization algorithm to obtain optimized performance prediction data.

[0087] By using decision tree algorithms, such as the CART (Classification and Regression Tree) algorithm, the performance correlation results are divided into different categories; specifically, at each node, the observation value is assigned according to a specific threshold, and finally a tree structure is generated to quickly identify the most influential and most common performance patterns.

[0088] For example, by constructing a decision tree containing multiple branches, we can enhance our understanding of the performance changes of batteries under high and low temperature conditions.

[0089] The optimized performance prediction data is input into a preset remote artificial intelligence monitoring platform. In the remote artificial intelligence monitoring platform, the optimized performance prediction data is identified using a deep learning algorithm to obtain an identification result, and a battery life prediction is generated based on the identification result; wherein the identification result includes the discharge and charging behavior of the battery under different usage conditions.

[0090] By combining convolutional neural networks (CNN) and recurrent neural networks (RNN), we can better grasp the characteristics and patterns of time series data; specifically, CNN can identify the spatial patterns of performance data, while RNN processes the time-varying relationships in these patterns to form an accurate prediction of battery life.

[0091] For example, the CNN-RNN combined model is used to simulate the performance of the battery under rapid discharge conditions, thereby accurately predicting its life in this state.

[0092] In step S500, the reliability of battery life prediction is evaluated by a preset risk analysis model to obtain battery life reliability data, and the influencing factors of the battery life reliability data are identified by a sensitivity analysis method to obtain battery life influencing factor data and battery life influence degree data.

[0093] By applying uncertainty analysis methods such as Monte Carlo simulation, the impact of various influencing factors on the reliability of life prediction is quantified; specifically, by performing multiple random simulations on battery life estimation, the system's sensitivity and vulnerability to variations are determined.

[0094] For example, multiple simulations have found that current fluctuations are more sensitive than voltage, which indicates that current is an indicator that needs to be monitored closely.

[0095] A comprehensive health status assessment is performed on the battery life influencing factor data and the battery life influencing degree data to obtain the battery health status.

[0096] Through the weighted comprehensive index method, the scores of different battery indicators can be evaluated in advance and classified into grades; specifically, the weight and importance of influencing factors are combined to form an overall battery health score so that users can understand the battery condition more intuitively.

[0097] For example, when the health status score drops below 75, the system automatically triggers a notification to remind the user to maintain or replace the battery.

[0098] In step S600, the battery health status is trend predicted by a preset prediction model to obtain the battery's future health trend data and potential risk data, and the battery's future health trend data is deeply analyzed using a data analysis platform to obtain performance improvement strategies and maintenance recommendation data.

[0099] Through time series analysis and machine learning algorithms (such as support vector machines, SVM), reliable models about future trends are generated; specifically, by combining historical health data, potential risk warnings and optimization suggestions are provided to users by analyzing the time-varying trends of health indicators.

[0100] For example, trend prediction can detect that batteries deteriorate rapidly under high usage frequency and recommend reducing the load to extend life.

[0101] The performance improvement strategy and maintenance recommendation data are optimized through the decision optimization algorithm to obtain the optimized maintenance recommendation.

[0102] The proposed solutions are optimized through simulated annealing or genetic algorithms, and the optimal strategy is selected to extend battery life and improve performance; specifically, the algorithm improves the strategy through repeated iterations, explores the global optimal solution, and ensures the feasibility and economy of the strategy by weighing different factors.

[0103] For example, optimization recommendations include advising users to keep batteries warm in low-temperature environments to reduce charging losses.

[0104] The optimized maintenance suggestions are sent to the user interface of the terminal through the communication interface.

[0105] By using the Internet of Things (IoT) platform, optimization information is sent quickly and securely to end devices; specifically, with the help of cloud-based processing capabilities, users are notified in real time through mobile applications or embedded panels, ensuring personalized and efficient support.

[0106] For example, the notifications received include detailed maintenance schedules and specific operation steps provided in the form of applications, allowing users to take timely actions.

[0107] In this embodiment, multivariate and multivariate regression analysis is used to correct the complex nonlinear relationship of the data and identify key performance factors and their potential impacts. Further, data mining and machine learning methods are combined to mine the deep structure and generate reliable performance evaluation and life prediction. The system integrates the use of deep learning algorithms to significantly improve the prediction accuracy and user feedback quality for risk assessment and comprehensive analysis of health status. Personalized optimization suggestions fed back to the user interface provide more intuitive and convenient operations, helping users to achieve comprehensive control of the battery status, effectively reduce risks, and achieve optimal allocation of resources.

[0108] Embodiment 3: like Figure 2 As shown, the present application also provides a battery data acquisition, analysis and management system 10 based on remote AI monitoring, including an acquisition unit 11, a detection unit 12, a mining unit 13, an analysis unit 14, an evaluation unit 15 and a generation unit 16.

[0109] The acquisition unit 11 is mainly used to acquire the voltage data, current data and temperature data of the battery, perform standardization processing on the voltage data to obtain a standardized voltage value, and perform correlation analysis on the standardized voltage value and the current data to obtain voltage-current correlation data.

[0110] The detection unit 12 is mainly used to detect abnormal points of temperature data through voltage-current correlation data to obtain abnormal temperature data and normal temperature data, and to correct the standardized voltage value using the abnormal temperature data to obtain the corrected voltage value.

[0111] The mining unit 13 is mainly used to input the corrected voltage value and normal temperature data into a preset battery management system, generate a battery status report, perform data mining on the battery status report, and obtain key performance indicators and performance prediction data.

[0112] The analysis unit 14 is mainly used to perform trend analysis on key performance indicators to obtain performance trend analysis results, use the performance trend analysis results to optimize the performance prediction data to obtain optimized performance prediction data, input the optimized performance prediction data into a preset remote artificial intelligence monitoring platform for analysis, and obtain a battery life prediction.

[0113] The evaluation unit 15 is mainly used to evaluate the battery life prediction, obtain the evaluation result, and generate the battery health status based on the evaluation result.

[0114] The generating unit 16 is mainly used to generate optimization suggestions based on the battery health status, and send the optimization suggestions to the user interface of the terminal.

[0115] In this embodiment, through a highly integrated battery data acquisition, analysis and management system, the automation and intelligence of the entire process from data acquisition to analysis, evaluation and optimization are realized. The acquisition unit 11 uses sensor technology to capture the voltage, current and temperature data of the battery in real time, and performs standardized processing on the voltage data to eliminate measurement errors and improve data consistency. Subsequently, the detection unit 12 applies advanced algorithms such as anomaly detection technology to identify and correct the impact of temperature anomalies to ensure the accuracy of basic data. The mining unit 13 extracts key performance indicators from the generated battery status report through deep data mining, which lays a solid foundation for subsequent trend analysis and performance prediction. The analysis unit 14 uses trend analysis methods to optimize performance prediction data and realizes accurate battery life prediction through comprehensive analysis of the remote AI monitoring platform. The evaluation unit 15 provides a health status assessment based on the prediction results, providing users with an intuitive battery status display. Finally, the optimization suggestions output by the generation unit 16 ensure that users obtain timely and practical strategy improvement suggestions on the terminal interface, thereby realizing optimized management of battery use.

[0116] It should be noted that technicians in the relevant technical field can clearly understand that for the convenience and conciseness of description, the specific working process of the above-described device and each module can refer to the corresponding process of the battery data acquisition, analysis and management method based on remote AI monitoring in the aforementioned embodiment 1, and will not be repeated here.

[0117] Embodiment 4: like Figure 3 As shown, the present application also provides an electronic device 20, including a memory 21 and a processor 22, the memory 21 stores a computer program that can be run on the processor 22, and when the processor 22 executes the computer program, the battery data collection, analysis and management method based on remote AI monitoring of Example 1 is implemented.

[0118] In this embodiment, a battery data collection, analysis and management method based on remote AI monitoring is implemented through the memory 21 and the processor 22 on the electronic device 20. The computer program stored in the memory 21 not only supports the execution of complex algorithms in each step, but also ensures the stability and efficiency of data processing. Driven by the high-speed computing power of the processor 22, the device can collect and correct data in real time, improve data quality through standardization and anomaly detection, and use intelligent algorithms to complete the mining and trend analysis of key performance indicators. By executing the preset program, the system can realize intelligent battery life prediction and health status assessment with the support of the device's own hardware, ensuring that users can accurately grasp and effectively manage the battery status. In the end, this improved function of the electronic device 20 not only improves the user experience, but also significantly increases the overall operating efficiency and safety of the device.

[0119] Embodiment 5: The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the processor executes the battery data collection, analysis and management method based on remote AI monitoring as described in Example 1.

[0120] In this embodiment, a complex battery data acquisition, analysis and management method is converted into an easy-to-use computer program through a computer-readable storage medium, which enables wide application on various programmable devices. When the program runs on the processor, it realizes a fully automated process from data acquisition to analysis, evaluation and recommendation output. By relying on the portability and compatibility of storage media, various types of equipment can enable complex analysis programs and enjoy artificial intelligence-driven battery health monitoring functions. With the help of efficient algorithm processing, the system performs data management operations involving multiple steps such as standardized processing, anomaly detection, and performance analysis in real time to ensure continuous performance improvement and reliability maintenance. This not only simplifies the user's operating process, but also provides the battery management industry with a solution that can be quickly integrated and applicable to various terminal devices.

[0121] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. 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 embodiments of the present invention.

Claims

1. A battery data collection, analysis and management method based on remote AI monitoring, characterized in that: include: Acquire voltage data, current data and temperature data of the battery, perform standardization processing on the voltage data to obtain a standardized voltage value, and perform correlation analysis between the standardized voltage value and the current data to obtain voltage-current correlation data; Perform abnormal point detection on the temperature data through voltage-current correlation data to obtain abnormal temperature data and normal temperature data, and use the abnormal temperature data to correct the standardized voltage value to obtain a corrected voltage value; Inputting the corrected voltage value and the normal temperature data into a preset battery management system to generate a battery status report, and performing data mining on the battery status report to obtain key performance indicators and performance prediction data; Performing trend analysis on the key performance indicators to obtain performance trend analysis results, optimizing the performance prediction data using the performance trend analysis results to obtain optimized performance prediction data, and inputting the optimized performance prediction data into a preset remote artificial intelligence monitoring platform for analysis to obtain a battery life prediction; Evaluate the battery life prediction to obtain an evaluation result, and generate a battery health status based on the evaluation result; Generate optimization suggestions based on the battery health status, and send the optimization suggestions to a user interface of the terminal.

2. The battery data collection, analysis and management method based on remote AI monitoring according to claim 1 is characterized in that: The step of acquiring voltage data, current data and temperature data of the battery, performing standardization processing on the voltage data to obtain a standardized voltage value, and performing correlation analysis between the standardized voltage value and the current data to obtain voltage-current correlation data includes: The battery monitoring sensor is used to obtain the battery voltage data, the current measuring device is used to obtain the battery current data, and the temperature sensor is used to obtain the battery temperature data; Performing noise filtering on the voltage data by a digital signal processor to obtain purified voltage data and residual noise data, performing linear regression analysis on the purified voltage data by a data analysis algorithm to obtain a linear regression voltage value, and performing error correction on the residual noise data by using the linear regression voltage value to obtain a standardized voltage value; The standardized voltage value and the current data are subjected to Pearson correlation analysis using statistical analysis software to obtain voltage-current correlation data.

3. The battery data acquisition, analysis and management method based on remote AI monitoring according to claim 1 is characterized in that: The step of detecting abnormal points of the temperature data by using the voltage-current correlation data to obtain abnormal temperature data and normal temperature data, and correcting the standardized voltage value by using the abnormal temperature data to obtain the corrected voltage value comprises: Clustering the voltage and current correlation data through a machine learning model to obtain a current high load interval and a current low load interval, and performing anomaly detection on the current high load interval through an anomaly detection algorithm to obtain abnormal current data; Dynamic time bending technology is used to compare and analyze the abnormal current data with historical data to obtain current trend data, the current trend data is matched with the current low load interval through a nonlinear optimization model, and a current correction coefficient is calculated based on the matching result; Perform abnormal point detection on the current correction coefficient and the temperature data through a neural network model to obtain abnormal temperature data and normal temperature data; A Gaussian process regression method is used to perform nonlinear regression analysis on the temperature anomaly data to obtain a temperature anomaly correction coefficient, and a dynamic adjustment algorithm is used to update the temperature anomaly correction coefficient in real time to obtain an updated temperature anomaly correction coefficient; The standardized voltage value is corrected using the updated temperature anomaly correction coefficient to obtain a corrected voltage value.

4. The battery data acquisition, analysis and management method based on remote AI monitoring according to claim 1 is characterized in that: The step of inputting the corrected voltage value and the normal temperature data into a preset battery management system, generating a battery status report, and performing data mining on the battery status report to obtain key performance indicators and performance prediction data includes: Using an integrated algorithm to perform data fusion on the corrected voltage value and the temperature data to obtain comprehensive voltage-temperature data, performing feature extraction on the comprehensive voltage-temperature data through a data preprocessing algorithm to obtain a voltage-temperature feature vector, and using the voltage-temperature feature vector to perform regression analysis on the current data to obtain current feature data; Inputting the current characteristic data and the voltage-temperature characteristic vector into a preset battery management system, wherein a pattern recognition algorithm is used in the battery management system to perform pattern matching on the current characteristic data and the voltage-temperature characteristic vector to obtain a battery status report; Performing an in-depth analysis on the battery status report through a data mining tool to obtain a battery performance evaluation result and a trend prediction result, and performing a comprehensive analysis on the battery performance evaluation result and the trend prediction result to obtain key performance indicators; The key performance indicators are analyzed in time series by using a time series prediction algorithm to obtain performance stability data, performance fluctuation data and performance trend data. The performance stability data and performance fluctuation data are compared and analyzed, and performance prediction data is generated based on the analysis results.

5. The battery data acquisition, analysis and management method based on remote AI monitoring according to claim 1 is characterized in that: The step of performing trend analysis on the key performance indicators to obtain performance trend analysis results, optimizing the performance prediction data using the performance trend analysis results to obtain optimized performance prediction data, and inputting the optimized performance prediction data into a preset remote artificial intelligence monitoring platform for analysis to obtain a battery life prediction comprises: Performing multivariate analysis on the key performance indicators through a statistical analysis algorithm to obtain performance correlation data, performance impact factor data, and performance impact weight data, and performing regression analysis on the performance correlation data using a preset multivariate regression model to obtain performance correlation results; Using a decision tree algorithm to classify the performance correlation results to obtain comprehensive performance analysis results, and optimizing the comprehensive performance analysis results through an optimization algorithm to obtain optimized performance prediction data; The optimized performance prediction data is input into a preset remote artificial intelligence monitoring platform, and in the remote artificial intelligence monitoring platform, the optimized performance prediction data is identified by a deep learning algorithm to obtain an identification result, and a battery life prediction is generated based on the identification result; wherein the identification result includes the discharge and charging behavior of the battery under different usage conditions.

6. The battery data collection, analysis and management method based on remote AI monitoring according to claim 1 is characterized in that: The step of evaluating the battery life prediction to obtain an evaluation result, and generating a battery health status based on the evaluation result includes: Performing a reliability assessment on the battery life prediction by using a preset risk analysis model to obtain battery life reliability data, and identifying influencing factors on the battery life reliability data by using a sensitivity analysis method to obtain battery life influencing factor data and battery life influence degree data; A comprehensive health status assessment is performed on the battery life influencing factor data and the battery life influencing degree data to obtain a battery health status.

7. The battery data collection, analysis and management method based on remote AI monitoring according to claim 1 is characterized in that: The step of generating an optimization suggestion based on the battery health status and sending the optimization suggestion to a user interface of a terminal includes: The battery health status is trend predicted by a preset prediction model to obtain the battery's future health trend data and potential risk data, and the battery's future health trend data is deeply analyzed by a data analysis platform to obtain performance improvement strategies and maintenance recommendation data; Optimizing the performance improvement strategy and the maintenance suggestion data by using a decision optimization algorithm to obtain optimized maintenance suggestions; The optimized maintenance suggestion is sent to the user interface of the terminal through the communication interface.

8. A battery data collection, analysis and management system based on remote AI monitoring, characterized in that: include: an acquisition unit, used to acquire voltage data, current data and temperature data of the battery, perform standardization processing on the voltage data to obtain a standardized voltage value, and perform correlation analysis between the standardized voltage value and the current data to obtain voltage-current correlation data; a detection unit, configured to detect abnormal points of the temperature data by using the voltage-current correlation data to obtain abnormal temperature data and normal temperature data, and to correct the standardized voltage value by using the abnormal temperature data to obtain a corrected voltage value; A mining unit, used for inputting the corrected voltage value and the normal temperature data into a preset battery management system, generating a battery status report, and performing data mining on the battery status report to obtain key performance indicators and performance prediction data; an analysis unit, configured to perform trend analysis on the key performance indicators to obtain performance trend analysis results, optimize the performance prediction data using the performance trend analysis results to obtain optimized performance prediction data, and input the optimized performance prediction data into a preset remote artificial intelligence monitoring platform for analysis to obtain a battery life prediction; An evaluation unit, configured to evaluate the battery life prediction, obtain an evaluation result, and generate a battery health state based on the evaluation result; A generating unit is used to generate an optimization suggestion based on the battery health status, and send the optimization suggestion to a user interface of a terminal.

9. An electronic device, characterized in that: It includes a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, it implements the battery data collection, analysis and management method based on remote AI monitoring as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the processor executes the battery data collection, analysis and management method based on remote AI monitoring as described in any one of claims 1 to 7.

Images