Storage battery on-line monitoring method and system for self-adaptive charging and discharging compensation

Abstract

The invention discloses a storage battery on-line monitoring method and system for self-adaptive charging and discharging compensation, and relates to the technical field of battery management, and the method comprises the steps: building a hybrid sensor architecture, collecting a multi-dimensional battery working data flow in real time, carrying out defect recognition in combination with check charging and discharging data, determining a battery working defect feature set, and carrying out fault diagnosis on the battery working defect feature set; and training a battery life prediction network to perform life prediction, obtaining potential safety hazard diagnosis information in combination with the battery potential safety hazard database, analyzing a life prediction result and the potential safety hazard diagnosis information, determining charge and discharge compensation parameters, and performing adaptive charge and discharge compensation control based on the charge and discharge compensation parameters. The technical problems that an existing storage battery monitoring means is single, defects and hidden dangers are difficult to accurately recognize, a dynamically optimized charging and discharging strategy is lacked, and the service life and operation safety of the battery are affected are solved, and the purposes of achieving dynamic self-adaptive charging and discharging control through multi-dimensional monitoring and intelligent analysis and improving the safety of the storage battery are achieved. The service life of the battery is prolonged; and the operation safety is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery management, and particularly to an online monitoring method and system for a storage battery with adaptive charge and discharge compensation. Background Art

[0002] Storage batteries are widely used in fields such as power systems, communication base stations, and electric vehicles. Their performance and reliability directly affect the normal operation and safety of equipment. However, existing storage battery monitoring methods mostly rely on the monitoring of a single parameter (such as voltage or current), which is difficult to comprehensively reflect the battery state and cannot accurately identify potential defects and safety hazards. In addition, existing technologies have deficiencies in battery life prediction and charge and discharge strategy optimization, lacking the ability of dynamic adjustment and being difficult to meet the actual application requirements. Summary of the Invention

[0003] This application provides an online monitoring method and system for a storage battery with adaptive charge and discharge compensation, which is used to solve the technical problems that existing storage battery monitoring means are single, it is difficult to accurately identify defects and potential hazards, and there is a lack of a dynamically optimized charge and discharge strategy, which affects battery life and operation safety.

[0004] In the first aspect of this application, an online monitoring method for a storage battery with adaptive charge and discharge compensation is provided. The method includes: building a hybrid sensor architecture, and collecting and obtaining a multi-dimensional battery operating data stream during the operation of a target storage battery through the hybrid sensor architecture; extracting and constructing a charge and discharge feature fingerprint matrix of the multi-dimensional battery operating data stream, and combining verification charge and discharge data to identify defects in the charge and discharge feature fingerprint matrix to determine a battery operating defect feature set; training and building a battery life prediction network, and using the battery life prediction network to perform life prediction for a preset time window on the battery operating defect feature set to obtain a target battery life prediction result; establishing a battery safety hazard database, comparing and matching the battery operating defect feature set with the battery safety hazard database to obtain battery safety hazard diagnosis information; performing charge and discharge strategy analysis on the target battery life prediction result and the battery safety hazard diagnosis information to determine target charge and discharge compensation parameters, and performing adaptive compensation control on the target storage battery based on the target charge and discharge compensation parameters.

[0005] In the second aspect of the present application, an on-line monitoring system for a storage battery with adaptive charge and discharge compensation is provided. The system includes: a battery operation data stream acquisition module, which is used to build a hybrid sensor architecture and collect and acquire multi-dimensional battery operation data streams during the operation of the target storage battery through the hybrid sensor architecture; a working defect identification module, which is used to extract and construct a charge and discharge characteristic fingerprint matrix of the multi-dimensional battery operation data stream, and identify defects in the charge and discharge characteristic fingerprint matrix in combination with verification charge and discharge data to determine a battery operation defect characteristic set; a battery life prediction module, which is used to train and construct a battery life prediction network, and use the battery life prediction network to perform life prediction on the battery operation defect characteristic set for a preset time window to obtain a target battery life prediction result; a potential safety hazard diagnosis module, which is used to establish a battery potential safety hazard database, compare and match the battery operation defect characteristic set with the battery potential safety hazard database to obtain battery potential safety hazard diagnosis information; an adaptive compensation module, which is used to perform charge and discharge strategy analysis on the target battery life prediction result and the battery potential safety hazard diagnosis information, determine target charge and discharge compensation parameters, and perform adaptive compensation control on the target storage battery based on the target charge and discharge compensation parameters.

[0006] One or more technical solutions provided in the present application have at least the following technical effects or advantages: An on-line monitoring method and system for a storage battery with adaptive charge and discharge compensation provided in the present application relate to the technical field of battery management. By using a hybrid sensor architecture to collect multi-dimensional battery data in real time, constructing a charge and discharge characteristic fingerprint matrix, identifying defects in combination with verification charge and discharge data, training a battery life prediction network for life prediction, and comparing with a potential safety hazard database to obtain potential safety hazard diagnosis information, and determining charge and discharge compensation parameters and performing adaptive compensation control based on the prediction result and hazard diagnosis, it solves the technical problems that the existing storage battery monitoring means are single, it is difficult to accurately identify defects and potential hazards, and there is a lack of dynamic optimized charge and discharge strategies, which affect the battery life and operation safety, and realizes the technical effect of realizing dynamic adaptive charge and discharge control through multi-dimensional monitoring and intelligent analysis, and improving the battery life and operation safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0008] Figure 1Schematic flow diagram of an on - line monitoring method for a storage battery with adaptive charge - discharge compensation provided by an embodiment of the present application; Figure 2 Schematic structural diagram of an on - line monitoring system for a storage battery with adaptive charge - discharge compensation provided by an embodiment of the present application.

[0009] Explanation of reference numerals: Battery operating data stream acquisition module 11, operating defect identification module 12, battery life prediction module 13, potential safety hazard diagnosis module 14, adaptive compensation module 15. Detailed implementation manners

[0010] The present application provides an on - line monitoring method and system for a storage battery with adaptive charge - discharge compensation, which is used to solve the technical problems that the existing storage battery monitoring means are single, it is difficult to accurately identify defects and potential hazards, and there is a lack of dynamic optimized charge - discharge strategies, which affect the battery life and operation safety.

[0011] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0012] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above - mentioned accompanying drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non - exclusive inclusion. For example, a process, method, system, product or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices.

[0013] Embodiment 1, as Figure 1 shown, the present application provides an on - line monitoring method for a storage battery with adaptive charge - discharge compensation, and the method includes: P10: Build a hybrid sensor architecture, and collect and obtain multi - dimensional battery operating data streams during the operation of the target storage battery through the hybrid sensor architecture.

[0014] Specifically, it is first necessary to build a hybrid sensor architecture to comprehensively and real-time monitor the operation process of the target battery. The hybrid sensor architecture refers to the organic combination of multiple different types and functions of sensors to form a sensor network that works collaboratively for comprehensively monitoring the operating state of the battery. The design of this architecture needs to comprehensively consider factors such as the type, quantity, layout method of the sensors, and the synchronization of data acquisition to ensure that the multi-dimensional working data stream of the battery can be accurately and comprehensively obtained.

[0015] These sensors usually include voltage sensors, current sensors, temperature sensors, etc. Each type of sensor is responsible for monitoring different working parameters of the battery. For example, the voltage sensor monitors the voltage fluctuations of the battery in real time. Voltage changes are an important indicator of battery health and can reflect whether the battery is in a state of charge-discharge imbalance or aging, etc.; the current sensor is used to detect the charge and discharge current of the battery, which is crucial for judging whether the battery is in an overcharge or over-discharge state and whether there may be problems such as internal short circuit of the battery; the temperature sensor is used to monitor the temperature of the battery. Temperature changes are directly related to the chemical reaction rate of the battery. Excessive temperature may cause the battery capacity to decay rapidly and may even cause safety problems.

[0016] In some special cases, if the battery application environment is relatively complex, vibration sensors and environmental humidity sensors can also be introduced. The vibration sensor can monitor the vibrations generated during the operation of the device, which may affect the physical state and internal structure of the battery, while the humidity sensor helps to monitor the humidity changes in the battery working environment. These factors may all affect the service life and safety of the battery. All the data collected by the sensors is transmitted to the data processing platform through wireless or wired communication methods to ensure the real-time and stable transmission of the data.

[0017] Through the above hybrid sensor architecture, the multi-dimensional battery working data stream during the operation of the target battery is collected in real time. The multi-dimensional data stream refers to a dynamic data sequence containing multiple dimensions (such as voltage, current, temperature, etc.). These data have continuity and relevance in the time series. According to the actual application scenario and monitoring requirements of the battery, the data acquisition frequency is reasonably set. For the dynamic charge and discharge process, the acquisition frequency should be relatively high. For example, dozens to hundreds of data are collected per second to capture the rapidly changing signals; for static monitoring, the acquisition frequency can be appropriately reduced, such as collecting data once per minute. By building a hybrid sensor architecture and collecting the multi-dimensional battery working data stream in real time, the operating state of the battery can be comprehensively monitored, and accurate data support can be provided for subsequent battery health diagnosis, life prediction, and charge and discharge strategy adjustment, etc., to ensure the safety and reliability of the battery under various working conditions.

[0018] P20: Extract and construct the charge-discharge characteristic fingerprint matrix of the multi-dimensional battery operation data stream, identify defects in the charge-discharge characteristic fingerprint matrix in combination with verification charge-discharge data, and determine the battery operation defect feature set.

[0019] Further, to extract and construct the charge-discharge characteristic fingerprint matrix of the multi-dimensional battery operation data stream, step P20 of the embodiment of the present application further includes: P21: Analyze the noise characteristics of the multi-dimensional battery operation data stream to obtain the battery operation data noise characteristics, and determine the battery noise data threshold according to the battery operation data noise characteristics; P22: Identify noise data in the multi-dimensional battery operation data stream according to the battery noise data threshold to obtain the battery operation noise data set; P23: Initialize the Kalman filter to perform filtering preprocessing on the battery operation noise data set to obtain the denoised multi-dimensional battery operation data stream; P24: Align and extract features from the denoised multi-dimensional battery operation data stream based on the sensor data sampling time, and construct the charge-discharge characteristic fingerprint matrix.

[0020] It should be understood that in order to improve the quality of data and the accuracy of subsequent analysis, it is necessary to perform noise processing and feature extraction on the multi-dimensional battery operation data stream, construct the charge-discharge characteristic fingerprint matrix, and perform defect identification based on this.

[0021] First, analyze the noise characteristics of the collected multi-dimensional battery operation data stream. The purpose of noise characteristic analysis is to identify the noise components in the data and determine their statistical characteristics, such as the amplitude distribution and frequency characteristics of the noise. By analyzing the noise characteristics, a reasonable battery noise data threshold can be determined for subsequent noise data identification. Specifically, methods such as statistical analysis, frequency domain analysis, and time domain analysis can be used to identify the noise characteristics. For example, by calculating statistical quantities such as the mean, variance, and standard deviation of the data, analyze the fluctuation range and distribution characteristics of the data; analyze its spectral characteristics through Fourier transform to identify the distribution range of high-frequency noise; observe the time domain waveform of the data to identify abnormal signals such as mutations and spikes. Based on these analysis results, determine the battery noise data threshold, and the setting of this threshold should be adjusted according to the actual application scenario and noise characteristics to ensure that normal data and noise data can be effectively distinguished.

[0022] Next, according to the determined battery noise data threshold, identify the noise data in the multi-dimensional battery operating data stream. Exemplarily, compare each data point with the threshold, and identify the data points outside the threshold range as noise data, thus obtaining a set containing all the noise data, that is, the battery operating noise data set. The process of noise data identification can be implemented by algorithms such as threshold comparison or sliding window method. For example, for each data point x, if ∣x−μ∣> threshold, then x is determined as noise data, where μ is the mean of the data. In order to remove the noise data and improve the quality and usability of the data, a Kalman filter can be used to perform filtering preprocessing on the battery operating noise data set. The Kalman filter is an efficient recursive filter that can estimate and filter the noisy measurement data according to the dynamic model and observation model of the system. In the filter initialization stage, it is necessary to set the state vector, covariance matrix, system model, and observation model parameters. After initialization, use the Kalman filter to perform filtering processing on the battery operating noise data set. Through two steps of prediction and update, gradually correct the estimated value of the state variable, remove the influence of the noise data, and thus obtain the multi-dimensional battery operating data stream after denoising. This data stream removes the interference of the noise and more accurately reflects the actual operating state of the battery.

[0023] After obtaining the multi-dimensional battery operating data stream after denoising, perform data alignment and feature extraction on it to construct a charge and discharge feature fingerprint matrix. Since there may be slight differences in the sampling times of different sensors, these data may not be completely aligned in time. Therefore, first, according to the sampling times of the sensor data, perform time alignment on the multi-dimensional battery operating data stream after denoising. Interpolation methods (such as linear interpolation or spline interpolation) can be used to perform time alignment on the data to ensure that the data in different dimensions have corresponding values at the same time point. On this basis, perform feature extraction on the multi-dimensional battery operating data stream, and extract the key features from the original data that can reflect the charge and discharge characteristics of the battery. Feature extraction methods can include time-domain features, frequency-domain features, statistical features, and custom features, etc. For example, the mean, variance, and extreme values of voltage, the change rate and fluctuation coefficient of current, the cumulative distribution function of temperature, the correlation coefficient between voltage and current, etc. Combine the extracted feature values into a matrix, that is, the charge and discharge feature fingerprint matrix. This matrix can comprehensively and accurately describe the behavior characteristics of the storage battery during the charge and discharge process, similar to the "fingerprint" of the battery, with uniqueness and identifiability.

[0024] Finally, defect identification is performed on the charge-discharge characteristic fingerprint matrix in combination with the verification charge-discharge data. The verification charge-discharge data refers to the data obtained by performing charge-discharge tests on the battery under standard conditions, and these data reflect the performance characteristics of the battery under ideal conditions. By comparing the characteristic values in the charge-discharge characteristic fingerprint matrix with the standard characteristic values in the verification charge-discharge data, potential defects of the battery can be identified. By calculating characteristic differences, threshold judgment, defect classification, etc., for example, calculating the differences between the characteristic values in the charge-discharge characteristic fingerprint matrix and the standard characteristic values in the verification charge-discharge data, such as Euclidean distance, cosine similarity, etc.; according to the set threshold, judging whether the characteristic difference exceeds the normal range. Finally, the identified defect characteristics are summarized to form a battery operation defect characteristic set, which details the potential problems of the battery in the current operating state and provides an important basis for subsequent life prediction and safety hazard diagnosis.

[0025] Further, step P24 of the embodiment of the present application further includes: P24-1: Arrange the denoised multi-dimensional battery operation data stream in sequence according to the sensor data sampling time to obtain a multi-dimensional battery operation time-series data stream; P24-2: Perform linear interpolation and time alignment processing on the multi-dimensional battery operation time-series data stream according to the data time-series information to obtain a multi-dimensional battery operation aligned data stream; P24-3: Extract charge characteristics and discharge characteristics from the multi-dimensional battery operation aligned data stream in sequence to obtain a charge characteristic data set and a discharge characteristic data set; P24-4: Arrange the charge characteristic data set and the discharge characteristic data set according to the number of charge-discharge cycles for feature vector arrangement and matrix combination construction to generate the charge-discharge characteristic fingerprint matrix.

[0026] Optionally, the construction process of the charge-discharge characteristic fingerprint matrix can be further refined.

[0027] First, arrange the denoised multi-dimensional battery operation data stream in sequence according to the sampling time of the sensor data. The main purpose of this process is to sort the data in terms of time, so that the data conforms to the actual acquisition order in the time dimension. Since the time points of sensor data acquisition may be asynchronous, sorting by time can make the data more chronological in subsequent processing, thus ensuring the chronology and consistency of the data.

[0028] Next, linear interpolation and time alignment processing are performed on the battery working time-series data stream through data time-series information. Since the data sampling times of different sensors may not be exactly the same, direct analysis may lead to data mismatches. Therefore, it is necessary to use the linear interpolation method to fill the data sampling gaps, achieve smooth data transition, and generate a multi-dimensional battery working aligned data stream. Through time alignment processing, the data of different sensors are aligned in time, ensuring data consistency in subsequent analysis and eliminating errors caused by sampling time differences.

[0029] Furthermore, the charging characteristics and discharging characteristics are extracted from the multi-dimensional battery working aligned data stream in sequence. Charging characteristic extraction refers to extracting key characteristic data related to the battery charging state from the battery charging process, such as charging voltage, charging current, charging rate, etc.; discharging characteristic extraction includes extracting characteristic data related to the discharging state from the battery discharging process, such as discharging voltage, discharging current, discharging time, etc. These charge and discharge characteristic data sets contain important information about the battery during charge and discharge processes and are the key to constructing the charge and discharge characteristic fingerprint matrix.

[0030] Finally, the charging characteristic data set and the discharging characteristic data set are arranged in feature vectors and combined into a matrix according to the number of charge and discharge cycles to generate a charge and discharge characteristic fingerprint matrix. The number of charge and discharge cycles is an important parameter during the battery usage process, which reflects the usage degree and aging condition of the battery. By arranging and combining the characteristic data according to the number of charge and discharge cycles, the charging and discharging characteristic data under different cycle numbers can be integrated into a matrix to form a charge and discharge characteristic fingerprint matrix. This matrix can display the characteristic changes of the battery under different charge and discharge cycles in a structured and systematic way, providing important data support for subsequent battery state evaluation, life prediction, and defect identification, etc., and helping to more accurately evaluate the health state of the battery.

[0031] Furthermore, to determine the battery working defect characteristic set, step P20 of the embodiment of the present application further includes: P25: Perform working state annotation on the verification charge and discharge data to obtain battery charge and discharge state data, where the battery charge and discharge state data includes normal working, minor defect, and serious defect state data; P26: Use a support vector machine to perform defect identification training based on the battery charge and discharge state data to generate a battery working defect identification network; P27: Based on the verification charge and discharge data as a reference, perform associated alignment fusion on the charge and discharge characteristic fingerprint matrix to obtain a charge and discharge fusion characteristic data set; P28: Based on the battery working defect identification network, perform defect identification on the charge and discharge fusion characteristic data set to determine the battery working defect characteristic set.

[0032] In a possible embodiment of the present application, to further determine the battery working defect feature set, the embodiment of the present application further introduces the processing and analysis of verification charge and discharge data.

[0033] First, label the working state of the verification charge and discharge data to obtain the charge and discharge state data of the battery. The core of this process lies in correlating the collected charge and discharge data with the actual working state of the battery. Specifically, it includes classifying and labeling the working state of the battery, usually divided into three categories: normal operation, minor defects, and serious defects. Through such labeling, the state of the battery at different working stages can be clarified, providing clear identification and training data for subsequent defect identification. Under normal working conditions, the charge and discharge characteristics of the battery are close to the ideal state; in the minor defect state, the battery may show some abnormalities, but has little impact on the overall performance; while the serious defect state indicates that the battery has significant problems, which may affect its use safety or lifespan.

[0034] Next, use the support vector machine (SVM) for defect identification training to generate a battery working defect identification network based on the battery charge and discharge state data. The support vector machine is a machine learning algorithm commonly used for classification problems, which can find the optimal separation hyperplane in a high-dimensional space to distinguish data in different states. In this step, the SVM trains on the charge and discharge state data to identify different battery working states, thereby generating a battery working defect identification network. This network can analyze the charge and discharge data of the battery to determine whether the battery is in a defective state and provide accurate defect identification capabilities for subsequent steps.

[0035] Furthermore, taking the verification charge and discharge data as a reference, perform the correlation alignment and fusion of the charge and discharge feature fingerprint matrix. Since the charge and discharge feature fingerprint matrix is extracted from the actual operation data, while the verification charge and discharge data is obtained under standard conditions, there may be certain differences in data format and time series between the two. Therefore, it is necessary to correlate and align the charge and discharge feature fingerprint matrix with the verification charge and discharge data to ensure their consistency in time series and feature dimensions. Through the correlation alignment, ensure that the data in the fingerprint matrix can correspond to the actual charge and discharge process, and through data fusion technology, generate a charge and discharge fusion feature data set. This fusion data set integrates more battery working features, can more comprehensively reflect the health state of the battery, and provides richer data support for defect identification.

[0036] Finally, based on the generated battery working defect recognition network, defect recognition is performed on the charge-discharge fusion feature dataset, and finally the battery working defect feature set is determined. At this stage, the defect recognition network will analyze the fused feature dataset, identify potential defects in the battery, classify them according to their severity, and generate the working defect feature set of the battery. These feature sets contain the defect patterns of the battery under different working conditions and can be used for subsequent battery health management, life prediction, and safety hazard diagnosis.

[0037] Through the above steps, the finally obtained battery working defect feature set provides accurate basic data for the health assessment and fault prediction of the battery, can effectively identify potential problems of the battery, and provides strong support for the intelligent management and safety control of the battery.

[0038] P30: Train and construct a battery life prediction network, and use the battery life prediction network to predict the life of the battery working defect feature set within a preset time window to obtain the target battery life prediction result.

[0039] Furthermore, step P30 of the embodiment of the present application further includes: P31: Collect and obtain a battery working life database, where the battery working life database includes historical battery working data and corresponding battery life data; P32: Divide, sort, and label the battery working life database according to a preset time window to obtain a battery working life time series dataset; P33: Perform multi-logical layer training and merge connection on the battery working life time series dataset to construct a battery life prediction network.

[0040] Optionally, train and construct a battery life prediction network, and based on the battery working defect feature set, use the battery life prediction network to predict the future life of the battery.

[0041] First, collect and obtain a battery working life database, which includes historical battery working data and corresponding battery life data. The historical working data refers to multiple monitoring data such as voltage, current, and temperature of the battery under different working conditions, and these data provide basic information for analyzing the battery life. The corresponding battery life data records the actual service life of the battery under specific conditions, including information such as the continuous use time, number of uses, and occurrence of faults of the battery. By associating these data with the actual working state of the battery, a complete battery working life database is formed, providing rich training data for the subsequent life prediction network.

[0042] Next, the battery operating life database is partitioned, sorted, and labeled according to a preset time window to obtain a battery operating life time series dataset. The preset time window refers to the time range selected for life prediction, which can be several hours, days, months, or years, depending on the battery usage cycle. By partitioning, sorting, and labeling the database, historical data can be organized in a time series to form an ordered dataset. The data within each time window contains information on the battery's operating status during that period, such as average voltage, cumulative charge and discharge, maximum temperature, etc., as well as the remaining battery life time corresponding to it. This way of constructing the time series dataset helps the model capture the law of battery life changing over time and provides structured data support for subsequent training.

[0043] Then, multi-logical layer training and merging connections are performed on the battery operating life time series dataset to construct a battery life prediction network. Multi-logical layer training refers to using a multi-layer neural network structure in deep learning to extract and learn features from data through multiple logical layers. Each logical layer can extract features at different levels, from the original data features at the bottom layer to the abstract features at the high layer, gradually constructing a feature space that can effectively represent the characteristics of battery life. During the training process, the weights and bias parameters of the network are adjusted through optimization algorithms (such as gradient descent method) so that the network can learn the complex non-linear relationship between the input data (battery operating data) and the output data (battery life). Merging connections means fusing the features extracted by different logical layers. Through operations such as fully connected layers or convolutional layers, the multi-layer features are merged into a comprehensive feature vector as the final prediction basis. The battery life prediction network constructed in this way can make full use of the information in historical data to accurately predict the battery life.

[0044] Finally, the trained battery life prediction network is used to predict the life of the battery operating defect feature set within a preset time window to obtain the target battery life prediction result. The battery operating defect feature set is input into the prediction network, and the network will output the predicted remaining life value of the battery within the preset time window according to the rules it has learned. This prediction result can provide a scientific basis for battery maintenance, replacement, and management, helping users plan the use and maintenance strategies of the battery in advance and reducing the risks and costs brought by battery failure.

[0045] Furthermore, step P33 of the embodiment of the present application further includes: P33-1: Identify the operating status data and service life data of the battery operating life time series dataset to obtain the battery operating status characteristic dataset and the corresponding battery service life dataset; P33-2: Use the LSTM network structure to evaluate and train the battery operating status characteristic dataset to generate a battery operating status characteristic evaluation layer; P33-3: Use a deep neural network structure to predict and train the battery service life dataset to obtain a battery life prediction logic layer; P33-4: Combine and connect the battery operating status characteristic evaluation layer and the battery life prediction logic layer to construct the battery life prediction network.

[0046] Specifically, to further improve the accuracy and reliability of the battery life prediction network, the construction process of the battery life prediction network can be further refined to ensure that the finally constructed battery life prediction network can effectively predict the battery life.

[0047] First, identify each time window data in the battery operating life time series dataset to obtain the battery operating status characteristic dataset and the corresponding battery service life dataset respectively. The identification of operating status data refers to classifying the operating status of the battery in each time window. For example, the operating status of the battery can be divided into different levels such as normal operation, mild aging, and severe aging. This classification can be achieved by analyzing the change trends of parameters such as the voltage, current, and temperature of the battery. The identification of service life data refers to quantifying the remaining service life of the battery corresponding to each time window. For example, it is expressed by the remaining cycle times or remaining usage time of the battery. Through this identification method, complex battery operating data can be transformed into a characteristic dataset with clear physical meanings, providing clear data input for subsequent model training.

[0048] Next, use the long short-term memory (LSTM) network structure to evaluate and train the battery operating status characteristic dataset to generate a battery operating status characteristic evaluation layer. The LSTM network is a deep learning algorithm commonly used to process time series data and can effectively capture long-term dependencies in sequential data. Since the operating status characteristics of the battery have obvious time series characteristics, for example, the performance change of the battery is a process that gradually develops over time, the LSTM network structure is very suitable for evaluating the operating status of the battery. During the training process, input the battery operating status characteristic dataset into the LSTM network. By learning the time series characteristics in the data, the network can accurately evaluate the operating status of the battery and generate a battery operating status characteristic evaluation layer. This evaluation layer can output the operating status level of the battery in each time window, providing important status information for subsequent life prediction.

[0049] Then, use a deep neural network structure to perform predictive training on the battery service life dataset to obtain a battery life prediction logic layer. The deep neural network has powerful feature learning and non-linear fitting capabilities and can model complex input-output relationships. In this step, the battery service life dataset is input into the deep neural network. Through multi-layer feature extraction and learning, the network can learn the complex relationship between the battery service life and the operating state. Finally, the obtained battery life prediction logic layer can predict the remaining service life of the battery based on the input battery operating state features. This logic layer is the core part of the battery life prediction network, and its accuracy and reliability directly determine the performance of the entire prediction system.

[0050] Finally, merge and connect the battery operating state characteristic evaluation layer and the battery life prediction logic layer to construct a battery life prediction network. Merge and connect means fusing the outputs of the two network layers to form a complete prediction model. Exemplarily, the output of the battery operating state characteristic evaluation layer can be used as one of the input features of the battery life prediction logic layer, or the outputs of the two network layers can be weighted and summed through a fully connected layer for merging. The battery life prediction network constructed in this way can make full use of the correlation between the battery operating state characteristics and the service life data to achieve accurate prediction of the battery life. This network can not only consider the current operating state of the battery but also combine its historical service life data to provide more accurate and reliable prediction results.

[0051] P40: Establish a battery safety hazard database, compare and match the battery working defect feature set with the battery safety hazard database to obtain battery safety hazard diagnosis information.

[0052] It should be understood that by establishing a battery safety hazard database and comparing and matching the battery working defect feature set with this database, the diagnosis of battery safety hazards is achieved.

[0053] First, establishing a battery safety hazard database is the starting point of this step. The battery safety hazard database should include various safety problems and defect modes that the battery may face, such as overcharging, over-discharging, abnormal temperature, internal short circuit of the battery, leakage of liquid, swelling, etc. The sources of these safety hazards can be factors such as the material characteristics of the battery, the use environment, and improper operations during the charge and discharge process. This database should record all the fault and hazard characteristics that different battery types may appear under different working conditions and provide standard data for subsequent comparison. The construction of the database needs to be based on a large amount of historical data, experimental results, and industry standards and be continuously updated to ensure its timeliness and accuracy.

[0054] Next, compare and match the battery working defect feature set with the battery safety hazard database. The battery working defect feature set includes the charge-discharge feature fingerprint matrix extracted through the previous steps and the defect recognition results, recording the abnormal patterns that may occur during the actual operation of the battery. These abnormal patterns may indicate potential safety hazards in the battery. By comparing these feature sets with the known defect patterns in the safety hazard database, the system can identify the potential hazards that may occur in the current working state of the battery. This comparison process can use technologies such as machine learning and pattern recognition to automatically match the working features of the battery with the standard hazard features in the database, quickly and accurately identifying potential problems.

[0055] Finally, through the comparison and matching, generate battery safety hazard diagnosis information. These diagnosis information can provide the current safety risks of the battery, such as whether there is overheating, overcharging and over-discharging of the battery, whether there may be short circuits or other safety failures. In addition, the diagnosis information can also be classified according to the severity of the hazards, providing a basis for subsequent early warning and maintenance. If there are major safety hazards in the battery, the system can automatically issue a warning, prompting the operator to check and handle it in time to avoid accidents.

[0056] Through the above steps, it is possible to effectively identify potential safety hazards in the battery during use, and provide decision-making support for the safety management of the battery through the battery safety hazard diagnosis information, reducing the occurrence of battery failures and safety accidents, and improving the safety and reliability of the battery.

[0057] P50: Analyze the charge-discharge strategy for the target battery life prediction result and the battery safety hazard diagnosis information, determine the target charge-discharge compensation parameter, and perform adaptive compensation control on the target storage battery based on the target charge-discharge compensation parameter.

[0058] Furthermore, step P50 of the embodiment of the present application further includes: P51: Build a charge-discharge compensation strategy library for the storage battery, where the charge-discharge compensation strategy library for the storage battery includes historical battery life data, battery safety hazard diagnosis data, and corresponding charge-discharge compensation strategy parameters; P52: Obtain the battery charge-discharge optimization target, and establish a charge-discharge compensation strategy fitness function according to the battery charge-discharge optimization target; P53: Obtain a charge-discharge compensation particle swarm space according to the charge-discharge compensation strategy library for the storage battery; P54: Use the charge-discharge compensation strategy fitness function to perform global search and optimization in the charge-discharge compensation particle swarm space to determine the target charge-discharge compensation parameter with the maximum particle fitness.

[0059] Optionally, to achieve adaptive charge-discharge compensation control for the target battery, charge-discharge strategy analysis is performed on the battery life prediction results and battery safety hazard diagnosis information of the target battery, and then the target charge-discharge compensation parameters are determined, thereby optimizing the service life and safety of the battery.

[0060] First, build a battery charge-discharge compensation strategy library. This strategy library should include three key elements: historical battery life data, battery safety hazard diagnosis data, and corresponding charge-discharge compensation strategy parameters. Among them, historical battery life data provides the life performance of the battery under different working conditions, helping to understand the degradation process of the battery in a specific state; battery safety hazard diagnosis data records the potential safety risks that may occur during the actual operation of the battery, such as overcharging, over-discharging, overheating, etc.; charge-discharge compensation strategy parameters include compensation schemes and adjustment strategies formulated for these life and safety issues. By integrating these data, the battery charge-discharge compensation strategy library provides rich reference information for subsequent compensation strategy optimization.

[0061] Next, obtain the battery charge-discharge optimization goals and establish a charge-discharge compensation strategy fitness function based on these goals. The charge-discharge optimization goals usually include extending battery life, improving battery efficiency, reducing energy loss, ensuring safety, etc. On this basis, the fitness function is used to evaluate the effects of different charge-discharge compensation strategies. Specifically, the fitness function can quantify the impacts of different strategies on battery life and safety, and provide a quantitative basis for strategy selection. By establishing the fitness function, the charge-discharge process can be optimized more precisely to ensure that each charge-discharge meets the battery health and safety requirements as much as possible.

[0062] Furthermore, obtain the charge-discharge compensation particle swarm space according to the battery charge-discharge compensation strategy library. Particle Swarm Optimization (PSO) is an optimization algorithm based on swarm intelligence. In this step, the charge-discharge compensation strategy parameters are regarded as particles in the particle swarm, and each particle represents a set of possible charge-discharge compensation parameters. The particle swarm space is the set of all possible combinations of charge-discharge compensation parameters. By initializing the particle swarm and assigning an initial position (i.e., a set of initial charge-discharge compensation parameters) and velocity to each particle, the charge-discharge compensation particle swarm space is constructed.

[0063] Finally, the fitness function of the charge-discharge compensation strategy is used to perform global search and optimization within the charge-discharge compensation particle swarm space, so as to determine the target charge-discharge compensation parameters with the maximum particle fitness. Exemplarily, through the particle swarm optimization algorithm, each particle adjusts its position and velocity according to its own fitness value and the information of other particles in the group, and gradually approaches the optimal solution. In each iteration, the fitness value of each particle is calculated, and the individual optimal position and the global optimal position are updated. Through multiple iterations, the target charge-discharge compensation parameters with the maximum particle fitness are finally determined. These parameters can maximize the satisfaction of the battery charge-discharge optimization objectives, such as extending the battery life and reducing safety risks.

[0064] Finally, based on the determined target charge-discharge compensation parameters, adaptive compensation control is performed on the target storage battery. This control process is dynamic, and according to the real-time working state of the battery, life prediction, and safety hazard diagnosis information, the charge-discharge strategy is continuously adjusted to ensure that the battery always operates in the optimal state.

[0065] In summary, the embodiments of the present application have at least the following technical effects: In the present application, by building a hybrid sensor architecture, multi-dimensional battery working data streams are collected in real time, a charge-discharge characteristic fingerprint matrix is constructed, and defect identification is performed in combination with verification charge-discharge data to determine the battery working defect feature set. Based on this feature set, a battery life prediction network is trained for life prediction, and compared with the battery safety hazard database to obtain safety hazard diagnosis information. Finally, by analyzing the life prediction results and the safety hazard diagnosis information, the charge-discharge compensation parameters are determined, and adaptive charge-discharge compensation control is performed based on this to optimize the battery performance and safety.

[0066] The technical effect of realizing dynamic adaptive charge-discharge control, improving the battery life and operation safety through multi-dimensional monitoring and intelligent analysis is achieved.

[0067] Embodiment 2, based on the same inventive concept as the online monitoring method for a storage battery with adaptive charge-discharge compensation in the foregoing embodiment, as Figure 2 shown, the present application provides an online monitoring system for a storage battery with adaptive charge-discharge compensation. The system and method embodiments in the present application are based on the same inventive concept. Among them, the system includes: A battery working data stream acquisition module 11, which is used to build a hybrid sensor architecture and collect and obtain multi-dimensional battery working data streams during the operation of the target storage battery through the hybrid sensor architecture.

[0068] The working defect identification module 12 is used to extract a charge-discharge feature fingerprint matrix for constructing the multi-dimensional battery working data stream, identify defects in the charge-discharge feature fingerprint matrix in combination with verification charge-discharge data, and determine a battery working defect feature set.

[0069] The battery life prediction module 13 is used to train and construct a battery life prediction network, and use the battery life prediction network to perform life prediction for a preset time window on the battery working defect feature set to obtain a target battery life prediction result.

[0070] The potential safety hazard diagnosis module 14 is used to establish a battery potential safety hazard database, compare and match the battery working defect feature set with the battery potential safety hazard database to obtain battery potential safety hazard diagnosis information.

[0071] The adaptive compensation module 15 is used to perform charge-discharge strategy analysis on the target battery life prediction result and the battery potential safety hazard diagnosis information, determine target charge-discharge compensation parameters, and perform adaptive compensation control on the target storage battery based on the target charge-discharge compensation parameters.

[0072] Furthermore, the working defect identification module 12 is further used to perform the following steps: Perform noise characteristic analysis on the multi-dimensional battery working data stream to obtain battery working data noise characteristics, and determine a battery noise data threshold according to the battery working data noise characteristics; identify noise data in the multi-dimensional battery working data stream according to the battery noise data threshold to obtain a battery working noise data set; initialize a Kalman filter to perform filtering preprocessing on the battery working noise data set to obtain a denoised multi-dimensional battery working data stream; perform data alignment and feature extraction on the denoised multi-dimensional battery working data stream based on the sensor data sampling time to construct a charge-discharge feature fingerprint matrix.

[0073] Furthermore, the working defect identification module 12 is further used to perform the following steps: Arrange the denoised multi-dimensional battery working data stream in sequence according to the sensor data sampling time to obtain a multi-dimensional battery working time-series data stream; perform linear interpolation and time alignment processing on the multi-dimensional battery working time-series data stream according to the data time-series information to obtain a multi-dimensional battery working aligned data stream; sequentially extract charge characteristics and discharge characteristics from the multi-dimensional battery working aligned data stream to obtain a charge characteristic data set and a discharge characteristic data set; arrange and combine the charge characteristic data set and the discharge characteristic data set according to the number of charge-discharge cycles to generate the charge-discharge feature fingerprint matrix.

[0074] Further, the working defect identification module 12 is further configured to perform the following steps: Label the working status of the verification charge and discharge data to obtain battery charge and discharge status data, where the battery charge and discharge status data includes normal working, minor defect, and serious defect status data; use a support vector machine to perform defect identification training based on the battery charge and discharge status data to generate a battery working defect identification network; based on the verification charge and discharge data, perform correlation alignment fusion on the charge and discharge feature fingerprint matrix to obtain a charge and discharge fusion feature data set; based on the battery working defect identification network, perform defect identification on the charge and discharge fusion feature data set to determine the battery working defect feature set.

[0075] Further, the battery life prediction module 13 is further configured to perform the following steps: Collect and obtain a battery life database, where the battery life database includes historical battery working data and corresponding battery life data; divide, sort, and label the battery life database according to a preset time window to obtain a battery life time series data set; perform multi-logical layer training and merge connection on the battery life time series data set to construct a battery life prediction network.

[0076] Further, the battery life prediction module 13 is further configured to perform the following steps: Label the operation status data and service life data of the battery life time series data set to obtain a battery operation status characteristic data set and a corresponding battery service life data set; use an LSTM network structure to perform evaluation training on the battery operation status characteristic data set to generate a battery operation status characteristic evaluation layer; use a deep neural network structure to perform prediction training on the battery service life data set to obtain a battery life prediction logic layer; merge and connect the battery operation status characteristic evaluation layer and the battery life prediction logic layer to construct the battery life prediction network.

[0077] Further, the adaptive compensation module 15 is further configured to perform the following steps: Build a battery charge and discharge compensation strategy library, where the battery charge and discharge compensation strategy library includes historical battery life data, battery safety hazard diagnosis data, and corresponding charge and discharge compensation strategy parameters; obtain the battery charge and discharge optimization target, and establish a charge and discharge compensation strategy fitness function according to the battery charge and discharge optimization target; according to the battery charge and discharge compensation strategy library, obtain a charge and discharge compensation particle swarm space; use the charge and discharge compensation strategy fitness function to perform global search and optimization in the charge and discharge compensation particle swarm space to determine the target charge and discharge compensation parameter with the maximum particle fitness.

[0078] It should be noted that the above order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the specific embodiments of this specification have been described. Moreover, the processes depicted in the drawings do not necessarily require the specific order or consecutive order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0079] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the protection scope of the present application.

[0080] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A method for online monitoring of batteries with adaptive charge and discharge compensation, characterized in that: The method comprises: Building a hybrid sensor architecture, through which a multi-dimensional battery operation data stream is acquired during the operation of a target battery; Extracting and constructing the charge and discharge feature fingerprint matrix of the multi-dimensional battery operation data stream, and performing defect identification on the charge and discharge feature fingerprint matrix in combination with the verification charge and discharge data to determine the battery operation defect feature set; Training and constructing a battery life prediction network, using the battery life prediction network to perform life prediction for the battery working defect feature set in a preset time window to obtain a target battery life prediction result; Establishing a battery safety hazard database, comparing and matching the battery operating defect feature set with the battery safety hazard database to obtain battery safety hazard diagnosis information; A charge and discharge strategy analysis is performed on the target battery life prediction result and the battery safety hazard diagnosis information to determine a target charge and discharge compensation parameter, and adaptive compensation control is performed on the target battery based on the target charge and discharge compensation parameter.

2. A battery online monitoring method with adaptive charge and discharge compensation as claimed in claim 1, characterized in that: The extracting and constructing the charge and discharge feature fingerprint matrix of the multi-dimensional battery operation data stream includes: Performing noise characteristic analysis on the multi-dimensional battery operation data stream to obtain the battery operation data noise characteristic, and determining the battery noise data threshold according to the battery operation data noise characteristic; Performing noise data identification on the multi-dimensional battery operation data stream according to the battery noise data threshold to obtain a battery operation noise data set; Initialize the Kalman filter to perform filtering preprocessing on the battery operation noise data set to obtain a denoised multi-dimensional battery operation data stream; Data alignment and feature extraction are performed on the denoised multi-dimensional battery operation data stream based on the sensor data sampling time to construct a charge and discharge feature fingerprint matrix.

3. A battery online monitoring method with adaptive charge and discharge compensation as claimed in claim 2, characterized in that: The construction of the charge-discharge characteristic fingerprint matrix includes: Arranging the denoised multidimensional battery operation data stream in sequence according to the sensor data sampling time to obtain a multidimensional battery operation timing data stream; Performing linear interpolation and time alignment processing on the multi-dimensional battery operation timing data stream according to the data timing information to obtain a multi-dimensional battery operation alignment data stream; Sequentially extracting charging features and discharging features from the multi-dimensional battery operation alignment data stream to obtain a charging feature data set and a discharging feature data set; The charging feature data set and the discharging feature data set are constructed by performing feature vector arrangement and matrix combination according to the number of charge and discharge cycles to generate the charging and discharging feature fingerprint matrix.

4. The method for online monitoring of a battery with adaptive charge and discharge compensation as claimed in claim 1, characterized in that: The step of determining a battery operation defect feature set comprises: Marking the verification charge and discharge data for working status to obtain battery charge and discharge status data, wherein the battery charge and discharge status data includes normal working, slight defect and severe defect status data; Using a support vector machine to perform defect recognition training based on the battery charge and discharge state data to generate a battery working defect recognition network; Based on the verification charge and discharge data, the charge and discharge feature fingerprint matrix is ​​associated, aligned and fused to obtain a charge and discharge fusion feature data set; Defect recognition is performed on the charge-discharge fusion feature data set based on the battery operation defect recognition network to determine the battery operation defect feature set.

5. The method for online monitoring of a battery with adaptive charge and discharge compensation as claimed in claim 1, characterized in that: The training constructs a battery life prediction network, including: Collecting and acquiring a battery working life database, wherein the battery working life database includes historical battery working data and corresponding battery life data; Divide and sort the battery service life database according to a preset time window to obtain a battery service life time series data set; The battery service life time series data set is trained at multiple logical layers and merged and connected to construct a battery service life prediction network.

6. A method for online monitoring of a battery with adaptive charge and discharge compensation as claimed in claim 5, characterized in that: The method of constructing a battery life prediction network includes: Performing operation status data identification and service life data identification on the battery operation life time series data set to obtain a battery operation status characteristic data set and a corresponding battery service life data set; The battery operation status characteristic data set is evaluated and trained using an LSTM network structure to generate a battery operation status characteristic evaluation layer; Use a deep neural network structure to perform prediction training on the battery life data set to obtain a battery life prediction logic layer; The battery operation status characteristic evaluation layer and the battery life prediction logic layer are merged and connected to construct the battery life prediction network.

7. The method for online monitoring of a battery with adaptive charge and discharge compensation as claimed in claim 1, characterized in that: The determining of the target charge-discharge compensation parameter comprises: Building a battery charge and discharge compensation strategy library, which includes historical battery life data, battery safety hazard diagnosis data, and corresponding charge and discharge compensation strategy parameters; Obtaining a battery charge and discharge optimization target, and establishing a charge and discharge compensation strategy fitness function according to the battery charge and discharge optimization target; According to the battery charge and discharge compensation strategy library, obtaining a charge and discharge compensation particle swarm space; The charge-discharge compensation strategy fitness function is used to perform a global search and optimization in the charge-discharge compensation particle swarm space to determine the target charge-discharge compensation parameter with the maximum particle fitness.

8. An online monitoring system for batteries with adaptive charge and discharge compensation, characterized in that: The system comprises: A battery operation data stream acquisition module, which is used to build a hybrid sensor architecture and collect and acquire multi-dimensional battery operation data streams during the operation of the target battery through the hybrid sensor architecture; A working defect recognition module, the working defect recognition module is used to extract the charge and discharge feature fingerprint matrix for constructing the multi-dimensional battery working data stream, perform defect recognition on the charge and discharge feature fingerprint matrix in combination with the verification charge and discharge data, and determine the battery working defect feature set; A battery life prediction module, the battery life prediction module is used to train and construct a battery life prediction network, and use the battery life prediction network to perform life prediction for the battery working defect feature set in a preset time window to obtain a target battery life prediction result; A safety hazard diagnosis module, which is used to establish a battery safety hazard database, compare and match the battery operating defect feature set with the battery safety hazard database, and obtain battery safety hazard diagnosis information; An adaptive compensation module is used to perform a charge and discharge strategy analysis on the target battery life prediction result and the battery safety hazard diagnosis information, determine a target charge and discharge compensation parameter, and perform adaptive compensation control on the target battery based on the target charge and discharge compensation parameter.

Images