Method, system and terminal for judging charging electric quantity abnormity in lithium battery charging process

Abstract

The invention provides a method, a system and a terminal for judging charging electric quantity abnormity in a lithium battery charging process, and the method comprises the steps: firstly collecting charging data of a plurality of lithium batteries in a specific time period, then cleaning the data to remove abnormal values, then extracting effective charging cycle data from the cleaned data, and carrying out the battery feature analysis; then grouping the characteristic data according to battery types, constructing an exclusive training data set for each type of battery, and then training a regression model network by using the data sets to form an anomaly judgment model for different battery types. And analyzing the charging data of the to-be-detected battery through the constructed model, and automatically predicting whether the electric quantity is abnormal or not. According to the method, the big data and machine learning technologies are utilized, historical data are combined, and through data cleaning, feature extraction and model training, high-accuracy anomaly prediction is achieved, manual participation is reduced, economical efficiency and efficiency are improved, electric quantity anomalies in the charging process are effectively prevented, and the situation that the health of the battery is affected due to overcharge of the battery is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium batteries, and in particular to a method, system and terminal for judging abnormal charging quantity during the charging process of a lithium battery. Background Art

[0002] Lithium-ion battery charging abnormality warning can avoid potential safety problems in the battery system, such as overcharging. At the same time, it can also determine abnormal battery health and charging equipment failures. Timely processing can avoid danger.

[0003] To determine whether a lithium-ion battery has abnormal power during the charging process, the existing technology generally uses the following methods: Method for detecting voltage and current abnormalities: During the charging process, the voltage and current of the battery are monitored in real time and compared with the predetermined charging curve. If the voltage or current is found to deviate from the normal range, there may be abnormal power. Method for detecting temperature abnormalities: The temperature of the battery should be kept within a certain range during the charging process. If the temperature is too high or too low, it may indicate that there is a problem with the battery, resulting in abnormal power. Method for detecting charging time abnormalities: Based on the battery capacity and charging rate, calculate the expected charging time. If the actual charging time is significantly longer or shorter than the expected time, there may be abnormal power.

[0004] However, the current method of judging abnormal power levels of lithium-ion batteries during charging not only has relatively simple logic rules, but also requires a lot of manual judgment and is unable to identify data with small differences. In addition, for batteries of different capacities, multiple inspection methods need to be designed, and there is no way to accurately and efficiently identify abnormal battery charging levels, especially in scenarios where a large number of batteries need to be monitored, and there is no way to achieve full promotion. Summary of the invention

[0005] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a method, system and terminal for judging abnormal charging capacity during the charging process of a lithium battery, so as to solve the technical problems that the prior art method for judging abnormal charging capacity of a lithium-ion battery during the charging process is heavily dependent on manpower, has low accuracy, is difficult to apply on a large scale, and has complex rule design and single judgment logic when facing different battery models.

[0006] To achieve the above-mentioned purpose and other related purposes, the present invention provides a method for judging abnormal charging capacity during the charging process of a lithium battery, the method comprising: obtaining battery charging data of multiple lithium batteries for each charging within a preset time period; wherein the battery charging data comprises: the battery capacity of the lithium battery and the charging process data of the current charging; performing data cleaning on the charging data of each battery to eliminate data with obvious abnormalities; extracting effective charging cycle data based on each charging process data to obtain effective charging process data; extracting battery features for each effective charging process data to obtain corresponding battery feature data; grouping each battery feature data according to battery type, and using each group of data to construct a training data set separately; using each regression model based on the gradient boosting algorithm for each training data set to construct a charging capacity abnormality judgment model corresponding to different battery types; using the constructed charging capacity abnormality judgment model to obtain the corresponding charging capacity abnormality judgment result according to the battery charging data to be detected.

[0007] In one embodiment of the present invention, the charging process data includes: SOC, current, temperature and cell voltage during the charging process.

[0008] In one embodiment of the present invention, the extraction of effective charging cycle data based on each charging process data to obtain effective charging process data includes: screening charging process data that meets the effective charging cycle data extraction conditions from each charging process data as effective charging process data; wherein the effective charging cycle data extraction conditions include: SOC range conditions, including: SOC at least included in a preset power range; charging data collection sample quantity conditions, including: the number of charging data collection samples collected at a set collection time interval is not less than a preset quantity threshold.

[0009] In one embodiment of the present invention, the extracted battery characteristics include: initial SOC, final SOC, average current during the complete charging period, average current in segments of multiple charge stages, initial temperature, maximum charging temperature, final temperature, initial voltage, maximum voltage in segments, and final voltage.

[0010] In one embodiment of the present invention, the battery characteristic data are grouped according to battery type and a training data set is constructed separately using each group of data, including: grouping the battery types according to the battery capacity in each battery characteristic data, so that battery characteristic data with the same battery capacity are grouped into the same battery type group; and dividing the battery characteristic data of each battery type group as a separate training data set.

[0011] In one embodiment of the present invention, the use of a regression model based on a gradient boosting algorithm for each training data set to construct a charging capacity abnormality judgment model corresponding to different battery types includes: based on the same model training mechanism, using each training data set to train the gradient boosting model respectively to construct a charging capacity abnormality judgment model corresponding to different battery types.

[0012] In one embodiment of the present invention, the model training mechanism includes: using a training data set to perform a first training on the gradient boosting model, and using the 2σ principle to remove abnormal data samples to form a new training data set; using the new training data set to perform a second training on the gradient boosting model to obtain a charging power abnormality judgment model.

[0013] In one embodiment of the present invention, the method of utilizing the constructed charging power abnormality judgment model to obtain the corresponding charging power abnormality judgment result according to the battery charging data to be detected includes: based on the charging power abnormality judgment model of the corresponding charging type selected by the battery charging data to be detected, obtaining the corresponding charging power abnormality judgment result according to the battery charging data to be detected.

[0014] In order to achieve the above-mentioned purpose and other related purposes, the present invention provides a system for judging abnormal charging capacity during the charging process of a lithium battery, the system comprising: a data acquisition module, used to obtain battery charging data of multiple lithium batteries each time they are charged within a preset time period; wherein the battery charging data comprises: the battery capacity of the lithium battery and the charging process data of the current charge; a data cleaning module, connected to the data acquisition module, used to clean the charging data of each battery to eliminate data with obvious abnormalities; an effective data extraction module, connected to the data cleaning module, used to extract effective charging cycle data based on each charging process data to obtain effective charging process data; a battery feature extraction module, connected to the effective data An extraction module is used to extract battery features from each valid charging process data to obtain corresponding battery feature data; a feature data grouping module is connected to the battery feature extraction module and is used to group each battery feature data according to the battery type, and use each group of data to separately construct a training data set; a model training module is connected to the feature data grouping module and is used to use the regression model based on the gradient boosting algorithm for each training data set to construct a charging power abnormality judgment model corresponding to different battery types; a charging power abnormality judgment module is connected to the model training module and is used to use the constructed charging power abnormality judgment model to obtain the corresponding charging power abnormality judgment result according to the battery charging data to be detected.

[0015] To achieve the above-mentioned purpose and other related purposes, the present invention provides an electronic terminal, comprising: one or more memories and one or more processors; the one or more memories are used to store computer programs; the one or more processors are connected to the memories and are used to run the computer program to execute the method for judging abnormal charging capacity during the charging process of the lithium battery.

[0016] As described above, the present invention is a method, system and terminal for judging abnormal charging capacity during the charging process of a lithium battery, which has the following beneficial effects: the present invention first collects charging data of multiple lithium batteries in a specific time period, then cleans the data to remove abnormal values, then extracts effective charging cycle data from the cleaned data, and performs battery feature analysis; then groups these feature data according to the battery type, builds a dedicated training data set for each battery, and then uses these data sets to train the regression model network separately to form an abnormality judgment model for different battery types. The constructed model will be used to analyze the charging data of the battery to be tested and automatically predict whether there is an abnormality in charge. The present invention uses big data and machine learning technology, combined with historical data, to achieve high-accuracy abnormality prediction through data cleaning, feature extraction and model training, reduce manual participation, improve economy and efficiency, effectively prevent abnormal charge during the charging process, and avoid the situation where the battery health is affected by overcharging. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Shown is a flow chart of a method for determining abnormal charging capacity during charging of a lithium battery in one embodiment of the present invention.

[0018] Figure 2 Shown is a schematic diagram of the structure of a gradient boosting model in one embodiment of the present invention.

[0019] Figure 3 Shown is a structural schematic diagram of a system for judging abnormal charging capacity during charging of a lithium battery in one embodiment of the present invention.

[0020] Figure 4 Shown is a schematic structural diagram of an electronic terminal in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0022] It should be noted that in the following description, reference is made to the accompanying drawings, which describe several embodiments of the present invention. It should be understood that other embodiments may also be used, and that mechanical composition, structural, electrical and operational changes may be made without departing from the spirit and scope of the present invention. The following detailed description should not be considered restrictive, and the scope of the embodiments of the present invention is limited only by the claims of the published patents. The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. Spatially related terms, such as "upper", "lower", "left", "right", "below", "below", "lower", "above", "upper", etc., may be used in the text to facilitate the description of the relationship between an element or feature shown in the figure and another element or feature.

[0023] Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case of "direct connection" but also the case of "indirect connection" by placing other elements therebetween. In addition, when a part is said to "include" a certain constituent element, unless otherwise stated, it does not exclude other constituent elements, but means that other constituent elements may be included.

[0024] The terms first, second and third mentioned herein are used to describe various parts, components, regions, layers and / or segments, but are not limited thereto. These terms are only used to distinguish a certain part, component, region, layer or segment from other parts, components, regions, layers or segments. Therefore, the first part, component, region, layer or segment described below may refer to the second part, component, region, layer or segment within the scope of the present invention.

[0025] Furthermore, as used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless there is an indication to the contrary in the context. It should be further understood that the terms "comprise", "include" indicate the presence of the described features, operations, elements, components, items, kinds, and / or groups, but do not exclude the presence, occurrence or addition of one or more other features, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Therefore, "A, B or C" or "A, B and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B and C". Exceptions to this definition will only occur when the combination of elements, functions or operations is inherently mutually exclusive in some way.

[0026] The present invention provides a method for judging abnormal charging power during the charging process of a lithium battery. First, the charging data of multiple lithium batteries in a specific time period are collected, and then the data is cleaned to remove abnormal values. Then, effective charging cycle data is extracted from the cleaned data, and battery feature analysis is performed; these feature data are grouped according to the battery type, and a dedicated training data set is constructed for each battery. Then, these data sets are used to train regression model networks respectively to form abnormal judgment models for different battery types. The constructed model will be used to analyze the charging data of the battery to be tested and automatically predict whether there is abnormal power. The present invention uses big data and machine learning technology, combined with historical data, through data cleaning, feature extraction and model training, to achieve high-accuracy abnormal prediction, reduce manual participation, improve economy and efficiency, effectively prevent abnormal power during the charging process, and avoid the situation where the battery health is affected by overcharging.

[0027] The following is a detailed description of the embodiments of the present invention with reference to the accompanying drawings so that those skilled in the art can easily implement the present invention. The present invention can be embodied in many different forms and is not limited to the embodiments described herein.

[0028] like Figure 1 A schematic flow chart showing a method for determining abnormal charging power during charging of a lithium battery in an embodiment of the present invention.

[0029] The lithium battery used in the present invention is a power battery, which is widely used in occasions that require continuous, stable and high energy density power supply, such as electric vehicles (EVs), power tools (such as electric drills, electric saws, etc.), energy storage systems (for grid energy storage, home energy storage, etc.) and other equipment that requires high power and long-term operation. Compared with other types of lithium batteries, power batteries usually have a larger capacity, usually around 200Ah, with a larger discharge power and higher specific energy, and can release a large amount of electricity in a short time. And support fast charging technology to meet the needs of rapid energy replenishment. Have a long cycle life to meet the needs of frequent charging and discharging.

[0030] The method comprises:

[0031] Step S1: Obtain battery charging data of multiple lithium batteries during each charging within a preset time period.

[0032] Specifically, the battery capacity of each lithium battery and the charging process data of each lithium battery during each charging within a preset time period are obtained. The preset time period is a specific time range set according to demand, such as the past week, month or longer time period.

[0033] In one embodiment, the charging process data includes:

[0034] SOC (State of Charge): reflects the percentage of the current remaining power of the battery and is a key parameter in the battery management system.

[0035] Current: The current value during the charging process reflects the battery charging rate. It can be measured by a shunt or Hall sensor, where the shunt uses Ohm's law (U=I*R) to convert the current into a voltage signal for measurement.

[0036] Temperature: The temperature of the battery during the charging process is critical to preventing overheating and ensuring battery performance. It can be measured by temperature sensors such as thermocouples, RTDs, or thermistors.

[0037] Cell voltage: The voltage of each battery cell during the charging process, which is very important for monitoring the battery status and ensuring balanced charging of the battery.

[0038] The battery capacity of each lithium battery and the SOC, current, temperature and cell voltage training model during the charging process are used to judge charging anomalies, which can fully capture the state of the battery during the charging process and improve the accuracy of anomaly detection. Among them, real-time monitoring of SOC and current helps to timely discover safety hazards such as overcharging or over-discharging, while temperature monitoring can predict and prevent overheating and protect battery performance. The analysis of cell voltage can identify inconsistencies within the battery and warn of potential failures in advance. In addition, using machine learning algorithms, charging patterns and abnormal characteristics can be learned from a large amount of historical data to provide data-driven decision support for the battery management system (BMS). And due to the introduction of the corresponding battery capacity, the model can adapt to different types of batteries and has good versatility and flexibility.

[0039] Step S2: Clean the charging data of each battery to remove obviously abnormal data.

[0040] Specifically, each battery charging data is checked for abnormal data, such as null values ​​or missing values, and the abnormal data is removed.

[0041] Step S3: Based on each charging process data, extract effective charging cycle data to obtain effective charging process data.

[0042] In one embodiment, step S3 includes:

[0043] Selecting charging process data that meets the effective charging cycle data extraction condition from each charging process data as effective charging process data;

[0044] Among them, the effective charging cycle data extraction conditions include:

[0045] The SOC range condition includes: an SOC at least included in a preset power range; it should be noted that the preset power range can be set according to demand, such as 20% to 80%, which is usually a range that is more sensitive to battery performance analysis.

[0046] The charging data collection sample quantity condition includes: the number of charging data collection samples collected in the battery charging data according to the set collection time interval is not less than the preset number threshold; the main purpose of this condition is to ensure that the data contains the entire charging process from low power to high power, so that the behavior of the battery at different power levels can be more comprehensively understood and analyzed. It should be noted that a suitable data collection time interval can be set according to the charging characteristics and analysis requirements of the battery, such as collecting once every minute or every second; the preset number threshold can be set according to the needs, such as 100, to ensure that there are enough data points for each charging cycle to facilitate reliable statistical analysis.

[0047] Step S4: extracting battery characteristics from each valid charging process data to obtain corresponding battery characteristic data.

[0048] In one embodiment, the extracted battery features include:

[0049] Initial SOC indicates the remaining battery power when charging begins; this feature can be used to understand the battery status before charging, which helps to evaluate the charging efficiency and the initial health status of the battery.

[0050] End SOC indicates the remaining power of the battery when charging is completed; this feature reflects the overall current output capability of the charging system and is related to charging efficiency and battery heating.

[0051] The average current during the complete charging period indicates the average current during the entire charging process from the beginning to the end of charging. This feature reflects the overall current output capability of the charging system and is related to the charging efficiency and battery heating.

[0052] The segmented average current of multiple power stages divides the charging process into multiple stages according to the power change, and calculates the current average of each stage. The current changes in different stages can reflect the dynamic characteristics of the battery's internal resistance, polarization phenomenon, etc., which helps to understand the behavior of the battery during the charging process. Preferably, the segmented average current of three power stages can be included, namely, the segmented average current of [0, 0.5], the segmented average current of [0.5, 0.8], and the segmented average current of [0.8, 1].

[0053] The initial temperature refers to the temperature of the battery when charging begins. Battery temperature is an important factor affecting charging performance and safety. The initial temperature can be used as a benchmark for temperature management.

[0054] The maximum charging temperature indicates the highest temperature reached by the battery during the charging process. Since excessively high temperatures will accelerate battery aging, reduce battery life, and even cause safety issues, it is crucial to monitor the maximum temperature.

[0055] The end temperature indicates the temperature of the battery at the end of charging. This feature combined with the initial temperature can evaluate the impact of the charging process on the battery temperature and guide the design and optimization of the cooling system.

[0056] Initial voltage refers to the voltage of the battery before charging begins. Voltage is an important indicator of battery status. This feature helps determine the health of the battery and whether pretreatment (such as precharging) is required.

[0057] The maximum voltage of each segment means that the maximum voltage reached by the battery in a specific power stage is recorded. The change in voltage can reflect the intensity of the chemical reaction inside the battery, which is helpful for understanding battery characteristics and optimizing charging strategies. Preferably, the maximum voltage reached by the battery in [0.8, 1] is selected.

[0058] The end voltage indicates the voltage of the battery at the end of charging. This feature combined with the initial voltage can evaluate the impact of the charging process on the battery voltage, especially when it is close to full charge. The change in voltage can reflect the charging limit conditions of the battery.

[0059] One of the advantages of selecting the above-mentioned battery feature combination for abnormal charging power judgment compared to existing solutions is that it realizes more comprehensive battery status monitoring. This combination not only covers the power before and after charging (initial SOC and final SOC), ensuring the basis for charging efficiency evaluation, but also accurately captures current fluctuations and changes in internal battery characteristics through complete charging period and multi-stage segmented average current analysis. At the same time, the temperature changes and voltage responses throughout the charging process are recorded in detail to effectively identify overheating risks and signs of battery performance degradation. The comprehensive application of these multiple feature data can build a more accurate charging power anomaly detection model, improve the accuracy and sensitivity of detection, and adapt to a wide range of power battery types and charging scenarios, showing strong versatility and flexibility.

[0060] Step S5: Group the battery characteristic data according to the battery type, and use each group of data to construct a training data set separately.

[0061] In one embodiment, the capacity of the battery cell is a key factor that directly affects the battery charging capacity. Batteries with different capacities will exhibit different characteristics during the charging process. Therefore, an independent training data set is constructed based on the grouping strategy of battery capacity, and then the model structures of different battery capacities are constructed respectively.

[0062] Therefore, step S5:

[0063] The battery types are grouped according to the battery capacity in each battery characteristic data, so that the battery characteristic data with the same battery capacity are grouped into the same battery type group; specifically, each battery characteristic data is traversed, and the battery characteristic data is allocated to the corresponding battery capacity group according to the battery capacity of each battery characteristic data.

[0064] The battery characteristic data of each battery type group is divided as a separate training data set. Specifically, for each battery capacity group, all battery characteristic data in the group are extracted to form an independent training data set. In the training data set of each battery capacity group, it is further divided into a training set, a validation set, and a test set. The proportion of these data sets can be adjusted according to actual conditions, but the common proportion is 70% training set, 20% validation set, and 10% test set.

[0065] Step S6: using the regression model based on the gradient boosting algorithm for each training data set to construct a charging capacity abnormality judgment model corresponding to different battery types.

[0066] In one embodiment, step S6 includes:

[0067] Based on the same model training mechanism, each training data set is used to train the gradient boosting model respectively to build charging capacity abnormality judgment models corresponding to different battery types.

[0068] Specifically, the regression model network used in this solution is a gradient boosting model; Figure 2 As shown in Figure 1, Gradient Boosting is an ensemble learning algorithm that combines multiple weak learners (usually decision trees) into a strong learner.

[0069] It performs well in many machine learning tasks and has several advantages:

[0070] 1. High prediction performance: Gradient boosting has been widely used in many machine learning competitions and practical applications, and has performed well in prediction performance. It is usually able to produce accurate prediction models with low bias and high generalization ability.

[0071] 2. Handling complex relationships: Gradient boosting can capture complex nonlinear relationships between features. By combining multiple decision tree models, it can learn more complex feature interactions and nonlinear relationships, thereby improving the model's expressiveness.

[0072] 3. Robustness: Gradient boosting is robust to outliers and noisy data. Due to the integration of multiple weak learners, it has relatively little impact on individual erroneous predictions and can reduce the risk of overfitting of the model.

[0073] In one embodiment, the model training mechanism is a two-stage optimization process, which aims to filter out abnormal data in the first training and retrain to improve the performance of the gradient boosting model in the charging power abnormality judgment task.

[0074] The model training mechanism includes:

[0075] The training data set is used to train the gradient boosting model for the first time, and the 2σ principle is used to remove abnormal data samples to form a new training data set; specifically, the training data set is used to train the gradient boosting model for the first time. After the first training is completed, the trained model is used to predict each data sample in the training data set to obtain the predicted value of each sample, the error between the predicted value and the true value is calculated, and the 2σ principle is applied to identify abnormal data samples. That is, data points whose errors exceed two standard deviations of the mean error are regarded as outliers, and these data samples identified as abnormal are removed to form a new training data set.

[0076] The gradient boosting model is trained for a second time using the new training data set to obtain a charging power abnormality judgment model. Specifically, after the first stage of abnormality detection screening, those data samples that are considered normal and representative in the original training data set are retained to form a new training data set. The gradient boosting model is trained for a second time using this new training data set. After the second training, an optimized gradient boosting model is obtained, which is specifically used for the charging power abnormality judgment task. The model has high prediction accuracy and robustness, and can effectively identify charging power abnormalities in practical applications.

[0077] Step S7: using the constructed charging power abnormality judgment model, obtain the corresponding charging power abnormality judgment result according to the battery charging data to be detected.

[0078] In one embodiment, step S7 includes:

[0079] Based on the charging power abnormality judgment model of the corresponding charging type selected by the battery charging data to be detected, the corresponding charging power abnormality judgment result is obtained according to the battery charging data to be detected.

[0080] Specifically, a charging capacity abnormality judgment model of the corresponding charging type is selected according to the battery charging data to be detected; battery features are extracted from the battery charging data to be detected to obtain corresponding battery feature data, and a charging capacity abnormality judgment result is obtained according to the selected charging capacity abnormality judgment model of the charging type.

[0081] Similar to the principle of the above-mentioned embodiment, the present invention provides a system for judging abnormal charging capacity during the charging process of a lithium battery.

[0082] The following provides specific embodiments in conjunction with the accompanying drawings:

[0083] like Figure 3 A structural schematic diagram showing a system for judging abnormal charging capacity during charging of a lithium battery in an embodiment of the present invention.

[0084] The system comprises:

[0085] The data acquisition module 1 is used to acquire the battery charging data of each charging of multiple lithium batteries within a preset time period; wherein the battery charging data includes: the battery capacity of the lithium battery and the charging process data of the charging;

[0086] A data cleaning module 2, connected to the data acquisition module 1, is used to clean the charging data of each battery to remove obviously abnormal data;

[0087] An effective data extraction module 3, connected to the data cleaning module 2, is used to extract effective charging cycle data based on each charging process data to obtain effective charging process data;

[0088] A battery feature extraction module 4, connected to the effective data extraction module 3, is used to extract battery features from each effective charging process data to obtain corresponding battery feature data;

[0089] A feature data grouping module 5, connected to the battery feature extraction module 4, is used to group the battery feature data according to the battery type, and use each group of data to construct a training data set separately;

[0090] A model training module 6, connected to the feature data grouping module 5, is used to use a regression model based on a gradient boosting algorithm for each training data set to construct a charging power abnormality judgment model corresponding to different battery types;

[0091] The charging power abnormality judgment module 7 is connected to the model training module 6 and is used to use the constructed charging power abnormality judgment model to obtain the corresponding charging power abnormality judgment result according to the battery charging data to be detected.

[0092] It should be understood that Figure 3 The division of each module in the system embodiment is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or physically separated. And these units can be all implemented in the form of software calling through processing elements; or all implemented in the form of hardware; or some units can be implemented in the form of processing elements calling software, and some units can be implemented in the form of hardware.

[0093] Since the implementation principle of the system for judging abnormal charging capacity during the lithium battery charging process has been described in the aforementioned embodiment, it will not be repeated here.

[0094] The method for judging abnormal charging power during the charging process of a lithium battery provided in the embodiment of the present invention can be implemented on the terminal side or the server side. As for the hardware structure of the electronic terminal, please refer to Figure 4 , is an optional hardware structure diagram of a terminal 1000 for judging abnormal charging capacity during lithium battery charging provided in an embodiment of the present invention. The terminal 1000 may be a mobile phone, a computer device, a tablet device, a personal digital processing device, a factory background processing device, etc. The terminal 1000 includes: at least one processor 1001, a memory 1002, at least one network interface 10010 and a user interface 1009. The various components in the device are coupled together through a bus system 1005. It can be understood that the bus system 1005 is used to realize connection and communication between these components. In addition to the data bus, the bus system 1005 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, in Figure 4 In the specification, various buses are labeled as bus systems.

[0095] The user interface 1009 may include a display, a keyboard, a mouse, a trackball, a click gun, keys, buttons, a touch pad or a touch screen.

[0096] It is understood that the memory 1002 can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), which is used as an external cache. By way of exemplary but not limiting explanation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM). The memory described in the embodiments of the present invention is intended to include but is not limited to these and any other suitable categories of memory.

[0097] The memory 1002 in the embodiment of the present invention is used to store various categories of data to support the operation of the terminal 1000. Examples of these data include: any executable program used to operate on the terminal 1000, such as an operating system 10021 and an application 10022; the operating system 10021 includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks. The application 10022 may include various applications, such as a media player (MediaPlayer), a browser (Browser), etc., for implementing various application services. The method for determining abnormal charging power during lithium battery charging provided in the embodiment of the present invention may be included in the application 10022.

[0098] The method disclosed in the above embodiment of the present invention can be applied to the processor 1001, or implemented by the processor 1001. The processor 1001 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 1001 or the instruction in the form of software. The above processor 1001 may be a general processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor 1001 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiment of the present invention. The general processor 1001 can be a microprocessor or any conventional processor, etc. In combination with the steps of the accessory optimization method provided in the embodiment of the present invention, it can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, which is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0099] In an exemplary embodiment, the terminal 1000 may be implemented by one or more application specific integrated circuits (ASIC), DSP, programmable logic device (PLD), complex programmable logic device (CPLD) to execute the aforementioned method.

[0100] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

[0101] In the embodiments provided in the present application, the computer readable and writable storage medium may include a read-only memory, a random access memory, an EEPROM, a CD-ROM or other optical disk storage device, a disk storage device or other magnetic storage device, a flash memory, a USB flash drive, a mobile hard disk, or any other medium that can be used to store a desired program code in the form of an instruction or data structure and can be accessed by a computer. In addition, any connection can be appropriately referred to as a computer-readable medium. For example, if the instruction is sent from a website, a server or other remote source using a coaxial cable, an optical fiber cable, a twisted pair, a digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwaves, the coaxial cable, optical fiber cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwaves are included in the definition of the medium. However, it should be understood that computer readable and writable storage media and data storage media do not include connections, carriers, signals, or other temporary media, but are intended to be non-temporary, tangible storage media. Disk and disc, as used in this application, includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers.

[0102] This patent introduces a regression model into the charging power judgment process of actual data, aiming to significantly improve the accuracy and efficiency of power anomaly detection. The regression model can automatically learn and adapt to the unique charging characteristics of each type of battery, including voltage changes, current fluctuations, temperature response, and charging time, so as to build a more realistic power anomaly detection model. The advantage of this solution lies in its high flexibility and scalability, so that for different types of batteries, it is only necessary to identify their basic capacity parameters without the tedious distinction of batches, manufacturers and other details to achieve accurate judgment of charging power data. Through this method, it can also effectively overcome the problems of simple logical rules, heavy manual intervention, and insufficient ability to identify differentiated data in traditional detection methods. In addition, the implementation of this solution will greatly improve the automation level of the production line, reduce manual participation, reduce the risk of human error, and realize a large-scale and efficient production model.

[0103] In summary, the method, system and terminal for judging abnormal charging power during the charging process of a lithium battery of the present invention first collect the charging data of multiple lithium batteries in a specific time period, then clean the data to remove abnormal values, then extract the effective charging cycle data from the cleaned data, and perform battery feature analysis; then group these feature data according to the battery type, build a dedicated training data set for each battery, and then use these data sets to train the regression model network respectively to form an abnormal judgment model for different battery types. The constructed model will be used to analyze the charging data of the battery to be tested and automatically predict whether there is abnormal power. The present invention uses big data and machine learning technology, combined with historical data, through data cleaning, feature extraction and model training, to achieve high-accuracy abnormal prediction, reduce manual participation, improve economy and efficiency, effectively prevent abnormal power during charging, and avoid the situation where the battery health is affected by overcharging. Therefore, the present invention effectively overcomes the various shortcomings of the prior art and has a high industrial utilization value.

[0104] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for judging abnormal charging quantity during the charging process of a lithium battery, characterized in that: The method comprises: Obtaining battery charging data of each charge of multiple lithium batteries within a preset time period; wherein the battery charging data includes: the battery capacity of the lithium battery and the charging process data of the charge; Clean the charging data of each battery to remove obviously abnormal data; Based on each charging process data, effective charging cycle data is extracted to obtain effective charging process data; Extract battery characteristics for each valid charging process data to obtain corresponding battery characteristic data; The battery characteristic data are grouped according to the battery type, and a training data set is constructed separately using each group of data; Using a regression model based on a gradient boosting algorithm for each training data set, a charging capacity abnormality judgment model corresponding to different battery types is constructed; The constructed charging power abnormality judgment model is used to obtain the corresponding charging power abnormality judgment result according to the battery charging data to be detected.

2. The method for judging abnormal charging quantity during charging of a lithium battery according to claim 1, characterized in that: The charging process data includes: SOC, current, temperature and cell voltage during the charging process.

3. The method for judging abnormal charging quantity during the charging process of a lithium battery according to claim 2, characterized in that: The extracting effective charging cycle data based on each charging process data to obtain effective charging process data includes: Selecting charging process data that meets the effective charging cycle data extraction condition from each charging process data as effective charging process data; Among them, the effective charging cycle data extraction conditions include: SOC range conditions include: an SOC at least within a preset power range; The charging data collection sample quantity condition includes: the number of charging data collection samples collected at a set collection time interval is not less than a preset number threshold.

4. The method for judging abnormal charging quantity during the charging process of a lithium battery according to claim 2, characterized in that: The extracted battery features include: initial SOC, final SOC, average current during the complete charging period, average current in multiple power stages, initial temperature, maximum charging temperature, final temperature, initial voltage, maximum voltage in each stage, and final voltage.

5. The method for judging abnormal charging quantity during the charging process of a lithium battery according to claim 2, characterized in that: The step of grouping the battery characteristic data according to the battery type and constructing a training data set using each group of data separately includes: Grouping the battery types according to the battery capacity in each battery characteristic data, so that the battery characteristic data with the same battery capacity are grouped into the same battery type group; The battery characteristic data of each battery type group is divided into a separate training data set.

6. The method for judging abnormal charging quantity during the charging process of a lithium battery according to claim 5, characterized in that: The method of using each training data set to construct a charging capacity abnormality judgment model corresponding to different battery types by using a regression model based on a gradient boosting algorithm includes: Based on the same model training mechanism, each training data set is used to train the gradient boosting model respectively to build charging capacity abnormality judgment models corresponding to different battery types.

7. The method for judging abnormal charging quantity during the charging process of a lithium battery according to claim 6, characterized in that: The model training mechanism includes: Use the training data set to train the gradient boosting model for the first time, and use the 2σ principle to remove abnormal data samples to form a new training data set; The gradient boosting model is trained for a second time using a new training data set to obtain a charging power abnormality judgment model.

8. The method for judging abnormal charging quantity during charging of a lithium battery according to claim 7, characterized in that: The method of using the constructed charging power abnormality judgment model to obtain the corresponding charging power abnormality judgment result according to the battery charging data to be detected includes: Based on the charging power abnormality judgment model of the corresponding charging type selected by the battery charging data to be detected, the corresponding charging power abnormality judgment result is obtained according to the battery charging data to be detected.

9. A system for judging abnormal charging quantity during lithium battery charging, characterized in that: The system comprises: A data acquisition module, used to acquire battery charging data of multiple lithium batteries each time they are charged within a preset time period; wherein the battery charging data includes: the battery capacity of the lithium battery and the charging process data of the charging; A data cleaning module, connected to the data acquisition module, for cleaning the charging data of each battery to remove obviously abnormal data; An effective data extraction module, connected to the data cleaning module, for extracting effective charging cycle data based on each charging process data to obtain effective charging process data; A battery feature extraction module, connected to the effective data extraction module, for extracting battery features from each effective charging process data to obtain corresponding battery feature data; A feature data grouping module, connected to the battery feature extraction module, for grouping the battery feature data according to the battery type, and using each group of data to construct a training data set separately; A model training module, connected to the feature data grouping module, is used to use a regression model based on a gradient boosting algorithm for each training data set to construct a charging power abnormality judgment model corresponding to different battery types; The charging power abnormality judgment module is connected to the model training module and is used to use the constructed charging power abnormality judgment model to obtain the corresponding charging power abnormality judgment result according to the battery charging data to be detected.

10. An electronic terminal, characterized in that: include: one or more memories and one or more processors; The one or more memories are used to store computer programs; The one or more processors, connected to the memory, are configured to run the computer program to perform the method according to any one of claims 1 to 8.

Images