Battery detection method, device, equipment, medium and product

Abstract

The embodiment of the invention provides a battery detection method, device and equipment, a medium and a product. The method comprises the following steps: determining monitoring data of different monitoring areas on a battery to be detected in a preset time, wherein the monitoring data comprises one or more of temperature monitoring data, pressure monitoring data, voltage monitoring data and thickness monitoring data; according to the monitoring data of different monitoring areas on the to-be-detected battery in the preset time, determining characteristic parameters representing the change condition of the corresponding monitoring data on the monitoring areas along with time; inputting the characteristic parameters into a recognition model to obtain a quality category label of the to-be-detected battery; wherein the identification model is obtained through training according to quality category labels of the sample battery and sample characteristic parameters corresponding to different quality category labels, and the sample characteristic parameters represent the change condition of monitoring sample data corresponding to different monitoring areas on the sample battery along with time. The method is used for achieving the effect of improving the detection accuracy of the battery.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery detection, and in particular to a battery detection method, device, equipment, medium and product. BACKGROUND

[0002] With the rapid development of new energy vehicles, energy storage systems and consumer electronics, the safety, reliability and service life of lithium ion batteries as the core energy carrier are of great concern. During the charging and discharging process of the battery, heat is generated and accompanied by slight structural deformation, resulting in uneven internal temperature and pressure distribution, which in turn affects the performance of the battery, and even causes safety risks such as thermal runaway.

[0003] Currently, when detecting the battery, a single-point or multi-point thermistor (such as NTC) or thermocouple is generally arranged on the surface or the tab of the battery to obtain the temperature of the battery, and a sensor is installed on the battery pack to obtain the voltage of the battery, and then the quality category of the battery is determined by manual judgment.

[0004] However, the existing method has the technical problem of inaccurate detection accuracy when detecting. SUMMARY

[0005] The battery detection method, device, equipment, medium and product provided by the embodiments of the present application can improve the accuracy of detecting the battery.

[0006] In a first aspect, the embodiments of the present application provide a battery detection method, comprising:

[0007] determining the monitoring data of different monitoring areas on the to-be-detected battery within a preset time, the monitoring data including one or more of temperature monitoring data, pressure monitoring data, voltage monitoring data and thickness monitoring data;

[0008] determining a feature parameter representing the change of the corresponding monitoring data on the monitoring area with time according to the monitoring data of different monitoring areas on the to-be-detected battery within the preset time;

[0009] inputting the feature parameter into an identification model to obtain a quality category label of the to-be-detected battery; wherein the identification model is trained according to the quality category label of the sample battery and the sample feature parameter corresponding to the different quality category labels, and the sample feature parameter represents the change of the monitoring sample data corresponding to different monitoring areas on the sample battery with time.

[0010] In a possible implementation, when the monitoring data is temperature monitoring data, determining the monitoring data of different monitoring areas on the to-be-detected battery at different times comprises:

[0011] According to the preset charging and discharging condition, thermal imaging data of the monitoring surface of the battery to be tested at different time points within a preset time is obtained; the charging and discharging condition includes a charging and discharging rate and a voltage range of the battery to be tested;

[0012] According to the preset area division rule and the thermal imaging data, temperature monitoring data of different monitoring areas on the battery to be tested at different times is determined.

[0013] In a possible implementation, when the monitoring data is pressure monitoring data, determining the monitoring data of different monitoring areas on the battery to be tested at different times includes:

[0014] According to the preset pressure monitoring condition, pressure sensor data of the monitoring surface of the battery to be tested at different time points within a preset time is obtained;

[0015] According to the preset area division rule and the pressure sensor data, pressure monitoring data of different monitoring areas on the battery to be tested at different times is determined.

[0016] In a possible implementation, before the feature parameter representing the change of the corresponding monitoring data on the monitoring area over time is input into the identification model to obtain the quality category of the battery to be tested, the method further includes:

[0017] Monitoring sample data of different monitoring areas on the sample battery within a preset time is obtained;

[0018] When the obtained monitoring sample data does not meet the preset data requirement, the monitoring sample data is marked;

[0019] In response to a user's judgment operation on the marked monitoring sample data, a quality category label of the sample battery is determined;

[0020] When the monitoring sample data corresponding to each quality category label is greater than a preset sample data amount, the initial model is trained according to the monitoring sample data and the quality category label of the sample battery, to obtain the identification model.

[0021] In a possible implementation, the initial model is trained according to the monitoring sample data and the quality category label of the sample battery, to obtain the identification model, including:

[0022] According to the monitoring sample data of different monitoring areas on the sample battery within a preset time, a feature sample parameter representing the change of the corresponding monitoring sample data on the monitoring area over time is determined;

[0023] The feature sample parameter is normalized to obtain a normalized sample parameter;

[0024] According to the normalized sample parameters and the quality category labels of the corresponding sample batteries, the initial model is trained to obtain an initial identification model; the quality category labels include a quality category label representing a normal sample battery and a quality category label representing an abnormal sample battery;

[0025] The thermal runaway label is filtered out from the quality category label representing the abnormal sample battery.

[0026] According to the thermal runaway label in the quality category label and the normalized sample parameters corresponding to the thermal runaway label, the initial identification model is trained to obtain an identification model.

[0027] In a possible implementation, the method further includes:

[0028] According to the quality category of the to-be-tested battery and the monitoring data of different monitoring regions on the to-be-tested battery within a preset time, the identification model is updated and trained to obtain an updated and trained identification model.

[0029] In a possible implementation, after the feature parameter representing the change of the corresponding monitoring data on the monitoring region over time is input into the identification model to obtain the quality category of the to-be-tested battery, the method further includes:

[0030] According to the monitoring data of different monitoring regions on the to-be-tested battery within a preset time, a change curve of the monitoring region and a global change curve representing all monitoring regions on the to-be-tested battery are generated.

[0031] The quality category of the to-be-tested battery, the change curve of the monitoring region and the global change curve representing all monitoring regions on the to-be-tested battery are displayed.

[0032] In a second aspect, an embodiment of the present application provides a battery detection device, including:

[0033] The determination module is configured to determine the monitoring data of different monitoring regions on the to-be-tested battery within a preset time, and the monitoring data includes one or more of temperature monitoring data, pressure monitoring data, voltage monitoring data and thickness monitoring data.

[0034] The extraction module is configured to determine, according to the monitoring data of different monitoring regions on the to-be-tested battery within a preset time, a feature parameter representing the change of the corresponding monitoring data on the monitoring region over time.

[0035] The identification module is configured to input the feature parameter into an identification model to obtain a quality category label of the to-be-tested battery; the identification model is trained according to the quality category labels of the sample batteries and the sample feature parameters corresponding to different quality category labels, and the sample feature parameters represent the change of the corresponding monitoring sample data on different monitoring regions of the sample battery over time.

[0036] In a third aspect, an electronic device is provided, comprising: a memory, a processor;

[0037] The memory stores computer-executable instructions.

[0038] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the first aspect and / or various possible implementation manners of the first aspect.

[0039] In a fourth aspect, a computer-readable storage medium is provided, and the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed by a processor, the computer-executable instructions are used to implement the first aspect and / or various possible implementation manners of the first aspect.

[0040] In a fifth aspect, a computer program product is provided, comprising a computer program. When the computer program is executed by a processor, the computer program implements the first aspect and / or various possible implementation manners of the first aspect.

[0041] The battery detection method, device, equipment, medium and product provided by the embodiments of the present application can obtain the monitoring data of different monitoring areas on the battery to be tested within a preset time when the battery to be tested is tested, and determine the characteristic parameters representing the change of temperature, pressure, voltage and thickness of different monitoring areas on the battery to be tested according to the monitoring data. The quality category label of the battery to be tested is determined by inputting the characteristic parameters of different monitoring areas into the identification model. Since the identification model is obtained by training and learning the quality category label of the sample battery and the sample characteristic parameters of the corresponding monitoring area, the deep correlation between the distribution characteristics of the monitoring data and the battery quality can be automatically mined, so that different types of performance defects and potential risks can be effectively identified, thereby improving the accuracy of the battery quality determination. BRIEF DESCRIPTION OF DRAWINGS

[0042] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0043] Figure 1 A scene schematic diagram of the battery detection method provided by the present application is provided.

[0044] Figure 1a A schematic diagram of a test instrument applied to the battery detection method provided by the embodiments of the present application is provided.

[0045] Figure 2 A flowchart of the battery detection method provided by the present application is provided. Figure 1

[0046] Figure 3 A flowchart of the battery detection method provided by the present application is provided.​Figure 2 ;

[0047] Figure 3a A schematic diagram of a comparison curve of voltage and temperature of a battery provided by an embodiment of the present application changing with time in different test states;

[0048] Figure 4 A structural schematic diagram of a battery detection device provided by the present application;

[0049] Figure 5 A structural schematic diagram of an electronic device provided by the present application.

[0050] The specific embodiments of the present application have been shown by the above-described drawings, and will be described in more detail hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of the present application by any means, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0051] The exemplary embodiments will be described in detail herein with reference to the attached drawings. In the following description, the same numbers are used to indicate the same or similar components. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the present application as detailed in the appended claims.

[0052] In the prior art, the battery temperature is monitored by using a single-point temperature sensor, which can only obtain the local temperature information of the battery and cannot comprehensively reflect the temperature distribution of the battery, making it difficult to capture potential problems such as local overheating of the battery. For the monitoring of the battery pressure, a single-point pressure detection is also used, which cannot accurately grasp the pressure distribution and force value of the whole battery, cannot effectively evaluate the structural changes and safety of the battery during the charging and discharging process, and cannot perform regional division to test abnormal data and perform failure analysis.

[0053] In terms of battery good or bad determination, the traditional method mainly relies on manual judgment based on limited test data, which is not only inefficient but also subjective, and the accuracy of the determination result is difficult to guarantee. At the same time, the existing system lacks autonomous learning ability and cannot optimize the determination standard as the test data accumulates, making it difficult to adapt to the detection needs of different types and different batches of batteries.

[0054] The battery detection method, device, equipment, medium and product provided by the embodiments of the present application can obtain the monitoring data of different monitoring areas on the battery to be detected within a preset time when the battery to be detected is tested, determine the characteristic parameters representing the change of temperature and / or pressure on the different monitoring areas of the battery to be detected according to the monitoring data, input the characteristic parameters of the different monitoring areas into the identification model, and thus determine the quality category of the battery to be detected, so as to solve the problems in the prior art, such as incomplete temperature monitoring, single pressure monitoring, low battery determination efficiency, strong subjectivity, lack of autonomous learning ability, and the like.

[0055] Figure 1 The scene schematic diagram of the battery detection method provided by the present application is shown in FIG. 1. Figure 1 As shown in FIG. 1, the specific application scenario of the present application is a battery detection system, wherein the system comprises a pressure array sensor, an infrared thermal imager and a processor. The battery to be detected can be placed in a clamp, the pressure array sensor is placed on the surface of the battery to be detected, and is used to collect the surface pressure distribution of the battery to be detected in real time. The infrared thermal imager is located below the surface of the battery to be detected and is aligned with the battery, and is used to obtain the surface temperature distribution of the battery to be detected. The pressure array sensor and the infrared thermal imager can determine the monitoring data of different monitoring areas on the battery to be detected within a preset time.

[0056] In order to improve the infrared imaging quality, the lower pressing plate can preferably be made of a material with high infrared transmittance, low thermal conductivity and high compressive strength, such as infrared transparent quartz glass or transparent ceramic. At the same time, the infrared thermal imager can adopt a long-wave infrared thermal imager. This wave band has better penetration ability for non-metal materials (such as battery aluminum plastic film and plastic clamp), and is less disturbed by environmental light, thereby ensuring the accuracy and stability of the monitoring. In addition, in order to obtain voltage monitoring data and thickness monitoring data, the system can further comprise an integrated voltage acquisition sensor and a thickness measuring device. The voltage acquisition sensor is electrically connected to the positive and negative electrodes of the battery to be detected, and records the voltage change of the battery to be detected in real time during the test process. The thickness measuring device can adopt a high-precision displacement sensor or a laser range finder, and non-contact monitoring of the thickness change of the battery caused by expansion or deformation during charging and discharging or thermal runaway.

[0057] After obtaining the monitoring data of different monitoring areas on the battery to be detected within a preset time, the processor can determine the characteristic parameters representing the change of the corresponding monitoring data on the monitoring area with time according to the monitoring data of different monitoring areas on the battery to be detected within a preset time, and input the characteristic parameters representing the change of the corresponding monitoring data on the monitoring area with time into the identification model in the processor, so as to obtain the quality category of the battery to be detected.

[0058] In some embodiments, the processor can be a server, which can be a mobile phone, a computer, a tablet computer or the like.

[0059] Figure 1a The schematic diagram of the test instrument applied to the battery detection method provided by the embodiment of the application is shown in Figure 1a As shown in the figure, the battery (soft package battery cell to be tested) is placed in a thermostat to ensure stable temperature of the test environment; the upper pressing plate is driven by the lifting motor to move up and down, so as to apply controllable pressure to the battery cell to be tested below; the pressure is collected in real time by the pressure array sensor installed between the upper and lower pressing plates; the lower pressing plate is made of transparent ceramic material, so that the infrared imager can penetrate from the bottom to observe the temperature distribution of the surface of the battery cell; at the same time, the infrared range finder is arranged on the side to measure the thickness change of the battery cell in a non-contact manner; the infrared imager is located at the lower part to obtain thermal imaging data of the battery cell during charging and discharging or under pressure. Thus, the key parameters such as pressure, thickness and temperature of the battery cell under constant temperature conditions can be monitored in high precision and multi-dimension synchronously.

[0060] The technical solutions of the application and how the technical solutions of the application solve the above technical problems will be described in detail in specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the application will be described in detail below with reference to the drawings.

[0061] Figure 2 The flowchart of the battery detection method provided by the application is shown in Figure 1 As shown in the figure, the method comprises: Figure 2

[0062] S201, determining the monitoring data of different monitoring areas on the battery to be tested within a preset time, the monitoring data comprising one or more of temperature monitoring data, pressure monitoring data, voltage monitoring data and thickness monitoring data.

[0063] The monitoring area on the battery to be tested can refer to a plurality of spatial units divided on the surface of the battery according to a preset rule or physical characteristics, which are used to collect and analyze temperature or pressure and other state parameters in different areas. The monitoring area can be a grid unit uniformly divided based on geometry (for example, the surface of the battery is divided into MxN rectangular areas), or a functional area defined according to the structural characteristics of the battery (for example, the tab area, the center area, the edge area, etc.).

[0064] ​In the embodiments of the present application, since battery detection has the characteristics of large batch and high automation, in order to realize efficient and unified data collection and analysis, a grid cell based on geometric uniform division is generally used as a monitoring area. That is, the surface of the battery to be detected is divided into a plurality of rectangular or square regions of equal size and regular arrangement, and each grid corresponds to an independent monitoring unit. In this way, it is not dependent on the specific model or structural details of the battery, has good universality and scalability, and is convenient for standardized deployment in large-scale, multi-batch battery detection production lines.

[0065] The monitoring data can include one or more of temperature monitoring data, pressure monitoring data, voltage monitoring data, and thickness monitoring data. The temperature monitoring data can refer to the temperature time series data of the surface or key positions of each monitoring area of the battery to be detected collected by an infrared thermal imager or a distributed temperature sensor within a preset time period. The pressure monitoring data can refer to the dynamic change data of the pressure of each monitoring area on the surface of the battery collected by a flexible pressure array sensor or a force-sensitive element within a preset time period. The voltage monitoring data refers to the change of the terminal voltage between the positive and negative electrodes of the battery to be detected recorded by the voltage acquisition module in real time within the same preset time period, reflecting the electrochemical state and possible internal short circuit, polarization anomaly and other behaviors. It should be noted that the voltage monitoring data of different monitoring areas on the battery to be detected within the preset time can be the same. The thickness monitoring data refers to the dynamic change information of the overall or local area thickness of the battery to be detected obtained by a high-precision displacement sensor, a laser range finder or an optical thickness measuring device within a preset time period. These data reflect the electrode material expansion and contraction behavior caused by lithium ion intercalation / deintercalation when the battery to be detected is subjected to charge and discharge cycles.

[0066] In the embodiments of the present application, the surface temperature field and pressure field data of the battery to be detected within a preset time can be synchronously collected by an infrared thermal imager and a pressure array sensor. The infrared thermal imager can continuously take thermal images of the battery surface at a certain frame rate (such as 5s / time), and the pressure array sensor can record the force change of each sensing unit in real time. Then, the surface of the battery is divided into a plurality of monitoring areas, and the corresponding temperature or pressure values of each area in the time sequence are extracted respectively, so as to obtain the monitoring data of different monitoring areas on the battery to be detected within a preset time.

[0067] As for the voltage monitoring data and the thickness monitoring data, the voltage monitoring data can be continuously recorded by a high-precision voltage acquisition sensor at a set sampling frequency (such as 1s / time) to form a voltage time sequence. Then, the time sequence is aligned with the preset time period, and the monitoring data of the voltage change of the battery with time during the entire test process can be obtained.

[0068] The thickness monitoring data can be continuously monitored by a non-contact thickness measuring device (such as a laser displacement sensor or an optical range finder) at a fixed sampling interval (such as 2 s / time) to monitor the size change of different monitoring areas on the battery to be tested in the vertical direction; the collected thickness values are arranged in chronological order, and synchronized with the preset time period, so that the time series data of the thickness evolution of the battery during the test period can be obtained.

[0069] In S202, a feature parameter representing the change of the corresponding monitoring data on the monitoring area over time is determined according to the monitoring data of different monitoring areas on the battery to be tested within a preset time.

[0070] The feature parameter can refer to a key indicator extracted from the temperature time series data and / or pressure time series data of each monitoring area within a preset time, which can quantify the dynamic change rule. These feature parameters include but are not limited to: maximum value, minimum value, average value, change range (range), temperature / pressure rising rate, temperature / pressure falling slope, temperature / pressure difference standard deviation, cumulative change amount (such as area integral), number of inflection points and occurrence time, etc.

[0071] In the embodiments of the present application, after determining the monitoring data of different monitoring areas on the battery to be tested within a preset time, the temperature time series data and / or pressure time series data of each monitoring area can be traversed by programming and the preset mathematical calculation and logical judgment rules can be applied for feature extraction. For example, the maximum value, minimum value, mean value, standard deviation, etc. in the time series data are obtained to obtain the basic distribution characteristics, and the change rate is obtained by using numerical differentiation or sliding window slope calculation.

[0072] In the embodiments of the present application, the feature parameter can be data representing the change amount, change rate, and uniformity. Wherein:

[0073] When the feature parameter can represent the change amount, the feature parameter can be the monitoring data at the current time minus the monitoring data at the last time; for example, the temperature change amount, pressure change amount, total temperature rise amplitude, and pressure growth amount of each monitoring area within a preset time, to reflect the overall strength of the thermal and force response of the battery during the working condition process.

[0074] When the feature parameter can represent the change rate, the feature parameter can be the change rate of the monitoring data at the current time relative to the monitoring data at the last time; for example, the maximum temperature rising rate, average temperature rising rate, pressure rising rate, and inverse of the time required for the temperature / pressure to reach the peak value of each area, or the local slope at the key stage (such as the end of charging), which describes the dynamic speed of the battery thermal and force response, and helps to identify risks such as abnormal heating or rapid expansion.

[0075] When the characteristic parameter can be used to represent uniformity, the characteristic parameter can be the maximum temperature difference, the maximum pressure difference, the temperature / pressure standard deviation, the range, and the coefficient of variation between each monitoring area at the same time. These parameters reflect the consistency of the distribution of the battery surface state to determine structural defects such as local overheating and swelling eccentricity.

[0076] In S203, the characteristic parameter is input into the identification model to obtain the quality category label of the battery under test. The identification model is trained according to the quality category label of the sample battery and the sample characteristic parameter corresponding to the different quality category labels. The sample characteristic parameter represents the change of the monitoring sample data corresponding to different monitoring areas on the sample battery over time.

[0077] The identification model can refer to a classification model that can determine the quality category of the battery under test according to the characteristic parameters on different monitoring areas on the battery under test. The classification model can be constructed based on machine learning or deep learning algorithms, that is, by learning the quality category label of a large number of sample batteries and the sample characteristic parameters corresponding to different quality category labels, the internal relationship between temperature and pressure change characteristics and battery health status is automatically mined.

[0078] In the embodiments of the present application, the identification model can include support vector machine (SVM), random forest (RandomForest), gradient boosting tree (GBDT), and neural network (such as fully connected network or graph neural network) models.

[0079] The quality category can refer to multiple types classified according to battery performance, safety status, and manufacturing defects, such as good products, qualified products, and non-defective categories. The quality category can be marked by artificial after testing.

[0080] In the embodiments of the present application, the characteristic parameters extracted from each monitoring area to represent the time sequence change of temperature or pressure are standardized and organized into a feature vector or a feature matrix according to a predetermined structure as input data into the trained identification model. The model automatically analyzes the combination mode of the input features based on the nonlinear mapping relationship between the learned features and the quality categories through internal weight calculation and classification decision mechanism, determines which known quality category it best fits, and finally outputs the determination result of the battery under test, realizing end-to-end intelligent identification from multi-region dynamic features to the overall quality category of the battery.

[0081] The battery detection method provided by the embodiments of the present application can fully capture the local thermal changes of the battery in the running process by dividing a plurality of monitoring areas on the surface of the battery to be detected, combining infrared thermal imaging and a pressure array sensor to collect high spatiotemporal resolution temperature field and pressure field data, automatically extracting the time sequence characteristics of the monitoring data of each area and constructing a multi-dimensional feature vector, and using an intelligent recognition model trained based on a large number of samples to perform fusion analysis and pattern recognition on the features, thereby effectively mining the internal correlation between temperature, pressure distribution abnormalities and battery quality, overcoming the limitations of one-sided traditional single-point monitoring information and strong subjectivity of manual discrimination, improving the sensitivity and classification accuracy of battery defect recognition, and thus realizing high-precision, automated and intelligent detection of battery quality.

[0082] In the embodiments of the present application, when the monitoring data is temperature monitoring data, the monitoring data of different monitoring areas on the battery to be detected at different times is determined, including:

[0083] According to the preset charging and discharging conditions, thermal imaging data of the monitoring surface of the battery to be detected at different time points within a preset time is obtained; the charging and discharging conditions include the charging and discharging rate and voltage range of the battery to be detected.

[0084] According to the preset area division rule and the thermal imaging data, the temperature monitoring data of different monitoring areas on the battery to be detected at different times is determined.

[0085] The charging and discharging conditions can refer to the combination of electrical load parameters and environmental control parameters applied to the battery during the test, which can include the size of the charging current, the size of the discharging current, the charging and discharging rate, the upper and lower limits of the voltage, the charging and discharging mode, the number of cycles, the temperature environment and the standing time, etc. These conditions can simulate the working state of the battery to be detected in actual use, directly affect the internal electrochemical reaction rate, heat generation behavior and structural stress change, and then affect the temperature and pressure distribution characteristics.

[0086] The monitoring surface of the battery to be detected can refer to the surface area of the battery used for installing sensors or non-contact measurement during the test, which can select the outer surface that can effectively reflect the internal state change of the battery and is convenient for data collection, such as the positive side, the negative side or the large-area flat shell surface of the battery.

[0087] The thermal imaging data can refer to the two-dimensional visual image sequence of the temperature distribution on the surface of the battery to be detected collected by the infrared thermal imager, each frame of image contains the mapping relationship of the pixel points and the corresponding temperature value, and can form complete thermal field information with spatial and temporal dimensions.

[0088] The area division rule can refer to a logic or method for dividing the battery monitoring surface to be tested into multiple monitoring areas, which can be set according to the test target, the battery structure and the sensor characteristics. In the embodiments of the present application, the monitoring surface can be equally divided into several sub-areas according to geometric shapes such as rectangles or hexagons, which is suitable for automatic batch detection and facilitates standardized feature extraction.

[0089] The temperature value of the monitoring area can be determined according to the average value of the temperature corresponding to each pixel point.

[0090] In some embodiments, the temperature monitoring data of the monitoring area can also be determined by calculating the heat energy contribution of each pixel in the area, combined with spatial continuity verification and heat conduction characteristic analysis, in a heat flux density weighted manner to determine the temperature value representing the monitoring area. The process can be:

[0091] 1. Heat flux density map generation;

[0092] Convert the thermal imaging data of one monitoring area into a heat flux density map, and consider the temperature value of each pixel point and its thermal influence range on the surrounding area, wherein a high temperature point is given a greater thermal influence weight.

[0093] For example: The thermal imaging data of the monitoring area can be a thermal imaging data set with a resolution of 1024x1024 pixels. Each pixel point has a corresponding temperature value. Among them, for a hot spot of 65℃, the thermal influence radius is set to 5 pixels; for a temperature zone of 45℃, the thermal influence radius is set to 2 pixels; for a low temperature zone of 35℃, the thermal influence radius is set to 1 pixel. Thus, using a Gaussian function, the original temperature data is converted into a heat flux density map, so that each high temperature point has a corresponding influence on its surrounding pixels, forming a "thermal halo".

[0094] 2. Heat energy contribution calculation;

[0095] Based on the heat flux density map, the heat energy contribution of each pixel to the entire area is calculated, considering both temperature value and spatial position.

[0096] For example: After obtaining the heat flux density map, the heat energy contribution of each pixel to the entire area is calculated. Among them, the center hot spot (65℃) covers 15 pixels, contributing 42% of the total heat energy of the contribution area; the main working area (45℃) occupies 120 pixels, contributing 48%; the edge area (35℃) contains 50 pixels, contributing the remaining 10%. Thus, by analyzing the temperature of each pixel point and its effective influence area, the contribution proportion of each pixel to the overall heat energy can be calculated.

[0097] 3. Heat conduction path analysis;

[0098] Analyze the heat conduction path in the monitoring area, identify the main heat source and heat dissipation path, and ensure that the representative temperature reflects the true thermal state rather than measurement anomalies.

[0099] For example: heat radiates outward from the 65℃ hotspot in the center, indicating that this is the main heat source; the 70℃ single point at the edge is considered a measurement noise rather than an actual heat source due to the lack of adjacent points.

[0100] 4. Thermal energy weighted temperature calculation;

[0101] Based on the thermal energy contribution of each region in the monitoring area, calculate the weighted representative temperature, and give higher weight to high temperature and high density regions.

[0102] For example: based on the thermal energy contribution of each region calculated above, calculate the weighted representative temperature. The formula satisfies: representative temperature = (65℃ x 42%) + (45℃ x 48%) + (35℃ x 10%) = 51.8℃.

[0103] 5. Thermal state consistency verification;

[0104] Verify whether the calculated representative temperature is consistent with the overall thermal state of the monitoring area, and output the temperature value if it is consistent.

[0105] For example: compare the representative temperature calculated in 4 with the original thermal image. Check if the representative temperature corresponds to the color of most areas in the image (i.e. temperature interval). If the main color of the image corresponds to the temperature range of 50-55℃, then 51.8℃ is the temperature value of the monitoring area.

[0106] If it is not consistent, the temperature value of the monitoring area can be determined according to the average value of the temperature corresponding to each pixel point.

[0107] In the embodiments of the present application, when the monitoring data is pressure monitoring data, the monitoring data of different monitoring areas on the battery to be tested at different times is determined, including:

[0108] According to the preset pressure monitoring condition, the pressure sensor data of the monitoring surface of the battery to be tested at different time points within the preset time is obtained;

[0109] According to the preset region division rule and the pressure sensor data, the pressure monitoring data of different monitoring areas on the battery to be tested at different times is determined.

[0110] The pressure monitoring condition can refer to a sum of physical environments and measurement parameters set in the process of collecting pressure data of the battery, which can include a mounting mode of the sensor (such as a fixed pressure plate, a flexible array adhesion or a clamp integration), a pre-tightening force applied, a contact surface material and flatness, a sampling frequency, a test temperature environment and a constraint state of the battery in the charging and discharging process (such as free expansion or limited expansion).

[0111] The pressure sensor data can refer to pressure change information of each contact point or area on the surface of the battery collected in real time by the flexible pressure array sensor or the distributed force sensitive element during the charging and discharging process of the battery, which is represented as a multi-channel pressure numerical set with time sequence characteristics. These data reflect the mechanical deformation behavior of the battery caused by the expansion, contraction and cyclic stress accumulation of the electrode material, and can include pressure data of each monitoring position.

[0112] In the embodiments of the present application, the method further includes:

[0113] According to the quality category of the battery to be tested and the monitoring data of different monitoring areas on the battery to be tested within a preset time, the identification model is updated and trained to obtain the updated and trained identification model.

[0114] In the embodiments of the present application, the method further includes:

[0115] In this way, by continuously introducing new data in actual detection to update and train the identification model, the generalization ability and adaptability of the identification model can be enhanced, and new defects or edge cases not covered in traditional samples can be effectively identified. At the same time, the identification threshold and weight distribution of the identification model to the temperature and pressure characteristics are closer to the real working conditions, reducing misjudgment and omission and improving the classification accuracy. Through self-optimization and knowledge accumulation of the identification model, the development of battery quality determination from static rules to dynamic evolution is realized, and the accuracy of monitoring the battery to be tested is improved.

[0116] In the embodiments of the present application, after the feature parameter representing the change of the monitoring data on the monitoring area with time is input into the identification model to obtain the quality category of the battery to be tested, the method further includes:

[0117] According to the monitoring data of different monitoring areas on the battery to be tested within a preset time, a change curve of the monitoring area is generated, and a global change curve representing all monitoring areas on the battery to be tested is generated.

[0118] The quality category of the battery to be tested, the change curve of the monitoring area, and the global change curve representing all monitoring areas on the battery to be tested are displayed.

[0119] The change curve can refer to a trend curve reflecting the dynamic evolution of the state of the monitoring area within a preset time, which is plotted with time as the horizontal axis and monitoring data (such as temperature or pressure) as the vertical axis. For example, for each grid area such as the tab area and the center area divided on the surface of the battery, the time series data of the temperature or pressure of the area is plotted into an independent curve, which can intuitively show the temperature rising / cooling process, the pressure rising rate, the peak occurrence time, and the recovery condition of the area.

[0120] The global change curve can refer to a summary curve representing the overall state evolution trend of the battery to be tested, which is generated by aggregating the data of all monitoring areas. In the embodiments of the present application, there are multiple monitoring areas on the battery to be tested, and the change curves corresponding to each monitoring area all have a common time axis. Therefore, by statistically aggregating the data of all monitoring areas at the same time, a value representing the overall state of the battery at that time can be obtained, and then these values are connected in time sequence to form the global change curve.

[0121] The change curve and the global change curve can effectively monitor the performance change in the whole cycle by continuously recording and analyzing the data of different monitoring areas of the battery to be tested during the monitoring process. The change curve of each monitoring area shows the parameter change trend of the position over time, and the global change curve is calculated based on the data of all monitoring areas and is used to reflect the overall state evolution of the battery. When the battery to be tested has an abnormality, such as a sudden temperature rise or voltage drop in a monitoring area, the corresponding change curve will deviate significantly from the normal trend. By comparing the change curves of the monitoring areas and the global change curve, the precise position and cycle period of the abnormality can be locked. Further, combined with the abnormal data and its background, such as specific operating conditions or time periods, a clear direction can be provided for subsequent process improvement to optimize the battery design or production process and prevent similar failures from occurring again.

[0122] For example, if the temperature of the tab area of a batch of batteries rises sharply at the end of the 50th cycle and is accompanied by a sudden increase in local pressure, the area can be locked as the starting point of failure through the change curve. Further investigation found that the loose welding of the tab led to an increase in internal resistance, resulting in overheating and local expansion during high-current charging. For another example, if the global pressure curve shows that the center area of the battery continues to swell during high-temperature cycle testing, while the pressure of the edge area is released with a lag, it indicates that the packaging structure design is unreasonable or the glue material is not uniformly solidified, causing stress concentration. For another example, if the bottom area of multiple batteries has a low temperature and a sharp voltage fluctuation during low-temperature discharge, combined with the cycle conditions, it can be determined that the bottom of the module is too strong, leading to lithium precipitation risk. Based on these locatable and traceable abnormal patterns, the welding process can be optimized, the packaging structure can be improved, or the thermal management strategy can be adjusted, thereby improving product consistency and long-term reliability.

[0123] In some embodiments, the system can also achieve automatic judgment and alarm of the battery monitoring process by analyzing the change curve and the global change curve. For example, when the system is running, the change curve and the global change curve of each monitoring area are collected in real time, and key features are automatically extracted, such as whether the local temperature rise rate exceeds the threshold, whether the pressure of a certain area increases abnormally, whether the standard deviation of the temperature difference or pressure difference between regions continues to increase, etc. The system has a multi-level decision rule (e.g., single-point temperature > 50°C for 10 seconds, or pressure non-uniformity index rises by more than 30%). Once the monitoring data triggers the threshold, the system will mark the abnormality and start the alarm process.

[0124] At the same time, the current feature pattern is classified in real time by combining the trained recognition model. If the recognition model determines that it is a poor quality category, the system will automatically pop up an alarm message, mark the specific area, cycle number, and possible causes of the abnormality, and respond through sound and light prompts, message pushing, or automatically pausing the test equipment. In this way, the whole process automation from data collection, feature analysis, abnormality recognition to alarm execution is realized, significantly improving the detection efficiency and safety.

[0125] Figure 3 Flowchart of the battery detection method provided in the present application Figure 2 As shown in Figure 3 the embodiment, the battery detection method is described in detail based on the Figure 2 embodiment. The method comprises:

[0126] S301, obtaining monitoring sample data of different monitoring areas on a sample battery within a preset time;

[0127] S302, marking the monitoring sample data when the obtained monitoring sample data does not meet the preset data requirements;

[0128] S303, in response to the user's decision operation on the marked monitoring sample data, determining the quality category label of the sample battery;

[0129] S304, when the monitoring sample data corresponding to each quality category label is greater than the preset sample data amount, training the initial model according to the monitoring sample data and the quality category label of the sample battery to obtain the recognition model.

[0130] The preset data requirement can refer to a discrimination rule of the monitoring sample data preset according to the battery type and process characteristics, which is used for preliminary screening and classification of the sample data.

[0131] The preset sample data amount can refer to the minimum sample number of the monitoring sample data to ensure model training. In the embodiment of the present application, the sample number corresponding to each quality category label needs to be greater than a preset threshold. For example, when there are 4 quality category labels, the sample number corresponding to each quality category label is greater than 200, and thus the entire sample number needs to be at least greater than 800.

[0132] In the test process of the sample battery, the temperature field and pressure field data of the sample battery under standard charge and discharge conditions are synchronously collected by the infrared thermal imager and the pressure array sensor, and the battery monitoring surface is divided into multiple monitoring areas according to a preset area division rule (such as uniform grid or functional partition). For each monitoring area, the temperature time series data and / or pressure time series data thereof within a preset time (such as one complete cycle or an aging test period) are extracted to form a structured monitoring sample data set.

[0133] For the sample data marked as abnormal or critical state by the system, the engineers perform rejudgment in the human-computer interaction interface. The operator can view the change curve of each monitoring area, the global change curve and the thermal map / pressure distribution map of the battery, and comprehensively evaluate combined with the actual test performance (such as whether to smoke, swell, voltage drop, etc.), and manually confirm the final quality category label (such as “good product”, “qualified product”, “inferior product” or specific defect type such as “local overheating”, “uneven swelling” etc.).

[0134] When the number of labeled samples accumulated by a certain quality category label reaches a preset threshold (such as no less than 200 groups), it indicates that the data of this category has statistical representativeness and training feasibility. At this time, the model training process is started: first, the feature parameters of each monitoring area are extracted from all qualified samples, such as the maximum value, mean value, rate of change, spatial variance, and inter-regional difference of temperature / pressure, and normalization processing is performed; then the feature vectors and the corresponding quality category labels constitute a training set, which is input into the initial model (such as random forest, XGBoost or neural network) for supervised learning. By continuously optimizing the model parameters, the model learns the thermal-power distribution patterns under different quality states, and finally generates an identification model that can automatically identify the quality category of new batteries. As more labeled data accumulates, the model can also be updated regularly to continuously improve the discrimination accuracy and adaptability, realizing the evolution from empirical rules to intelligent identification.

[0135] In the embodiments of the present application, the initial model is trained according to the monitoring sample data and the quality category label of the sample battery to obtain the identification model, which includes:

[0136] According to the monitoring sample data of different monitoring areas on the sample battery within a preset time, a feature sample parameter representing the change of the corresponding monitoring sample data on the monitoring area over time is determined;

[0137] The feature sample parameter is normalized to obtain a normalized sample parameter;

[0138] The initial model is trained according to the normalized sample parameter and the quality category label of the corresponding sample battery to obtain an initial identification model; the quality category label includes a quality category label representing a normal sample battery and a quality category label representing an abnormal sample battery;

[0139] A thermal runaway label is selected from the quality category label representing the abnormal sample battery;

[0140] The initial identification model is trained according to the thermal runaway label in the quality category label and the normalized sample parameter corresponding to the thermal runaway label to obtain the identification model.

[0141] In the model training process, the quality category label can be divided into a quality category label representing a normal sample battery and a quality category label representing an abnormal sample battery. The initial model is trained by the quality category label representing the normal sample battery and the normalized sample parameter corresponding thereto, and the quality category label representing the abnormal sample battery and the normalized sample parameter corresponding thereto, to obtain an initial identification model. After obtaining the initial identification model, a thermal runaway label representing thermal runaway of the sample battery is selected from the quality category label, and the initial identification model is trained according to the label and the corresponding normalized sample parameter to obtain the identification model.

[0142] In some embodiments, after obtaining the monitoring sample data of each monitoring area on the sample battery, feature extraction can be performed, that is, key indicators capable of representing the dynamic change rule are calculated from the time series data of each monitoring area.

[0143] Among them, the monitoring area on the sample battery can be divided into three large areas: the tab area, the edge area, and the core functional area; wherein:

[0144] The tab area is the area connected with the tab;

[0145] The edge area is the part of the four surrounding areas on the sample battery except the tab area;

[0146] The core functional area is the middle area surrounded by the tab area and the edge area.

[0147] Among them, the failure mode of the tab area is:

[0148] The stress repeatedly acts on the virtual welding or welding itself, and the battery capacity is abnormal, the internal circuit is broken, and the battery is failed.

[0149] The failure mode of the edge area is:

[0150] Sealing failure: improper heat sealing process parameters or defects in the aluminum plastic film itself, air and moisture enter the battery, causing electrolyte hydrolysis and electrode oxidation; if the electrolyte is a sulfide system, it may also leak toxic gas, while the battery capacity rapidly decays;

[0151] Structural deformation: more sensitive to external mechanical impact and extrusion, easy to displace the electrode; internal circuit is broken or short-circuited, the battery is completely failed, and serious accidents may occur.

[0152] The failure mode of the core functional area is:

[0153] Interface impedance surge: electrode material volume expansion / contraction, leading to interface peeling, interface side reaction, causing battery internal resistance to rise sharply, charge-discharge rate performance to deteriorate, capacity to rapidly decay, and even "capacity cliff drop";

[0154] Lithium dendrite growth: uneven interface contact forms locally, low temperature or large current charging, lithium ion transport rate is lower than deposition rate, accelerating lithium dendrite precipitation through solid-state electrolyte, triggering internal short circuit of positive and negative electrodes, leading to local overheating of the battery, soft package bulging, and serious heat runaway.

[0155] Therefore, the model can be divided according to the existing test data or experience, such as dividing the battery into normal type and abnormal type (i.e. quality class label representing normal sample battery and quality class label representing abnormal sample battery).

[0156] The normal range of the pressure P of all test points of the normal battery under constant spacing is (P1, P2), the normal range of the temperature T of all test points during normal cycle is (T1, T2), the thickness d under constant pressure changes in the range of (d1, d2) during the cycle, and the charge and discharge voltage values are all smooth curves of slow rise or fall. The batteries within the range can be determined as normal batteries, and any one of the four conditions can be determined as an abnormal battery.

[0157] The abnormal battery is a battery whose surface pressure of any point exceeds (P1, P2) under constant spacing test conditions, or the surface temperature of any point exceeds (T1, T2), or the thickness exceeds (d1, d2) under constant pressure, or the voltage appears a pressure drop in any cycle during the charge and discharge process, whether it is recovered later or not.

[0158] The abnormal battery can accurately lock the abnormal area according to the abnormal pressure or temperature or thickness or voltage, and thus analyze the failure mode.

[0159] In order to further screen abnormal batteries, thermal runaway is generally considered as a serious battery failure mode,

[0160] Here, thermal runaway can be preset as one of the abnormal batteries. The main characteristics of the thermal runaway battery are that the temperature change rate is greater than or equal to 1℃ / s, lasts for more than 3s, and even the battery smokes, burns, etc. At the same time, it will also be accompanied by a large pressure drop and a large thickness change rate.

[0161] When the thermal runaway battery appears in the mode, the pressure, temperature, voltage or thickness data can be extracted, normalized, and the pressure change rate, pressure change value, temperature change rate, temperature change value, voltage change rate, thickness change rate or thickness change value of each monitoring point of the battery from the start of the test to the failure can be calculated. The data of multiple batches of thermal runaway batteries are sorted, and the preset thermal runaway battery mode is enriched and improved.

[0162] When the mode is further improved, the test battery can be pre-judged according to the real-time pressure, temperature, voltage or thickness data test, so as to avoid the occurrence of thermal runaway phenomenon, so as to avoid causing damage to the equipment, environmental hazards and economic losses.

[0163] Figure 3a The schematic diagram of the voltage and temperature change curves of the battery provided by the embodiment of the application under different test states over time is shown in the following figure. Figure 3aAs shown, the left side represents a normal test, where the voltage remains stable and the temperature fluctuates slightly but remains around 28°C, indicating that the battery is operating smoothly. The middle side represents thermal runaway, where the voltage drops sharply to 0 after about 35 seconds, while the temperature rises rapidly to over 600°C, accompanied by intense heat release, which is a typical characteristic of thermal runaway. The right side represents an abnormal test that can continue, where the voltage drops briefly and then recovers, and the temperature rises slightly before stabilizing, indicating a local abnormality (such as an internal short circuit or poor contact) but without causing serious failure, and the test can still continue.

[0164] Specifically, if a battery has defects, a short circuit will occur at the defect location during the compression process, causing a voltage drop and a temperature increase. Furthermore, different types of short circuits produce different voltage and temperature changes. For example, in some defective batteries, the voltage drops directly to near 0V and the temperature rises linearly after a short circuit; while in other defective batteries, the voltage drops first, then rebounds, then drops again, and the temperature rises linearly twice.

[0165] Therefore, the battery detection method provided in this application embodiment can:

[0166] Comprehensive monitoring: Through a high-precision infrared thermal imager and pressure array sensor, the temperature and pressure distribution of the battery can be monitored comprehensively and accurately, overcoming the limitations of traditional single-point monitoring.

[0167] Full process recording: Temperature and pressure changes can be accumulated and stored. After the test is completed, the complete temperature and pressure change process can be viewed, providing detailed evidence for analyzing changes in battery performance.

[0168] Self-optimization: Possesses self-learning capabilities, continuously optimizing judgment criteria as test data accumulates, improving the accuracy and adaptability of judgments, and meeting the testing needs of different types and batches of batteries.

[0169] Automatic determination: It realizes the automatic determination of battery quality, reduces manual intervention, improves determination efficiency, and reduces the influence of subjectivity on the determination results.

[0170] Figure 4 This is a schematic diagram of the battery testing device provided in this application, as shown below. Figure 4 As shown, the battery detection device 40 provided in this embodiment includes:

[0171] The determination module 401 is used to determine the monitoring data of different monitoring areas on the battery under test within a preset time. The monitoring data includes one or more of the following: temperature monitoring data, pressure monitoring data, voltage monitoring data, and thickness monitoring data.

[0172] The extraction module 402 is configured to determine a characteristic parameter representing a change of the corresponding monitoring data on the monitoring area over time according to the monitoring data of different monitoring areas on the battery to be tested within a preset time.

[0173] The identification module 403 is configured to input the characteristic parameter into an identification model to obtain a quality category label of the battery to be tested.

[0174] In a possible implementation, when the monitoring data is temperature monitoring data, the determination module 401 can be specifically configured to:

[0175] According to a preset charging and discharging condition, obtain thermal imaging data of a monitoring surface of the battery to be tested at different time points within a preset time; the charging and discharging condition includes a charging and discharging rate and a voltage range of the battery to be tested.

[0176] According to a preset area division rule and the thermal imaging data, determine temperature monitoring data of different monitoring areas on the battery to be tested at different times.

[0177] In a possible implementation, when the monitoring data is pressure monitoring data, the determination module 401 can be specifically configured to:

[0178] According to a preset pressure monitoring condition, obtain pressure sensor data of the monitoring surface of the battery to be tested at different time points within a preset time.

[0179] According to a preset area division rule and the pressure sensor data, determine pressure monitoring data of different monitoring areas on the battery to be tested at different times.

[0180] In a possible implementation, the identification module 403 can be specifically configured to:

[0181] Obtain monitoring sample data of different monitoring areas on the sample battery within a preset time.

[0182] When the obtained monitoring sample data does not meet a preset data requirement, mark the monitoring sample data.

[0183] In response to a judgment operation of a user on the marked monitoring sample data, determine a quality category label of the sample battery.

[0184] When the monitoring sample data corresponding to each quality category label is greater than a preset sample data amount, train an initial model according to the monitoring sample data and the quality category label of the sample battery to obtain the identification model.

[0185] In a possible implementation, the identification module 403 can be further specific to:

[0186] According to the monitoring sample data of different monitoring areas on the sample battery within a preset time, a characteristic sample parameter is determined, which represents a change of the corresponding monitoring sample data on the monitoring area over time.

[0187] The characteristic sample parameter is normalized to obtain a normalized sample parameter.

[0188] According to the normalized sample parameter and the quality category label of the corresponding sample battery, the initial model is trained to obtain an initial identification model; the quality category label includes a quality category label representing a normal sample battery and a quality category label representing an abnormal sample battery.

[0189] The thermal runaway label is selected from the quality category label representing the abnormal sample battery.

[0190] According to the thermal runaway label in the quality category label and the normalized sample parameter corresponding to the thermal runaway label, the initial identification model is trained to obtain an identification model.

[0191] In a possible implementation, the identification module 403 can be further specific to:

[0192] According to the quality category of the battery to be tested and the monitoring data of different monitoring areas on the battery to be tested within a preset time, the identification model is updated and trained to obtain an updated and trained identification model.

[0193] In a possible implementation, the identification module 403 can be further specific to:

[0194] According to the monitoring data of different monitoring areas on the battery to be tested within a preset time, a change curve of the monitoring area and a global change curve representing all monitoring areas on the battery to be tested are generated.

[0195] The quality category of the battery to be tested, the change curve of the monitoring area and the global change curve representing all monitoring areas on the battery to be tested are displayed.

[0196] The battery detection device provided in this embodiment can execute the method provided in the method embodiment, and has similar implementation principles and technical effects. Details are not described herein.

[0197] Figure 5 The structure of the electronic device provided in this application is shown in FIG. 1. Figure 5 As shown in FIG. 1, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, the memory 502 and the communication component 503 are connected through a bus 504.

[0198] In the implementation process, the at least one processor 501 executes the computer execution instructions stored in the memory 502, so that the at least one processor 501 executes the above-mentioned method.

[0199] The specific implementation process of the processor 501 can refer to the method embodiments described above, which have similar implementation principles and technical effects, and will not be described here in detail.

[0200] In the above embodiments, it should be understood that the processor can be a central processing unit (English: Central Processing Unit, CPU for short), and can also be other general-purpose processors, digital signal processors (English: Digital Signal Processor, DSP for short), application specific integrated circuits (English: Application Specific Integrated Circuit, ASIC for short), etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the application can be directly embodied as hardware processor execution completion, or executed by hardware and software modules in the processor.

[0201] The memory can contain a random access memory (RAM), and can also include a non-volatile memory (NVM), such as at least one disk memory.

[0202] The bus can be an industry standard architecture (ISA) bus, a peripheral component (PCI) bus, or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, the bus in the drawings of the present application does not limit only one bus or one type of bus.

[0203] The present application also provides a computer program product, comprising a computer program, which is executed by the processor to realize the above-mentioned method.

[0204] The present application also provides a computer readable storage medium, which stores computer execution instructions, and when the processor executes the computer execution instructions, the above-mentioned method is realized.

[0205] The above-mentioned readable storage medium can be realized by any type of volatile or nonvolatile storage devices or their combinations, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0206] An exemplary readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in the device.

[0207] The division of units is only a logical functional division, and in actual implementation, there can be another division manner, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0208] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment scheme.

[0209] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.

[0210] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the method of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0211] It can be understood by those skilled in the art that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned 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 various media that can store program codes.

[0212] Finally, it should be noted that: those skilled in the art will easily think of other embodiments of the present application after considering the specification and practicing the application disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application that follow the general principles of the present application and include common knowledge or conventional technical means in the art that are not disclosed in the present application, and is not limited to the precise structure described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present application is only limited by the appended claims.

Claims

1. A battery testing method, characterized in that, include: The monitoring data of different monitoring areas on the battery under test are determined within a preset time period. The monitoring data includes one or more of the following: temperature monitoring data, pressure monitoring data, voltage monitoring data, and thickness monitoring data. Based on the monitoring data of different monitoring areas on the battery under test within a preset time period, characteristic parameters characterizing the changes of the corresponding monitoring data on the monitoring areas over time are determined; The feature parameters are input into the recognition model to obtain the quality category label of the battery under test; wherein, the recognition model is trained based on the quality category label of the sample battery and the sample feature parameters corresponding to different quality category labels, and the sample feature parameters characterize the changes of monitoring sample data corresponding to different monitoring areas on the sample battery over time.

2. The method according to claim 1, characterized in that, When the monitoring data is temperature monitoring data, determining the monitoring data of different monitoring areas on the battery under test at different times includes: According to preset charging and discharging conditions, thermal imaging data of the monitoring surface of the battery under test at different time points within a preset time period are obtained; the charging and discharging conditions include the charging and discharging rate and voltage range of the battery under test. Based on the preset area division rules and the thermal imaging data, the temperature monitoring data of different monitoring areas on the battery under test at different times are determined.

3. The method according to claim 1, characterized in that, When the monitoring data is pressure monitoring data, determining the monitoring data of different monitoring areas on the battery under test at different times includes: According to the preset pressure monitoring conditions, the pressure sensor data of the monitoring surface of the battery under test at different time points within a preset time period are obtained. Based on the preset area division rules and the pressure sensor data, the pressure monitoring data of different monitoring areas on the battery under test at different times are determined.

4. The method according to claim 1, characterized in that, Before inputting the feature parameters into the recognition model to obtain the quality category of the battery under test, the method further includes: Acquire monitoring sample data of different monitoring areas on the sample battery within a preset time period; When the acquired monitoring sample data does not meet the preset data requirements, the monitoring sample data is marked. In response to a user's judgment operation on the labeled monitoring sample data, a quality category label for the sample battery is determined; When the monitoring sample data corresponding to each quality category label is greater than the preset sample data volume, the initial model is trained based on the monitoring sample data and the quality category label of the sample battery to obtain the recognition model.

5. The method according to claim 4, characterized in that, The step of training an initial model based on the monitored sample data and the quality category label of the sample battery to obtain an identification model includes: Based on the monitoring sample data of different monitoring areas on the sample battery within a preset time period, characteristic sample parameters that characterize the changes of the corresponding monitoring sample data on the monitoring area over time are determined. The feature sample parameters are normalized to obtain normalized sample parameters; Based on the normalized sample parameters and the corresponding quality category labels of the sample batteries, the initial model is trained to obtain the initial recognition model; the quality category labels include quality category labels that represent the normality of the sample batteries and quality category labels that represent the abnormality of the sample batteries. Thermal runaway labels are selected from the quality category labels that characterize the abnormalities of the sample batteries; The initial recognition model is trained based on the thermal runaway label in the quality category label and the normalized sample parameters corresponding to the thermal runaway label to obtain the recognition model.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: Based on the quality category of the battery under test and the monitoring data of different monitoring areas on the battery under test within a preset time, the recognition model is updated and trained to obtain the updated recognition model.

7. The method according to any one of claims 1 to 5, characterized in that, After inputting the feature parameters characterizing the changes in monitoring data over time in the monitoring area into the recognition model to obtain the quality category of the battery under test, the method further includes: Based on the monitoring data of different monitoring areas on the battery under test within a preset time period, a change curve of the monitoring area and a global change curve representing all monitoring areas on the battery under test are generated. The quality category of the battery under test, the change curve of the monitoring area, and the global change curve representing all monitoring areas on the battery under test are displayed.

8. A battery testing device, characterized in that, include: The determination module is used to determine the monitoring data of different monitoring areas on the battery under test within a preset time. The monitoring data includes one or more of the following: temperature monitoring data, pressure monitoring data, voltage monitoring data, and thickness monitoring data. The extraction module is used to determine feature parameters that characterize the change of monitoring data in different monitoring areas on the battery under test over a preset time, based on the monitoring data in different monitoring areas on the battery under test. The identification module is used to input the feature parameters into the identification model to obtain the quality category label of the battery under test; wherein, the identification model is trained based on the quality category label of the sample battery and the sample feature parameters corresponding to different quality category labels, and the sample feature parameters characterize the changes of monitoring sample data corresponding to different monitoring areas on the sample battery over time.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

11. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-7.

Images