Leakage current testing method and system based on low-frequency EIS and transverse current pulse

Abstract

The invention provides a leakage current testing method and system based on a low-frequency EIS and a transverse current pulse, and relates to the technical field of leakage current testing. The method comprises the following steps: firstly, determining use information and low-frequency EIS test parameters of a target battery to be tested, then testing the battery by using electrochemical working equipment according to the parameters, obtaining impedance data under different frequencies, and further obtaining first dynamic capacitance information. And controlling the power supply equipment to apply a preset transverse pulse current to the battery, determining a voltage variation cut-off condition, continuously monitoring the voltage variation, and recording the test time variation when the cut-off condition is reached. And finally, the leakage current value of the battery is determined by integrating the first dynamic capacitance information, the transverse current pulse current, the test time variation and the voltage variation. Compared with a traditional method, the leakage current value can be obtained only by applying the direct current once after the low-frequency EIS test without changing the magnitude and the duration time of the current for many times, the test process is simplified, the test efficiency is improved, and the leakage current value can be determined more accurately.

Description

Technical Field

[0001] This application relates to the technical field of leakage current testing, and particularly to a leakage current testing method and system based on low-frequency EIS and constant-current pulses. Background Art

[0002] The leakage current of a battery refers to a tiny, continuous current that still exists even when the battery is not powering an external load. This current is usually caused by internal factors such as defects in battery materials, decomposition of the electrolyte, self-discharge mechanisms, or internal micro-shorts. The leakage current gradually consumes the battery over time, reducing its overall capacity and service life. Therefore, monitoring and minimizing the leakage current are crucial for maintaining battery performance and extending its life.

[0003] In traditional battery testing, to obtain certain performance parameters of the battery, such as internal resistance, polarization resistance, etc., current pulse signals of different magnitudes and durations are usually applied, and by measuring the voltage response of the battery under different current pulses and analyzing the relationship between voltage changes and current changes, the relevant performance parameters can be obtained. Additionally, the battery static self-discharge test method is also used, that is, the battery is left stationary at a specific temperature for a period of time, and the leakage current magnitude is indirectly calculated by measuring the battery voltage drop.

[0004] However, traditional methods require changing the current magnitude and duration multiple times, and the testing process is relatively complex and time-consuming. Summary of the Invention

[0005] This application provides a leakage current testing method and system based on low-frequency EIS and constant-current pulses, which are used to efficiently and accurately measure the battery leakage current, and effectively solve the problems of cumbersome testing process, long time consumption, and inaccurate measurement results existing in traditional testing methods.

[0006] In a first aspect, this application provides a leakage current testing method based on low-frequency EIS and constant-current pulses, which is applied to a leakage current testing system. The method includes: determining the battery usage information of the target battery to be tested; determining the test parameters of the low-frequency EIS according to the battery usage information; controlling the electrochemical working device to apply a test to the target battery to be tested according to the test parameters, and obtaining the impedance data of the target battery to be tested at different frequencies; obtaining the first dynamic capacitance information according to the imaginary part of the impedance corresponding to the impedance data and the test parameters; controlling the power supply device to apply a constant-current pulse current of a preset magnitude to the target battery to be tested, and determining the voltage change amount cut-off condition; continuously monitoring the voltage change amount of the target battery to be tested, and recording the test time change amount when the voltage change amount reaches the voltage change amount cut-off condition; determining the leakage current value according to the first dynamic capacitance information, the constant-current pulse current, the test time change amount, and the voltage change amount.

[0007] By adopting the above technical solution, the battery usage information of the target battery to be tested is first determined, which provides basic data for subsequent tests. According to the battery usage information, the low-frequency EIS test parameters are determined to make the test more targeted. The electrochemical working device is used to test according to the parameters to obtain impedance data, and the first dynamic capacitance information is obtained by combining the imaginary part of the impedance and the test parameters. Then, a constant current pulse is applied and the voltage change amount and time change amount are monitored. Finally, the leakage current value is determined by synthesizing this information. The entire process combines the low-frequency EIS and constant current pulse technologies. Compared with the traditional method, it is not necessary to change the current magnitude and duration multiple times. Only a direct current needs to be applied once after the low-frequency EIS test to obtain the leakage current value, which simplifies the test process, improves the test efficiency, and can more accurately determine the leakage current value, providing a more reliable basis for battery performance evaluation.

[0008] Combined with some embodiments of the first aspect, in some embodiments, the step of obtaining the first dynamic capacitance information according to the imaginary part of the impedance corresponding to the impedance data and the test parameters specifically includes: determining the first dynamic capacitance information DNC1, DNC1 = - , is the frequency in the test parameters of the low-frequency EIS, is the imaginary part of the impedance of the target battery to be tested, represents the impedance of the target battery to be tested at a frequency of f.

[0009] By adopting the above technical solution, based on the impedance characteristics of the battery at different frequencies, key information is directly extracted from the impedance data obtained by the test to calculate the dynamic capacitance, which can quickly and accurately reflect the dynamic capacitance characteristics of the battery, provides an important intermediate parameter for subsequent calculation of the leakage current, makes the leakage current calculation more accurate and scientific, and helps to accurately evaluate the battery performance.

[0010] Combined with some embodiments of the first aspect, in some embodiments, after the step of determining the first dynamic capacitance information DNC1, it further includes: determining the second dynamic capacitance information DNC2, DNC2 = , where is the current value applied for the first time during the test of applying the constant current pulse, is the current value applied for the second time during the test of applying the constant current pulse, is the voltage change amount across the battery corresponding to , is the voltage change amount across the battery corresponding to , is the time interval experienced by the voltage change process corresponding to , is the voltage change amount across the battery corresponding to The time interval experienced during the voltage change process; obtaining the number of battery cycles of the target battery to be tested through the battery management system, and obtaining the current test environment temperature through the temperature sensor; inputting the number of battery cycles and the previous test environment temperature into the dynamic capacitance weight fusion model to determine the first weight of the first dynamic capacitance information DNC1 and the second weight of the second dynamic capacitance information DNC2 under the current test environment. The dynamic capacitance weight fusion model is obtained by prior machine learning training based on battery sample data under different numbers of battery cycles and different test environment temperatures. The battery sample data includes DNC1, DNC2, and weight combination labels with cycle number and temperature conditions; adjusting the first dynamic capacitance information DNC1 according to the first dynamic capacitance information DNC1 and the first weight, and the second dynamic capacitance information DNC2 and the second weight.

[0011] By adopting the above technical solution, the number of battery cycles and the test environment temperature will affect the battery performance. The weight fusion model is trained based on a large amount of sample data and can reflect the relationship between these factors and the dynamic capacitance weight. Adjusting the first dynamic capacitance information according to the weight can more accurately consider the characteristics of the battery under different usage states and environments, make the first dynamic capacitance information more in line with the actual situation, and then improve the accuracy of subsequent leakage current calculation, providing strong support for more accurate assessment of the battery leakage condition.

[0012] Combined with some embodiments of the first aspect, in some embodiments, the step of determining the leakage current value according to the first dynamic capacitance information, the constant current pulse current, the test time change amount, and the voltage change amount specifically includes: determining the leakage information according to the leakage current formula , and the leakage current formula is =DNC1× , where DNC1 is the first dynamic capacitance information, is the voltage change amount corresponding to the time change amount when the voltage change amount reaches the voltage change amount cut-off condition, is the voltage change amount matching the voltage change amount cut-off condition at both ends of the battery during the test process, and I is the pulse current value applied during the test process.

[0013] By adopting the above technical solution, integrating and calculating the previously obtained key information, based on physical principles and mathematical relationships, directly obtaining the leakage current value. Its calculation method is scientific and reasonable, making full use of the data obtained from the previous tests, and can intuitively and accurately reflect the leakage current situation of the battery, providing a quantitative basis for judging whether the battery is leaking and the degree of leakage.

[0014] In some embodiments in combination with some embodiments of the first aspect, in the step of determining the test parameters of the low-frequency EIS according to the battery usage information, it specifically includes: determining the environmental data of the current low-frequency EIS test; inputting the battery usage information and the environmental data into a preset frequency determination model to obtain a matching low-frequency range. The battery usage information includes the battery type. The construction and training process of the frequency determination model is as follows: obtaining the actual operation data of multiple different types of batteries under different environmental conditions, where the actual operation data includes impedance data under different frequency EIS tests; using the battery type as a classification label, and obtaining the frequency determination model through machine learning in combination with the actual operation data.

[0015] By adopting the above technical solution, first determine the environmental data of the current low-frequency EIS test, and then input the battery usage information and the environmental data into a preset frequency determination model. The frequency determination model is trained based on the actual operation data of different types of batteries under different environments, with the battery type as the classification label. This method combines the battery type and the test environment factors, and can find the most suitable low-frequency test range according to the actual situation. The appropriate test frequency can improve the accuracy of the test results, avoid test data deviation caused by improper frequency selection, ensure that the impedance data obtained subsequently can better reflect the true characteristics of the battery, and lay a good foundation for accurately calculating the dynamic capacitance and leakage current.

[0016] In some embodiments in combination with some embodiments of the first aspect, before the step of controlling the electrochemical workstation to apply a test to the target battery to be tested according to the test parameters, it further includes: obtaining the temperature curve data of the target battery to be tested before the test; inputting the temperature curve data into a pre-trained temperature feature recognition model to obtain a temperature recognition output result. The temperature feature recognition model is trained through machine learning using the temperature change data of multiple normal batteries and leakage batteries; according to the temperature recognition output result, adjusting the test parameters of the low-frequency EIS in a dynamic adjustment manner. The dynamic adjustment manner includes when it is recognized that the temperature curve data has a leakage feature, reducing the test frequency according to a set value to improve the impedance measurement accuracy; when it is recognized that the temperature curve data does not have a leakage feature, increasing the test frequency according to a set value to shorten the test time.

[0017] By adopting the above technical solution, the temperature curve data of the target battery to be tested before the test is obtained and input into a pre-trained temperature feature recognition model. This model is trained with the temperature change data of normal batteries and leaking batteries and can identify whether the temperature curve has leakage characteristics. According to the recognition result, the low-frequency EIS test parameters are dynamically adjusted. If there are leakage characteristics, the test frequency is reduced to improve the impedance measurement accuracy; if not, the test frequency is increased to shorten the test time. In this way, it not only ensures more accurate test data can be obtained in case of possible leakage, but also improves the test efficiency under normal conditions, realizes the intelligentization and optimization of the test process, and improves the performance of the entire leakage current test system.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of determining the leakage current value according to the first dynamic capacitance information, the cross-flow pulse current, the test time variation amount and the voltage variation amount, the method further includes: determining the leaking battery, and inputting the remaining capacity, health state and usage environment data of the obtained leaking battery into a pre-constructed intelligent leakage assessment model for risk quantification assessment to obtain a risk assessment result. The intelligent leakage assessment model is established in advance through machine learning training based on battery leakage data and their corresponding risk level labels under multiple groups of different remaining capacities, different health states and different usage environment conditions; prioritizing the batteries according to the risk level based on the risk assessment result and the magnitude of the leakage current value to obtain a sorting result; attaching an electronic label containing information on the leakage degree and risk level with different color markings to the battery according to the sorting result; and sending the electronic label to the visual end in the order of the sorting result.

[0019] By adopting the above technical solution, after determining the leaking battery, its remaining capacity, health state and usage environment data are input into the intelligent leakage assessment model. This model is established through training with a large amount of battery leakage data under different conditions. The model outputs a risk assessment result, combines the leakage current value to sort the batteries according to the risk level, manufactures and sends an electronic label with information on the leakage degree and risk level to the visual end. This process realizes the quantitative assessment and intuitive display of the risk of leaking batteries, facilitates the staff to quickly understand the battery leakage risk situation, and promptly process high-risk batteries, such as repair or replacement, to ensure the safety and reliability of battery use and improve the efficiency and quality of battery management.

[0020] In a second aspect, the present application provides a leakage current test system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code includes computer instructions. The one or more processors call the computer instructions to enable the leakage current test system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] In a third aspect, the present application provides a computer-readable storage medium including instructions that, when running on a leakage current test system, cause the leakage current test system to execute the method described in the first aspect and any possible implementation manner of the first aspect.

[0022] In a fourth aspect, the present application provides a computer program product that, when running on a leakage current test system, causes the leakage current test system to execute the method described in the first aspect and any possible implementation manner of the first aspect.

[0023] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Since technical means of determining low-frequency EIS test parameters by combining battery usage information of a target battery to be tested, applying a constant current pulse, and monitoring relevant parameters, and finally comprehensively determining the leakage current value are adopted, the technical problems of cumbersome leakage current testing process, low efficiency, and insufficient accuracy in the prior art are effectively solved. Furthermore, the technical effects of simplifying the test process, improving the test efficiency, and more accurately determining the leakage current value, and providing a reliable basis for battery performance evaluation are achieved.

[0024] 2. Since technical means of determining second dynamic capacitance information, combining the number of battery cycles and the test environment temperature, and adjusting the first dynamic capacitance information using a dynamic capacitance weight fusion model are adopted, the technical problem in the prior art that the influence of battery usage status and environmental factors on dynamic capacitance is not fully considered, resulting in inaccurate leakage current calculation, is effectively solved. Furthermore, the technical effects of making the first dynamic capacitance information more in line with the actual situation, improving the accuracy of subsequent leakage current calculation, and providing strong support for accurately evaluating the battery leakage condition are achieved.

[0025] 3. Since technical means of obtaining the temperature curve data of a target battery to be tested before testing, inputting it into a temperature feature recognition model, and dynamically adjusting the low-frequency EIS test parameters according to the recognition result are adopted, the technical problem in the prior art that the low-frequency EIS test frequency is fixed and cannot balance test accuracy and efficiency is effectively solved. Furthermore, the technical effects of improving the impedance measurement accuracy in the case of possible leakage, shortening the test time under normal conditions, optimizing the test process, and improving the performance of the leakage current test system are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flowchart of a leakage current test method based on low-frequency EIS and constant current pulse in an embodiment of the present application; Figure 2 is another flowchart of a leakage current test method based on low-frequency EIS and constant current pulse in an embodiment of the present application; Figure 3It is a schematic structural diagram of an entity device in the leakage current test system according to an embodiment of the present application. Detailed implementation manners

[0027] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. As used in the specification and appended claims of the present application, the singular forms "a", "an", "the", "above-mentioned", "this" and "this one" are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term " / and" used in the present application refers to and includes any or all possible combinations of one or more of the listed items.

[0028] Hereinafter, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0029] For ease of understanding, the method provided in this embodiment is described in terms of a process below. Please refer to Figure 1 , which is a schematic flowchart of a leakage current test method based on low-frequency EIS and cross-flow pulse in an embodiment of the present application.

[0030] S101. Determine the battery usage information of the target battery to be tested; The leakage current test system will first collect the battery usage information of the target battery to be tested through various channels. The system will perform data interaction with the battery management system (BMS). The battery management system is responsible for monitoring and managing various states of the battery, which contains rich battery usage information. The battery usage information may include the charge and discharge times of the battery and the battery type. For example, in the battery management scenario of an electric vehicle, the BMS monitors various state information of the power battery in real time. The leakage current test system uses the CAN protocol to periodically send data request instructions to the BMS. After receiving the instructions, the BMS will package and send the stored battery charge and discharge times data to the test system.

[0031] For the usage information of the battery type, the leakage current test system will determine the battery type by reading the identification code of the battery. During the production process, each battery is given a unique identification code, which contains key information such as the type and production batch of the battery.

[0032] S102. Determine the test parameters of the low-frequency EIS according to the battery usage information; When determining the test parameters of low-frequency EIS (electrochemical impedance spectroscopy), the leakage current test system needs to accurately set various factors to ensure the accuracy and effectiveness of subsequent tests. The system first determines the environmental data of the current low-frequency EIS test by setting the temperature sensor and air pressure sensor within the battery setting range in the test experiment. The environmental data can include the temperature data and air pressure data during the test. The reasons are as follows: The chemical reaction rate inside the battery is closely related to the temperature. When the temperature rises, the chemical reaction rate accelerates. This is because the temperature increase increases the thermal motion energy of the molecules, making it easier for the reactant molecules to overcome the activation energy of the reaction, thereby accelerating the reaction. Taking lithium-ion batteries as an example, the increase in temperature will accelerate the insertion and deinsertion rate of lithium ions in the positive and negative electrode materials, resulting in a decrease in the internal resistance of the battery, an increase in the self-discharge rate, and an increase in the leakage current. On the contrary, a decrease in temperature will slow down the chemical reaction rate, increase the internal resistance of the battery, and reduce the leakage current. If the temperature factor is not considered during the test, the leakage current test results obtained will not truly reflect the actual use of the battery, and may overestimate or underestimate the leakage degree of the battery, affecting the accurate evaluation of the battery performance. For the electrode materials of the battery, their crystal structure and electronic conductivity will change at different temperatures. In a high temperature environment, the crystal structure of the electrode material may be distorted to a certain extent, resulting in changes in the diffusion path and diffusion rate of lithium ions, which in turn affects the charge and discharge performance and leakage current of the battery. In low-frequency EIS testing, temperature will affect the performance of the test equipment and the accuracy of the test data. Temperature changes may cause the performance of the electronic components of the test equipment to drift, causing deviations in the impedance data obtained from the test. If the temperature factor is not taken into account, the first dynamic capacitance information calculated based on these inaccurate impedance data and the final leakage current value will also have errors.

[0033] In addition, air pressure will change the partial pressure of gas inside the battery. In some batteries that use gas to participate in the reaction, such as metal-air batteries, oxygen enters the battery through the air to participate in the chemical reaction. When the air pressure changes, the partial pressure of oxygen changes, affecting the rate at which oxygen diffuses to the electrode surface. At high altitudes, the air pressure is lower, the oxygen diffusion rate decreases, and the reaction rate of the battery decreases accordingly, which causes the impedance characteristics of the battery to change. Considering air pressure when determining low-frequency EIS test parameters can more accurately reflect the internal reaction of the battery under different air pressure environments and obtain more realistic impedance data.

[0034] The system inputs the battery usage information and environmental data into a preset frequency determination model to obtain the matching low-frequency range in the test parameters. The construction and training process of this frequency determination model is based on a large amount of actual operation data. The system retrieves the actual operation data of multiple different types of batteries under different environmental conditions from the database. These data cover common battery types such as lead-acid batteries, lithium-ion batteries, nickel-metal hydride batteries, etc., as well as the situations under different temperatures (-20°C - 60°C), humidities (20%RH - 80%RH), and air pressures (90 kPa - 110 kPa). The actual operation data contains impedance data under different frequency EIS tests. The frequency range is set from 0.01 Hz to 100 Hz, and the impedance data is tested and recorded at a step of 0.01 Hz.

[0035] Using the battery type as the classification label and combining with the actual operation data, the system uses a machine learning algorithm for model training. During the training process, the support vector machine (SVM) algorithm is used to classify and perform regression analysis on the data. Taking the battery type as the classification target, the environmental data (temperature, humidity, air pressure) and the impedance data at different frequencies are input into the model as feature vectors. After multiple iterative trainings and adjusting the model parameters, the model can accurately predict the most suitable low-frequency test range according to the input battery usage information (battery type) and environmental data.

[0036] Suppose the battery usage information obtained by the system shows that the target battery to be tested is a certain model of lithium-ion battery, the current environmental temperature is 28°C, the humidity is 55%RH, and the air pressure is 102 kPa. After inputting this information into the frequency determination model, the output low-frequency test range of the model is 0.1 Hz - 1 Hz. The system determines the test frequency parameters of the low-frequency EIS according to this result and selects 0.1 Hz, 0.3 Hz, 0.5 Hz, 0.7 Hz, 0.9 Hz as the specific test frequency points. At the same time, the amplitude of the AC excitation signal is determined to be 5 mV to ensure accurate impedance data is obtained without affecting the internal electrochemical process of the battery.

[0037] It should be noted that for the test voltage of the test parameters of the low-frequency EIS, 1 mV - 10 mV can be selected to avoid non-linearity. For the test frequency range, 10 mHz - 10 kHz can be selected to extract the dynamic capacitance in the low-frequency band. For the AC amplitude selected for the test, 1 - 10 mV can be selected to avoid activating side reactions.

[0038] S103. Control the electrochemical working equipment to apply the test to the target battery to be tested according to the test parameters, and obtain the impedance data of the target battery to be tested at different frequencies; The leakage current test system transmits the determined low-frequency EIS test parameters to the electrochemical working device by establishing a communication connection with it. After receiving the test parameters, the electrochemical working device generates corresponding test signals based on these parameters. Taking a common electrochemical workstation as an example, the signal generator inside it generates a sinusoidal AC excitation signal with a frequency range of 0.1 Hz - 1 Hz (assuming the low-frequency test range obtained according to the frequency determination model), and the signal amplitude is 5 mV. This AC excitation signal is applied to the target battery under test through a three-electrode system composed of a working electrode, a reference electrode, and a counter electrode. During the signal application process, the electrochemical working device monitors the output of the signal in real time to ensure the stability and accuracy of the signal. If signal fluctuations or abnormalities occur, the device will automatically adjust or report an error, notifying the leakage current test system to re-check the test parameters or device connections.

[0039] To obtain the impedance data of the target battery under test at different frequencies, the electrochemical working device will conduct tests sequentially according to the set test frequency points. Each time a new test frequency is switched to, the device will wait for a period of time to allow the electrochemical process inside the battery to reach a stable state. The setting of this waiting time is based on the characteristics of the battery and relevant electrochemical theories, generally ranging from several seconds to dozens of seconds. For example, for a lithium-ion battery, during low-frequency testing, it may be necessary to wait for 10 - 20 seconds to ensure the stability of the ion diffusion and charge transfer processes inside the battery and avoid obtaining inaccurate impedance data due to premature testing.

[0040] After stabilizing at each test frequency point, the electrochemical working device measures the current response and voltage response of the battery at that frequency. Using the measured voltage and current data, and applying Ohm's law and complex number operations, the impedance of the battery is calculated. The specific calculation process is as follows: First, record the voltage amplitude V across the battery and the current amplitude I passing through the battery under the action of the AC excitation signal, as well as the phase difference . Then, according to the definition of impedance Z = V / I, calculate the amplitude of the impedance |Z|. At the same time, using the concept of complex numbers, represent the impedance as Z = |Z| ( ), where j is the imaginary unit. In this way, the impedance data of the battery at that frequency is obtained, including the real part and the imaginary part of the impedance. During the entire test process, the electrochemical working device transmits the measured impedance data to the leakage current test system in real time. After receiving the impedance data, the leakage current test system performs preliminary filtering and denoising on the data to remove abnormal data caused by measurement noise or other interference factors. Common filtering methods include mean filtering, median filtering, etc. By statistically analyzing multiple measurement data, data points that deviate significantly from the normal range are removed to improve the quality and reliability of the data.

[0041] In some embodiments, before the step S103, the leakage current test system can also obtain the temperature curve data of the target battery under test before the test through a built-in or external high-precision temperature sensor array. These temperature sensors are closely attached to different positions of the battery, such as both ends of the positive and negative electrodes, the center and edges of the battery housing, etc., to ensure that the temperature changes of all parts of the battery can be comprehensively captured. Taking the power battery pack of an electric vehicle as an example, the battery pack is composed of multiple battery cells connected in series or in parallel. The system will install temperature sensors at key positions of each battery cell to monitor the temperature distribution of the entire battery pack. The system samples the temperature data at fixed time intervals, and the sampling frequency can be flexibly adjusted according to the characteristics of the battery and the test requirements. For batteries with relatively rapid temperature changes, such as lithium-ion batteries during high-rate charge and discharge, the sampling frequency can be set to 1 Hz, that is, the temperature data is collected once per second; for batteries with relatively slow temperature changes, the sampling frequency can be reduced to 0.1 Hz. During the sampling process, the system will record the temperature data of each sampling point and the corresponding timestamp in real time to form temperature-time series data. In actual tests, assuming that before the leakage current test of a mobile phone battery, the system starts collecting temperature data from time 0, and within the next 5 minutes, 300 temperature data points are collected at a sampling frequency of 1 Hz. These data points constitute the initial temperature curve data of the battery before the test. The leakage current test system extracts the temperature curve data of the target battery under test stored locally and preprocesses it according to the format required by the temperature feature recognition model. The preprocessing process includes steps such as data normalization, denoising, and feature extraction. After preprocessing, the system inputs the temperature curve data into a pre-trained temperature feature recognition model, which is obtained through machine learning training using the temperature change data of a large number of normal batteries and leaking batteries. Taking the convolutional neural network (CNN) in deep learning as an example, during the training process, temperature curve data of thousands of different types of batteries (including lithium-ion batteries, lead-acid batteries, nickel-metal hydride batteries, etc.) are collected, half of which are data of normal batteries and the other half are data of leaking batteries. These data are divided into a training set, a validation set, and a test set. The training set is used for model training, the validation set is used to adjust the hyperparameters of the model, and the test set is used to evaluate the performance of the model. During the training process of the model, through components such as convolutional layers, pooling layers, and fully connected layers, it automatically learns the feature patterns in the temperature curve data. The convolutional layer is used to extract the local features of the data, the pooling layer reduces the dimension of the features to reduce the computational amount, and the fully connected layer comprehensively analyzes the extracted features and outputs the classification results. After multiple iterations of training, the model gradually learns the differential features between the temperature curves of normal batteries and leaking batteries, thus having the ability to accurately identify whether the temperature curve has leakage characteristics.

[0042] After the system inputs the temperature curve data of the target battery to be tested into the model, the model outputs a temperature recognition result after calculation and analysis. For example, for a certain temperature curve data, the result output by the model is "with leakage characteristics", which indicates that the temperature change pattern of this battery is similar to the temperature curve characteristics of the leaking batteries in the training set, and the possibility of leakage is relatively high; if the model outputs "without leakage characteristics", it means that the temperature curve of this battery conforms to the characteristics of a normal battery. The system will store the temperature recognition output result locally and display it on the operation interface in real time so that the tester can timely understand the temperature status of the battery.

[0043] After the leakage current test system obtains the output result of the temperature feature recognition model, it adjusts the test parameters of the low-frequency EIS in a dynamically adjusted manner according to this result. When it is recognized that the temperature curve data has leakage characteristics, the system will reduce the test frequency according to the set value to improve the impedance measurement accuracy. This is because in the case of leakage, the electrochemical process inside the battery becomes more complex, and the impedance characteristics will also change. A lower test frequency can detect the impedance information inside the battery more deeply. Taking a certain type of lithium-ion battery as an example, the set low-frequency EIS test frequency range is 0.1Hz - 1Hz under normal conditions. When the temperature feature recognition model determines that the temperature curve of this battery has leakage characteristics, the system adjusts the test frequency range to 0.01Hz - 0.1Hz. During the process of adjusting the frequency, the system will synchronously adjust other relevant test parameters, such as the amplitude of the AC excitation signal. To ensure that no additional damage is caused to the battery during the test process and to ensure that the impedance can be accurately measured. After reducing the test frequency, the system will correspondingly extend the test time for each test frequency point. Because the time required for the electrochemical process inside the battery to reach a stable state is longer during low-frequency testing, in order to obtain accurate impedance data, it is necessary to wait long enough. For example, when testing at normal frequency, the test time for each test frequency point is set to 10 seconds, while after reducing the frequency, the test time may be extended to 30 seconds. The system will monitor the impedance response of the battery at different frequencies in real time to ensure the accuracy and reliability of the test data.

[0044] When it is recognized that the temperature curve data does not have leakage characteristics, the system will increase the test frequency according to the set value to shorten the test time. In this case, since the battery is in a normal state, a higher test frequency can obtain sufficient impedance information in a shorter time, improving the test efficiency. This not only ensures more accurate test data in the case of possible leakage but also improves the test efficiency under normal conditions, realizing the intelligence and optimization of the test process and improving the performance of the entire leakage current test system.

[0045] S104. Obtain the first dynamic capacitance information according to the imaginary part of the impedance corresponding to the impedance data and the test parameters; After receiving impedance data at different frequencies transmitted by the electrochemical working device, the leakage current test system extracts the imaginary part information of the impedance from these data. The system correlates and organizes the imaginary part data of the impedance corresponding to each frequency point according to the previously set test frequency parameters. For example, if the test frequency points are 0.1Hz, 0.3Hz, 0.5Hz, 0.7Hz, 0.9Hz, the system will extract the imaginary part values of the battery impedance at these frequencies respectively. According to the formula DNC1 = - , for each test frequency point, the system will substitute the frequency value f and the corresponding imaginary part value of the impedance into the formula for calculation.

[0046] S105. Control the power supply device to apply a preset constant current pulse current to the target battery under test, and determine the cut-off condition of the voltage change amount; The leakage current test system is connected to the power supply device through a control interface, sends instructions to the power supply device to make it apply a preset constant current pulse current to the target battery under test. Before sending the instructions, the system checks the size of the preset constant current pulse current to ensure that it meets the test requirements and the safe operating range of the battery. For example, for a certain type of lithium-ion battery, according to its specification sheet and test experience, the size of the preset constant current pulse current may be set to 100mA. After receiving the instructions from the system, the power supply device generates a stable constant current pulse current signal. The current control circuit inside the power supply device will precisely control the size of the current and the width of the pulse through a feedback adjustment mechanism. For example, using a PID (Proportional-Integral-Derivative) control algorithm, it monitors the size of the output current in real time, compares it with the preset value, and adjusts the current output according to the deviation to ensure that the stability of the constant current pulse current is within ±1%. During the generation of the constant current pulse current, the power supply device monitors its own working state, such as parameters like temperature and voltage. If an abnormal situation occurs, it will send an alarm message to the leakage current test system in time to prevent damage to the battery.

[0047] While applying the constant current pulse current, the leakage current test system needs to determine the cut-off condition of the voltage change amount. The determination of this cut-off condition is based on the characteristics of the battery and the test purpose. First, the system refers to the battery specification sheet and relevant industry standards to obtain the voltage change range of the battery under normal working conditions. In this application, the voltage change amount needs to be greater than 0.5mV (signal-to-noise ratio) and less than 10mV (linear region). The cut-off condition of the voltage change amount in this application can be 2mv. In order to monitor the voltage change amount and the voltage change rate in real time, the leakage current test system will sample the voltage across the battery in real time through a data acquisition module. The sampling frequency of the data acquisition module can be set according to the test requirements.

[0048] In some embodiments, to more accurately determine the constant-current pulse current applied to the target battery under test, a current determination model can be constructed using a machine learning-based method. Specifically, the system first collects a large amount of battery test data. The response data of different types of batteries (such as lithium-ion batteries, lead-acid batteries, nickel-metal hydride batteries, etc.) under different current stimuli are collected. These response data include various parameters such as voltage changes, internal resistance changes, and temperature changes across the battery terminals. At the same time, each set of data is labeled with the corresponding battery type tag to distinguish different battery types. Machine learning algorithms are used to train the collected battery response data with different type tags. Taking the neural network algorithm as an example, the collected battery response data is used as the input layer data, and the battery type tag is used as the output layer data. During the training process, the neural network continuously adjusts the internal weights and biases. Through multiple iterations, the predicted battery type results output by the network are made to match the actual tags as closely as possible. After training with a large amount of data, the neural network gradually learns the characteristic patterns of the response data of different types of batteries under different current stimuli, thus constructing a current determination model. This model can understand the relationship between the characteristics of different types of batteries and the response to current stimuli. When a leakage current test needs to be performed on the target battery under test, the type of the target battery under test is input into the trained current determination model. The model outputs the constant-current pulse current value suitable for this type of battery according to the characteristic patterns learned previously. For example, if the target battery under test is a certain model of lithium-ion battery, the current determination model will give a suitable constant-current pulse current value, such as 80 mA, based on the characteristics of lithium-ion batteries it has learned. This current value is obtained by comprehensively considering the characteristics of this type of battery and the optimal stimulation current situation in previous test data, and is more scientific and accurate than the traditional empirically set current value. S106. Continuously monitor the voltage change amount of the target battery under test, and record the test time change amount when the voltage change amount reaches the voltage change amount cut-off condition. The leakage current test system performs high-frequency real-time sampling of the voltage across the target battery under test through the data acquisition module. The sampling frequency is set according to the battery characteristics and test accuracy requirements. The system analyzes the collected voltage data in real time, calculates the voltage difference between adjacent sampling points, and thus obtains the voltage change amount. When it is monitored that the voltage change amount gradually approaches the preset voltage change amount cut-off condition, the system starts the high-precision timing module. The timing module continuously records the time until the voltage change amount reaches the cut-off condition. At this time, the obtained time is the test time change amount.

[0049] S107. Determine the leakage current value according to the first dynamic capacitance information, the constant-current pulse current, the test time change amount, and the voltage change amount.

[0050] Specifically, according to the leakage current formula, =DNC1× To determine the leakage current information , where DNC1 = - , f is the frequency in the low-frequency EIS test parameters, represents the impedance of the target battery under test at the frequency f, is the imaginary part of the impedance of the target battery under test. During the test, the electrochemical working device applies tests to the battery according to the set parameters to obtain impedance data at different frequencies. The system extracts the imaginary part of the impedance from these data, substitutes it into the formula to calculate DNC1, which reflects the dynamic capacitance characteristics of the battery and provides important parameters for leakage current calculation. Among them is the voltage change amount is the time change amount corresponding to reaching the cut-off condition of the voltage change amount, is the voltage change amount matching the cut-off condition of the voltage change amount at both ends of the battery during the test, and I is the pulsed current value applied during the test.

[0051] In the embodiments of the present application, by determining the battery usage information of the target battery under test, the low-frequency EIS test parameters are determined specifically, the electrochemical working device and the power supply device are controlled to perform tests and obtain relevant data, and the leakage current value is determined by integrating the first dynamic capacitance information, the constant-current pulsed current, the test time change amount, and the voltage change amount. This not only simplifies the test process, avoids the complex operations of changing the current magnitude and duration multiple times in the traditional method, improves the test efficiency, but also fully considers the battery characteristics and test environment factors, can more accurately determine the leakage current value, and provides a reliable basis for battery performance evaluation.

[0052] In some embodiments, after determining that there is a battery leakage situation, a series of operations will be carried out to evaluate the leakage risk, classify and manage the batteries, and visually present them to the operator. Specifically, the first step is to identify the leaking batteries, which is determined by the previous leakage current test method based on the set leakage current threshold. When the detected leakage current of the battery exceeds the threshold, it is identified as a leaking battery. For these leaking batteries, it is necessary to collect their remaining capacity, health status, and usage environment data. The remaining capacity can be obtained through the battery management system, the health status can be comprehensively judged according to various parameters such as the number of charge and discharge cycles and the change of internal resistance of the battery, and the usage environment data includes temperature, humidity, air pressure, etc., which are collected through corresponding sensors. Then, these data are input into a pre-constructed intelligent leakage assessment model. This model is constructed based on a large amount of experimental data and machine learning techniques. In the model training stage, a multi-group of battery leakage data under different remaining capacities, different health statuses, and different usage environment conditions are collected, and each group of data is labeled with a corresponding risk level label. The risk level can be divided into low risk, medium risk, high risk, etc. Using these data, through machine learning algorithms such as decision trees and neural networks, the model is trained. During the training process, the model continuously learns the features and laws in the data and adjusts its own parameters, and finally has the ability to evaluate the leakage risk according to the input battery data. When the relevant data of the leaking battery are input, the model can output the corresponding risk assessment result.

[0053] After obtaining the risk assessment results, the leakage current values obtained from the previous tests are combined to prioritize the risk of the leaking batteries. The basis for the prioritization is a comprehensive consideration of the risk assessment results and the leakage current values. Batteries with a high-risk assessment result and a large leakage current value will be ranked in a higher priority position; while batteries with a low-risk assessment result and a small leakage current value will be ranked in a lower priority position. In this way, all the leaking batteries are sorted in an orderly manner according to the level of risk, forming a sorting result. Such sorting helps to quickly distinguish the risk levels of different batteries, facilitating targeted subsequent processing. According to the sorting result, an electronic label is made for each battery. The electronic label contains information about the degree of leakage, which can be divided into different levels such as mild leakage, moderate leakage, and severe leakage according to the size of the leakage current value; at the same time, it also carries risk level information with different color markings, for example, green represents low risk, yellow represents medium risk, and red represents high risk. Through this intuitive color marking and leakage degree identification, the operator can quickly understand the risk status of the battery. The system sends the made electronic labels to the visual terminal in the order of the sorting result. The visual terminal can be a computer screen, a mobile device display screen, or a dedicated industrial monitoring screen, etc. At the visual terminal, the operator can clearly see the electronic label information of all the leaking batteries, arranged and displayed in order of risk level from high to low. This enables the operator to have a clear understanding of the battery leakage situation at a glance and take measures for high-risk batteries in a timely manner, such as repair, replacement, or further testing, thereby effectively managing the batteries and ensuring the safety and stability of battery use.

[0054] In some embodiments, after confirming the first dynamic capacitance information and before confirming the leakage current value, the first dynamic capacitance information can be corrected in advance. The following details the process description of this aspect. After combining the above content, the following further provides a more specific process description of the method provided in this embodiment. Please refer to Figure 2 , which is another schematic flowchart of the leakage current testing method based on low-frequency EIS and cross-flow pulse in the embodiments of the present application.

[0055] S201. Determine the first dynamic capacitance information DNC1; After the leakage current testing system obtains the impedance data of the target battery to be tested at different frequencies transmitted by the electrochemical working device, it begins to determine the first dynamic capacitance information DNC1. The system will accurately extract the imaginary part information of the impedance from these impedance data and, according to the test frequency parameters, orderly associate and organize the imaginary part data of the impedance corresponding to each frequency point. More specifically, DNC1 = - , is the frequency in the test parameters of low-frequency EIS, is the imaginary part of the impedance of the target battery under test. Taking a certain type of ternary lithium battery as an example, assume the test frequency points are set to 0.05Hz, 0.1Hz, 0.2Hz, 0.5Hz, and 1Hz. The system first targets the frequency point of 0.05Hz and finds the corresponding imaginary part value of the impedance from the corresponding impedance data among them . Then, according to the formula DNC1 = - , substitute the frequency f = 0.05Hz and the imaginary part value of the impedance at this frequency into the formula for calculation

[0056] To ensure the reliability of the calculation, the system will perform multiple verifications on the calculation process. On the one hand, the system will check whether the input frequency value and the imaginary part value of the impedance are accurate and within a reasonable measurement range. If it is found that the imaginary part value of the impedance at a certain frequency point shows abnormal fluctuations, such as being too different from the imaginary part value of the same type of battery under similar test conditions, the system will automatically mark this data point and recheck the measurement process of the electrochemical working equipment at this frequency. This may involve checking the stability of the test signal, the connection between the electrode and the battery, etc. On the other hand, the system will use different calculation methods to verify the results. For example, in addition to directly calculating according to the formula, the system will also use numerical approximation algorithms for approximate calculation. Compare the two calculation results. If the deviation between the two is within the allowable error range (such as 0.5%), the calculation result is considered reliable

[0057] S202. Determine the second dynamic capacitance information DNC2 When the leakage current test system determines the second dynamic capacitance information DNC2, it needs to obtain relevant data during the application of the constant current pulse current test. First, the system will extract the current value applied for the first time and the current value applied for the second time during the application of the constant current pulse current from the test record and . For example, in a test on a lithium iron phosphate battery, the current value applied for the first time is set to 50mA, and the current value applied for the second time is set to 80mA. At the same time, the system will obtain the voltage change amount across the battery within and the voltage change amount across the battery within , where and both match the pre-set voltage change cut-off condition. In this application, the voltage change cut-off condition is 2mv, so and can both be 2mv and are both within the range of the pre-set voltage change cut-off condition. In this application, the voltage change cut-off condition is 2mv, so and can both be 2mv

[0058] Then, through DNC2 = obtain the second dynamic capacitance information, where is the current value applied for the first time during the cross - current pulse current test, is the current value applied for the second time during the cross - current pulse current test, is the corresponding voltage change across the battery within is the corresponding voltage change across the battery within is the corresponding time interval experienced during the voltage change process, is the corresponding time interval experienced during the voltage change process, where is in line with the voltage change amount cut - off condition.

[0059] S203. Obtain the number of battery cycles of the target battery to be tested through the battery management system, and obtain the current test environment temperature through the temperature sensor; The leakage current test system establishes a stable data transmission link with the battery management system (BMS) through a standard communication protocol, such as the Controller Area Network (CAN) protocol. In the electric vehicle battery management scenario, the BMS is responsible for real - time monitoring of key information such as the charge - discharge state and remaining power of the power battery, which includes the number of battery cycles. The system sends a request data packet containing a specific instruction code to the BMS. After receiving the request, the BMS reads the number of battery cycle data from its internal storage unit and encapsulates the data into a response data packet according to the protocol format and sends it back to the leakage current test system.

[0060] In terms of obtaining the current test environment temperature, the leakage current test system uses a high - precision thermistor temperature sensor. The resistance value of this sensor changes precisely with the ambient temperature, and its sensitivity can reach 0.01 °C. The sensor is installed near the target battery to be tested to ensure accurate measurement of the actual temperature of the environment where the battery is located.

[0061] S204. Input the number of battery cycles and the previous test environment temperature into the dynamic capacitance weight fusion model to determine the first weight of the first dynamic capacitance information DNC1 and the second weight of the second dynamic capacitance information DNC2 under the current test environment. The dynamic capacitance weight fusion model is obtained by prior machine learning training based on battery sample data under different numbers of battery cycles and different test environment temperatures. The battery sample data includes DNC1, DNC2, and weight combination labels with cycle number and temperature conditions; After obtaining the number of battery cycles and the current test environment temperature of the target battery to be tested, the leakage current test system organizes these two sets of data into a specific input format and inputs them into the dynamic capacitance weight fusion model. This model is trained through machine learning algorithms based on a large number of battery sample data with different numbers of battery cycles and different test environment temperatures. Taking the support vector machine (SVM) algorithm to train the model as an example, in the training stage, thousands of groups of sample data of different types of batteries (such as lithium-ion batteries, lead-acid batteries, nickel-metal hydride batteries, etc.) under various cycle numbers (from 0 cycles of a new battery to 800 cycles close to the end of the service life) and test environment temperatures (-20°C - 60°C) are collected. Each group of sample data contains the corresponding first dynamic capacitance information DNC1, the second dynamic capacitance information DNC2, and the weight combination label with cycle number and temperature conditions determined in advance according to experiments and experience.

[0062] When the system inputs the cycle number and test environment temperature of the current battery into the model, the algorithms inside the model will perform feature extraction and analysis on the input data. For example, for the cycle number, the model will analyze the degree of its influence on the internal structure and performance of the battery; for the temperature, the model will consider its influence on the battery chemical reaction rate and capacitance characteristics. Through these analyses, the model outputs the first weight of the first dynamic capacitance information DNC1 and the second weight of the second dynamic capacitance information DNC2 under the current test environment.

[0063] Suppose when testing a new type of lithium-ion battery, the number of battery cycles is 250 times and the test environment temperature is 28°C. After inputting these data into the dynamic capacitance weight fusion model, the model outputs that the first weight of the first dynamic capacitance information DNC1 is 0.6, and the second weight of the second dynamic capacitance information DNC2 is 0.4. This indicates that under the current battery usage state and test environment, the first dynamic capacitance information is relatively more important in subsequent calculations, but the second dynamic capacitance information cannot be ignored.

[0064] The system will verify the rationality of the weights output by the model. On the one hand, referring to the weight data of the same type of battery in the past under similar cycle numbers and temperature conditions, check whether the current weights are within a reasonable fluctuation range; on the other hand, by changing the small values of the input data (such as increasing or decreasing the cycle number by 10 times and changing the temperature by 1°C), input them into the model again and observe the changes in the weights. If the weight changes conform to the expected trend and the change range is small, it is considered that the weights output by the model are reliable; if the weights show abnormal changes, the system will recheck the input data and the training situation of the model, and re-train or adjust the model if necessary.

[0065] S205. Adjust the first dynamic capacitance information DNC1 according to the first dynamic capacitance information DNC1 and the first weight, the second dynamic capacitance information DNC2 and the second weight.

[0066] After the leakage current test system obtains the first dynamic capacitance information DNC1, the first weight, the second dynamic capacitance information DNC2 and the second weight, it starts to adjust the first dynamic capacitance information DNC1. The purpose of the adjustment is to comprehensively consider the influence of different factors on the dynamic capacitance, so that the first dynamic capacitance information can more accurately reflect the true characteristics of the battery in the current state, thereby improving the accuracy of the subsequent leakage current calculation.

[0067] The adjustment formula can be: the adjusted DNC1 = the initial DNC1 × the first weight + DNC2 × the second weight. Taking the battery of a certain laptop as an example, the previously calculated first dynamic capacitance information DNC1 is 20 μF, the first weight obtained from the dynamic capacitance weight fusion model is 0.7, the second dynamic capacitance information DNC2 is 15 μF, and the second weight is 0.3. According to the adjustment formula, the calculation is as follows: the adjusted DNC1 = 20 μF × 0.7 + 15 μF × 0.3 = 14 μF + 4.5 μF = 18.5 μF.

[0068] The system will record and compare and analyze the first dynamic capacitance information before and after the adjustment. By drawing a chart, the change of DNC1 before and after the adjustment is visually displayed, which is convenient for technicians to understand the influence degree of different factors on the dynamic capacitance. If the adjusted DNC1 changes greatly compared with that before the adjustment, the system will further analyze the reason. It may be that the actual use state of the battery and the test environment are quite different from the previous expectations, or there may be errors in obtaining the DNC1, DNC2 or weight data.

[0069] In the embodiment of the present application, the technical solution of determining the first dynamic capacitance information, the second dynamic capacitance information, and adjusting the first dynamic capacitance information in combination with the battery cycle number and the test environment temperature, by accurately calculating the dynamic capacitance information, obtaining the key data of the battery and using the trained model to determine the weight, can more accurately reflect the capacitance characteristics of the battery in the current state, not only improve the accuracy of the subsequent leakage current calculation, effectively solve the problem in the prior art that the influence of the battery use state and environmental factors on the dynamic capacitance is not fully considered, resulting in inaccurate leakage current calculation, and thus realize providing strong support for accurately evaluating the battery leakage condition, but also can more reliably evaluate the battery performance based on more accurate leakage current data, improving the safety and stability of battery management and use.

[0070] The following describes the leakage current test system in the embodiment of the present invention application from the perspective of hardware processing. Please refer to Figure 3 , which is a schematic structural diagram of an entity device of the leakage current test system in the embodiment of the present application.

[0071] It should be noted that Figure 3 the structure of the leakage current test system shown is only an example and should not impose any limitations on the functions and scope of use of the embodiments of the present invention.

[0072] As Figure 3 shown, the leakage current test system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage section 308 into the random access memory (RAM) 303, such as executing the method described in the above embodiments. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0073] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD), an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read from it can be installed into the storage section 308 as needed.

[0074] Specifically, according to the embodiments of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments of the present invention include a computer program product that includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are executed.

[0075] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fibers, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium that contains or stores a program, and this program can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0076] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings.

[0077] Specifically, the leakage current test system of this embodiment includes a processor and a memory, and a computer program is stored on the memory. When the computer program is executed by the processor, the leakage current test method based on low-frequency EIS and constant-current pulse provided in the above-mentioned embodiment is implemented.

[0078] On the other hand, the present invention also provides a computer-readable storage medium, which may be included in the leakage current test system described in the above-mentioned embodiment; or it may exist alone without being assembled into the leakage current test system. The above storage medium carries one or more computer programs. When the above one or more computer programs are executed by a processor of the leakage current test system, the leakage current test system implements the leakage current test method based on low-frequency EIS and constant-current pulse provided in the above-mentioned embodiment.

[0079] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present application.

[0080] As used in the foregoing embodiments, depending on the context, the term "when" may be interpreted to mean "if" or "after" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "upon determining" or "if (the stated condition or event) is detected" may be interpreted to mean "if determined" or "in response to determining" or "when (the stated condition or event) is detected" or "in response to detecting (the stated condition or event)".

[0081] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented. The processes can be completed by relevant hardware instructed by a computer program, which can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the foregoing method embodiments. The foregoing storage medium includes various media that can store program codes, such as ROM or random access memory RAM, magnetic disks, or optical discs.

Claims

1. A leakage current testing method based on low-frequency EIS and cross-flow pulse, applied to a leakage current testing system, characterized in that, The method includes: Determining the battery usage information of the target battery to be tested; Determining the test parameters of the low-frequency EIS according to the battery usage information; Controlling the electrochemical working device to apply a test to the target battery to be tested according to the test parameters, and obtaining impedance data of the target battery to be tested at different frequencies; Obtaining first dynamic capacitance information according to the imaginary part of the impedance corresponding to the impedance data and the test parameters; Controlling the power supply device to apply a constant current pulse current of a preset magnitude to the target battery to be tested, and determining the cut-off condition of the voltage change amount; Continuously monitoring the voltage change amount of the target battery to be tested, and recording the change amount of the test time when the voltage change amount reaches the cut-off condition of the voltage change amount; Determining the leakage current value according to the first dynamic capacitance information, the constant current pulse current, the change amount of the test time, and the voltage change amount.

2. The method according to claim 1, wherein The step of obtaining first dynamic capacitance information according to the imaginary part of the impedance corresponding to the impedance data and the test parameters specifically includes: Determine the first dynamic capacitance information DNC1, DNC1 = - , is the frequency among the test parameters of the low-frequency EIS, is the imaginary part of the impedance of the target battery under test, represents the impedance of the target battery under test at a frequency of f.

3. The method according to claim 2, wherein After the step of determining the first dynamic capacitance information DNC1, it further includes: Determine the second dynamic capacitance information DNC2, DNC2 = , where is the current value applied for the first time during the cross-flow pulse current test, is the current value applied for the second time during the cross-flow pulse current test, is the voltage change across the battery corresponding to within, is the voltage change across the battery corresponding to within, is the time interval experienced during the voltage change process corresponding to , is the time interval experienced during the voltage change process corresponding to ; Obtaining the battery cycle count of the target battery to be tested through the battery management system, and obtaining the current test environment temperature through the temperature sensor; Inputting the battery cycle count and the previous test environment temperature into the dynamic capacitance weight fusion model to determine the first weight of the first dynamic capacitance information DNC1 and the second weight of the second dynamic capacitance information DNC2 under the current test environment. The dynamic capacitance weight fusion model is obtained by prior machine learning training based on battery sample data under different battery cycle counts and different test environment temperatures. The battery sample data includes DNC1, DNC2, and weight combination labels under cycle count and temperature conditions; Adjusting the first dynamic capacitance information DNC1 according to the first dynamic capacitance information DNC1 and the first weight, the second dynamic capacitance information DNC2 and the second weight.

4. The method according to claim 1, wherein The step of determining the leakage current value according to the first dynamic capacitance information, the constant current pulse current, the change amount of the test time, and the voltage change amount specifically includes: Determine the leakage current information according to the leakage current formula , and the leakage current formula is =DNC1× , where DNC1 is the first dynamic capacitance information, is the voltage change amount is the time change amount corresponding to reaching the cut-off condition of the voltage change amount, is the voltage change amount matching the cut-off condition of the voltage change amount at both ends of the battery during the test, and I is the pulsed current value applied during the test.

5. The method according to claim 1, characterized in that In the step of determining the test parameters of the low-frequency EIS according to the battery usage information, specifically includes: Determining the environmental data of the current low-frequency EIS test; Inputting the battery usage information and the environmental data into a preset frequency determination model to obtain a matching low-frequency range. The battery usage information includes the battery type. The construction and training process of the frequency determination model is as follows: Obtaining the actual operation data of multiple different types of batteries under different environmental conditions. The actual operation data includes impedance data under different frequency EIS tests; Using the battery type as a classification label, and obtaining the frequency determination model through machine learning in combination with the actual operation data.

6. The method according to claim 1, characterized in that, Before the step of controlling the electrochemical workstation to apply a test to the target battery to be tested according to the test parameters, it further includes: Obtaining the temperature curve data of the target battery to be tested before the test; Inputting the temperature curve data into a pre-trained temperature feature recognition model to obtain a temperature recognition output result. The temperature feature recognition model is obtained by machine learning training with temperature change data of multiple normal batteries and leakage batteries; Based on the output result of the temperature recognition, adjust the test parameters of the low-frequency EIS in a dynamic adjustment manner. The dynamic adjustment manner includes: when it is recognized that the temperature curve data has a leakage feature, reduce the test frequency according to a set value to improve the impedance measurement accuracy; when it is recognized that the temperature curve data does not have a leakage feature, increase the test frequency according to a set value to shorten the test time.

7. The method according to claim 1, characterized in that After the step of determining the leakage current value according to the first dynamic capacitance information, the cross-flow pulse current, the test time variation amount, and the voltage variation amount, it further includes: Determine the leakage battery, and input the remaining capacity, health status, and usage environment data of the obtained leakage battery into a pre-constructed intelligent leakage assessment model for risk quantification assessment to obtain a risk assessment result. The intelligent leakage assessment model is established in advance through machine learning training based on multiple groups of battery leakage data and their corresponding risk level labels under different remaining capacities, different health statuses, and different usage environment conditions. Priority-sort the batteries according to the risk level based on the risk assessment result and the magnitude of the leakage current value to obtain a sorting result. Attach an electronic label containing information on the leakage degree and risk level marked with different colors to the battery according to the sorting result. Send the electronic labels to the visual end in the order of the sorting result.

8. A leakage current testing system, characterized in that, The leakage current test system includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code. The computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the leakage current test system to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium, comprising instructions, characterized in that, When the instruction runs on the leakage current test system, enable the leakage current test system to execute the method described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product runs on the leakage current test system, enable the leakage current test system to execute the method described in any one of claims 1-7.

Images