Battery testing system and method
By combining deep learning models to analyze battery data and defects in battery testing, the problem of battery testing methods relying on empirical formulas in the existing technology is solved, and the detailed evaluation and division of battery performance is achieved, which improves the safety and reliability of the battery .
Patent Information
- Application Number
- CN202411420199.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-12
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-10-12
AI Technical Summary
The existing battery testing methods rely on empirical formulas and simple calculations, lack the analysis and identification of multiple battery defect data through deep learning models, and lack the comprehensive evaluation value of battery performance obtained by combining multiple calculation data to divide battery performance.
By obtaining the current, voltage and internal resistance parameter data of the battery in real time under constant temperature conditions, using deep learning models to analyze defect data on the battery surface, combining the internal resistance increase rate and static discharge rate, the comprehensive defect value and performance evaluation value of the battery are constructed. , achieve a comprehensive evaluation of battery performance.
It improves the accuracy and comprehensiveness of battery performance evaluation, can more effectively identify battery defects and performance degradation, provides a detailed division of battery performance, and improves the safety and reliability of the battery.
Smart Images

Figure CN118914868B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of battery testing methods, and in particular relates to a battery testing system and method. Background Art
[0002] With the continuous development of battery technology, batteries have gradually become an indispensable energy supply method in modern life. The performance of batteries is very important for the normal operation of equipment. The various data collected from the battery in real time will be pre-processed, and the data will be trained and analyzed through deep learning models to identify the various data. The battery performance evaluation value is obtained by integrating various data to analyze the battery performance and improve the safety and reliability of the battery.
[0003] Although existing battery testing methods have achieved battery testing to a certain extent, traditional battery testing methods mainly rely on empirical formulas and simple calculations. They lack the ability to analyze and identify a variety of battery defect data through deep learning models, and lack the ability to classify battery performance by integrating a variety of calculated data to obtain a comprehensive battery performance degradation evaluation value. Summary of the invention
[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a battery testing system and method for solving the following technical problems:
[0005] Although existing battery testing methods have achieved battery testing to a certain extent, traditional battery testing methods mainly rely on empirical formulas and simple calculations. They lack the ability to analyze and identify a variety of battery defect data through machine learning models, and lack the ability to classify battery performance by integrating a variety of calculated data to obtain a comprehensive battery performance degradation evaluation value.
[0006] To solve the above problems, a first aspect of the present invention provides a battery testing method, comprising the following steps:
[0007] S1: Under constant temperature conditions, obtain current, voltage and internal resistance parameter data from the battery in real time;
[0008] S2: Set an initial constant temperature value, and record the highest temperature heating value and the lowest temperature heating value of the internal resistance value of the battery by continuously heating the initial constant temperature value;
[0009] S3: Obtain the starting internal resistance value, starting temperature value, highest temperature heating value and lowest temperature heating value on the battery chip, record the highest temperature value as the first resistance temperature, record the lowest temperature value as the second resistance temperature, and obtain the internal resistance correction calculation value by combining the difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial internal resistance value. During the measurement discharge process, the initial internal resistance value and the internal resistance correction calculation value are used to obtain the battery internal resistance increase rate;
[0010] S4: Obtain image data of the battery through an image acquisition device, identify the convex and concave defect data that appear on the surface of the battery during use, use a deep learning model to train and analyze historical convex and concave defect data, and construct a defect factor based on the defect area and defect depth, combined with the defect quantity factor to obtain a comprehensive defect value of the battery;
[0011] S5: Under the same temperature and time, in each cycle, the battery is kept at rest when fully charged, and the remaining power of the battery after the preset rest time is detected. The battery static discharge rate is obtained by the initial power and the remaining power of the battery. The comprehensive evaluation value of battery performance degradation is obtained by combining the calculated value of the comprehensive defects of the battery, the increase rate of the internal resistance of the battery and the static discharge rate of the battery, and the battery performance is classified.
[0012] As a further solution of the present invention: Step S1 comprises the following steps:
[0013] Randomly select 60% of the total from each battery batch as battery test samples, and measure the current, voltage and internal resistance parameter data of the experimental samples by measuring equipment at a constant temperature and time and record them;
[0014] Among them, the various measuring devices include voltmeters, current clamps, internal resistance meters and temperature sensors.
[0015] As a further solution of the present invention: the step S2 comprises the following steps:
[0016] A known initial constant temperature is set as the initial temperature value, the temperature of the battery is increased at a temperature interval of 2 degrees Celsius, the internal resistance value of the battery is measured and recorded at each temperature point, the temperature point with the largest resistance value is obtained as the highest temperature heating value, and the temperature point with the smallest resistance value is obtained as the lowest temperature heating value.
[0017] As a further solution of the present invention: in step S3, the difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial internal resistance value are combined to obtain an internal resistance correction calculation value, including the following steps:
[0018] Internal resistance correction calculation formula:
[0019] ,
[0020] in, is the calculated value for internal resistance correction, is the initial internal resistance value, is the initial temperature value, is the current temperature value, is the temperature value when the internal resistance is the highest, is the temperature value when the internal resistance is the lowest, and is the temperature coefficient of resistance.
[0021] As a further solution of the present invention: the method of measuring the initial internal resistance value and the internal resistance correction calculation value to obtain the battery internal resistance increase rate during the discharge process includes the following steps:
[0022] The calculation formula of the battery internal resistance increase rate is:
[0023] ,
[0024] in, is the initial internal resistance value, For the The internal resistance value measured is the calculated value for internal resistance correction, is the number of cycles, is the rate of increase of internal resistance of the battery.
[0025] As a further solution of the present invention: the step S4 comprises the following steps:
[0026] Use a high-resolution camera to capture images of the battery surface, obtain images of raised and sunken defects, divide the boundaries of the defect area through manual annotation tools, separate the defect area from the battery background according to image processing technology, perform image enhancement and grayscale processing, segment the grayscale data values of the defect area and the battery surface area through a set grayscale value threshold, extract area features from the segmented defect area as input for training the deep learning model, train the model using a training data set with defect area features, use the trained model to predict the defect area ratio in the battery, and obtain the defect area factor value based on the ratio;
[0027] The defect area factor calculation formula is obtained by the ratio of defect area to total battery area:
[0028] ,
[0029] in, is the calculated value of defect area factor, is the area value of battery defects detected in the image, Detect the total battery area value for the image;
[0030] The distance between each pixel in the defective area and the camera in the collected image is used. Images at the same position are taken from different angles by two cameras. The depth data of the defective area is calculated using the principle of visual difference. The depth value of the defective area after denoising is converted to a uniform range through normalization for deep feature extraction. The extracted data is annotated, and the model is trained using a training data set with annotated information. The depth ratio of defects in the battery is predicted using the trained deep learning model, and the defect depth factor value is calculated by the ratio of the defect depth to the total battery depth:
[0031] The defect depth factor calculation formula is obtained by the ratio of defect depth to total battery depth:
[0032] ,
[0033] in, Calculate the value for the defect depth factor, The depth value of the battery defect detected in the image, Detect the total depth value of the battery for the image;
[0034] The defect factor calculation formula is obtained by the defect area factor and the defect depth factor:
[0035] ,
[0036] in, is the defect factor calculation value, is the calculated value of defect area factor, Calculate the value for the defect depth factor, and is the corresponding weight coefficient;
[0037] The battery surface is photographed with a high-resolution camera, and the image is converted into a digital signal for denoising and contrast enhancement. The processed image is subjected to feature extraction through a deep learning model, and the extracted features are classified and annotated as input for training the deep learning model. The number of various defective areas is calculated through the trained deep learning model to obtain the defect quantity factor value;
[0038] The calculation formula for the comprehensive defect value of the battery is obtained based on the defect factor and defect quantity factor:
[0039] ,
[0040] in, is the calculated value of comprehensive battery defects, is the defect factor calculation value, is the defect quantity factor value, and is the corresponding weight coefficient.
[0041] As a further solution of the present invention: in step S5, obtaining the static discharge rate of the battery by using the initial power and the remaining power of the battery includes the following steps:
[0042] The calculation formula of battery static discharge rate is:
[0043] ,
[0044] in, is the calculated value of the battery static discharge rate, is the initial power value, The remaining power value.
[0045] As a further solution of the present invention: the comprehensive battery performance degradation evaluation value is obtained by combining the battery comprehensive defect calculation value, the battery internal resistance increase rate and the battery static discharge rate, and the battery performance is classified, including the following steps:
[0046] The calculation formula for the comprehensive evaluation value of battery performance degradation is as follows:
[0047] ,
[0048] in, is the comprehensive degradation evaluation value of battery performance, is the calculated value of the battery static discharge rate, is the calculated value of the battery internal resistance increase rate, is the calculated value of comprehensive battery defects, , , are the weights corresponding to the data, is 0.2, is 0.3, is 0.5;
[0049] The battery performance is divided according to the comprehensive degradation evaluation value of the battery performance:
[0050] If 0%< <20%, the battery performance is excellent;
[0051] If 20%≤ <40%, the battery performance is good;
[0052] If 40%≤ <60%, the battery performance is average;
[0053] If 60%≤ <80%, the battery performance is poor;
[0054] If 80%≤ , the battery performance is very poor.
[0055] Beneficial effects of the present invention:
[0056] The present invention collects the difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial resistance value during the discharge process of the battery, and constructs an internal resistance correction calculation value, thereby improving the accuracy of the result. The internal resistance increase rate of the battery can be obtained by combining the initial resistance value and the internal resistance correction calculation value, which is of great significance to the evaluation of battery performance.
[0057] The present invention obtains data of the battery defect area and the battery surface area through an image acquisition device, and constructs a battery comprehensive defect value calculation value obtained by constructing the area of the defect area, the depth combined with the quantitative factor and the area of the battery surface area, the depth combined with the quantitative factor ratio according to different weights, so as to realize the evaluation and judgment of the battery performance;
[0058] The present invention assigns corresponding weights according to the role in the battery through the battery comprehensive defect calculation value, the battery internal resistance increase rate and the battery static discharge rate, and comprehensively obtains the battery performance comprehensive degradation evaluation value to analyze the battery performance, thereby realizing a comprehensive evaluation of the battery performance by combining multiple calculation data, which is of great significance for the use of the battery. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0060] Figure 1 It is a schematic diagram of the system framework of the present invention;
[0061] Figure 2 It is a schematic diagram of the system module of the present invention. DETAILED DESCRIPTION
[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0063] See also Figure 1 As shown, the present invention is a battery testing method, comprising the following steps:
[0064] S1: Under constant temperature conditions, obtain current, voltage and internal resistance parameter data from the battery in real time;
[0065] S2: Set an initial constant temperature value, and record the highest temperature heating value and the lowest temperature heating value of the internal resistance value of the battery by continuously heating the initial constant temperature value;
[0066] S3: Obtain the starting internal resistance value, starting temperature value, highest temperature heating value and lowest temperature heating value on the battery chip, record the highest temperature value as the first resistance temperature, record the lowest temperature value as the second resistance temperature, and obtain the internal resistance correction calculation value by combining the difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial internal resistance value. During the measurement discharge process, the initial internal resistance value and the internal resistance correction calculation value are used to obtain the battery internal resistance increase rate;
[0067] S4: Obtain image data of the battery through an image acquisition device, identify the convex and concave defect data that appear on the surface of the battery during use, use a deep learning model to train and analyze historical convex and concave defect data, and construct a defect factor based on the defect area and defect depth, combined with the defect quantity factor to obtain a comprehensive defect value of the battery;
[0068] S5: Under the same temperature and time, in each cycle, the battery is kept at rest when fully charged, and the remaining power of the battery after the preset rest time is detected. The battery static discharge rate is obtained by the initial power and the remaining power of the battery. The comprehensive evaluation value of battery performance degradation is obtained by combining the calculated value of the comprehensive defects of the battery, the increase rate of the internal resistance of the battery and the static discharge rate of the battery, and the battery performance is classified.
[0069] Specifically, the voltage value in the battery is measured in real time by a voltmeter, the current value in the battery is measured in real time by a current clamp, and the internal resistance value in the battery is measured in real time by an internal resistance meter, and the detected voltage, current and internal resistance data are recorded; the temperature of the battery is increased at a temperature interval of 2 degrees Celsius, and the internal resistance value of the battery is measured and recorded at each temperature point, and the temperature point with the largest resistance value is obtained as the highest temperature heating value, and the temperature point with the smallest resistance value is obtained as the lowest temperature heating value; the internal resistance correction calculation value is obtained by combining the difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial internal resistance value, and the battery internal resistance increase rate is obtained by the initial battery internal resistance value and the internal resistance correction calculation value; the defect area and defect depth are obtained according to the three deep learning models, and the defect factor is constructed to obtain the battery comprehensive defect value combined with the defect number factor; the battery comprehensive performance degradation evaluation value is obtained by combining the battery comprehensive defect calculation value, the battery internal resistance increase rate and the battery static discharge rate, and the battery performance is divided.
[0070] In one embodiment of the present invention, the step S1 comprises the following steps:
[0071] Randomly select 60% of the total from each battery batch as battery test samples, and measure the current, voltage and internal resistance parameter data of the experimental samples by measuring equipment at a constant temperature and time and record them;
[0072] Among them, the various measuring devices include voltmeters, current clamps, internal resistance meters and temperature sensors.
[0073] Specifically, under the constant temperature set initially, sixty percent of the total battery batch is randomly selected as test samples. The voltage value at both ends of the battery is directly measured by a voltmeter, and the current clamp converts the induced current through the wire into a readable electrical signal. The internal resistance meter measures the initial internal resistance of the battery, and the voltage, current and internal resistance data are obtained. The data is recorded and analyzed in real time to help users better understand the current working status and changes of the battery.
[0074] In one embodiment of the present invention, the step S2 comprises the following steps:
[0075] A known initial constant temperature is set as the initial temperature value, the temperature of the battery is increased at a temperature interval of 2 degrees Celsius, the internal resistance value of the battery is measured and recorded at each temperature point, the temperature point with the largest resistance value is obtained as the highest temperature heating value, and the temperature point with the smallest resistance value is obtained as the lowest temperature heating value.
[0076] Specifically, an initial constant temperature is set as the starting point of the experiment. The internal resistance of the battery is measured at this temperature as the initial internal resistance value. The temperature of the battery is gradually increased at a temperature interval of 2 degrees Celsius. After waiting for a period of time at each new temperature point to ensure that the battery reaches thermal equilibrium, the temperature point value and the internal resistance value corresponding to the internal resistance of the battery are recorded. The temperature point with the largest resistance value is obtained as the highest temperature heating value, and the temperature point with the smallest resistance value is obtained as the lowest temperature heating value, which are important components of the internal resistance correction calculation value.
[0077] In one embodiment of the present invention, the step S3 obtains the internal resistance correction calculation value by combining the difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial internal resistance value, including the following steps:
[0078] Internal resistance correction calculation formula:
[0079] ,
[0080] in, is the calculated value for internal resistance correction, is the initial internal resistance value, is the initial temperature value, is the current temperature value, is the temperature value when the internal resistance is the highest, is the temperature value when the internal resistance is the lowest, and is the temperature coefficient of resistance.
[0081] Specifically, the internal resistance value of the battery under the initial condition, the initial temperature value, the current temperature value, the highest temperature heating value of the temperature point with the largest resistance value, and the lowest temperature heating value of the temperature point with the smallest resistance value are recorded. The internal resistance correction calculation value can be obtained by combining multiple values and the resistance temperature coefficient. Among them, the resistance temperature coefficient The value is 0.03, The value is 0.02. The calculated internal resistance correction value can solve potential problems that may occur in the battery during the detection process before they become failures, reduce unexpected risks, and improve the accuracy of test results.
[0082] In one embodiment of the present invention, the method of measuring the initial internal resistance value and the internal resistance correction calculation value to obtain the battery internal resistance increase rate during the discharge process includes the following steps:
[0083] The calculation formula of the battery internal resistance increase rate is:
[0084] ,
[0085] in, is the initial internal resistance value, For the The internal resistance value measured is the calculated value for internal resistance correction, is the number of cycles, is the rate of increase of internal resistance of the battery.
[0086] Specifically, during the discharge measurement process, the internal resistance of the battery is measured at intervals. The subsequent measured internal resistance values are compared with the initial internal resistance values to calculate the rate of change of the internal resistance. The internal resistance increase rate of the battery is obtained by combining the internal resistance correction calculation values. The increase in the battery internal resistance rate will directly affect the power output and discharge efficiency of the battery, and have a negative impact on the overall performance of the battery. It is the first important indicator for evaluating battery performance.
[0087] In one embodiment of the present invention, the step S4 comprises the following steps:
[0088] Use a high-resolution camera to capture images of the battery surface, obtain images of raised and sunken defects, divide the boundaries of the defect area through manual annotation tools, separate the defect area from the battery background according to image processing technology, perform image enhancement and grayscale processing, segment the grayscale data values of the defect area and the battery surface area through a set grayscale value threshold, extract area features from the segmented defect area as input for training the deep learning model, train the model using a training data set with defect area features, use the trained model to predict the defect area ratio in the battery, and obtain the defect area factor value based on the ratio;
[0089] The defect area factor calculation formula is obtained by the ratio of defect area to total battery area:
[0090] ,
[0091] in, is the calculated value of defect area factor, is the area value of battery defects detected in the image, Detect the total battery area value for the image;
[0092] The distance between each pixel in the defective area and the camera in the collected image is used. Images at the same position are taken from different angles by two cameras. The depth data of the defective area is calculated using the principle of visual difference. The depth value of the defective area after denoising is converted to a uniform range through normalization for deep feature extraction. The extracted data is annotated, and the model is trained using a training data set with annotated information. The depth ratio of defects in the battery is predicted using the trained deep learning model, and the defect depth factor value is calculated by the ratio of the defect depth to the total battery depth:
[0093] The defect depth factor calculation formula is obtained by the ratio of defect depth to total battery depth:
[0094] ,
[0095] in, Calculate the value for the defect depth factor, The depth value of the battery defect detected in the image, Detect the total depth value of the battery for the image;
[0096] The defect factor calculation formula is obtained by the defect area factor and the defect depth factor:
[0097] ,
[0098] in, is the defect factor calculation value, is the calculated value of defect area factor, Calculate the value for the defect depth factor, and is the corresponding weight coefficient;
[0099] The battery surface is photographed with a high-resolution camera, and the image is converted into a digital signal for denoising and contrast enhancement. The processed image is subjected to feature extraction through a deep learning model, and the extracted features are classified and annotated as input for training the deep learning model. The number of various defective areas is calculated through the trained deep learning model to obtain the defect quantity factor value;
[0100] The calculation formula for the comprehensive defect value of the battery is obtained based on the defect factor and defect quantity factor:
[0101] ,
[0102] in, is the calculated value of comprehensive battery defects, is the defect factor calculation value, is the defect quantity factor value, and is the corresponding weight coefficient.
[0103] Specifically, a high-resolution camera is used to collect data on the battery image, and the data after corresponding processing and feature extraction is used as input data to be input into respective deep learning models for training. The trained model is used to predict the defect area ratio in the battery, and the defect area factor value is obtained according to the ratio; the number of various defect areas is calculated through the trained deep learning model to obtain the defect quantity factor value; the number of various defect areas is calculated through the trained deep learning model to obtain the defect quantity factor value; according to the defect area factor value and the defect quantity factor value combined with the corresponding weights, and The weight values are all 0.5, and the defect factor value is constructed. According to the defect factor value and the defect quantity factor value, the defect factor weight value is calculated according to the corresponding weight. is 0.7, the defect quantity factor weight The calculated value of the comprehensive defect value of the battery is 0.3, which reflects the overall quality of the battery during use and serves as the second important indicator for evaluating battery performance.
[0104] In one embodiment of the present invention, the step S5 of obtaining the static discharge rate of the battery by using the initial power and the remaining power of the battery includes the following steps:
[0105] The calculation formula of battery static discharge rate is:
[0106] ,
[0107] in, is the calculated value of the battery static discharge rate, is the initial power value, The remaining power value.
[0108] Specifically, under constant temperature conditions, the fully charged battery is left to stand for 7 days without any charging and discharging operations. The battery capacity is tested regularly, and the fixed time of each measurement and the corresponding capacity are recorded. The battery capacity in the initial state and the remaining capacity during the test are recorded. The battery static discharge rate can be obtained by taking the ratio of the consumed capacity to the initial capacity obtained by the difference between the initial capacity and the remaining capacity. The change in the battery static discharge rate can be used to predict the battery performance and remaining life. The battery static discharge rate is the third important indicator for evaluating battery performance.
[0109] In one embodiment of the present invention, the comprehensive degradation evaluation value of battery performance is obtained by combining the calculated value of comprehensive battery defects, the rate of increase of internal resistance of the battery and the static discharge rate of the battery, and the performance of the battery is classified, including the following steps:
[0110] The calculation formula for the comprehensive evaluation value of battery performance degradation is as follows:
[0111] ,
[0112] in, is the comprehensive degradation evaluation value of battery performance, is the calculated value of the battery static discharge rate, is the calculated value of the battery internal resistance increase rate, is the calculated value of comprehensive battery defects, , , are the weights corresponding to the data, is 0.2, is 0.3, is 0.5;
[0113] The battery performance is divided according to the comprehensive degradation evaluation value of the battery performance:
[0114] If 0%< <20%, the battery performance is excellent;
[0115] If 20%≤ <40%, the battery performance is good;
[0116] If 40%≤ <60%, the battery performance is average;
[0117] If 60%≤ <80%, the battery performance is poor;
[0118] If 80%≤ , the battery performance is very poor.
[0119] Specifically, corresponding weights are assigned according to the calculated value of comprehensive defects of the battery, the rate of increase of internal resistance of the battery and the static discharge rate of the battery, and a comprehensive evaluation value of the comprehensive degradation of battery performance is obtained. The degradation evaluation value can intuitively reflect the degree of attenuation of battery performance and analyze the status of the battery.
[0120] See also Figure 2 As shown, the present invention is a battery testing system, comprising the following modules:
[0121] Parameter acquisition module: used to obtain current, voltage and internal resistance parameter data from the battery in real time under constant temperature conditions;
[0122] Battery internal resistance temperature characteristic test module: used to set an initial constant temperature value, and record the highest temperature heating value and the lowest temperature heating value of the battery internal resistance value by continuously heating the initial constant temperature value;
[0123] Battery internal resistance correction and increase module: used to obtain the starting internal resistance value, starting temperature value, highest temperature heating value and lowest temperature heating value on the battery chip, record the highest temperature value as the first resistance temperature, record the lowest temperature value as the second resistance temperature, and obtain the internal resistance correction calculation value by combining the difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial internal resistance value. During the discharge measurement process, the initial internal resistance value and the internal resistance correction calculation value are used to obtain the battery internal resistance increase rate;
[0124] Image acquisition defect factor module: used to obtain battery image data through image acquisition equipment, identify the convex and concave defect data that appear on the surface of the battery during use, use the deep learning model to train and analyze the historical convex and concave defect data, and construct a defect factor based on the defect area and defect depth, combined with the defect quantity factor to obtain the comprehensive defect value of the battery;
[0125] Battery performance comprehensive evaluation and classification module: It is used to keep the battery in a static state when it is fully charged at the same temperature and time in each cycle, detect the remaining power of the battery after the preset static time, and obtain the battery static discharge rate through the initial power and remaining power of the battery. The comprehensive battery performance degradation evaluation value is obtained by combining the battery comprehensive defect calculation value, the battery internal resistance increase rate and the battery static discharge rate, and the battery performance is classified.
[0126] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A battery testing method, characterized in that: The following steps are involved: S1: Under constant temperature conditions, obtain current, voltage and internal resistance parameter data from the battery in real time; S2: Set an initial constant temperature value, and record the highest temperature heating value and the lowest temperature heating value of the internal resistance value of the battery by continuously heating the initial constant temperature value; S3: Obtain the starting internal resistance value, starting temperature value, highest temperature heating value and lowest temperature heating value on the battery chip, record the highest temperature value as the first resistance temperature, record the lowest temperature value as the second resistance temperature, and obtain the internal resistance correction calculation value by combining the difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial internal resistance value. During the measurement discharge process, the initial internal resistance value and the internal resistance correction calculation value are used to obtain the battery internal resistance increase rate; S4: Obtain image data of the battery through an image acquisition device, identify the convex and concave defect data that appear on the surface of the battery during use, use a deep learning model to train and analyze historical convex and concave defect data, and construct a defect factor based on the defect area and defect depth, combined with the defect quantity factor to obtain a comprehensive defect value of the battery; S5: Under the same temperature and time, in each cycle, the battery is kept in a static state when fully charged, and the remaining power of the battery after the preset static time is detected. The static discharge rate of the battery is obtained by the initial power and the remaining power of the battery. The comprehensive degradation evaluation value of the battery performance is obtained by combining the calculated value of the battery comprehensive defects, the increase rate of the battery internal resistance and the static discharge rate of the battery, and the performance of the battery is classified; The difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial internal resistance value are combined to obtain an internal resistance correction calculation value, including the following steps: Internal resistance correction calculation formula: , in, is the calculated value for internal resistance correction, is the initial internal resistance value, is the initial temperature value, is the current temperature value, is the temperature value when the internal resistance is the highest, is the temperature value when the internal resistance is the lowest, and is the temperature coefficient of resistance; During the discharge measurement, the initial internal resistance value and the internal resistance correction calculation value are used to obtain the battery internal resistance increase rate, including the following steps: The calculation formula of the battery internal resistance increase rate is: , in, is the initial internal resistance value, For the The internal resistance value measured is the calculated value for internal resistance correction, is the number of cycles, is the rate of increase of internal resistance of the battery; Use a high-resolution camera to capture images of the battery surface, obtain images of raised and sunken defects, divide the boundaries of the defect area through manual annotation tools, separate the defect area from the battery background according to image processing technology, perform image enhancement and grayscale processing, segment the grayscale data values of the defect area and the battery surface area through a set grayscale value threshold, extract area features from the segmented defect area as input for training the deep learning model, train the model using a training data set with defect area features, use the trained model to predict the defect area ratio in the battery, and obtain the defect area factor value based on the ratio; The defect area factor calculation formula is obtained by the ratio of defect area to total battery area: , in, is the calculated value of defect area factor, is the area value of battery defects detected in the image, Detect the total battery area value for the image; The distance between each pixel in the defective area and the camera in the collected image is used. Images at the same position are taken from different angles by two cameras. The depth data of the defective area is calculated using the principle of visual difference. The depth value of the defective area after denoising is converted to a uniform range through normalization for deep feature extraction. The extracted data is annotated, and the model is trained using a training data set with annotated information. The depth ratio of defects in the battery is predicted using the trained deep learning model, and the defect depth factor value is calculated by the ratio of the defect depth to the total battery depth: The defect depth factor calculation formula is obtained by the ratio of defect depth to total battery depth: , in, Calculate the value for the defect depth factor, The depth value of the battery defect detected in the image, Detect the total depth value of the battery for the image; The defect factor calculation formula is obtained by the defect area factor and the defect depth factor: , in, is the defect factor calculation value, is the calculated value of defect area factor, Calculate the value for the defect depth factor, and is the corresponding weight coefficient; Use a high-resolution camera to photograph the battery surface, convert the image into a digital signal, and then perform denoising and contrast enhancement. Use a deep learning model to extract features from the processed image. Classify and annotate the extracted features and use them as input for training the deep learning model. Use the trained deep learning model to calculate the number of various defect areas and obtain the defect quantity factor value. The calculation formula for the comprehensive defect value of the battery is obtained based on the defect factor and defect quantity factor: , in, is the calculated value of comprehensive battery defects, is the defect factor calculation value, is the defect quantity factor value, and is the corresponding weight coefficient.
2. A battery testing method according to claim 1, characterized in that: The step S1 comprises the following steps: Randomly select 60% of the total from each battery batch as battery test samples, and measure the current, voltage and internal resistance parameter data of the experimental samples by measuring equipment at a constant temperature and time and record them; Among them, the various measuring devices include voltmeters, current clamps, internal resistance meters and temperature sensors.
3. A battery testing method according to claim 1, characterized in that: The step S2 comprises the following steps: A known initial constant temperature is set as the initial temperature value, the temperature of the battery is increased at a temperature interval of 2 degrees Celsius, the internal resistance value of the battery is measured and recorded at each temperature point, the temperature point with the largest resistance value is obtained as the highest temperature heating value, and the temperature point with the smallest resistance value is obtained as the lowest temperature heating value.
4. A battery testing method according to claim 1, characterized in that: The step S5 obtains the static discharge rate of the battery by using the initial power and the remaining power of the battery, including the following steps: The calculation formula of battery static discharge rate is: , in, is the calculated value of the battery static discharge rate, is the initial power value, The remaining power value.
5. A battery testing method according to claim 1, characterized in that: The method of obtaining a comprehensive battery performance degradation evaluation value based on the battery comprehensive defect calculation value, the battery internal resistance increase rate and the battery static discharge rate, and classifying the battery performance, includes the following steps: The calculation formula for the comprehensive evaluation value of battery performance degradation is as follows: , in, is the comprehensive degradation evaluation value of battery performance, is the calculated value of the battery static discharge rate, is the calculated value of the battery internal resistance increase rate, is the calculated value of comprehensive battery defects, , , are the weights corresponding to the data, is 0.2, is 0.3, is 0.5; The battery performance is divided according to the comprehensive degradation evaluation value of the battery performance: If 0%< <20%, the battery performance is excellent; If 20%≤ <40%, the battery performance is good; If 40%≤ <60%, the battery performance is average; If 60%≤ <80%, the battery performance is poor; If 80%≤ , the battery performance is very poor.
6. A battery testing system, using a battery testing method as claimed in any one of claims 1 to 5, characterized in that: Includes the following modules: Parameter acquisition module: used to obtain current, voltage and internal resistance parameter data from the battery in real time under constant temperature conditions; Battery internal resistance temperature characteristic test module: used to set an initial constant temperature value, and record the highest temperature heating value and the lowest temperature heating value of the battery internal resistance value by continuously heating the initial constant temperature value; Battery internal resistance correction and increase module: used to obtain the starting internal resistance value, starting temperature value, highest temperature heating value and lowest temperature heating value on the battery chip, record the highest temperature value as the first resistance temperature, record the lowest temperature value as the second resistance temperature, and obtain the internal resistance correction calculation value by combining the difference between the first resistance temperature and the second resistance temperature, the difference between the current temperature and the initial temperature, and the initial internal resistance value. During the discharge measurement process, the initial internal resistance value and the internal resistance correction calculation value are used to obtain the battery internal resistance increase rate; Image acquisition defect factor module: used to obtain battery image data through image acquisition equipment, identify the convex and concave defect data that appear on the surface of the battery during use, use the deep learning model to train and analyze the historical convex and concave defect data, and construct a defect factor based on the defect area and defect depth, combined with the defect quantity factor to obtain the comprehensive defect value of the battery; Battery performance comprehensive evaluation and classification module: It is used to keep the battery in a static state when it is fully charged at the same temperature and time in each cycle, detect the remaining power of the battery after the preset static time, and obtain the battery static discharge rate through the initial power and remaining power of the battery. The comprehensive battery performance degradation evaluation value is obtained by combining the battery comprehensive defect calculation value, the battery internal resistance increase rate and the battery static discharge rate, and the battery performance is classified.
