Solid-state battery thermal fault diagnosis and online early warning method

Abstract

This invention discloses a method for thermal fault diagnosis and online early warning of solid-state batteries, belonging to the field of battery fault detection technology. The method includes: acquiring temperature distribution images of the solid-state battery, generating image feature parameters and a temperature gradient map; obtaining a solid-state battery usage status assessment value based on the quantized image feature parameters and environmental parameters; setting a screening temperature, and executing corresponding strategies according to the distribution areas of high-temperature images. The solid-state battery thermal fault diagnosis and online early warning method of this invention uses thermal imaging to detect image feature parameters under battery usage conditions and analyzes them to obtain the battery status assessment value (Bauy). t Secondly, the state evaluation value Bauy t By combining the test results obtained from subsequent testing strategies, the battery's adaptability can be comprehensively predicted, making it easier to formulate corresponding battery usage modes based on the battery's tolerance, thereby making full use of the battery's remaining lifespan.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery fault detection, in particular to a solid-state battery thermal fault diagnosis and online early warning method. BACKGROUND

[0002] Lithium-ion batteries are one of the most widely used rechargeable batteries. Its high energy density, long life and lightweight advantage makes it widely used in electric vehicles, portable electronic devices and energy storage systems. However, the normal operation of lithium-ion batteries will be affected by temperature. Overcharging, internal and external short circuit, mechanical damage and long-term work in high temperature environment can cause battery thermal runaway, and even cause fire or explosion. Therefore, equipping lithium-ion batteries with appropriate thermal management system and conducting real-time online temperature diagnosis can avoid irreversible thermal runaway and ensure the safe and reliable operation of lithium-ion batteries.

[0003] At present, the Chinese patent with patent application number: "CN202310041018.5" discloses a battery thermal runaway fault detection method and system. At high temperature, the decomposition of battery materials can cause gas generation and battery swelling. The battery structure is designed to rupture at the critical gas pressure and discharge the accumulated electrolytic gas to prevent explosive force. Although this patent can detect the type of battery internal short circuit fault by detecting the battery swelling force and gas, it is not accurate to judge only by "swelling". Taking a mobile phone as an example, although the ions inside the existing mobile phone battery are constantly changing during use, the gas and swelling degree generated during battery operation are very small. When we find that the mobile phone battery has "bulge" and other faults, it means that the battery damage degree has reached more than 60%, and it needs to be stopped using. For electric vehicles, it is not possible to stop using immediately when a fault occurs, which leads to the user still needs to drive a vehicle with potential safety hazards for a period of time. In addition, the Chinese patent with patent application number: "CN202211467164.6" discloses a power lithium battery thermal runaway fault classification and risk prediction method and system, including: a module-level power battery model fault injection method, a random fault generation and labeling method, a deep learning-based power lithium-ion battery fault multi-classification model, and a migration learning method for applying the model to real vehicles. Although this application can diagnose the battery by simulating the use of the vehicle, in actual use, it often needs to be predicted in advance, and because the battery heating conditions are different, if the same detection method is used for detection, it is easy to deviate.

[0004] However, during the implementation of the above technical solution, at least the following technical problems are found:

[0005] The existing battery thermal fault detection system mostly adopts the form of overall operation for calculation, that is, it needs to be executed from beginning to end according to a predetermined program, and during the operation process, the normal battery or circuit is also detected, which leads to a large amount of time for operation for each detection, and secondly, since all influencing factors during the battery generation process cannot be guaranteed to remain consistent, and the actual use environment is also different, therefore the bearing capacity of each battery core is also dynamically changing, therefore only the data obtained during the actual use process of the battery can best reflect the condition of the battery;

[0006] Common battery thermal faults are divided into two kinds, one is battery core failure, during the use process of the battery, the battery core continuously performs chemical reactions, leading to the loss of electric energy in the form of heat; the other is circuit failure, when the circuit appears short circuit, not only the circuit is damaged, but also a large amount of heat energy is generated, and no matter which kind of thermal fault, finally it will be manifested in the form of heat, that is, a large amount of heat energy will be generated at the position of the fault, therefore the position of the thermal fault can be preliminarily judged by detecting the distribution of heat, and the subsequent inspection is facilitated, for this purpose, the solid-state battery thermal fault diagnosis and online early warning method is proposed. SUMMARY

[0007] (1) Technical problems solved

[0008] In view of the deficiencies of the prior art, the solid-state battery thermal fault diagnosis and online early warning method is provided, which detects the indication temperature, temperature change rate and temperature diffusion speed of the battery in use by adopting the way of thermal imaging, analyzes to obtain a state evaluation value Bauy t which can directly judge the actual bearing capacity of the battery, and according to the positional relationship between the state evaluation value Bauy t and the state threshold interval, the corresponding solution is retrieved, thereby providing a basis for the replacement and adjustment of the battery, and when the replacement period is missed, the service life of the battery can be prolonged, providing sufficient time for battery replacement, and secondly, the state evaluation value Bauy t is combined with the detection result obtained by the subsequent detection strategy to obtain an adaptive estimated value Bauy x , thereby comprehensively predicting the adaptability of the battery, facilitating the formulation of a corresponding battery use mode according to the bearing capacity of the battery, thereby the remaining service life of the battery can be fully utilized, and the technical problems of low operation efficiency and inaccurate prediction result of the existing solid-state battery thermal fault diagnosis method in use are solved.

[0009] (2) Technical solutions

[0010] In order to achieve the above purposes, the technical solutions are as follows:

[0011] The method comprises the following steps:

[0012] After the collected solid-state battery temperature distribution image is normalized, the image is input into a feature extraction network to generate image feature parameters and a temperature gradient image, wherein the image feature parameters include the indicative temperature, temperature change rate and temperature diffusion speed of each region.

[0013] According to the quantized image feature parameters and the environmental parameters of the location where the solid-state battery is located, a solid-state battery use state evaluation value is obtained, and the use state of the current solid-state battery is evaluated according to the position of the solid-state battery use state evaluation value in the preset state threshold interval.

[0014] A screening temperature is set, the image below the screening temperature in the temperature gradient image is deleted, the remaining image is recorded as a high-temperature image, the location of the high-temperature image is marked in the pre-recorded solid-state battery internal structure distribution image, and the corresponding strategy is executed according to the distribution area of the high-temperature image.

[0015] When the high-temperature image is distributed at the location of the battery core, a maintenance warning is issued, and a battery detection strategy is executed.

[0016] When the high-temperature image is distributed at the location of the line, a maintenance warning is issued, and a line detection strategy is executed.

[0017] When the high-temperature image is distributed at the location of the solid-state battery other than the location of the line or the location of the battery core, the solid-state battery use state evaluation value obtained by re-collecting the temperature distribution image of the solid-state battery during use is compared, and when the regions of the two collected results are consistent, no response is made, otherwise a verification warning is issued.

[0018] The adaptability estimation value of the solid-state battery under the current environmental parameters and the solid-state battery thermal fault prediction value corresponding to the battery adaptability estimation value are calculated in combination with the health estimation coefficient obtained in the detection strategy process and the solid-state battery use state evaluation value.

[0019] Further, the process of obtaining the image feature parameters is as follows:

[0020] According to the pre-set image segmentation size, the temperature distribution image is segmented into n image blocks of the same size, and m image points are randomly selected in the same image block. The RGB values of the image points are called, and the temperature proportion matching the RGB values of the image points is obtained from the database. The obtained value is recorded as the indicative temperature Qc of the corresponding image block.

[0021] The formula for generating the indicative temperature Qc of the corresponding image block is as follows:

[0022]

[0023] In the formula, Del represents the average color vision coefficient of all image points in the same image block, and Qed represents the preset temperature ratio. x Orn In the formula, Orn represents the index of the temperature ratio, and δ represents the correction factor of the average color vision coefficient. Del is the battery temperature influence coefficient during detection, and λ Del is the ambient temperature influence coefficient during detection, and ε1 and ε2 represent the weights of the preset battery temperature influence coefficient and the ambient temperature influence coefficient, respectively, ε1+ε2=1, and R i , G i , B i represent the RGB values of the i-th image point, Rcl j , Glid j , Bibo j represent the weights corresponding to the RGB values of the i-th image point, 0

[0024]

[0025] In the formula, W q represents the temperature change rate of the same image block, T Q represents the preset detection time, and Qc t represents the indication temperature at the end of the detection time, and Qc0 represents the indication temperature at the start of the detection time.

[0026] After the indication temperature calculation is completed, a plurality of image blocks with the same indication temperature are randomly selected and marked as starting image blocks, and the temperature at this time is recorded as the approved temperature. The indication temperatures of the image blocks around the starting image blocks are obtained, and the image blocks with indication temperatures lower than the indication temperature are recorded as to-be-detected image blocks. The time T L at which the indication temperature of the to-be-detected image block reaches the approved temperature is recorded.

[0027]

[0028] In the formula, B t represents the temperature diffusion speed, and L represents the distance between adjacent two image blocks.

[0029] Further, the environmental parameters of the position where the solid-state battery is located include air temperature, air humidity, and air thermal conductivity. The influence of the external environment of the solid-state battery is analyzed, and the specific analysis process is as follows:

[0030] The environmental adaptability information of the solid-state battery stored in the database is extracted to obtain the environmental parameter range that the solid-state battery can withstand, and further obtain the temperature range, humidity range, and air thermal conductivity range that the solid-state battery can withstand.

[0031] The temperature range tolerated by the solid-state battery is compared with the maximum indicated temperature obtained in the actual detection process, and the adaptation coefficient of the solid-state battery tolerance temperature and the maximum indicated temperature is analyzed, which is denoted as

[0032] Similarly, according to the analysis method of the fitting coefficient of the solid-state battery tolerance temperature and the maximum indicated temperature, the adaptation coefficients of the solid-state battery tolerance humidity and the tolerance air heat conductivity with the humidity and the air heat conductivity in the actual detection process are obtained, which are denoted as

[0033] Through the analysis formula The environmental parameter influence value Eltq of the solid-state battery is obtained t , wherein φ represents the correction factor of the preset environmental parameter influence value, Int represents the down rounding, W t , S t , Dr t respectively represent the weight factors of the preset temperature, humidity and air heat conductivity, and W t > S t > Dr t > 0.

[0034] Further, the solid-state battery usage state evaluation value Bauy t is generated according to the following formula:

[0035]

[0036] In the formula, Qc max represents the maximum value of the indicated temperature of all the indicated temperatures in the temperature distribution image, Qc x represents the indicated temperature of the xth image block, x is a positive integer, a1, a2 and a3 respectively represent the weights of the indicated temperature, the temperature diffusion coefficient and the environmental parameter influence coefficient, a1+a2+a3=1, and G1 represents the correction coefficient.

[0037] Further, the state threshold interval includes the state threshold upper limit Bauy ↑ and the state threshold lower limit Bauy ↓ .

[0038] When the state evaluation value Bauy t ≥ the state threshold upper limit Bauy ↑ , no response is made;

[0039] When the state threshold upper limit Bauy ↑ > the state evaluation value Bauy t ≥ the state threshold lower limit Bauy ↓ , a replacement instruction is issued;

[0040] When the state threshold lower limit Bauy↓ state evaluation value Bauy t then a replacement instruction is issued, and the state evaluation value Bauy t corresponding to the low-power output mode, the maximum output voltage and the maximum output current of the solid-state battery are limited.

[0041] Further, the specific process of executing the battery detection strategy is as follows:

[0042] The length of the monitoring period is set, the maximum and minimum values of the current and voltage in the monitoring period of the solid-state battery are obtained, and the change rates of the current and voltage in the monitoring period of the solid-state battery are obtained.

[0043] In the circuit, a preset detection load is connected, and the change rates of the current and voltage in the monitoring period are recorded according to the preset monitoring period.

[0044] According to the data obtained in the two monitoring periods, the battery core health estimation coefficient K of the solid-state battery is obtained by analyzing the formula 电池 where maxI0, minI0, maxI1, and minI1 respectively represent the maximum and minimum values of the current in the monitoring period before and after connecting the load, maxU0, minU0, maxU1, and minU1 respectively represent the maximum and minimum values of the voltage in the monitoring period before and after connecting the load, T d represents the length of the preset monitoring period, Ies and Upg respectively represent the weight factors of the current change rate and the voltage change rate, Ies>0, Upg>0, and G2 represents a correction coefficient.

[0045] Further, the specific process of executing the line detection strategy is as follows:

[0046] A preset current and voltage are input to the line, and the average temperature of the region where the line is located in the solid-state battery is recorded after T s time.

[0047] The line health estimation coefficient C of the solid-state battery is obtained by analyzing the formula 线路 where θ represents a correction factor of the preset line health estimation value, I q and I b respectively represent the line input current before and after adjustment, U q and U b respectively represent the line input voltage before and after adjustment, Ice and Uce respectively represent the weight factors of the current and voltage, and Ice+Uce=1.

[0048] ​​Further, when receiving the check warning, the shooting frequency of the solid-state battery temperature distribution image is adjusted, and the adjusted shooting frequency is greater than 5 times or more of the original shooting frequency; the coordinates of the region block corresponding to the maximum temperature in the solid-state battery temperature distribution image are obtained, and all the coordinates are substituted into the two-dimensional coordinate axis to generate a coordinate change curve; the obtained coordinate change curve is compared with the historical coordinate change curve in the database; when the coincidence degree of the two curves reaches 80% or more, no response is made; otherwise, a maintenance instruction is issued.

[0049] Further, the formula for generating adaptive estimated value is as follows:

[0050]

[0051] In the formula, Bauy x represents the adaptive estimated value, Bauy x ≥ 0, when Bauy x < 0, the solid-state battery temperature distribution image is collected;

[0052] The formula for generating the solid-state battery thermal fault prediction value is as follows:

[0053] Waty t = Bauy x × γ - Eltq t × μ

[0054] In the formula, Waty t represents the solid-state battery thermal fault prediction value, γ represents the proportion coefficient of the adaptive estimated value, μ represents the proportion coefficient of the environmental parameter influence value, and is compared with the last obtained solid-state battery thermal fault prediction value; when the difference is less than the prediction difference interval, the result is output; otherwise, the solid-state battery temperature distribution image is collected again.

[0055] Further, the evaluation threshold interval of the thermal fault prediction value is set, and the evaluation threshold interval includes the qualified threshold interval, the risk threshold interval and the cooling threshold interval;

[0056] When the thermal fault prediction value is located in the qualified threshold interval, no response is made, and the result is output;

[0057] When the fault prediction value is located in the risk threshold interval, a maintenance instruction is issued, and the auxiliary cooling system is started to cool the solid-state battery;

[0058] When the thermal fault prediction value is located in the cooling threshold interval, a danger warning is issued, the auxiliary cooling system is started to cool the solid-state battery, and the output power of the solid-state battery is lowered until the indicated temperature on the surface of the solid-state battery no longer rises.

[0059] (Three) beneficial effects

[0060] 1、The application detects the indication temperature, temperature change rate and temperature diffusion speed of the battery in use by adopting the way of thermal imaging, analyzes to obtain a state evaluation value Bauy which can intuitively judge the actual bearing capacity of the battery t , according to the position relationship between the state evaluation value Bauy t and the state threshold interval, the corresponding solution is called to provide the basis for the replacement and adjustment of the battery, and when the replacement period is missed, the service life of the battery can be extended to provide sufficient time for the replacement of the battery, secondly, the state evaluation value Bauy t is combined with the detection result obtained by the subsequent detection strategy to obtain an adaptive estimated value Bauy x , so as to comprehensively predict the adaptability of the battery, and facilitate the formulation of the corresponding battery use mode according to the bearing capacity of the battery, so that the remaining service life of the battery can be fully utilized.

[0061] 2、The application detects the distribution of the indication temperature, finds the heat source of the battery, and according to the position of the heat source, the corresponding detection strategy is called to execute, which can not only improve the detection efficiency, but also improve the accuracy of the calculation result, in addition, the result obtained by the detection strategy and the state evaluation value Bauy t are combined to comprehensively predict the service life of the battery.

[0062] 3、The application obtains the tolerance range of the battery to temperature, humidity and air heat conductivity, calculates the corresponding adaptability coefficient and environmental parameter influence value Eltq t , so as to judge the influence degree of the environment on the battery and the adaptability of the battery in the current environment, which not only facilitates the calculation of the state evaluation value Bauy t at the time of excluding the interference of the environment, but also combines the historical record of the environmental parameter influence value Eltq t to remind the user of the influence of the current environment on the battery, so that the user can intuitively understand, and can cooperate with the vehicle-mounted system to control, thereby increasing the service life of the battery.

[0063] In summary, the system designed in the application can realize the comprehensive evaluation and analysis of the battery pack, improve the energy utilization efficiency, the health management and maintenance ability of the battery, and the adaptability evaluation ability of the battery pack. BRIEF DESCRIPTION OF DRAWINGS

[0064] The above description is only a summary of the technical scheme of the application, in order to more clearly understand the technical means of the application, and can be implemented according to the content of the specification, the following is a preferred embodiment of the application and the detailed description of the drawings.

[0065] Fig. 1The overall flowchart of the embodiment of the present application is shown in the figure;

[0066] Fig. 2 The high-temperature image of the solid-state battery in the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0067] The technical problem of low operation efficiency and inaccurate prediction result of the existing solid-state battery thermal fault diagnosis method is solved by providing a solid-state battery thermal fault diagnosis and online early warning method. When the solid-state battery thermal fault diagnosis method is used, the present application detects the indication temperature, temperature change rate and temperature diffusion speed of the battery in use by using thermal imaging, and analyzes to obtain a state evaluation value Bauy t According to the positional relationship between the state evaluation value Bauy t and the state threshold interval, the corresponding solution is retrieved, thereby providing a basis for battery replacement and adjustment, and also prolonging the service life of the battery when the replacement cycle is missed, providing sufficient time for battery replacement. Secondly, the state evaluation value Bauy t is combined with the detection result obtained by subsequently executing the detection strategy to obtain an adaptive estimated value Bauy x , thereby comprehensively predicting the adaptability of the battery, facilitating the formulation of a corresponding battery use mode according to the bearing capacity of the battery, so that the remaining service life of the battery can be fully utilized. By detecting the distribution of the indication temperature to find the heat source of the battery, and retrieving and executing the corresponding detection strategy according to the position of the heat source, the detection efficiency can be improved while the accuracy of the operation result can also be improved. In addition, the service life of the battery can be comprehensively predicted according to the result obtained by executing the detection strategy and the state evaluation value Bauy t .

[0068] Embodiment 1: Please refer to Figs. 1-2 The technical solution in the embodiment of the present application is to solve the technical problem of low operation efficiency and inaccurate prediction result of the existing solid-state battery thermal fault diagnosis method in use, and the general idea is as follows:

[0069] In view of the problems in the prior art, the present application provides a solid-state battery thermal fault diagnosis and online early warning method. The method is divided into hardware (mainly thermal imaging instrument, other devices can use vehicle-mounted devices) and software two parts. The hardware and software cooperate with each other to complete the detection of the thermal fault of the battery, and can also retrieve the corresponding detection strategy according to the detection result, thereby improving the detection accuracy and facilitating subsequent maintenance. The specific structure is as follows:

[0070] Hardware:

[0071] The NV384 thermal imager is used as the collection device of the temperature distribution image of the solid-state battery. The NV384 thermal imager is a high-sensitivity uncooled detector, and has stable imaging circuit, low image noise, frame frequency, resolution, temperature measurement accuracy, power consumption, stable and reliable work, compact structure, convenient installation, external power supply and reserved RS232 communication interface, is convenient for system integration, is suitable for night auxiliary driving of various vehicles, and is connected with the vehicle-mounted system through the RS232 communication interface. The collected temperature distribution image of the solid-state battery is transmitted to the vehicle-mounted system for operation.

[0072] Installation: The NV384 thermal imager is installed at a position 20-30 cm away from the solid-state battery, and the lens of the thermal imager is tilted towards the direction of the solid-state battery. When the solid-state battery cannot be completely covered, the combined mode is adopted, that is, the number and position of the NV384 thermal imager are set to divide the solid-state battery into a corresponding number of regions, and the NV384 thermal imager is corresponded to the region to be photographed (the corresponding photographed region needs to be completely covered).

[0073] Software:

[0074] The image collected by the hardware (NV384 thermal imager) is analyzed to obtain the data brought by the image, and the condition of the battery is derived through the image data,

[0075] Image data collection: The thermal imager is used to monitor the surface temperature change of the battery in real time, and the thermal imaging image is saved at the same time interval. The saved surface thermal imaging image is processed in batches, and is cropped into images with consistent size as the input of the feature extraction network. The image needs to be normalized before input. The feature extraction network first extracts key point information from the image, converts it into a digital signal and inputs it into the network. Then, the extracted information is forward propagated to complete classification, identification and positioning. The key data in the temperature distribution image of the solid-state battery is extracted through the feature extraction network, and the image feature parameters reflecting the temperature characteristics of the battery and the temperature gradient image reflecting the temperature distribution are obtained. The image feature parameters include the indication temperature, temperature change rate and temperature diffusion speed of each region. The above feature extraction network architecture can adopt common image network architecture on the market, for example, the feature extraction network architecture is pointed out in the journal “Lithium-ion battery intelligent fault positioning technology based on thermal imaging” published by Tianjin University in February 2024.

[0076] Indicating temperature: According to the preset image segmentation size, the temperature distribution image of the solid-state battery collected by the thermal imager is divided into a plurality of small images (image blocks), then a plurality of points (image points) are randomly selected in an image block, and the RGB value of the corresponding point is extracted according to the image feature extraction network, and the temperature proportion coefficient matching the RGB value of the image point is called from the database, wherein the number of temperature distribution image segmentation is n, the number of image points randomly selected in the same image block is m, and the obtained value is imported as the indicating temperature Qc of the corresponding image block.

[0077] The average color vision coefficient Del of all image points in the same image block is obtained, and the average color vision coefficient Del is used as the color of the image block to participate in subsequent calculation. The formula based on the average color vision coefficient Del is as follows:

[0078]

[0079] In the formula, R i , G i , B i represent the RGB value of the i-th image point, Rcl j , Glid j , Bibo j represent the weight corresponding to the RGB value of the i-th image point, and 0

[0080] The obtained average color vision coefficient Del is combined with the preset temperature proportion Qed x Orn to obtain the indicating temperature Qc of the corresponding image block. The average color vision coefficient Del increases with the deepening of the color, and the value increases. The temperature proportion Qed x Orn is set according to the temperature represented by different average color vision coefficients Del. Since the image at the beginning of the temperature is cyan, it gradually changes to yellow and then to red as the temperature rises, but the color change is not obvious when the temperature continues to rise. It may change from light red to dark red. Therefore, the temperature proportion Qed x Orn is set to change in exponential form, so as to display the temperature under different colors. The specific formula is as follows:

[0081]

[0082] In the formula, Del represents the average color vision coefficient of all image points in the same image block, Qed x Orn represents the preset temperature proportion, Orn represents the index of the temperature proportion, δ represents the correction factor of the average color vision coefficient, and ψ Del is the battery temperature influence coefficient during detection, and λDel To detect the ambient temperature influence coefficient at the time, ε1 and ε2 represent the preset battery temperature influence coefficient and the weight of the ambient temperature influence coefficient respectively, ε1 + ε2 = 1, e is a natural constant.

[0083] Temperature change rate: by monitoring the temperature change of the same image block in the detection time, the temperature change of the image block in the detection process can be obtained, and the specific formula is as follows:

[0084]

[0085] In the formula, W q represents the temperature change rate of the same image block, T represents the preset detection time, Qc t represents the temperature at the end of the detection time, and Qc0 represents the temperature at the beginning of the detection time.

[0086] For example, the detection time T Q is set to 1s, the temperature Qc0 at the beginning is 23℃, the temperature Qc t at the end of the detection time is 25℃, then And by observing the positive and negative of W q , it can be judged whether the same image block is rising or falling, and the size of W q reflects the change of the battery surface temperature.

[0087] When W q ≥ 8℃ / s, the vehicle-mounted battery cooling system needs to be started to cool the solid-state battery;

[0088] When W q ≥ 14℃ / s, the vehicle-mounted battery cooling system needs to be started at the same time, and the output power of the solid-state battery needs to be reduced, and the reduction amplitude can be set by the user as needed, and the minimum needs to be reduced to 8.5% of the adjusted power, and an alarm instruction is sent to the driver.

[0089] Temperature diffusion speed: after the calculation of the temperature is completed, a plurality of image blocks with the same temperature are randomly selected and marked as starting image blocks, and the temperature at this time is recorded as the approved temperature. The temperature of the image blocks around the starting image blocks is obtained, and the image blocks with a temperature lower than the temperature are recorded as the image blocks to be measured, and the time T L at which the temperature of the image blocks to be measured reaches the approved temperature is recorded.

[0090]

[0091] In the formula, B trepresents a temperature diffusion speed, L represents a distance between two adjacent image blocks, the distance between the two adjacent image blocks is equal to a distance between two image block center points, that is, the coordinates of the two image block center points are (X0, Y0) and (X1, Y1), and then Since the size of the image segmentation is fixed, the distance between the center points of the two adjacent image blocks is also fixed, and only related to the preset segmentation size, which can be calculated by the above method.

[0092] When the starting image block is selected, a plurality of image blocks are randomly selected, and the indicative temperature of the surrounding image blocks of the selected image blocks is obtained, and the surrounding image blocks with an indicative temperature greater than that of the selected image blocks are discarded; the rest is the image block to be measured, and when a certain image block to be measured reaches the indicative temperature of the selected image block in the starting state, the counting is stopped, and the time period is T L .

[0093] In the process of solid-state battery detection, the environment around the solid-state battery also affects it, so the environment of the location where the solid-state battery is located also needs to be analyzed, and the temperature, humidity and air heat conductivity tolerance range of the solid-state battery can also be obtained., calculate the corresponding adaptation coefficient and environmental parameter influence value Eltq t , so as to judge the influence degree of the environment on the battery and the adaptation ability of the battery in the current environment, which is convenient for subsequent state evaluation value Bauy t , excluding the interference of the environment during calculation, and combining the historical record of the environmental parameter influence value Eltq t , reminding the user of the influence of the current environment on the battery, which is convenient for the user to intuitively understand, and can be matched with the vehicle-mounted system for regulation and control, thereby increasing the service life of the battery. The influence of the external environment on the solid-state battery is analyzed, and the specific analysis process is as follows:

[0094] Extract the environmental adaptability information of the solid-state battery stored in the database to obtain the environmental parameter range that the solid-state battery can tolerate, and further obtain the temperature range, humidity range and air heat conductivity range that the solid-state battery can tolerate;

[0095] Compare the temperature range that the solid-state battery can tolerate with the maximum indicative temperature obtained in the actual detection process, analyze the adaptation coefficient of the solid-state battery tolerance temperature and the maximum indicative temperature, and mark it as

[0096] Similarly, according to the analysis method of the fitting coefficient of the solid-state battery tolerance temperature and the maximum indicative temperature, the adaptation coefficients of the solid-state battery tolerance humidity and the tolerance air heat conductivity with the humidity and the air heat conductivity in the actual detection process are obtained, respectively. Marked as

[0097] Through the analysis formula Obtaining the environmental parameter influence value Eltq of the solid-state battery t , wherein φ represents a correction factor of the preset environmental parameter influence value, Int represents rounding down, W t , S t , Dr t respectively represent weight factors of preset temperatures, humidities and air thermal conductivities, and W t > S t > Dr t > 0.

[0098] Evaluating the use state of the solid-state battery: according to the quantified image feature parameters and the environmental parameters of the location where the solid-state battery is located, obtaining a solid-state battery use state evaluation value, and generating a solid-state battery use state evaluation value Bauy t The formula used is as follows:

[0099]

[0100] In the formula, Qc max represents the maximum value of the indicated temperature in the temperature distribution image, Qc x represents the indicated temperature of the xth image block, x is a positive integer, a1, a2 and a3 respectively represent the weights of the indicated temperature, the temperature diffusion coefficient and the environmental parameter influence coefficient, a1+a2+a3=1, G1 represents a correction coefficient, represents the temperature diffusion coefficient.

[0101] Since the use of the battery is affected by the environment, the environmental parameter influence value Eltq t is removed from the state evaluation value Bauy t , so that the state evaluation value Bauy t obtained can best reflect the current use state of the solid-state battery.

[0102] After obtaining the state evaluation value Bauy t reflecting the use of the battery, it needs to be compared with the preset state threshold interval, and according to the position of the solid-state battery use state evaluation value in the preset state threshold interval, the use of the current solid-state battery is evaluated, wherein the state threshold interval includes a state threshold upper limit Bauy ↑ (qualified line) and a state threshold lower limit Bauy ↓ (scrap line);

[0103] When the state evaluation value Bauy t is greater than or equal to the state threshold upper limit Bauy ↑ , it indicates that the service life and use state of the battery meet the use requirements, and it can continue to be used normally, and no response is made.

[0104] When the upper limit of the state threshold is Bauy ↑ > Status assessment value Bauy t ≥ lower limit of state threshold Bauy ↓ If the battery is close to reaching its end-of-life and needs to be replaced promptly, a replacement instruction will be issued if the battery can still be used at this stage.

[0105] When the lower limit of the state threshold is Bauy ↓ > Status assessment value Bauy t At this point, the battery's lifespan has reached its critical value, and failure to replace it in time could lead to danger. Therefore, immediate replacement is necessary, and the battery's condition assessment value (Bauy) should be retrieved. t The corresponding low-power output mode (which uses a preset method to set the upper limit of the solid-state battery's output so that when it is used in a low-power state, it must be reduced to at least 8.5% of the original power) limits the maximum output voltage and maximum output current of the solid-state battery, thereby ensuring that the solid-state battery does not operate under high load.

[0106] Thermal fault location filtering: Set a filtering temperature, delete all images in the temperature gradient map below the filtering temperature, and record the remaining images as high-temperature images, such as... Fig. 2 As shown, the location of the high-temperature image is marked on the pre-recorded internal structure distribution map of the solid-state battery, and the corresponding strategy is executed according to the distribution area of ​​the high-temperature image. The internal structure distribution map of the solid-state battery adopts a partitioning method, and the location of the circuit and the location of the battery cell are marked with different colored lines on the surface of the solid-state battery to facilitate subsequent differentiation. The corresponding strategy is executed according to the distribution area of ​​the high-temperature image in the following four cases:

[0107] Scenario 1: When the high-temperature images are distributed at the location of the battery cells, it indicates that the temperature rise is generated by the battery cells. Furthermore, the color differences in the images indicate that the temperature of battery cells at different locations varies. For battery cells with higher temperatures, a battery maintenance warning is issued, and a battery detection strategy is executed. The specific process of the battery detection strategy is as follows:

[0108] Set the duration of the monitoring period, obtain the maximum and minimum values ​​of current and voltage within the solid-state battery monitoring period, and obtain the rate of change of current and voltage within the solid-state battery monitoring period;

[0109] A preset detection load is connected to the circuit, and the circuit is re-monitored according to a preset monitoring cycle. The rate of change of current and voltage within the monitoring cycle is recorded.

[0110] Based on the data obtained from the two monitoring cycles, the formula was analyzed. Obtain the cell health prediction coefficient K of the solid-state battery.电池 wherein maxI0, minI0, maxI1, minI1 represent the maximum and minimum values of the current in the monitoring period before and after the access load, maxU0, minU0, maxU1, minU1 represent the maximum and minimum values of the voltage in the monitoring period before and after the access load, T d represents the length of the preset monitoring period, Ies, Upg represent the weight factors of the current change rate and the voltage change rate, Ies>0, Upg>0, G2 represents the correction coefficient.

[0111] Since the health estimation coefficient K 电池 and the state evaluation value Bauy t are obtained, the service life of the battery core and the solid-state battery is evaluated respectively.

[0112] Case two: when the high-temperature image is distributed at the position of the line, it indicates that the temperature rise point is caused by the line, such as short circuit, line aging, etc., and the line needs to be repaired in time, then a repair warning is issued, and a line detection strategy is executed, and the specific process of executing the line detection strategy is as follows:

[0113] a preset current and voltage are input to the line, and the average temperature of the area where the line is located in the solid-state battery is recorded after T s time;

[0114] The line health estimation coefficient C of the solid-state battery is obtained by using the analysis formula 线路 wherein θ represents the correction factor of the preset line health estimation value, I q , I b represent the input current of the line before and after adjustment, U q , U b represent the input voltage of the line before and after adjustment, Ice, Uce represent the weight factors of the current and voltage, Ice+Uce=1.

[0115] The result obtained is compared with the standard numerical value, including the upper limit threshold and the lower limit threshold of the loss, so as to know whether the line is damaged, and the specific process is as follows:

[0116] When the line is normal, a preset current and voltage are input to the line, and the line health estimation coefficient C′ 线路 is obtained after T s time, while when C′ 线路 -C 线路 is greater than the upper limit threshold of the loss, it indicates that the influence of temperature on the line can be ignored, and no response is made;

[0117] When the lower limit threshold of the loss is greater than C′ 线路 -C 线路≥ the lower threshold of loss, it means that the current temperature has caused interference to the line and the data calculation of the circuit board needs to be reduced;

[0118] When C' 线路 -C 线路 < the lower threshold of loss, it means that the line has been irreversibly damaged, and the use of solid-state batteries needs to be stopped for timely repair.

[0119] Case three: when the high temperature image is distributed in the line and the battery core position in the solid-state battery, the battery detection strategy and the line detection strategy are executed at the same time.

[0120] Case four: when the high temperature image is distributed in the non-line position or the battery core position in the solid-state battery, the solid-state battery temperature distribution image is reacquired, and the solid-state battery usage state evaluation values obtained by two acquisitions are compared. When the regions of the two acquisition results are consistent, no response is made, otherwise a verification warning is issued.

[0121] When the regions of the two acquisition results are consistent, the image needs to be sent to the background management system for manual review by the staff.

[0122] When receiving the verification warning, adjust the shooting frequency of the solid-state battery temperature distribution image, and the adjusted shooting frequency needs to be more than 5 times the original shooting frequency;

[0123] Obtain the coordinates of the region block corresponding to the maximum temperature in the solid-state battery temperature distribution image, and substitute all the coordinates into the two-dimensional coordinate axis to generate a coordinate change curve.

[0124] Compare the obtained coordinate change curve with the historical coordinate change curve in the database; when the coincidence degree of the two curves reaches 80% or more, no response is made, and the coordinate change curve obtained this time is stored in the database as a historical coordinate change curve; otherwise, an overhaul instruction is issued.

[0125] Battery adaptability prediction: combined with the health estimation coefficient obtained in the detection strategy process and the solid-state battery usage state evaluation value, the adaptability estimation value of the solid-state battery under the current environmental parameters is calculated. Since the detection strategy is different, the formula for calculating the adaptability estimation value will also be different. The formula for generating the adaptability estimation value is as follows:

[0126]

[0127] In the formula, Bauy x represents the adaptability estimation value, Bauy x ≥ 0, when Bauy x < 0, the solid-state battery temperature distribution image is collected.

[0128] wherein, when the battery detection strategy and the line detection strategy are performed, the average of the adaptability estimates obtained by the two strategies, for example: But when the Bauy x obtained by the battery detection strategy is greater than the Bauy x obtained by the line detection strategy by more than an error threshold, a maintenance instruction is issued.

[0129] Then, from the battery adaptability estimate, the solid-state battery thermal fault prediction value corresponding thereto is calculated, and the obtained solid-state battery thermal fault prediction value Waty t is compared with the evaluation threshold interval. The formula on which the solid-state battery thermal fault prediction value is generated is as follows:

[0130] Waty t = Bauy x × γ - Eltq t × μ

[0131] In the formula, Waty t represents the solid-state battery thermal fault prediction value, γ represents the proportional coefficient of the adaptability estimate, μ represents the proportional coefficient of the environmental parameter influence value, and is compared with the last obtained solid-state battery thermal fault prediction value, i.e. the solid-state battery thermal fault prediction value Waty t of this time - the last obtained solid-state battery thermal fault prediction value Waty t < the prediction difference interval, the result is output; otherwise, the solid-state battery temperature distribution image is reacquired.

[0132] The evaluation threshold interval of the thermal fault prediction value is set, and the evaluation threshold interval includes a qualified threshold interval, a risk threshold interval and a cooling threshold interval;

[0133] Wherein, when the thermal fault prediction value is located in the qualified threshold interval, no response is made, and the result is output;

[0134] When the fault prediction value is located in the risk threshold interval, a maintenance instruction is issued, and the auxiliary cooling system is started to cool the solid-state battery;

[0135] When the thermal fault prediction value is located in the cooling threshold interval, a danger warning is issued, the auxiliary cooling system is started to cool the solid-state battery, and the output power of the solid-state battery is lowered until the indication temperature on the surface of the solid-state battery no longer rises.

[0136] It should be noted that the above-mentioned embodiments are merely used to clearly illustrate the technical solutions of the present application, and should not be construed as limitations to the present application. Based on the above-mentioned embodiments, those skilled in the art can make other variations or modifications without departing from the spirit of the present application. The present application is not required to enumerate all of the embodiments, and the variations or modifications made without departing from the spirit of the present application should fall within the scope of the present application.

Claims

1. A method for solid-state battery thermal fault diagnosis and online early warning, characterized in that, The method comprises: After the collected solid-state battery temperature distribution image is normalized, it is input into a feature extraction network to generate an image feature parameter and a temperature gradient map, wherein the image feature parameter comprises an indicating temperature, a temperature change rate and a temperature diffusion speed of each region; According to the quantized image feature parameter and the environmental parameter of the position where the solid-state battery is located, a solid-state battery use state evaluation value is obtained, and then the use state of the current solid-state battery is evaluated according to the position of the preset state threshold interval where the solid-state battery use state evaluation value is located, wherein the environmental parameter of the position where the solid-state battery is located comprises air temperature, air humidity and air thermal conductivity; A screening temperature is set, the image below the screening temperature in the temperature gradient map is deleted, the remaining image is recorded as a high-temperature image, the position where the high-temperature image is located is marked in the pre-recorded solid-state battery internal structure distribution map, and the corresponding strategy is executed according to the distribution area of the high-temperature image; When the high-temperature image is distributed at the position of the battery core, a maintenance warning is issued, and a battery detection strategy is executed; When the high-temperature image is distributed at the position of the circuit, a maintenance warning is issued, and a circuit detection strategy is executed; When the high-temperature image is distributed at the position other than the circuit or the battery core in the solid-state battery, the solid-state battery use process temperature distribution image is re-collected, the solid-state battery use state evaluation values obtained by two times of collection are compared, when the regions where the results obtained by two times of collection are consistent, no response is made, otherwise a verification warning is issued; The adaptability estimation value of the solid-state battery under the current environmental parameter and the solid-state battery thermal failure prediction value corresponding to the battery adaptability estimation value are calculated in combination with the health estimation coefficient obtained in the detection strategy process and the solid-state battery use state evaluation value.

2. The solid-state battery thermal fault diagnosis and online early warning method of claim 1, wherein: The process of obtaining the image feature parameter is as follows: According to the pre-set image segmentation size, the temperature distribution image is segmented into n image blocks of the same size, m image points are randomly selected in the same image block, the RGB values of the image points are called, the temperature proportion matching the RGB values of the image points is obtained from the database, and the obtained value is recorded as the indicating temperature Qc of the corresponding image block; The formula for generating the indicating temperature Qc of the corresponding image block is as follows: wherein Del represents the average color vision coefficient of all image points in the same image block, Qed x Orn represents the preset temperature ratio, Orn represents the index of the temperature ratio, δ represents the correction factor of the average color vision coefficient, ψ Del is the battery temperature influence coefficient at detection, λ Del is the ambient temperature influence coefficient at detection, ε1 and ε2 respectively represent the weights of the preset battery temperature influence coefficient and the ambient temperature influence coefficient, ε1+ε2=1, R i , G i , B i represents the RGB value of the i-th image point, Rcl j , Glid j , Bibo j represents the weight corresponding to the RGB value of the i-th image point, 0 In the formula, W q represents the temperature change rate of the same image block, T Q represents the preset detection time, Qc t represents the indication temperature at the end of the detection time, Qc0represents the indication temperature at the start of the detection time; After the indication temperature calculation is completed, a plurality of image blocks with the same indication temperature are randomly selected and marked as starting image blocks, the temperature at this time is recorded as the approved temperature, the indication temperatures of the image blocks around the starting image blocks are obtained, the image blocks with the indication temperatures lower than the indication temperature are recorded as to-be-measured image blocks, and the time T at which the indication temperature of the to-be-measured image block reaches the approved temperature is recorded L ; In the formula, B t represents the temperature diffusion speed, and L represents the distance between two adjacent image blocks.

3. The solid-state battery thermal fault diagnosis and online early warning method of claim 2, wherein: The influence of the external environment of the solid-state battery is analyzed, and the specific analysis process is as follows: The environmental adaptability information of the solid-state battery stored in the database is extracted to obtain the environmental parameter range that the solid-state battery can withstand, and further obtain the temperature range, humidity range and air thermal conductivity range that the solid-state battery can withstand; The temperature range tolerated by the solid-state battery is compared with the maximum indicated temperature obtained during the actual detection process, and the adaptation coefficient of the temperature range tolerated by the solid-state battery and the maximum indicated temperature is analyzed, which is denoted as Similarly, according to the analysis method of the fitting coefficient of the temperature resistance of the solid-state battery and the maximum indicated temperature, the adaptation coefficients of the humidity resistance and the air heat conductivity resistance of the solid-state battery to the humidity and the air heat conductivity in the actual detection process are obtained, which are respectively denoted as By analyzing the formula obtain the environmental parameter influence value Eltq of the solid-state battery t wherein φ represents a correction factor of the preset environmental parameter influence value, Int represents a downward rounding, W t , S t , Dr t respectively represent the weight factors of the preset temperature, humidity and air thermal conductivity, and W t >S t >Dr t >0.

4. The solid-state battery thermal fault diagnosis and online early warning method of claim 3, wherein: Generating a solid-state battery usage state evaluation value Bauy t The formula on which this is based is as follows: In the formula, Qc max represents the maximum value of the indicated temperature in the temperature distribution image, Qc x represents the indicated temperature of the xth image block, x is a positive integer, a1, a2, and a3 respectively represent the weights of the indicated temperature, the temperature diffusion coefficient, and the environmental parameter influence coefficient, a1+a2+a3=1, and G1 represents a correction coefficient.

5. The solid-state battery thermal fault diagnosis and online early warning method of claim 4, wherein: The state threshold interval comprises a state threshold upper limit Bauy ↑ and a state threshold lower limit Bauy ↓ ; When the state assessment value Bauy t ≥ the upper state threshold Bauy ↑ then no response is made; When the state threshold upper limit Bauy ↑ the state assessment value Bauy t the state threshold lower limit Bauy ↓ a replacement instruction is issued; When the lower limit of the state threshold is Bauy ↓ > Status assessment value Bauy t If the condition is met, a replacement instruction is issued, and the status assessment value Bauy is retrieved. t The corresponding low-power output mode limits the maximum output voltage and maximum output current of the solid-state battery.

6. The solid-state battery thermal fault diagnosis and online early warning method of claim 4, wherein: The specific process of executing the battery detection strategy is as follows: The length of the monitoring period is set, the maximum and minimum values of the current and voltage in the solid-state battery monitoring period are obtained, and the change rates of the current and voltage in the solid-state battery monitoring period are obtained; A preset detection load is connected in the circuit, and the change rates of the current and voltage in the monitoring period are recorded according to the preset monitoring period; According to the data obtained in two monitoring periods, the battery core health estimation coefficient K of the solid-state battery is obtained by analyzing the formula According to the data obtained in two monitoring periods, the battery core health estimation coefficient K of the solid-state battery is obtained by analyzing the formula 电池 , wherein maxI0, minI0, maxI1, minI1 respectively represent the maximum and minimum values of the current in the monitoring period before and after the access load, maxU0, minU0, maxU1, minU1 respectively represent the maximum and minimum values of the voltage in the monitoring period before and after the access load, T d represents the length of the preset monitoring period, Ies, Upg respectively represent the weight factors of the current change rate and the voltage change rate, Ies>0, Upg>0, G2 represents the correction coefficient.

7. The solid-state battery thermal fault diagnosis and online early warning method of claim 6, wherein: The specific process of executing the circuit detection strategy is as follows: A preset current and voltage are input to the line, and this is continued for T s After a certain period of time, the average temperature of the area in which the line is located in the solid-state battery is recorded; By using the analysis formula The line health estimation coefficient C of the solid-state battery is obtained 线路 where θ represents a correction factor of a preset line health estimation value, I q , I b respectively represent the line input current before and after adjustment, U q , U b respectively represent the line input voltage before and after adjustment, Ice, Uce respectively represent weight factors of the current and the voltage, and Ice+Uce=1.

8. The solid-state battery thermal fault diagnosis and online early warning method of claim 1, wherein: Upon receiving the check warning, the shooting frequency of the solid-state battery temperature distribution image is adjusted, and the adjusted shooting frequency is greater than 5 times or more than the original shooting frequency; the coordinates of the region block corresponding to the maximum temperature in the solid-state battery temperature distribution image are obtained, and all the coordinates are substituted into the two-dimensional coordinate axis to generate a coordinate change curve; the obtained coordinate change curve is compared with the historical coordinate change curve in the database; wherein, when the coincidence degree of the two curves reaches 80% or more, no response is made; otherwise, a maintenance instruction is issued.

9. The solid-state battery thermal fault diagnosis and online early warning method of claim 7, wherein: The formula for generating adaptive estimated value is as follows: Bauy x represents an adaptive estimate, Bauy x ≥ 0, when Bauy x < 0, then a solid-state battery temperature distribution image is acquired; The formula for generating solid-state battery thermal fault prediction value is as follows: Waty t = Bauy x x γ - Eltq t x μ In the formula, Waty t The solid-state battery thermal failure prediction value is compared with the last obtained solid-state battery thermal failure prediction value, and when the difference is less than the prediction difference interval, the result is output; otherwise, the solid-state battery temperature distribution image is reacquired.

10. The solid-state battery thermal fault diagnosis and online early warning method of claim 9, wherein: Set the evaluation threshold interval of the thermal fault prediction value, which includes the qualified threshold interval, the risk threshold interval and the cooling threshold interval; Wherein, when the thermal fault prediction value is located in the qualified threshold interval, no response is made, and the result is output; When the fault prediction value is located in the risk threshold interval, a maintenance instruction is issued, and the auxiliary cooling system is started to cool the solid-state battery; When the thermal fault prediction value is located in the cooling threshold interval, a danger warning is issued, the auxiliary cooling system is started to cool the solid-state battery, and the output power of the solid-state battery is reduced until the indicated temperature on the surface of the solid-state battery no longer rises.

Images