A power automobile battery pack disassembly system and method based on thermal runaway early warning

Abstract

The present application relates to the technical field of battery pack, and specifically discloses a power automobile battery pack disassembly system and disassembly method based on thermal runaway early warning, which utilizes temperature field to locate high-temperature area, combines visual recognition of body surface bulging deformation slope and CO concentration and pressure parameters obtained by pre-embedded sensors, adopts grid segmentation to quantify thermal runaway risk, extracts CO concentration high-frequency features through wavelet packet decomposition, calculates pressure mutation gradient through Kalman filtering, generates gas dynamic mutation coefficient, normalizes and fuses temperature, deformation, gas and pressure parameters to form a comprehensive early warning coefficient, disassembles low-risk areas based on modular layout, marks no-disassembly path in high-temperature area, and forms a dynamic disassembly strategy, which can effectively solve the defects of artificial experience dependence, risk identification lag and fixed process rigidity in traditional disassembly, realize precise early warning of thermal runaway risk and optimization of disassembly path, and significantly improve the safety and efficiency of power automobile battery pack disassembly.

Description

Technical Field

[0001] This invention relates to the field of battery pack technology, and in particular to a power vehicle battery pack disassembly system and disassembly method based on thermal runaway early warning. Background Technology

[0002] With the popularization of new energy vehicles, the safety and recycling efficiency of power battery packs have become the focus of industry attention. Currently, power battery packs need to be disassembled after retirement or failure to recycle valuable materials. However, during the disassembly process, combustion or explosion may occur due to thermal runaway inside the battery, causing serious safety hazards.

[0003] Existing disassembly processes rely heavily on human experience, requiring operators to be in close contact with the battery pack. However, risks such as combustion and release of toxic gases caused by thermal runaway are difficult to avoid. For example, if a short circuit occurs in a certain area of ​​the battery during disassembly and the temperature rises suddenly, it is difficult for personnel to quickly identify and evacuate, which can easily lead to personal injury. A method for disassembling electric vehicle battery packs based on thermal runaway early warning is needed to solve the above problems. Summary of the Invention

[0004] The purpose of this invention is to provide a power vehicle battery pack disassembly system and disassembly method based on thermal runaway early warning, so as to solve the technical problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for disassembling a power vehicle battery pack based on thermal runaway early warning, comprising:

[0007] Acquire battery surface temperature field distribution data during the disassembly process of electric vehicle battery packs, and obtain local information of the battery pack based on the battery pack surface temperature field distribution data;

[0008] Local monitoring data of the battery pack is obtained based on the local information of the battery pack, wherein the local monitoring data includes surface feature data and internal feature data of the battery pack.

[0009] Multiple surface change rates are obtained based on the surface feature data of the battery pack, and the thermal runaway risk coefficient of the battery pack is obtained based on the multiple surface change rates.

[0010] Gas concentration data and internal pressure change rate are obtained based on the internal characteristic data of the battery pack, and the gas mutation coefficient of the battery pack is obtained based on the gas concentration data and the internal pressure change rate.

[0011] The early warning coefficient is obtained based on the battery pack thermal runaway risk coefficient and the battery pack gas mutation coefficient.

[0012] The battery pack of the electric vehicle was disassembled based on the aforementioned warning and alert coefficient, and the disassembly results were obtained.

[0013] Preferably, the step of obtaining local information of the battery pack based on the surface temperature field distribution data of the battery pack includes:

[0014] Obtain an initial infrared temperature distribution image of the battery pack surface;

[0015] The temperature extension direction and temperature change rate are obtained based on the temperature field distribution data on the surface of the battery pack.

[0016] The initial infrared temperature distribution image is segmented according to the temperature extension direction and temperature change rate to obtain multiple temperature-affected regions.

[0017] The temperatures of multiple regions are obtained based on the multiple temperature-affected regions.

[0018] Sequentially determine whether the temperature of multiple regions exceeds a preset threshold;

[0019] If the temperature of a region exceeds a preset threshold, the temperature-affected region corresponding to that temperature will be used as local information of the battery pack.

[0020] Preferably, the step of obtaining multiple surface change rates based on the battery pack surface feature data, and obtaining the battery pack thermal runaway risk coefficient based on the multiple surface change rates, includes:

[0021] The area of ​​a local region is obtained based on the local information of the battery pack, and the local region area is divided into grids according to a preset area to obtain multiple grid points;

[0022] Obtain multiple initial coordinates corresponding to multiple grid points;

[0023] Based on the surface feature data of the battery pack, obtain multiple changing coordinates of the battery surface within a preset working time.

[0024] Multiple deformation slopes are obtained between the multiple initial coordinates and the multiple changed coordinates, and the deformation slopes are used as the body surface change rate;

[0025] The surface change coefficient is obtained based on multiple surface change rates and a preset working time, and the surface change coefficient is used as the thermal runaway risk coefficient of the battery pack.

[0026] Preferably, the step of obtaining the battery pack gas mutation coefficient based on the gas concentration data and the internal pressure change rate includes:

[0027] Wavelet packet decomposition was performed on the gas concentration data to obtain multiple concentration fluctuation features in different frequency bands;

[0028] The internal pressure change rate is processed by Kalman filtering to obtain the pressure abrupt change gradient;

[0029] The battery pack gas mutation coefficient is obtained based on the pressure mutation gradient and multiple concentration fluctuation characteristics.

[0030] Preferably, the step of obtaining the early warning coefficient based on the battery pack thermal runaway risk coefficient and the battery pack gas mutation coefficient includes:

[0031] The thermal runaway risk coefficient of the battery pack is normalized to obtain the normalized value of the thermal runaway risk coefficient of the battery pack.

[0032] The gas mutation coefficient of the battery pack is normalized to obtain the normalized value of the gas mutation coefficient of the battery pack.

[0033] The warning coefficient is obtained by weighting the normalized value of the thermal runaway risk coefficient of the battery pack and the normalized value of the gas mutation coefficient of the battery pack.

[0034] Preferably, the step of disassembling the power vehicle battery pack according to the warning reminder coefficient to obtain the disassembly result includes:

[0035] Determine the relationship between the warning reminder coefficient and the preset warning reminder threshold range;

[0036] If the warning reminder coefficient is not within the preset warning reminder threshold range, and the warning reminder coefficient is greater than the maximum value of the preset warning reminder threshold range, then the local information of the battery pack is determined to be in a high-risk state, and a first warning reminder message is generated;

[0037] If the warning reminder coefficient is within the preset warning reminder threshold range, the local information of the battery pack is determined to be in a medium-risk state, and a second warning reminder message is generated;

[0038] If the warning reminder coefficient is not within the preset warning reminder threshold range, and the warning reminder coefficient is less than the minimum value of the preset warning reminder threshold range, then the local information of the battery pack is determined to be in a low-risk state, and a third warning reminder message is generated;

[0039] Obtain the disassembly path of the electric vehicle battery pack;

[0040] The disassembly path of the power vehicle battery pack is marked according to the first warning reminder information, the second warning reminder information and the third warning reminder information to obtain the marked disassembly path;

[0041] The battery pack of the electric vehicle is disassembled according to the marked disassembly path.

[0042] This application also provides a power vehicle battery pack disassembly system based on thermal runaway early warning, including:

[0043] The first acquisition module is used to acquire battery surface temperature field distribution data during the disassembly process of the electric vehicle battery pack, and to acquire local information of the battery pack based on the battery pack surface temperature field distribution data.

[0044] The second acquisition module is used to acquire local monitoring data of the battery pack based on the local information of the battery pack, wherein the local monitoring data includes surface feature data of the battery pack and internal feature data of the battery pack.

[0045] The third acquisition module is used to acquire multiple surface change rates based on the surface feature data of the battery pack, and to acquire the thermal runaway risk coefficient of the battery pack based on the multiple surface change rates.

[0046] The fourth acquisition module is used to acquire gas concentration data and internal pressure change rate based on the internal characteristic data of the battery pack, and to acquire the gas mutation coefficient of the battery pack based on the gas concentration data and the internal pressure change rate.

[0047] The fifth acquisition module is used to acquire the early warning and reminder coefficient based on the thermal runaway risk coefficient of the battery pack and the gas mutation coefficient of the battery pack;

[0048] The disassembly module is used to disassemble the power vehicle battery pack according to the warning reminder coefficient and obtain the disassembly results.

[0049] Preferably, the first acquisition module includes:

[0050] The first acquisition unit is used to acquire an initial infrared temperature distribution image of the battery pack surface;

[0051] The second acquisition unit is used to acquire the temperature extension direction and single battery pack chain reaction information based on the temperature field distribution data on the surface of the battery pack, and to acquire the temperature change rate based on the single battery pack chain reaction information.

[0052] The segmentation unit is used to segment the initial infrared temperature distribution image according to the temperature extension direction and temperature change rate to obtain multiple temperature-affected regions;

[0053] The third acquisition unit is used to acquire the temperature of multiple regions corresponding to the multiple temperature-affected regions.

[0054] A judgment unit is used to sequentially determine whether the temperature of the multiple regions exceeds a preset threshold.

[0055] If the temperature of a region exceeds a preset threshold, the temperature-affected region corresponding to that temperature will be used as local information of the battery pack.

[0056] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0057] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0058] The beneficial effects of this application are as follows: This invention identifies local high-temperature areas through temperature field distribution, and constructs a multi-dimensional monitoring system by combining surface deformation data obtained from visual sensors with internal gas concentration and pressure data collected from pre-embedded sensors. It quantifies the thermal runaway risk coefficient using grid segmentation and deformation slope calculation, extracts high-frequency fluctuation features of CO concentration using wavelet packet decomposition, and obtains abrupt change gradients by combining pressure data with Kalman filtering to comprehensively generate a gas abrupt change coefficient. Through normalization processing, temperature, deformation, gas, and pressure data are fused into a warning coefficient, and three levels of warning (red, yellow, and green) are established based on thresholds. According to the warning level and battery modular layout, low-risk areas are prioritized for disassembly, while prohibited disassembly paths are marked in high-temperature areas. This effectively solves the defects of traditional disassembly, such as reliance on manual experience, delayed risk identification, and rigid fixed processes, achieving accurate early warning of thermal runaway risk and optimization of disassembly paths, significantly improving the safety and efficiency of electric vehicle battery pack disassembly. Attached Figure Description

[0059] Figure 1 This is a schematic diagram of a method flow according to an embodiment of this application.

[0060] Figure 2 This is a schematic diagram of the system structure according to an embodiment of this application.

[0061] Figure 3 This is a schematic diagram of the internal structure of a computer device according to an embodiment of this application.

[0062] Figure 4 This is a schematic diagram of a battery pack disassembly device according to an embodiment of this application.

[0063] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0064] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0065] like Figure 1 As shown, this application provides a method for disassembling a power vehicle battery pack based on thermal runaway early warning, including:

[0066] S1. Obtain battery surface temperature field distribution data during the disassembly process of the power vehicle battery pack, and obtain local information of the battery pack based on the battery pack surface temperature field distribution data;

[0067] S2. Obtain local monitoring data of the battery pack based on the local information of the battery pack, wherein the local monitoring data includes surface feature data of the battery pack and internal feature data of the battery pack;

[0068] S3. Obtain multiple surface change rates based on the battery pack surface feature data, and obtain the battery pack thermal runaway risk coefficient based on the multiple surface change rates.

[0069] S4. Obtain gas concentration data and internal pressure change rate based on the internal characteristic data of the battery pack, and obtain the gas mutation coefficient of the battery pack based on the gas concentration data and the internal pressure change rate.

[0070] S5. Obtain the early warning coefficient based on the battery pack thermal runaway risk coefficient and the battery pack gas mutation coefficient;

[0071] S6. Disassemble the power vehicle battery pack according to the aforementioned warning and reminder coefficient, and obtain the disassembly results.

[0072] As described in steps S1-S6 above, the existing disassembly process relies heavily on human experience, requiring operators to be in close contact with the battery pack. Risks such as combustion and release of toxic gases caused by thermal runaway are difficult to avoid. For example, if a short circuit occurs in a certain area of ​​the battery during disassembly and the temperature rises suddenly, it is difficult for personnel to quickly identify and evacuate, which can easily lead to personal injury. This invention first obtains the battery surface temperature field distribution data during the disassembly of the power vehicle battery pack, and obtains local information of the battery pack based on the battery pack surface temperature field distribution data. By identifying local high-temperature areas of the battery pack through the temperature field distribution data, the risk of thermal runaway can be accurately located, avoiding the limitations of traditional single-point temperature measurement.

[0073] Simultaneously, local monitoring data of the battery pack is obtained based on the local information of the battery pack. The local monitoring data includes surface feature data and internal feature data of the battery pack. By combining surface deformation (such as bulging) and internal parameters (such as gas concentration), a more comprehensive risk basis is constructed, improving the accuracy of early warning. At the same time, surface features are obtained through visual sensors, and internal features are collected in real time through pre-embedded sensors (such as pressure sensors and gas sensors), without disassembling the battery pack. Furthermore, by simultaneously collecting surface and internal data, multi-dimensional monitoring of "temperature-deformation-gas-pressure" is achieved, allowing for the early detection of complex abnormal signals before thermal runaway.

[0074] Then, multiple surface change rates are obtained based on the surface feature data of the battery pack, and the thermal runaway risk coefficient of the battery pack is obtained based on the multiple surface change rates. By gridding and deformation slope calculation, the surface deformation is transformed into a quantifiable risk coefficient to avoid subjective judgment errors.

[0075] Secondly, gas concentration data and internal pressure change rate are obtained based on the internal characteristic data of the battery pack. The gas mutation coefficient of the battery pack is then obtained based on the gas concentration data and internal pressure change rate. Through wavelet packet decomposition and Kalman filtering, gas concentration fluctuation features and pressure mutation gradient are extracted to quantify the severity of the internal reaction. Specifically, wavelet packet decomposition is performed on the CO concentration data to extract high-frequency (>10Hz) and low-frequency (<1Hz) fluctuation features. The high-frequency component reflects sudden concentration changes (such as rapid gas release in the early stages of thermal runaway). Kalman filtering is used to remove noise from the pressure data, and the pressure mutation gradient (the rate of change of internal pressure in the battery pack per unit time) is calculated. Finally, by combining the high-frequency concentration fluctuation amplitude and the pressure mutation gradient, a weighted formula is used to obtain the gas mutation coefficient. This comprehensively reflects the changes inside the battery and quantifies the severity of the internal reaction.

[0076] Next, an early warning coefficient is obtained based on the battery pack thermal runaway risk coefficient and the battery pack gas mutation coefficient. Through normalization and weighted calculation, multi-dimensional data such as temperature, deformation, gas, and pressure are integrated into a single early warning indicator. For example, the thermal runaway risk coefficient (range 0-10) and the gas mutation coefficient (range 0-1) are normalized to the [0,1] interval (e.g., the thermal runaway risk coefficient 4 is normalized to 0.4, and the gas mutation coefficient 0.7 is normalized to 0.7). Then, according to the preset weights (e.g., thermal runaway risk weight 0.5, gas mutation weight 0.5), the early warning coefficient T(X) = 0.4 × 0.5 + 0.7 × 0.5 = 0.55 is calculated.

[0077] Finally, the battery pack of the electric vehicle is disassembled according to the aforementioned warning and reminder coefficients to obtain the disassembly results. The danger levels are divided according to the warning and reminder coefficients: High danger (coefficient > 0.8), generating a red warning (first warning information); Medium danger (0.5 ≤ coefficient (0.51-0.79) ≤ 0.8), generating a yellow warning (second warning information); Low danger (coefficient < 0.5), generating a green warning (third warning information). Combined with the battery pack structure design (such as a modular layout, obtained according to a preset battery pack structure design layout), "Disassembly Prohibited Priority" is marked on the disassembly path corresponding to high-risk areas (such as local high-temperature areas). Low-risk areas are prioritized to avoid triggering thermal runaway. This effectively solves the defects of traditional disassembly, such as reliance on manual experience, delayed risk identification, and rigid fixed processes. It achieves accurate warning of thermal runaway risk and optimization of the disassembly path, significantly improving the safety and efficiency of electric vehicle battery pack disassembly.

[0078] In one embodiment, step S1, which involves obtaining local information of the battery pack based on the surface temperature field distribution data of the battery pack, includes:

[0079] S101. Obtain the initial infrared temperature distribution image of the battery pack surface;

[0080] S102. Obtain the temperature extension direction and single-cell pack chain reaction information based on the temperature field distribution data on the surface of the battery pack, and obtain the temperature change rate based on the single-cell pack chain reaction information.

[0081] S103. The initial infrared temperature distribution image is segmented according to the temperature extension direction and temperature change rate to obtain multiple temperature-affected regions.

[0082] S104. Obtain the corresponding temperatures of multiple regions based on the multiple temperature-affected regions;

[0083] S105. Sequentially determine whether the temperature of the multiple regions exceeds a preset threshold.

[0084] If the temperature of a region exceeds a preset threshold, the temperature-affected region corresponding to that temperature will be used as local information of the battery pack.

[0085] As described in steps S101-S105 above, since the present invention first obtains an initial infrared temperature distribution image of the battery pack surface, and obtains the global temperature data of the battery pack surface through the initial infrared temperature distribution image (instead of single-point temperature measurement), it avoids missed detections caused by manual touch or visual inspection. It is especially suitable for risk pre-assessment before disassembly. At the same time, it can also identify abnormal temperature areas on the battery pack surface, segment temperature-affected areas, accurately locate potential thermal runaway risk points, and avoid the risk of manual contact with high-temperature components during disassembly. The global temperature distribution provides basic data for subsequent analysis.

[0086] Then, based on the temperature field distribution data on the battery pack surface, the temperature extension direction and single-cell pack chain reaction information are obtained, and the temperature change rate is obtained based on the single-cell pack chain reaction information. By analyzing the temperature field data, the diffusion direction of the high-temperature area (such as spreading from a single cell to an adjacent cell) and the heating rate are determined. At the same time, it is determined whether thermal runaway is in the diffusion stage, distinguishing between local overheating and global runaway, providing a dynamic basis for risk level assessment. Secondly, the continuous temperature field is divided into independent "temperature influence areas" (such as high-temperature core area, gradual diffusion area, and safe area), which facilitates subsequent targeted monitoring and dismantling planning, and replaces the subjectivity of manually delineating risk areas. Automated segmentation is achieved through preset rules (such as temperature difference threshold and regional connectivity), improving efficiency and consistency.

[0087] In the specific implementation process, the region growing algorithm uses temperature anomaly points (such as pixels with a temperature > 80℃) as seed points to expand to adjacent pixels. If the temperature difference between the adjacent pixels and the seed point is < 5℃, they are merged into the same region. For example, if the seed point temperature is 85℃, the surrounding pixels with a temperature of 80-84℃ are merged into a high temperature region, and the pixels with a temperature of 40-79℃ are classified as a medium temperature region.

[0088] Next, the initial infrared temperature distribution image is segmented according to the temperature extension direction and temperature change rate to obtain multiple temperature-affected regions. Based on the temperature extension direction and change rate, the infrared image is segmented into different temperature-affected regions (such as high-temperature core region, temperature diffusion region, and normal region). At the same time, the continuous temperature field is discretized into independent regions that can be analyzed, which facilitates targeted monitoring of the temperature evolution of each region. Furthermore, the battery pack is composed of multiple single cells connected in series / parallel. After segmentation, risk units can be accurately located, avoiding "one-size-fits-all" disassembly.

[0089] Next, based on the multiple temperature-affected areas, the corresponding temperatures of multiple areas are obtained. These multiple area temperatures can provide numerical basis for subsequent threshold determination and distinguish the risk levels of different areas.

[0090] Finally, it is determined whether the temperatures of multiple regions exceed preset thresholds. If a region temperature exceeds a preset threshold, the temperature-affected area corresponding to that region temperature is used as local information of the battery pack. The temperature of each region is compared with the preset threshold (e.g., 60°C), and regions with abnormal temperatures are selected as local information of the battery pack. This allows for the rapid identification of dangerous areas that require key monitoring, enabling focused resource analysis (e.g., subsequent collection of surface and internal feature data). It also allows for quantitative comparison of temperatures in multiple regions, optimizing the disassembly path of the battery pack.

[0091] In one embodiment, step S3, which involves obtaining multiple surface change rates based on the battery pack surface feature data and obtaining a thermal runaway risk coefficient for the battery pack based on the multiple surface change rates, includes:

[0092] S301. Obtain the area of ​​a local region based on the local information of the battery pack, and divide the local region area into multiple grid points according to a preset area.

[0093] S302. Obtain multiple initial coordinates corresponding to the multiple grid points;

[0094] S303. Obtain multiple changing coordinates of the battery surface within a preset working time based on the battery pack surface feature data;

[0095] S304. Obtain multiple deformation slopes between the multiple initial coordinates and the multiple changed coordinates, and use the deformation slopes as the body surface change rate;

[0096] S305. Obtaining the body surface change coefficient based on the multiple body surface change rates and the preset working time specifically involves: obtaining the start working time and the end working time based on the preset working time.

[0097] S306. Calculate the average body surface change rate based on the multiple stated body surface change rates, the start time of work, and the end time of work, wherein the calculation formula is:

[0098]

[0099] in, Let T(B) represent the average body surface change rate, t1 represent the start time of work, t2 represent the end time of work, T(B) represent the body surface change rate, and n represent the number of body surface change rates, where n = 1, 2, 3...n;

[0100] S307. Calculate the standard body surface change rate based on the multiple stated body surface change rates and the average body surface change rate, wherein the calculation formula is:

[0101]

[0102] Where B(Z) represents the standard body surface change rate, and T(B) represents the change rate of the standard body surface. l Let N represent the rate of change in the body surface of the l-th individual, N represent the total index of the rate of change in the body surface, and l represent the index of the rate of change in the body surface. This represents the average rate of change in body surface area.

[0103] S308. Calculate the envelope instantaneous coefficient based on the standard body surface change rate and the average body surface change rate, wherein the calculation formula is:

[0104]

[0105] Where X(s) represents the coefficient of change in body surface area. B(Z) represents the average rate of change in body surface area, and B(Z) represents the standard rate of change in body surface area.

[0106] The surface change coefficient is used as the thermal runaway risk coefficient of the battery pack.

[0107] As described in steps S301-S308 above, the present invention first obtains the area of ​​a local region based on the local information of the battery pack, and then divides the local region area into multiple grid points according to a preset area. This allows the local high-temperature areas of the battery pack to be divided according to a preset area (e.g., 1 cm²). 2 The grid is divided into multiple grid points to achieve spatial discretization of deformation monitoring. Secondly, through the refined analysis of grid points, local subtle deformations (such as bulging of a single battery cell) can be captured to avoid missing early risk signals. Moreover, after gridding, each grid point corresponds to an independent coordinate, which facilitates the spatiotemporal analysis of subsequent deformation data (such as the difference in deformation rate between different grid points).

[0108] Then, multiple initial coordinates corresponding to multiple grid points are obtained, and the spatial coordinates (x, y, z) of each grid point at the initial moment of disassembly are recorded to establish the benchmark data for deformation monitoring. By comparing the initial coordinates with the subsequent changes in coordinates, the displacement of each grid point is quantified to achieve quantitative analysis of deformation.

[0109] Next, based on the battery pack surface feature data, multiple changing coordinates of the battery surface within a preset working time are obtained. Within the preset working time (e.g., 30 seconds), grid point coordinates are collected in real time to capture the time sequence characteristics of deformation during disassembly (e.g., linear growth, sudden jumps). At the same time, it can also capture the dynamic deformation process and distinguish between slow deformation (normal aging) and rapid deformation (precursor to thermal runaway).

[0110] Next, multiple deformation slopes are obtained based on multiple initial coordinates and multiple changed coordinates, and the deformation slopes are used as the surface change rate to convert deformation into a quantifiable rate indicator, which facilitates horizontal comparison of risks in different areas.

[0111] Secondly, the process of obtaining the surface change coefficient based on multiple surface change rates and a preset working time is as follows: The start and end working times are obtained based on the preset working time. Then, the average surface change rate is calculated based on the multiple surface change rates, the start and end working times. Subsequently, the standard surface change rate is calculated based on the multiple surface change rates and the average surface change rate. The standard deviation of the surface change rate is then calculated to measure the dispersion of deformation at each grid point. Finally, the envelope instantaneous coefficient is calculated based on the standard surface change rate and the average surface change rate. Thus, the envelope instantaneous coefficient is obtained by comparing the average change rate with the standard deviation, comprehensively reflecting the deformation speed and dispersion. A larger ratio indicates a faster overall deformation speed and lower dispersion (uniform risk diffusion), while a smaller ratio may indicate a risk of localized abrupt changes.

[0112] Finally, the surface change coefficient is used as the thermal runaway risk coefficient of the battery pack. Specifically, the X(s) obtained by comprehensive calculation is used as the thermal runaway risk coefficient, which is mapped to a value in the 0-1 range (e.g., through normalization). The complex deformation data is transformed into a single risk indicator, which is convenient for fusion analysis with other parameters (e.g., gas mutation coefficient).

[0113] In one embodiment, step S4, which involves obtaining the battery pack gas mutation coefficient based on the gas concentration data and the internal pressure change rate, includes:

[0114] S401. Perform wavelet packet decomposition on the gas concentration data to obtain multiple concentration fluctuation features in different frequency bands;

[0115] S402. Perform Kalman filtering on the internal pressure change rate to obtain the pressure abrupt change gradient;

[0116] S403. Obtain the battery pack gas mutation coefficient based on the pressure mutation gradient and multiple concentration fluctuation characteristics.

[0117] As described in steps S401-S403 above, the present invention first performs wavelet packet decomposition on the gas concentration data to obtain multiple concentration fluctuation features in different frequency bands. Through wavelet packet decomposition, the gas concentration signal can be decomposed into different frequency ranges (such as low frequency <1Hz, mid frequency 1-10Hz, high frequency >10Hz). Among them, low frequency fluctuations reflect slow gas leakage or background interference, while high frequency fluctuations indicate sudden gas release (such as violent chemical reactions in the early stage of thermal runaway). Secondly, compared with traditional Fourier transform, wavelet packet decomposition has better time-frequency localization ability for non-stationary signals (such as concentration fluctuations caused by random vibrations during the decomposition process) and can accurately capture transient anomalies.

[0118] The specific process for realizing multiple concentration fluctuation characteristics in different frequency bands involves first removing the mean and normalizing the original gas concentration sequence (e.g., CO concentration, unit ppm, sampling frequency 10Hz) to eliminate zero drift and dimensional effects. Then, the db4 wavelet basis (suitable for abrupt signal detection) is selected to decompose the signal into 8 sub-bands (2 3 =8), covering 0-5Hz (when the sampling frequency is 10Hz, the Nyquist frequency is 5Hz), sub-band division example: band 0: 0-0.625Hz (very low frequency, environmental interference), band 3: 1.875-2.5Hz (medium frequency, normal gas diffusion), band 7: 4.375-5Hz (high frequency, sudden release), finally, calculate the energy ratio of each band (band energy / total energy) as the concentration fluctuation characteristic;

[0119] Then, the internal pressure change rate is processed by Kalman filtering to obtain the pressure mutation gradient. Kalman filtering, through the state prediction-update framework, effectively filters out random noise in the pressure data (such as fluctuations caused by the vibration of dismantling tools), retains the true trend component, and the output pressure mutation gradient (unit: kPa / s) reflects the instantaneous rate of pressure change. It can capture rapid pressure rise events (such as the explosive generation of gas during thermal runaway) that traditional low-pass filtering cannot identify. For example, Kalman filtering, through a recursive algorithm, in a noisy environment, the measured pressure data is [100, 101, 105, 104, 110] kPa (including impulse noise), and the filtered pressure change rate sequence is [0.5, 4, -1, 6] kPa / s. It accurately identifies the 4 kPa / s mutation at time 3 (traditional moving average would show 2 kPa / s, underestimating the risk).

[0120] Finally, the gas mutation coefficient of the battery pack is obtained based on the pressure mutation gradient and multiple concentration fluctuation characteristics. This combines the frequency characteristics of gas concentration (reflecting the release mode) and the rate of pressure change (reflecting the release intensity) to construct a more comprehensive risk indicator, avoiding misjudgment based on a single parameter. Furthermore, the gas mutation coefficient integrates the "gas release amount" (concentration) and "release rate" (pressure), and is positively correlated with the severity of thermal runaway. The larger the value, the more severe the internal reaction, and it provides a basis for subsequent risk quantification.

[0121] In one embodiment, step S5, which involves obtaining the early warning coefficient based on the battery pack thermal runaway risk coefficient and the battery pack gas mutation coefficient, includes:

[0122] S501. Normalize the thermal runaway risk coefficient of the battery pack to obtain the normalized value of the thermal runaway risk coefficient of the battery pack.

[0123] S502. Normalize the gas mutation coefficient of the battery pack to obtain the normalized value of the gas mutation coefficient of the battery pack.

[0124] S503. The normalized value of the battery pack thermal runaway risk coefficient and the normalized value of the battery pack gas mutation coefficient are weighted and calculated to obtain the early warning coefficient, wherein the calculation formula is:

[0125] T(X)=F(X)*w+B(X)*(1-w);

[0126] Where T(X) represents the warning and alert coefficient, F(X) represents the normalized value of the battery pack thermal runaway risk coefficient, W represents the weight value of the normalized value of the battery pack thermal runaway risk coefficient, and B(X) represents the normalized value of the battery pack gas mutation coefficient.

[0127] As described in steps S501-S503 above, the present invention first normalizes the thermal runaway risk coefficient of the battery pack to obtain the normalized value of the thermal runaway risk coefficient of the battery pack. The thermal runaway risk coefficient (F) is usually in the range of 0-10, which is very different in dimension and scale from the gas mutation coefficient (range 0-1). After normalization, it is unified to the [0,1] interval, which is convenient for weighted calculation. In addition, the absolute value of the risk coefficient of different battery packs may fluctuate due to factors such as detection accuracy and environmental conditions. After normalization, it is converted into a relative risk level, which improves the comparability across scenarios.

[0128] Subsequently, the gas mutation coefficient of the battery pack is normalized to obtain the normalized value of the gas mutation coefficient of the battery pack. Although the gas mutation coefficient is in the range of 0-1, it may be due to the difference in sensor accuracy, which may cause distribution shift (such as the actual effective range of 0.1-0.9). After normalization, it is forcibly aligned to [0,1] to ensure that the weight is balanced with the normalized value of the battery pack thermal runaway risk coefficient.

[0129] Finally, the normalized values ​​of the battery pack thermal runaway risk coefficient and the normalized values ​​of the battery pack gas mutation coefficient are weighted and calculated to obtain the early warning coefficient. The weight allocation reflects the importance of different risk factors (for example, in the disassembly scenario, surface deformation may be easier to observe directly than gas changes), improving the pertinence of the early warning. Moreover, the weighted calculation can comprehensively reflect the overall risk level and avoid misjudgment based on a single indicator.

[0130] In one embodiment, step S6, which involves disassembling the power vehicle battery pack according to the warning and alert coefficient to obtain the disassembly result, includes:

[0131] S601. Determine the relationship between the warning reminder coefficient and the preset warning reminder threshold range;

[0132] If the warning reminder coefficient is not within the preset warning reminder threshold range, and the warning reminder coefficient is greater than the maximum value of the preset warning reminder threshold range, then the local information of the battery pack is determined to be in a high-risk state, and a first warning reminder message is generated;

[0133] If the warning reminder coefficient is within the preset warning reminder threshold range, the local information of the battery pack is determined to be in a medium-risk state, and a second warning reminder message is generated;

[0134] If the warning reminder coefficient is not within the preset warning reminder threshold range, and the warning reminder coefficient is less than the minimum value of the preset warning reminder threshold range, then the local information of the battery pack is determined to be in a low-risk state, and a third warning reminder message is generated;

[0135] S602, Obtain the disassembly path of the power vehicle battery pack;

[0136] S603. Mark the disassembly path of the power vehicle battery pack according to the first warning reminder information, the second warning reminder information and the third warning reminder information to obtain the marked disassembly path;

[0137] S604. Disassemble the power vehicle battery pack according to the marked disassembly path.

[0138] As described in steps S601-S604 above, the present invention first determines the relationship between the warning reminder coefficient and the preset warning reminder threshold range. If the warning reminder coefficient is not within the preset warning reminder threshold range and the warning reminder coefficient is greater than the maximum value of the preset warning reminder threshold range, then the local information of the battery pack is determined to be in a high-risk state, and a first warning reminder message is generated. If the warning reminder coefficient is within the preset warning reminder threshold range, then the local information of the battery pack is determined to be in a medium-risk state, and a second warning reminder message is generated. If the warning reminder coefficient is not within the preset warning reminder threshold range and the warning reminder coefficient is less than the minimum value of the preset warning reminder threshold range, then the local information of the battery pack is determined to be in a low-risk state, and a third warning reminder message is generated. In this way, the quantified risk coefficient is transformed into an executable operation instruction to guide operators or automated equipment to take different coping strategies.

[0139] Then, the disassembly path of the power vehicle battery pack is obtained. This involves sorting local information from multiple battery packs based on the relationship between warning coefficients and preset warning threshold ranges, forming the disassembly path. After the path is formed, disassembly instructions are generated, and the battery pack disassembly device (e.g., ...) is controlled according to these instructions. Figure 4As shown), the battery pack disassembly device specifically includes: 1. X-axis slide, 2. Y-axis slide, 3. electric hoist bracket, 4. operating box, 5. lifting and tilting platform, 6. frame, 7. battery pack bridging device, and 8. tensioning device. The X-axis slide in the battery pack disassembly device drives the bridging device, which then moves to directly above the battery pack (X-axis alignment). Next, the Z-axis electric cylinder drives the bridging device to descend. The grippers are adapted to the battery pack flange via a hydraulic system, and the clamping force is monitored by a pressure sensor (ensuring ≤500N to avoid damage to the casing). The lifting and tilting platform is activated, raising the battery pack 500mm from the vehicle chassis, and then rotating it 45° to expose the top casing screws directly in front of the maintenance platform.

[0140] Then, based on the first, second, and third warning messages, the disassembly path of the electric vehicle battery pack is marked to obtain the marked disassembly path. This allows for visual marking of disassembly steps corresponding to high-risk areas on the maintenance device's operating interface (touchscreen). For example, high-risk steps (such as disassembling the wiring harness in the bulging area) are marked in red with the instruction "The tensioning device must be activated to secure the wiring harness first." Medium-risk steps (such as removing ordinary screws) are marked in yellow with the prompt "Operate slowly, monitor the gripper pressure." Low-risk steps (such as removing the outer casing cover) are marked in green, allowing for automated and rapid execution.

[0141] Finally, the power vehicle battery pack is disassembled according to the marked disassembly path. This marked disassembly path effectively solves the defects of traditional disassembly, such as reliance on manual experience, delayed risk identification, and rigid fixed processes. It realizes accurate early warning of thermal runaway risk and optimization of disassembly path, significantly improving the safety and efficiency of power vehicle battery pack disassembly.

[0142] like Figure 2 As shown, this application also provides a power vehicle battery pack disassembly system based on thermal runaway early warning, including:

[0143] The first acquisition module is used to acquire battery surface temperature field distribution data during the disassembly process of the electric vehicle battery pack, and to acquire local information of the battery pack based on the battery pack surface temperature field distribution data.

[0144] The second acquisition module is used to acquire local monitoring data of the battery pack based on the local information of the battery pack, wherein the local monitoring data includes surface feature data of the battery pack and internal feature data of the battery pack.

[0145] The third acquisition module is used to acquire multiple surface change rates based on the surface feature data of the battery pack, and to acquire the thermal runaway risk coefficient of the battery pack based on the multiple surface change rates.

[0146] The fourth acquisition module is used to acquire gas concentration data and internal pressure change rate based on the internal characteristic data of the battery pack, and to acquire the gas mutation coefficient of the battery pack based on the gas concentration data and the internal pressure change rate.

[0147] The fifth acquisition module is used to acquire the early warning and reminder coefficient based on the thermal runaway risk coefficient of the battery pack and the gas mutation coefficient of the battery pack;

[0148] The disassembly module is used to disassemble the power vehicle battery pack according to the warning reminder coefficient and obtain the disassembly results.

[0149] In one embodiment, the first acquisition module includes:

[0150] The first acquisition unit is used to acquire an initial infrared temperature distribution image of the battery pack surface;

[0151] The second acquisition unit is used to acquire the temperature extension direction and single battery pack chain reaction information based on the temperature field distribution data on the surface of the battery pack, and to acquire the temperature change rate based on the single battery pack chain reaction information.

[0152] The segmentation unit is used to segment the initial infrared temperature distribution image according to the temperature extension direction and temperature change rate to obtain multiple temperature-affected regions;

[0153] The third acquisition unit is used to acquire the temperature of multiple regions corresponding to the multiple temperature-affected regions.

[0154] A judgment unit is used to sequentially determine whether the temperature of the multiple regions exceeds a preset threshold.

[0155] If the temperature of a region exceeds a preset threshold, the temperature-affected region corresponding to that temperature will be used as local information of the battery pack.

[0156] like Figure 3 As shown, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above method.

[0157] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0158] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0159] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0160] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for disassembling a power vehicle battery pack based on thermal runaway early warning, characterized in that, include: Acquire surface temperature field distribution data of the battery pack during the disassembly process of the electric vehicle battery pack, and obtain local information of the battery pack based on the surface temperature field distribution data of the battery pack; Local monitoring data of the battery pack is obtained based on the local information of the battery pack, wherein the local monitoring data includes surface feature data and internal feature data of the battery pack. Multiple surface change rates are obtained based on the battery pack surface feature data, and a thermal runaway risk coefficient for the battery pack is obtained based on the multiple surface change rates, including: The area of ​​a local region is obtained based on the local information of the battery pack, and the local region area is divided into grids according to a preset area to obtain multiple grid points; Obtain multiple initial coordinates corresponding to multiple grid points; Based on the surface feature data of the battery pack, obtain multiple changing coordinates of the battery surface within a preset working time. Multiple deformation slopes are obtained between the multiple initial coordinates and the multiple changed coordinates, and the deformation slopes are used as the body surface change rate; The surface change coefficient is obtained based on multiple surface change rates and a preset working time, and the surface change coefficient is used as the thermal runaway risk coefficient of the battery pack. Gas concentration data and internal pressure change rate are obtained based on the internal characteristic data of the battery pack, and the gas mutation coefficient of the battery pack is obtained based on the gas concentration data and the internal pressure change rate. The early warning coefficient is obtained based on the battery pack thermal runaway risk coefficient and the battery pack gas mutation coefficient. The battery pack of the electric vehicle was disassembled based on the aforementioned warning and alert coefficient, and the disassembly results were obtained.

2. The method for disassembling a power vehicle battery pack based on thermal runaway early warning as described in claim 1, characterized in that, The step of obtaining local information of the battery pack based on the surface temperature field distribution data of the battery pack includes: Obtain an initial infrared temperature distribution image of the battery pack surface; The temperature extension direction and single-cell chain reaction information are obtained based on the surface temperature field distribution data of the battery pack, and the temperature change rate is obtained based on the single-cell chain reaction information. The initial infrared temperature distribution image is segmented according to the temperature extension direction and temperature change rate to obtain multiple temperature-affected regions. The temperatures of multiple regions are obtained based on the multiple temperature-affected regions. Sequentially determine whether the temperature of multiple regions exceeds a preset threshold; If the temperature of a region exceeds a preset threshold, the temperature-affected region corresponding to that temperature will be used as local information of the battery pack.

3. The method for disassembling a power vehicle battery pack based on thermal runaway early warning as described in claim 1, characterized in that, The step of obtaining the battery pack gas mutation coefficient based on the gas concentration data and the internal pressure change rate includes: Wavelet packet decomposition was performed on the gas concentration data to obtain multiple concentration fluctuation features in different frequency bands; The internal pressure change rate is processed by Kalman filtering to obtain the pressure abrupt change gradient; The battery pack gas mutation coefficient is obtained based on the pressure mutation gradient and multiple concentration fluctuation characteristics.

4. The method for disassembling a power vehicle battery pack based on thermal runaway early warning as described in claim 1, characterized in that, The step of obtaining the early warning coefficient based on the battery pack thermal runaway risk coefficient and the battery pack gas mutation coefficient includes: The thermal runaway risk coefficient of the battery pack is normalized to obtain the normalized value of the thermal runaway risk coefficient of the battery pack. The gas mutation coefficient of the battery pack is normalized to obtain the normalized value of the gas mutation coefficient of the battery pack. The warning coefficient is obtained by weighting the normalized value of the thermal runaway risk coefficient of the battery pack and the normalized value of the gas mutation coefficient of the battery pack.

5. The method for disassembling a power vehicle battery pack based on thermal runaway early warning according to claim 1, characterized in that, The step of disassembling the power vehicle battery pack according to the warning reminder coefficient and obtaining the disassembly result includes: Determine the relationship between the warning reminder coefficient and the preset warning reminder threshold range; If the warning reminder coefficient is not within the preset warning reminder threshold range, and the warning reminder coefficient is greater than the maximum value of the preset warning reminder threshold range, then the local information of the battery pack is determined to be in a high-risk state, and a first warning reminder message is generated; If the warning reminder coefficient is within the preset warning reminder threshold range, the local information of the battery pack is determined to be in a medium-risk state, and a second warning reminder message is generated; If the warning reminder coefficient is not within the preset warning reminder threshold range, and the warning reminder coefficient is less than the minimum value of the preset warning reminder threshold range, then the local information of the battery pack is determined to be in a low-risk state, and a third warning reminder message is generated; Obtain the disassembly path of the electric vehicle battery pack; The disassembly path of the power vehicle battery pack is marked according to the first warning reminder information, the second warning reminder information and the third warning reminder information to obtain the marked disassembly path; The battery pack of the electric vehicle is disassembled according to the marked disassembly path.

6. A power vehicle battery pack disassembly system based on thermal runaway early warning, characterized in that, include: The first acquisition module is used to acquire surface temperature field distribution data of the battery pack during the disassembly process of the electric vehicle battery pack, and to acquire local information of the battery pack based on the surface temperature field distribution data of the battery pack. The second acquisition module is used to acquire local monitoring data of the battery pack based on the local information of the battery pack, wherein the local monitoring data includes surface feature data of the battery pack and internal feature data of the battery pack. The third acquisition module is used to acquire multiple surface change rates based on the battery pack surface feature data, and to acquire a battery pack thermal runaway risk coefficient based on the multiple surface change rates, including: The area of ​​a local region is obtained based on the local information of the battery pack, and the local region area is divided into grids according to a preset area to obtain multiple grid points; Obtain multiple initial coordinates corresponding to multiple grid points; Based on the surface feature data of the battery pack, obtain multiple changing coordinates of the battery surface within a preset working time. Multiple deformation slopes are obtained between the multiple initial coordinates and the multiple changed coordinates, and the deformation slopes are used as the body surface change rate; The surface change coefficient is obtained based on multiple surface change rates and a preset working time, and the surface change coefficient is used as the thermal runaway risk coefficient of the battery pack. The fourth acquisition module is used to acquire gas concentration data and internal pressure change rate based on the internal characteristic data of the battery pack, and to acquire the gas mutation coefficient of the battery pack based on the gas concentration data and the internal pressure change rate. The fifth acquisition module is used to acquire the early warning and reminder coefficient based on the thermal runaway risk coefficient of the battery pack and the gas mutation coefficient of the battery pack; The disassembly module is used to disassemble the power vehicle battery pack according to the warning and reminder coefficient, and obtain the disassembly results.

7. A power vehicle battery pack disassembly system based on thermal runaway early warning as described in claim 6, characterized in that, The first acquisition module includes: The first acquisition unit is used to acquire an initial infrared temperature distribution image of the battery pack surface; The second acquisition unit is used to acquire the temperature extension direction and single battery pack chain reaction information based on the temperature field distribution data on the surface of the battery pack, and to acquire the temperature change rate based on the single battery pack chain reaction information. The segmentation unit is used to segment the initial infrared temperature distribution image according to the temperature extension direction and temperature change rate to obtain multiple temperature-affected regions; The third acquisition unit is used to acquire the temperature of multiple regions corresponding to the multiple temperature-affected regions. A judgment unit is used to sequentially determine whether the temperature of the multiple regions exceeds a preset threshold. If the temperature of a region exceeds a preset threshold, the temperature-affected region corresponding to that temperature will be used as local information of the battery pack.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images