CN118483596A - VOC-based lithium battery pack thermal runaway monitoring method and system - Google Patents
VOC-based lithium battery pack thermal runaway monitoring method and system Download PDFInfo
- Publication number
- CN118483596A CN118483596A CN202410902615.7A CN202410902615A CN118483596A CN 118483596 A CN118483596 A CN 118483596A CN 202410902615 A CN202410902615 A CN 202410902615A CN 118483596 A CN118483596 A CN 118483596A
- Authority
- CN
- China
- Prior art keywords
- battery pack
- concentration
- voc
- lithium battery
- thermal runaway
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 title claims abstract description 262
- 229910052744 lithium Inorganic materials 0.000 title claims abstract description 262
- 238000012544 monitoring process Methods 0.000 title claims abstract description 85
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000004458 analytical method Methods 0.000 claims abstract description 106
- 238000000605 extraction Methods 0.000 claims abstract description 23
- 238000013507 mapping Methods 0.000 claims abstract description 21
- 230000002159 abnormal effect Effects 0.000 claims description 232
- 230000004044 response Effects 0.000 claims description 74
- 238000012545 processing Methods 0.000 claims description 44
- 238000012502 risk assessment Methods 0.000 claims description 32
- 238000004088 simulation Methods 0.000 claims description 20
- 238000011002 quantification Methods 0.000 claims description 17
- 238000012937 correction Methods 0.000 claims description 14
- 238000000547 structure data Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 9
- 238000005065 mining Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 4
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 239000012855 volatile organic compound Substances 0.000 description 228
- 230000008859 change Effects 0.000 description 41
- 238000004422 calculation algorithm Methods 0.000 description 18
- 230000000007 visual effect Effects 0.000 description 13
- 238000011156 evaluation Methods 0.000 description 11
- 230000002349 favourable effect Effects 0.000 description 9
- 238000012800 visualization Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 238000003708 edge detection Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000004445 quantitative analysis Methods 0.000 description 5
- 230000010485 coping Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 238000007619 statistical method Methods 0.000 description 4
- 239000000126 substance Substances 0.000 description 4
- 238000012300 Sequence Analysis Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000017525 heat dissipation Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 238000002955 isolation Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000011158 quantitative evaluation Methods 0.000 description 2
- 238000012731 temporal analysis Methods 0.000 description 2
- 230000008646 thermal stress Effects 0.000 description 2
- 238000000700 time series analysis Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000001808 coupling effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000020169 heat generation Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Secondary Cells (AREA)
Abstract
The invention relates to the field of lithium battery thermal monitoring, in particular to a method and a system for monitoring thermal runaway of a lithium battery pack based on VOC. The method comprises the following steps: acquiring the internal VOC parameters of the real-time lithium battery pack based on the distributed sensor nodes; carrying out spatial concentration distribution analysis on VOC parameters in the real-time lithium battery pack to generate VOC concentration distribution fluctuation trend data; constructing a three-dimensional coordinate system based on the distributed sensor nodes; performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to obtain a smooth optimized concentration distribution network; performing global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph; acquiring a multi-azimuth image of the lithium battery pack; and carrying out three-dimensional point cloud reconstruction on the multi-azimuth image of the lithium battery pack, and constructing a three-dimensional point cloud model of the lithium battery pack. The invention realizes timely and high-precision thermal runaway monitoring of the lithium battery pack.
Description
Technical Field
The invention relates to the field of lithium battery thermal monitoring, in particular to a method and a system for monitoring thermal runaway of a lithium battery pack based on VOC.
Background
With the wide application of lithium batteries in the fields of energy storage, electric automobiles and the like, the problem of thermal runaway of a lithium battery pack has attracted attention. In the long-term use process of the lithium battery pack, thermal runaway phenomenon may occur due to the influence of chemical reaction inside the battery and external environmental factors, resulting in dangerous situations such as temperature rapid increase, hot spot formation and even explosion. The traditional lithium battery pack thermal runaway monitoring method mainly relies on monitoring of single parameters such as a temperature sensor, a pressure sensor and the like, and has the problems of low monitoring precision and untimely early warning. Therefore, an intelligent lithium battery pack thermal runaway monitoring method is needed to improve the monitoring accuracy and response speed.
Disclosure of Invention
The invention provides a thermal runaway monitoring method and a thermal runaway monitoring system for a lithium battery pack based on VOC (volatile organic compounds) for solving at least one of the technical problems.
In order to achieve the above object, the present invention provides a method for monitoring thermal runaway of a VOC-based lithium battery pack, comprising the steps of:
Step S1: acquiring the internal VOC parameters of the real-time lithium battery pack based on the distributed sensor nodes; carrying out spatial concentration distribution analysis on VOC parameters in the real-time lithium battery pack to generate VOC concentration distribution fluctuation trend data;
Step S2: constructing a three-dimensional coordinate system based on the distributed sensor nodes; performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to obtain a smooth optimized concentration distribution network; performing global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph;
step S3: acquiring a multi-azimuth image of the lithium battery pack; carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes; carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes to construct a three-dimensional point cloud model of the lithium battery pack;
Step S4: carrying out concentration situation evolution on the VOC concentration distribution scene graph to generate regional concentration situation evolution data; abnormal concentration identification is carried out on the regional concentration situation evolution data based on a preset VOC concentration distribution threshold value so as to obtain VOC abnormal concentration data; extracting an abnormal concentration region scene graph based on the VOC abnormal concentration data;
step S5: carrying out abnormal heat region positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration region scene graph so as to obtain an abnormal heat region model; performing thermal runaway response analysis on the abnormal thermal region model to generate thermal runaway response data of the battery assembly;
Step S6: performing risk quantification assessment on the battery assembly thermal runaway response data to generate thermal runaway risk assessment data; performing thermal runaway decision analysis based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy; and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
According to the invention, the Volatile Organic Compound (VOC) parameters in the lithium battery pack are obtained in real time through the distributed sensor nodes, so that the real-time monitoring of the internal environment of the battery pack is provided. And analyzing the VOC parameters acquired in real time to obtain the spatial distribution condition of the VOC concentration in the lithium battery pack, and knowing the concentration difference and trend change of different positions. And the fluctuation trend of the VOC concentration is analyzed to obtain the change mode of the VOC concentration in the lithium battery pack, so that basic data is provided for subsequent abnormal concentration identification and risk assessment. And constructing a three-dimensional coordinate system inside the lithium battery pack through the position information of the distributed sensor nodes, and providing a basis for subsequent space mapping. And mapping the VOC concentration fluctuation trend data into a constructed three-dimensional coordinate system to obtain a smooth and optimized concentration distribution network, and providing space visualization of the VOC concentration in the lithium battery pack. And carrying out global deep convolution extraction on the smooth optimized concentration distribution network, capturing higher-level characteristics, and improving the understanding and analysis capability of VOC concentration distribution scenes. And by acquiring multi-azimuth images of the lithium battery pack, the external form and component distribution information of the battery pack are comprehensively captured. And processing multi-azimuth images of the lithium battery pack, identifying and marking key nodes of the battery pack, such as connectors, cooling fins and the like, and providing positioning information for subsequent three-dimensional point cloud reconstruction. And the three-dimensional point cloud model of the lithium battery pack is obtained by reconstructing the three-dimensional point cloud of the battery pack assembly nodes, so that the visualization and analysis capability of the internal space structure of the battery pack are provided. And carrying out concentration situation evolution analysis on the VOC concentration distribution scene graph, observing the change trend of the concentration of different areas in the lithium battery pack, and knowing the time-space evolution rule of the concentration. Based on a preset VOC concentration distribution threshold value, analyzing the regional concentration situation evolution data, identifying a region with abnormal concentration, namely the VOC abnormal concentration data, and marking the region with the problem. And extracting a scene graph of the abnormal concentration region according to the VOC abnormal concentration data, highlighting the region with the abnormality, and providing a visual and quantitative analysis basis for the subsequent abnormal thermal region positioning. And (3) analyzing the scene graph of the abnormal concentration region and the three-dimensional point cloud model of the lithium battery pack, and positioning and identifying an abnormal thermal region, namely a region with a thermal runaway problem. Based on the abnormal thermal region positioning result, an abnormal thermal region model is generated, and the visualization and quantitative analysis capability of the region with the thermal runaway problem is provided. And carrying out thermal runaway response analysis on the abnormal thermal region model, knowing the degree and the influence range of the thermal runaway problem, generating thermal runaway response data of the battery assembly, and providing basis for subsequent risk assessment and decision analysis. And analyzing and evaluating the thermal runaway response data of the battery assembly, quantifying the thermal runaway risk, and knowing the potential threat of the thermal runaway to the safe and stable operation of the battery assembly. Based on the thermal runaway risk assessment data, thermal runaway decision analysis is carried out, corresponding measures and strategies are formulated to cope with the thermal runaway problem, and the safety and reliability of the lithium battery pack are guaranteed. And carrying out thermal runaway monitoring operation of the lithium battery pack based on a thermal runaway decision strategy, implementing targeted monitoring and control measures, timely coping with potential thermal runaway risks, and guaranteeing safe operation of the lithium battery pack.
Preferably, step S1 comprises the steps of:
Step S11: monitoring the real-time working state of the lithium battery pack based on the distributed sensor nodes, and acquiring the internal VOC parameters of the real-time lithium battery pack;
Step S12: performing time sequence concentration characteristic analysis on the VOC parameters in the real-time lithium battery pack to generate VOC time sequence concentration characteristic data;
step S13: performing spatial concentration distribution analysis on the VOC time sequence concentration characteristic data to obtain VOC time sequence concentration distribution data;
Step S14: and quantifying fluctuation trend of the VOC time sequence concentration distribution data to generate VOC concentration distribution fluctuation trend data.
According to the invention, the distributed sensor nodes are used for monitoring the lithium battery pack in real time to obtain the VOC parameters in the lithium battery pack, the VOC parameters are relevant information such as the concentration of volatile organic compounds in the lithium battery pack, the state and the change condition of chemical substances in the lithium battery pack can be provided, the change trend of the concentration of the volatile organic compounds in the lithium battery pack is revealed through carrying out time sequence concentration characteristic analysis on the VOC parameters in the lithium battery pack in real time, the generated VOC time sequence concentration characteristic data provide detailed information of the change of the concentration of the VOC in the lithium battery pack along with time, the dynamic behavior of the chemical substances in the lithium battery pack is understood, the spatial concentration distribution analysis is carried out on the VOC time sequence concentration characteristic data, the VOC concentration difference of different positions in the lithium battery pack is known, the spatial distribution mode of the concentration difference of VOC in different areas in the lithium battery pack is helped to be determined, the fluctuation trend of the concentration of the volatile organic compounds in the lithium battery pack is evaluated through carrying out fluctuation quantification on the VOC time sequence concentration distribution data, and the generated VOC time sequence concentration distribution data provide the fluctuation trend of the information which is relevant to the change trend of the concentration of the lithium battery pack, the fluctuation trend comprises the fluctuation trend of the concentration of the VOC, the fluctuation trend is favorable for judging the change of the concentration, and the fluctuation trend of the concentration of the VOC is favorable.
Preferably, step S2 comprises the steps of:
step S21: carrying out spatial position identification on the distributed sensor nodes to obtain the spatial positions of the sensor nodes;
step S22: constructing a three-dimensional coordinate system based on the spatial positions of the sensor nodes;
Step S23: performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to generate a VOC three-dimensional concentration distribution network;
step S24: performing iterative interpolation smooth optimization on the VOC three-dimensional concentration distribution network to obtain a smooth optimized concentration distribution network;
Step S25: and carrying out global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph.
According to the invention, the spatial position of each sensor node is identified by carrying out spatial position identification on the distributed sensor nodes, the specific position of each sensor node in the lithium battery pack is determined, so that the corresponding relation between the sensor nodes and the internal space of the lithium battery pack is established, accurate position information is provided for subsequent analysis and processing, a unified reference frame is provided for spatial distribution in the lithium battery pack by constructing a three-dimensional coordinate system based on the spatial position of the sensor nodes, thus the subsequent spatial mapping and analysis are convenient, the relation among the positions is clearer and easier to understand, the three-dimensional coordinate system is mapped to the three-dimensional coordinate system by carrying out spatial mapping on the fluctuation trend data of the VOC concentration distribution, the generated three-dimensional concentration distribution network of the VOC provides VOC concentration distribution conditions of different positions in the lithium battery pack, and the relevance among the positions is favorable for comprehensively understanding and analyzing the spatial distribution characteristics of the VOC, the noise and uncertainty in data are reduced by carrying out iterative smooth optimization on the three-dimensional concentration distribution network, the distribution network with higher spatial resolution is obtained, the subsequent spatial resolution is improved, the three-dimensional concentration distribution is more clearly and easily understood, the subsequent thermal concentration profile is more intuitively and more clearly understood by the user-perceived by carrying out the analysis on the graph, and the visual analysis is better shows the visual and the visual analysis result is better.
Preferably, step S25 comprises the steps of:
performing discrete voxel decomposition on the smooth optimized concentration distribution network to obtain a plurality of three-dimensional concentration voxel units;
Carrying out local convolution filtering treatment on a plurality of three-dimensional concentration voxel units, and extracting the parameter combination characteristics of the voxel units;
performing multi-scale deep convolution feature mining on the voxel unit parameter combination features to generate multi-scale global concentration structure information;
And carrying out global feature activation processing on the smooth optimized concentration distribution network by utilizing the multi-scale global concentration structure information so as to obtain a VOC concentration distribution scene graph.
According to the invention, the smooth optimized concentration distribution network is subjected to discrete voxel decomposition, the concentration distribution network is divided into a plurality of discrete three-dimensional voxel units, so that continuous concentration distribution is converted into the discrete voxel units, subsequent processing and analysis are facilitated, local convolution filtering processing is performed on the plurality of three-dimensional concentration voxel units, local characteristics of each voxel unit are extracted, thus concentration change modes and characteristic information in each voxel unit are captured, a basis is provided for subsequent characteristic mining, multi-scale deep convolution characteristic mining is performed on voxel unit parameter combination characteristics, global structure information of concentration distribution is extracted on different scales, local characteristics and global structures are combined, more comprehensive and rich concentration characteristics are obtained, the understanding and analysis capability of VOC concentration distribution in the lithium battery pack are facilitated to be improved, global characteristic activation processing is performed on the smooth optimized concentration distribution network by utilizing the multi-scale global concentration structure information, important concentration distribution characteristics are highlighted, the key information of VOC concentration distribution is highlighted, thus a VOC concentration distribution scene map is generated, the result of thermal runaway monitoring of the lithium battery pack is more visual and visual, and the quick judgment of the concentration distribution of VOC is facilitated to users.
Preferably, step S3 comprises the steps of:
Step S31: acquiring a multi-azimuth image of the lithium battery pack;
Step S32: carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes;
step S33: extracting image edges of multi-azimuth images of the lithium battery pack to obtain multi-azimuth image edge contour lines;
step S34: performing feature point overlapping matching on the battery pack assembly nodes based on the multi-azimuth image edge contour lines to generate an image overlapping region;
Step S35: and carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes according to the image overlapping region, and constructing a three-dimensional point cloud model of the lithium battery pack.
According to the invention, the multi-azimuth images of the lithium battery pack are acquired, the image information of the battery pack under different angles and visual angles is acquired, so that more comprehensive and three-dimensional visual data is provided, further analysis and processing of the battery pack are facilitated, the multi-azimuth images of the lithium battery pack are identified by key nodes, the component nodes of the battery pack are automatically identified and marked, the positions and boundaries of all components in the battery pack are accurately determined, accurate positioning information is provided for subsequent processing and reconstruction, the multi-azimuth images of the lithium battery pack are subjected to image edge extraction, the edge contour line of the battery pack component is extracted, the shape and boundary information of the battery pack are acquired, a basis is provided for subsequent feature point matching and three-dimensional reconstruction, the overlapping matching of feature points is carried out on the component nodes of the battery pack based on the edge contour line of the multi-azimuth images, the overlapping areas of the images are found, the corresponding relation between all the images is determined, more accurate and complete information of the battery pack components is provided, the three-dimensional point cloud reconstruction is carried out on the component nodes of the battery pack according to the overlapping areas of the images, the three-dimensional point cloud reconstruction is carried out on the battery pack, the three-dimensional point cloud model is generated, the three-dimensional model is further, the three-dimensional model of the lithium battery pack is realized, and the three-dimensional model is further analyzed, and the three-dimensional model is realized, and the three-dimensional model is further is realized.
Preferably, the specific steps of step S35 are:
performing three-line intersection method calculation on the battery pack assembly nodes according to the image overlapping areas to obtain initial three-dimensional node coordinates;
Extracting adjacent characteristic points of the battery pack assembly nodes according to the image overlapping areas to obtain characteristic points around the nodes;
performing three-dimensional back projection on the initial three-dimensional node coordinates based on the feature point degrees around the nodes to generate component node three-dimensional space position coordinates;
Carrying out fine correction processing on the three-dimensional space position coordinates of the component nodes to obtain corrected three-dimensional space position coordinates of the nodes;
carrying out three-dimensional structure analysis on the three-dimensional space position coordinates of the correction nodes to obtain three-dimensional structure data of the lithium battery pack;
And carrying out three-dimensional point cloud reconstruction on the three-dimensional structure data of the lithium battery pack, and constructing a three-dimensional point cloud model of the lithium battery pack.
The invention calculates initial three-dimensional node coordinates by applying a three-line intersection method to battery pack assembly nodes in an image overlapping area, wherein the three-line intersection method is a calculation method based on a multi-view geometric principle, calculates three-dimensional positions of the nodes by the node overlapping areas under a plurality of view angles, preliminarily determines space position information of the battery pack assembly nodes, extracts characteristic points of the battery pack assembly nodes in the image overlapping area to obtain characteristic point sets around the nodes, wherein the characteristic points are edge points, corner points or other points with obvious characteristics, thus providing additional characteristic information for subsequent three-dimensional back projection, improving the accuracy of node position calculation, and performing three-dimensional back projection calculation on the initial three-dimensional node coordinates by using the characteristic point sets around the nodes, the three-dimensional back projection is a process of back projecting characteristic points in an image into a three-dimensional space, three-dimensional space position coordinates corresponding to the characteristic points are obtained through calculation through positions and camera parameters of the characteristic points in the image, the accuracy and precision of the node positions are further improved, the accuracy of the node positions is further improved through refined correction processing on the three-dimensional space position coordinates of the component nodes, the correction processing comprises error compensation, application of an optimization algorithm or other mathematical models, errors caused by measurement errors or uncertainty in the calculation process are corrected, more accurate and reliable three-dimensional space position coordinates of the node are obtained, three-dimensional structure data of the lithium battery pack are obtained through three-dimensional structure analysis on the corrected three-dimensional space position coordinates of the node, the three-dimensional structure analysis comprises calculation of information such as distance, angle, connection relation and the like among the components, the method is characterized in that the internal structural characteristics and the topological relation of the lithium battery pack are revealed, so that more comprehensive and detailed space structure data are provided for thermal runaway monitoring and analysis of the lithium battery pack, three-dimensional point cloud reconstruction is carried out on the three-dimensional structure data of the lithium battery pack, discrete node coordinates and connection relations are converted into continuous three-dimensional point cloud models, the three-dimensional point cloud models are geometric models formed by a large number of three-dimensional points, the shape and the space structure of objects can be accurately represented, visual presentation of the lithium battery pack is achieved, further analysis and visual display are carried out by utilizing the three-dimensional point cloud models, the three-dimensional point cloud models of the lithium battery pack provide more visual and comprehensive space information, and the method is beneficial to the tasks of thermal runaway monitoring and analysis, space structure analysis, anomaly detection and the like.
Preferably, the specific steps of step S4 are:
Step S41: performing region segmentation processing on the VOC concentration distribution scene graph to obtain a plurality of concentration region scene graphs;
step S42: carrying out concentration situation evolution on the plurality of concentration area scene graphs to generate area concentration situation evolution data;
step S43: future concentration prediction is carried out on the regional concentration situation evolution data to obtain regional VOC concentration prediction data;
step S44: performing abnormal concentration identification on the regional VOC concentration prediction data based on a preset VOC concentration distribution threshold value to obtain VOC abnormal concentration data;
Step S45: and carrying out abnormal region positioning on the concentration region scene graph based on the VOC abnormal concentration data so as to extract the abnormal concentration region scene graph.
The invention divides the VOC concentration distribution scene graph into a plurality of concentration areas by carrying out area division processing on the scene graph, the area division is based on an image processing algorithm, such as clustering, segmentation algorithm and the like, adjacent pixel points are grouped according to the concentration values of the adjacent pixel points to form a plurality of concentration areas, thus the complex concentration distribution scene graph is converted into an area scene graph with more analyzability, the concentration trend of each area along with time is obtained by carrying out concentration trend evolution on the plurality of concentration area scene graphs, the concentration trend evolution comprises the information of calculating the average concentration, the concentration change rate, the concentration gradient and the like of each area, thereby the area concentration trend evolution data is generated, the concentration change of each area is analyzed and compared, the dynamic change situation of the VOC concentration is known, the VOC concentration in a period in the future is predicted by carrying out future concentration prediction on the area concentration trend evolution data, the VOC concentration in the future is predicted by utilizing methods of time sequence analysis, machine learning and the like, the prediction is carried out on the basis of historical concentration data and the trend, the basis of the subsequent abnormal concentration is provided with basis, the abnormal concentration is further accurately identified by carrying out the analysis on the abnormal concentration threshold value by carrying out the position determination on the area by setting the abnormal concentration of the abnormal concentration is more than the threshold value, the abnormal concentration is further determined by setting the abnormal concentration is more than the abnormal concentration is determined by the abnormal concentration is more than the threshold is determined by the abnormal concentration is more than the abnormal concentration is determined by the position is more than the abnormal concentration is determined by the abnormal concentration is more than the abnormal concentration is determined, abnormal region positioning determines specific region positions according to abnormal concentration data, and the abnormal concentration regions are extracted from the original scene graph to form an abnormal concentration region scene graph, so that the abnormal concentration regions are intuitively displayed, and further analysis of abnormal conditions and corresponding monitoring and control measures are conveniently carried out.
Preferably, the specific steps of step S5 are:
Step S51: performing lithium battery pack position matching on the abnormal concentration area scene graph based on the sensor node spatial position to obtain the abnormal concentration position of the lithium battery pack;
Step S52: performing abnormal heat area positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration position of the lithium battery pack so as to obtain an abnormal heat area model;
step S53: performing thermal runaway simulation on the abnormal thermal zone model to obtain abnormal zone thermal runaway simulation data;
step S54: and carrying out battery assembly response analysis on the abnormal region thermal runaway simulation data to generate battery assembly thermal runaway response data.
The invention carries out lithium battery pack position matching on the scene graph of the abnormal concentration area based on the space position of the sensor node, determines the position of the lithium battery pack corresponding to the abnormal concentration area, correlates the position information of the sensor node with the abnormal concentration area, thereby determining the lithium battery pack existing in the area, realizing the matching of the position of the lithium battery pack and the abnormal concentration area, providing accurate position information for the subsequent abnormal thermal area positioning, carrying out abnormal thermal area positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration position of the lithium battery pack, determining the abnormal thermal area existing in the lithium battery pack, wherein the three-dimensional point cloud model of the lithium battery pack comprises the geometric shape and space distribution information of the battery assembly, and positioning the abnormal thermal area by matching the abnormal concentration position of the lithium battery pack with the three-dimensional point cloud model, the abnormal thermal region in the lithium battery pack is accurately positioned, an accurate region range is provided for subsequent thermal runaway simulation, thermal behavior of the abnormal region under the condition of thermal runaway is simulated by performing thermal runaway simulation on an abnormal thermal region model, the thermal runaway simulation is based on a physical model and a thermal conduction theory, processes such as heat generation, transmission and heat dissipation of the abnormal thermal region are considered, temperature distribution and change conditions of the abnormal region are simulated, abnormal region thermal runaway simulation data are obtained, temperature information under the abnormal condition is provided, input is provided for subsequent battery pack response analysis, the response condition of the battery pack under the abnormal thermal condition is analyzed by performing the battery pack response analysis on the abnormal region thermal runaway simulation data, parameters such as heat capacity and heat conductivity of the battery are considered by the battery pack response analysis, and the temperature change of the battery pack is calculated by combining the thermal runaway simulation data, and the indexes such as thermal stress and the like generate thermal runaway response data of the battery assembly, provide thermal response information of the battery assembly under abnormal conditions, evaluate the thermal runaway risk of the battery assembly by analyzing the response data, and provide basis for thermal runaway monitoring and control.
Preferably, the specific steps of step S6 are:
Step S61: analyzing the thermal runaway fault type of the battery assembly according to the thermal runaway response data of the battery assembly to generate thermal runaway fault type data;
Step S62: performing risk quantification assessment on the thermal runaway fault type data to generate thermal runaway risk assessment data;
Step S63: performing thermal runaway decision analysis on the thermal runaway response data of the battery assembly based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy;
step S64: and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
According to the invention, the thermal runaway fault types of different components are determined by analyzing thermal runaway response data of the battery component, the thermal runaway fault types comprise overheat, overcharge, overdischarge, thermal runaway diffusion and the like, whether the battery component is out of control or not is judged by analyzing indexes such as temperature change, thermal stress and the like in response data, and a specific fault type is further determined, so that thermal runaway fault type data are generated, a basis is provided for subsequent risk assessment and decision analysis, risk of a thermal runaway event is assessed and quantified by carrying out risk quantification assessment on the thermal runaway fault type data, the risk quantification assessment considers factors such as severity, occurrence probability and loss caused by the thermal runaway fault type, influence degree of the thermal runaway event on the lithium battery pack is comprehensively assessed, thermal runaway risk assessment data is generated by quantification assessment, quantitative description of the thermal runaway risk is provided, basis is provided for subsequent decision analysis, the thermal runaway response data of the battery component are subjected to decision analysis based on the thermal runaway risk assessment data, corresponding thermal runaway decision strategy making decision analysis is carried out according to different thermal runaway risk decision results, different countermeasures and high-level emergency measures are taken, for example, thermal runaway stop measures are taken to enhance heat dissipation and the like; for low-risk thermal runaway events, daily monitoring and maintenance are carried out, an effective thermal runaway management strategy is formulated according to actual conditions through thermal runaway decision-making analysis, occurrence and influence of the thermal runaway risks are reduced, thermal runaway monitoring operation of the lithium battery pack is carried out based on the thermal runaway decision-making strategy, corresponding monitoring tasks and control measures are executed according to the formulated decision-making strategy, the thermal runaway conditions of the battery pack are monitored and managed in real time, the monitoring operation comprises real-time monitoring of a temperature sensor, operation of an abnormal alarm system, regulation and control of a heat dissipation system and the like, potential thermal runaway risks are timely found and processed through the thermal runaway monitoring operation, so that safe operation of the lithium battery pack is ensured, and meanwhile, the monitoring operation also provides real-time thermal runaway information for subsequent data analysis and decision-making, and support is provided for continuous improvement and optimization of the thermal runaway management.
In this specification, there is provided a VOC-based lithium battery thermal runaway monitoring system for performing the VOC-based lithium battery thermal runaway monitoring method as described above, comprising:
the concentration distribution module is used for acquiring the VOC parameters in the real-time lithium battery pack based on the distributed sensor nodes; carrying out spatial concentration distribution analysis on VOC parameters in the real-time lithium battery pack to generate VOC concentration distribution fluctuation trend data;
the concentration distribution scene module is used for constructing a three-dimensional coordinate system based on the distributed sensor nodes; performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to obtain a smooth optimized concentration distribution network; performing global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph;
The three-dimensional model module is used for acquiring a multi-azimuth image of the lithium battery pack; carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes; carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes to construct a three-dimensional point cloud model of the lithium battery pack;
The concentration situation evolution module is used for carrying out concentration situation evolution on the VOC concentration distribution scene graph to generate regional concentration situation evolution data; abnormal concentration identification is carried out on the regional concentration situation evolution data based on a preset VOC concentration distribution threshold value so as to obtain VOC abnormal concentration data; extracting an abnormal concentration region scene graph based on the VOC abnormal concentration data;
The thermal runaway response module is used for carrying out abnormal thermal area positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration area scene graph so as to obtain an abnormal thermal area model; performing thermal runaway response analysis on the abnormal thermal region model to generate thermal runaway response data of the battery assembly;
The thermal runaway decision module is used for performing risk quantification assessment on the thermal runaway response data of the battery assembly so as to generate thermal runaway risk assessment data; performing thermal runaway decision analysis based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy; and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
The invention acquires the VOC parameters in the lithium battery pack through the distributed sensor nodes, monitors the concentration of Volatile Organic Compounds (VOC) in the battery pack in real time, is favorable for knowing the chemical environment in the battery pack, identifying potential volatile matters leakage or abnormal conditions, generating VOC concentration distribution fluctuation trend data, providing detailed analysis on VOC concentration change, facilitating the discovery of abnormal conditions and taking corresponding measures, obtaining a smooth optimized concentration distribution network of VOC concentration in the lithium battery pack by constructing a three-dimensional coordinate system and carrying out space mapping on the VOC concentration fluctuation trend data, providing more accurate and comprehensive VOC concentration information, facilitating the understanding and analysis of the VOC concentration in the internal space of the lithium battery pack, and further improving the quality of a concentration distribution scene graph through global deep convolution extraction, providing more accurate VOC concentration distribution information, determining the positions of assembly nodes of a battery pack by acquiring multi-azimuth images of the lithium battery pack and carrying out key node identification and marking on the images, carrying out three-dimensional point cloud reconstruction on the assembly nodes of the battery pack, constructing a three-dimensional point cloud model of the lithium battery pack, providing three-dimensional representation of the internal structure of the battery pack, facilitating subsequent thermal runaway response analysis and abnormal hot area positioning identification, carrying out concentration situation evolution analysis on a VOC concentration distribution scene graph, generating regional concentration situation evolution data, providing trend and evolution situation of VOC concentration change in a region, helping to find the mode and trend of concentration change, carrying out abnormal concentration identification on the regional concentration situation evolution data based on a preset VOC concentration distribution threshold value, further obtaining VOC abnormal concentration data, the method is favorable for timely finding and positioning abnormal VOC concentration areas, carrying out abnormal thermal area positioning identification on a three-dimensional point cloud model of the lithium battery pack based on an abnormal concentration area scene graph, determining the position of the abnormal thermal area, carrying out thermal runaway response analysis on the abnormal thermal area model, generating thermal runaway response data of the battery pack, providing characteristics and attributes of the abnormal thermal area, facilitating understanding and evaluating the degree and influence of a thermal runaway event, carrying out risk quantification evaluation on the thermal runaway response data of the battery pack, carrying out quantitative evaluation on the thermal runaway risk, generating thermal runaway risk evaluation data, providing information on the risk level and the potential influence of the thermal runaway event, carrying out thermal runaway decision analysis based on the thermal runaway risk evaluation data, and making corresponding decision strategies, such as taking emergency maintenance measures or taking safety isolation measures to reduce the thermal runaway risk, wherein the decision strategies are favorable for timely coping with potential thermal runaway problems and guaranteeing safe operation of the lithium battery pack.
Drawings
FIG. 1 is a schematic flow chart of the step of the method for monitoring thermal runaway of a VOC-based lithium battery pack according to the present invention;
FIG. 2 is a detailed implementation step flow diagram of step S1;
FIG. 3 is a detailed implementation step flow diagram of step S2;
fig. 4 is a detailed implementation step flow diagram of step S3.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The embodiment of the application provides a thermal runaway monitoring method and system for a lithium battery pack based on VOC. The execution subject of the VOC-based lithium battery pack thermal runaway monitoring method and system includes, but is not limited to, the system being mounted: mechanical devices, data processing platforms, cloud server nodes, network uploading devices, etc. may be considered general purpose computing nodes of the present application, including but not limited to: at least one of an audio image management system, an information management system and a cloud data management system.
Referring to fig. 1 to 4, the invention provides a thermal runaway monitoring method of a lithium battery pack based on VOC, comprising the following steps:
Step S1: acquiring the internal VOC parameters of the real-time lithium battery pack based on the distributed sensor nodes; carrying out spatial concentration distribution analysis on VOC parameters in the real-time lithium battery pack to generate VOC concentration distribution fluctuation trend data;
Step S2: constructing a three-dimensional coordinate system based on the distributed sensor nodes; performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to obtain a smooth optimized concentration distribution network; performing global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph;
step S3: acquiring a multi-azimuth image of the lithium battery pack; carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes; carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes to construct a three-dimensional point cloud model of the lithium battery pack;
Step S4: carrying out concentration situation evolution on the VOC concentration distribution scene graph to generate regional concentration situation evolution data; abnormal concentration identification is carried out on the regional concentration situation evolution data based on a preset VOC concentration distribution threshold value so as to obtain VOC abnormal concentration data; extracting an abnormal concentration region scene graph based on the VOC abnormal concentration data;
step S5: carrying out abnormal heat region positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration region scene graph so as to obtain an abnormal heat region model; performing thermal runaway response analysis on the abnormal thermal region model to generate thermal runaway response data of the battery assembly;
Step S6: performing risk quantification assessment on the battery assembly thermal runaway response data to generate thermal runaway risk assessment data; performing thermal runaway decision analysis based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy; and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
According to the invention, the Volatile Organic Compound (VOC) parameters in the lithium battery pack are obtained in real time through the distributed sensor nodes, so that the real-time monitoring of the internal environment of the battery pack is provided. And analyzing the VOC parameters acquired in real time to obtain the spatial distribution condition of the VOC concentration in the lithium battery pack, and knowing the concentration difference and trend change of different positions. And the fluctuation trend of the VOC concentration is analyzed to obtain the change mode of the VOC concentration in the lithium battery pack, so that basic data is provided for subsequent abnormal concentration identification and risk assessment. And constructing a three-dimensional coordinate system inside the lithium battery pack through the position information of the distributed sensor nodes, and providing a basis for subsequent space mapping. And mapping the VOC concentration fluctuation trend data into a constructed three-dimensional coordinate system to obtain a smooth and optimized concentration distribution network, and providing space visualization of the VOC concentration in the lithium battery pack. And carrying out global deep convolution extraction on the smooth optimized concentration distribution network, capturing higher-level characteristics, and improving the understanding and analysis capability of VOC concentration distribution scenes. And by acquiring multi-azimuth images of the lithium battery pack, the external form and component distribution information of the battery pack are comprehensively captured. And processing multi-azimuth images of the lithium battery pack, identifying and marking key nodes of the battery pack, such as connectors, cooling fins and the like, and providing positioning information for subsequent three-dimensional point cloud reconstruction. And the three-dimensional point cloud model of the lithium battery pack is obtained by reconstructing the three-dimensional point cloud of the battery pack assembly nodes, so that the visualization and analysis capability of the internal space structure of the battery pack are provided. And carrying out concentration situation evolution analysis on the VOC concentration distribution scene graph, observing the change trend of the concentration of different areas in the lithium battery pack, and knowing the time-space evolution rule of the concentration. Based on a preset VOC concentration distribution threshold value, analyzing the regional concentration situation evolution data, identifying a region with abnormal concentration, namely the VOC abnormal concentration data, and marking the region with the problem. And extracting a scene graph of the abnormal concentration region according to the VOC abnormal concentration data, highlighting the region with the abnormality, and providing a visual and quantitative analysis basis for the subsequent abnormal thermal region positioning. And (3) analyzing the scene graph of the abnormal concentration region and the three-dimensional point cloud model of the lithium battery pack, and positioning and identifying an abnormal thermal region, namely a region with a thermal runaway problem. Based on the abnormal thermal region positioning result, an abnormal thermal region model is generated, and the visualization and quantitative analysis capability of the region with the thermal runaway problem is provided. And carrying out thermal runaway response analysis on the abnormal thermal region model, knowing the degree and the influence range of the thermal runaway problem, generating thermal runaway response data of the battery assembly, and providing basis for subsequent risk assessment and decision analysis. And analyzing and evaluating the thermal runaway response data of the battery assembly, quantifying the thermal runaway risk, and knowing the potential threat of the thermal runaway to the safe and stable operation of the battery assembly. Based on the thermal runaway risk assessment data, thermal runaway decision analysis is carried out, corresponding measures and strategies are formulated to cope with the thermal runaway problem, and the safety and reliability of the lithium battery pack are guaranteed. And carrying out thermal runaway monitoring operation of the lithium battery pack based on a thermal runaway decision strategy, implementing targeted monitoring and control measures, timely coping with potential thermal runaway risks, and guaranteeing safe operation of the lithium battery pack.
In the embodiment of the present invention, referring to fig. 1, a flow chart of steps of a method for monitoring thermal runaway of a VOC-based lithium battery pack according to the present invention is shown, where in this example, the method for monitoring thermal runaway of a VOC-based lithium battery pack includes the steps of:
Step S1: acquiring the internal VOC parameters of the real-time lithium battery pack based on the distributed sensor nodes; carrying out spatial concentration distribution analysis on VOC parameters in the real-time lithium battery pack to generate VOC concentration distribution fluctuation trend data;
In this embodiment, a plurality of distributed sensor nodes are installed at appropriate positions inside the lithium battery pack, the sensor nodes measure and record VOC parameters, such as VOC concentration, temperature, humidity, etc., in the lithium battery pack, the VOC parameter data inside the lithium battery pack are continuously acquired through the distributed sensor nodes, the sensor nodes should be capable of sampling at certain time intervals and transmitting the data to a central processing unit, the real-time VOC parameter data acquired by the sensor nodes are transmitted to the central processing unit for processing and analyzing, the spatial concentration distribution analysis is performed on the received real-time VOC parameter data, this involves associating the data with the sensor nodes at different positions inside the lithium battery pack to understand the VOC concentration differences and distribution conditions at different positions, and according to the analysis results, VOC concentration distribution fluctuation trend data are generated, which are time series data showing the variation conditions of VOC concentration at different positions and time points, and processing and analyzing the data using statistical methods or machine learning algorithms, and visualizing the generated concentration fluctuation trend data, such as graph, thermal map, presentation or three-dimensional graph.
Step S2: constructing a three-dimensional coordinate system based on the distributed sensor nodes; performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to obtain a smooth optimized concentration distribution network; performing global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph;
In this embodiment, a three-dimensional coordinate system is constructed according to the physical structure of the lithium battery pack and the positions of the distributed sensor nodes. The appropriate origin of coordinates and axes are determined to accurately represent the spatial relationship of the various locations within the lithium battery pack in subsequent analysis. And mapping the VOC concentration fluctuation trend data to a constructed three-dimensional coordinate system. And according to the fluctuation trend data obtained by analysis, the VOC concentration information at different positions is associated with the corresponding space coordinates to form a smooth and optimized concentration distribution network. And carrying out smooth optimization treatment on the constructed concentration distribution network to reduce the influence of noise and abnormal values and improve the spatial continuity of concentration distribution. The concentration data is smoothed using various signal processing and filtering techniques, such as gaussian filtering, mean filtering, and the like. And performing feature extraction on the smoothly optimized concentration distribution network by using a global deep Convolutional Neural Network (CNN). CNNs learn the local and global features of the concentration profile and encode them into a higher level representation. And extracting abstract features of the VOC concentration distribution scene through superposition of a plurality of convolution layers and a pooling layer. The features extracted based on the global deep convolution are converted into a VOC concentration profile scene graph, which is a two-dimensional image in which each pixel represents a location within the lithium battery pack and the gray value or color of the pixel represents the VOC concentration level of the corresponding location.
Step S3: acquiring a multi-azimuth image of the lithium battery pack; carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes; carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes to construct a three-dimensional point cloud model of the lithium battery pack;
In this embodiment, a suitable image acquisition device (such as a camera and a laser scanner) is used to acquire multi-azimuth images of the lithium battery pack from different angles and positions, so as to ensure that all sides and components of the lithium battery pack are covered, to acquire comprehensive information, to perform image processing and analysis on the acquired multi-azimuth images, to identify key nodes of the lithium battery pack, which are important positions of components, connectors, battery cells and the like of the battery pack, once the key nodes are identified, to perform marking on the images for subsequent processing and analysis, to reconstruct three-dimensional point clouds of the lithium battery pack by using marked key node information, to acquire point cloud data by performing three-dimensional reconstruction on images of multiple angles, or to use technologies such as laser scanning to acquire alignment and matching of the point clouds, to ensure accuracy and integrity of the point clouds, to combine the three-dimensional point cloud data obtained by reconstruction, to construct a three-dimensional point cloud model of the lithium battery pack, which accurately represents the shape and component distribution of the lithium battery pack, and to provide a high-fidelity model with spatial information.
Step S4: carrying out concentration situation evolution on the VOC concentration distribution scene graph to generate regional concentration situation evolution data; abnormal concentration identification is carried out on the regional concentration situation evolution data based on a preset VOC concentration distribution threshold value so as to obtain VOC abnormal concentration data; extracting an abnormal concentration region scene graph based on the VOC abnormal concentration data;
In the embodiment, the concentration situation evolution is performed on the VOC concentration distribution scene graph, and the regional concentration situation evolution data is generated. Analysis of the concentration change for each region in the concentration scene graph using time series analysis methods such as sliding window or filtering techniques will provide an evolution of the concentration profile for each region over time. And based on a preset VOC concentration distribution threshold value, carrying out abnormal concentration identification on the regional concentration situation evolution data. And setting a proper threshold value according to the characteristics and the safety requirements of the lithium battery pack to judge which areas have the concentration exceeding the normal range. And using a threshold detection method, such as a statistical method or a machine learning algorithm, to perform anomaly detection on the regional concentration situation evolution data. And extracting an abnormal concentration area scene graph based on the VOC abnormal concentration data. The areas with abnormal concentration are marked on the original VOC concentration distribution scene graph to form a new scene graph, wherein the areas with abnormal concentration are obviously visually distinguished from other areas. The abnormal density region is extracted and labeled using image processing and segmentation techniques such as threshold segmentation or edge detection.
Step S5: carrying out abnormal heat region positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration region scene graph so as to obtain an abnormal heat region model; performing thermal runaway response analysis on the abnormal thermal region model to generate thermal runaway response data of the battery assembly;
In this embodiment, based on an abnormal concentration area scene graph, an abnormal thermal area positioning and identifying is performed on a three-dimensional point cloud model of a lithium battery pack, the abnormal concentration area scene graph is mapped onto the three-dimensional point cloud model of the lithium battery pack, so that each abnormal concentration area has corresponding position information in the three-dimensional point cloud model, according to the shape, position and characteristics of the abnormal concentration area, the three-dimensional point cloud model is positioned and identified by using a point cloud analysis and area segmentation algorithm, the abnormal thermal area model is generated based on the abnormal thermal area obtained by positioning and identifying, the abnormal thermal area is marked and extracted in the three-dimensional point cloud model of the lithium battery pack, a new model is formed, wherein the abnormal thermal area has obvious differences from other areas, the abnormal thermal area is extracted and modeled by using a point cloud processing and segmentation technology such as a clustering analysis or voxel method, the abnormal thermal area model is subjected to thermal runaway response analysis, thermal runaway response data of a battery component are generated, the thermal runaway condition in the abnormal thermal area model is simulated by performing thermal conduction analysis and thermal runaway simulation on the abnormal thermal area model, physical characteristics, the thermal runaway condition and thermal runaway characteristics and the thermal curve are considered in the lithium battery pack are considered, the thermal runaway temperature response curve is calculated, and temperature change curve is calculated according to the temperature change curve, temperature change, and the temperature change curve is calculated, and the like.
Step S6: performing risk quantification assessment on the battery assembly thermal runaway response data to generate thermal runaway risk assessment data; performing thermal runaway decision analysis based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy; and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
In this embodiment, risk quantification evaluation is performed on thermal runaway response data of a battery assembly, according to information such as a temperature change curve and a temperature peak value in the thermal runaway response data, design specifications and safety requirements of a lithium battery pack are combined, a risk evaluation model and a method are adopted to quantitatively evaluate the thermal runaway risk, the steps of calculating a risk index, establishing a risk level or an evaluation model and the like are included to obtain the thermal runaway risk evaluation data, based on the thermal runaway risk evaluation data, thermal runaway decision analysis is performed, according to the result of the thermal runaway risk evaluation data, the thermal runaway risk is analyzed and judged by combining with preset decision rules and standards, the steps of determining a threshold value of the thermal runaway risk, formulating a response strategy, a decision rule and the like are included to construct a thermal runaway decision strategy, based on the thermal runaway decision strategy, thermal runaway monitoring operation of the lithium battery pack is performed, and according to the response strategy and decision rules determined in the thermal runaway decision strategy, a thermal runaway monitoring plan and an operation flow are formulated, and steps of periodically monitoring the temperature change, setting an alarm mechanism, an emergency treatment scheme and the like are included to ensure timely monitoring and treatment of potential thermal runaway conditions.
In this embodiment, referring to fig. 2, a detailed implementation step flow chart of the step S1 is shown, and in this embodiment, the detailed implementation step of the step S1 includes:
Step S11: monitoring the real-time working state of the lithium battery pack based on the distributed sensor nodes, and acquiring the internal VOC parameters of the real-time lithium battery pack;
Step S12: performing time sequence concentration characteristic analysis on the VOC parameters in the real-time lithium battery pack to generate VOC time sequence concentration characteristic data;
step S13: performing spatial concentration distribution analysis on the VOC time sequence concentration characteristic data to obtain VOC time sequence concentration distribution data;
Step S14: and quantifying fluctuation trend of the VOC time sequence concentration distribution data to generate VOC concentration distribution fluctuation trend data.
In this embodiment, distributed sensor nodes are deployed at appropriate positions inside the lithium battery pack, the sensor nodes collect various parameters inside the lithium battery pack in real time, including battery temperature, humidity, volatile Organic Compound (VOC) concentration, and the like, the sensor nodes transmit the collected data to a data processing center through wireless communication or wired connection, perform preprocessing on real-time VOC parameter data collected from the sensor nodes, including steps of noise removal, outlier processing, data correction, and the like, to ensure accuracy and reliability of the data, extract time-series concentration characteristics based on the preprocessed VOC parameter data, perform feature extraction on the VOC parameter data, such as average concentration, peak concentration, concentration variation rate, and the like, convert time-series dimensions of the VOC time-series concentration characteristic data into space dimensions, according to the structure and layout of the lithium battery pack, the VOC time sequence concentration characteristic data is mapped to the spatial position of the lithium battery pack to obtain VOC concentration characteristic data of different areas, spatial concentration distribution analysis is carried out based on the mapped VOC concentration characteristic data, the VOC concentration distribution condition of different positions is analyzed by using an interpolation method, a spatial statistical analysis or a machine learning algorithm to obtain VOC time sequence concentration distribution data, fluctuation trend calculation is carried out on the VOC time sequence concentration distribution data, the fluctuation condition of the VOC concentration distribution of different positions along with time is calculated by using a statistical method, a time sequence analysis or an image processing technology, the fluctuation trend is quantized to generate VOC concentration fluctuation trend data, the fluctuation trend is converted into comparable and analyzable data by using a numerical index, a fluctuation rate or other quantization methods, for subsequent thermal runaway risk assessment or decision analysis.
In this embodiment, referring to fig. 3, a detailed implementation step flow chart of the step S2 is shown, and in this embodiment, the detailed implementation step of the step S2 includes:
step S21: carrying out spatial position identification on the distributed sensor nodes to obtain the spatial positions of the sensor nodes;
step S22: constructing a three-dimensional coordinate system based on the spatial positions of the sensor nodes;
Step S23: performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to generate a VOC three-dimensional concentration distribution network;
step S24: performing iterative interpolation smooth optimization on the VOC three-dimensional concentration distribution network to obtain a smooth optimized concentration distribution network;
Step S25: and carrying out global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph.
In this embodiment, node positioning configuration and operation are performed on distributed sensor nodes to obtain spatial position information of the sensor nodes, a suitable origin position is selected as a reference point of a three-dimensional coordinate system, a position easy to reference and calculate is selected as an origin according to the structure and layout of a lithium battery pack, a coordinate axis direction of the three-dimensional coordinate system is determined according to the layout and spatial characteristics of the lithium battery pack, a direction related to a main axis or a key component of the lithium battery pack is generally selected as the coordinate axis direction, VOC concentration fluctuation trend data is associated with spatial position information of the sensor nodes, it is ensured that the spatial position of each sensor node corresponds to the corresponding VOC concentration fluctuation trend data, the associated VOC concentration fluctuation trend data is spatially mapped in the three-dimensional coordinate system according to the spatial position and coordinate system information of the sensor nodes, mapping the VOC concentration distribution data to corresponding spatial positions to form a VOC three-dimensional concentration distribution network, selecting a proper interpolation method for interpolating the VOC three-dimensional concentration distribution network, performing iterative interpolation smooth optimization on the VOC three-dimensional concentration distribution network by common interpolation methods including linear interpolation, spline interpolation, kriging interpolation and the like, interpolating a missing value or discontinuous region in the concentration distribution network according to the selected interpolation method, performing iterative smooth optimization to obtain a more continuous and accurate concentration distribution network, performing global feature extraction on the smooth optimized concentration distribution network by using a selected deep convolution model, extracting key features in the concentration distribution network through convolution operation and pooling operation to capture scene information of VOC concentration distribution, extracting features according to the deep convolution model, a VOC concentration profile scene graph is generated, and features are mapped into a visualized image or thermodynamic diagram to show the VOC concentration profile of different areas.
In this embodiment, the specific steps of step S25 are as follows:
performing discrete voxel decomposition on the smooth optimized concentration distribution network to obtain a plurality of three-dimensional concentration voxel units;
Carrying out local convolution filtering treatment on a plurality of three-dimensional concentration voxel units, and extracting the parameter combination characteristics of the voxel units;
performing multi-scale deep convolution feature mining on the voxel unit parameter combination features to generate multi-scale global concentration structure information;
And carrying out global feature activation processing on the smooth optimized concentration distribution network by utilizing the multi-scale global concentration structure information so as to obtain a VOC concentration distribution scene graph.
In the embodiment, a smooth optimized concentration distribution network is discretized, continuous concentration distribution is converted into discrete voxel units, the whole concentration distribution network is divided into grid voxel units, each voxel unit represents a discrete space region, the size and the position of each voxel unit are determined according to the discretized concentration distribution network, the size and the resolution of the voxel units are adjusted according to the needs so as to adapt to specific application scenes, a local convolution filter is applied to each voxel unit for feature extraction, the local convolution filter is one-dimensional, two-dimensional or three-dimensional, corresponding selection is carried out according to the dimension of the voxel units, the parameter combination features of each voxel unit are extracted through the local convolution filter, the characteristics comprise the mean value, the variance, the gradient and the like of the local concentration distribution and the associated characteristics of surrounding voxel units, the deep convolution feature mining is carried out on the parameter combination features of the voxel units by utilizing the convolution filter with different sizes, the characteristics of different sizes are gradually enlarged from local to global so as to obtain the characteristics of different sizes, the characteristics of the multiple extracted characteristics are fused so as to generate global concentration information, the global concentration information is enabled to be the global concentration information, the global concentration information is enabled to be activated through the global concentration map, the global concentration information is further processed by the global concentration information, the global concentration information is enabled to be the overall concentration information is mapped to be the overall, and the overall concentration information is enabled to be activated through the global concentration information, and the global concentration information is enabled to be the overall level is enabled to be activated, and the overall concentration information is enabled to be the global by the global level map is activated, the characterization representation is converted into a visual image or thermodynamic diagram to reveal scene information of the VOC concentration profile.
In this embodiment, as described with reference to fig. 4, a detailed implementation step flow diagram of the step S3 is shown, and in this embodiment, the detailed implementation step of the step S3 includes:
Step S31: acquiring a multi-azimuth image of the lithium battery pack;
Step S32: carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes;
step S33: extracting image edges of multi-azimuth images of the lithium battery pack to obtain multi-azimuth image edge contour lines;
step S34: performing feature point overlapping matching on the battery pack assembly nodes based on the multi-azimuth image edge contour lines to generate an image overlapping region;
Step S35: and carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes according to the image overlapping region, and constructing a three-dimensional point cloud model of the lithium battery pack.
In this embodiment, a suitable photographing device, such as a camera or a laser scanner, is used to obtain images of multiple directions of the lithium battery pack, ensure good image quality, fully cover each component of the lithium battery pack, and keep the change of perspective angle as much as possible to obtain more comprehensive information, perform image processing pretreatment, such as denoising, enhancing contrast, etc., on each multi-directional image to improve the subsequent node recognition effect, perform image analysis and processing using image processing and computer vision algorithms, such as edge detection, feature extraction, template matching, etc., identify key nodes of the lithium battery pack, such as battery cells, connecting pieces, insulating pads, etc., and mark these nodes on the image for subsequent processing use, perform edge extraction on multi-directional images of the lithium battery pack using image processing techniques, such as Canny edge detection algorithm, adjusting parameters of an edge detection algorithm to obtain clear and accurate edge contour lines, extracting edges of images of each azimuth to obtain edge contour lines of a plurality of azimuth, extracting key characteristic points, such as angular points, intersection points and the like, for the edge contour lines of each azimuth, comparing similarity between images of different azimuth by using descriptors of the characteristic points, performing characteristic point matching, determining an image overlapping region, namely an overlapping region of a battery pack assembly node, among the images of the plurality of azimuth according to the matched characteristic points, converting pixel points in the image overlapping region into point clouds in a three-dimensional space by using a three-dimensional reconstruction algorithm, such as three-dimensional matching, structured light scanning and the like, performing point cloud splicing and fusion on the image of each azimuth according to point cloud data of the overlapping region to generate a complete three-dimensional point cloud model of the lithium battery pack, post-processing, such as filtering, denoising, gridding, etc., is performed on the generated three-dimensional point cloud model to obtain a more accurate and smooth model.
In this embodiment, the specific steps of step S35 are as follows:
performing three-line intersection method calculation on the battery pack assembly nodes according to the image overlapping areas to obtain initial three-dimensional node coordinates;
Extracting adjacent characteristic points of the battery pack assembly nodes according to the image overlapping areas to obtain characteristic points around the nodes;
performing three-dimensional back projection on the initial three-dimensional node coordinates based on the feature point degrees around the nodes to generate component node three-dimensional space position coordinates;
Carrying out fine correction processing on the three-dimensional space position coordinates of the component nodes to obtain corrected three-dimensional space position coordinates of the nodes;
carrying out three-dimensional structure analysis on the three-dimensional space position coordinates of the correction nodes to obtain three-dimensional structure data of the lithium battery pack;
And carrying out three-dimensional point cloud reconstruction on the three-dimensional structure data of the lithium battery pack, and constructing a three-dimensional point cloud model of the lithium battery pack.
In this embodiment, three battery pack assembly nodes are selected as reference nodes in the image overlapping region. For each reference node, the coordinates of the characteristic points on the image are acquired from images in different directions. The initial three-dimensional coordinates of the reference node are calculated using a three-wire intersection method. The method is based on the principle of triangulation, and three-dimensional coordinates are calculated through the sight line direction of a camera and pixel coordinates of feature points. For each component node, extracting surrounding feature points in the image superposition area, wherein the feature points are points adjacent to the node, points on the edge or other points with obvious features. Image processing and computer vision algorithms, such as corner detection, feature extraction, etc., are used to identify and extract feature points around the nodes. For each component node, three-dimensional back projection is carried out on the initial three-dimensional coordinates and surrounding characteristic points. And calculating the position of the characteristic point in the three-dimensional space, and fusing the characteristic point with the initial three-dimensional coordinates of the node to obtain the more accurate three-dimensional space position coordinates of the component node. And according to the actual situation, selecting an appropriate correction method to carry out refined correction processing on the three-dimensional space position coordinates of the component nodes. To error correction, coordinate system conversion, or other correction methods. And correcting and adjusting the three-dimensional space position coordinates of the nodes according to the selected correction method so as to improve the accuracy and reliability of the coordinates. And carrying out three-dimensional structure analysis of the lithium battery pack based on the corrected three-dimensional space position coordinates of the nodes. And analyzing the distance, angle, connection relation and the like among the nodes by using a computer graphics and geometric analysis method so as to acquire the three-dimensional structure data of the lithium battery pack. And converting the three-dimensional structure data of the lithium battery pack into a three-dimensional point cloud form, wherein each point represents the three-dimensional space position of one component node. And performing point cloud reconstruction on the three-dimensional structure data of the lithium battery pack by using a point cloud processing and reconstruction algorithm, such as point cloud registration, surface reconstruction and the like. A three-dimensional point cloud model of the lithium battery pack is ultimately generated, which is used for further analysis, visualization, and other applications.
In this embodiment, step S4 includes the following steps:
Step S41: performing region segmentation processing on the VOC concentration distribution scene graph to obtain a plurality of concentration region scene graphs;
step S42: carrying out concentration situation evolution on the plurality of concentration area scene graphs to generate area concentration situation evolution data;
step S43: future concentration prediction is carried out on the regional concentration situation evolution data to obtain regional VOC concentration prediction data;
step S44: performing abnormal concentration identification on the regional VOC concentration prediction data based on a preset VOC concentration distribution threshold value to obtain VOC abnormal concentration data;
Step S45: and carrying out abnormal region positioning on the concentration region scene graph based on the VOC abnormal concentration data so as to extract the abnormal concentration region scene graph.
In this embodiment, the region of interest is extracted from the VOC concentration profile scene graph.
The VOC concentration scene graph is segmented into a plurality of image segments or segmented regions representing regions of different concentration using image processing and computer vision methods such as threshold segmentation, edge detection, or region growing algorithms. And calculating the change trend and evolution condition of the concentration value of each concentration region scene graph.
The concentration value of each concentration region is subjected to situation evolution analysis using a time series analysis method such as an average value, variance, regression analysis, etc., which generates data describing a concentration change such as a concentration curve, a concentration change rate, etc. Based on the regional concentration situation evolution data, a prediction model or a time sequence analysis method is used for predicting the future concentration.
Future concentration values for each region are predicted taking into account historical concentration data, evolution trends, and other relevant factors, which will generate predicted data for the region VOC concentration, such as a concentration prediction curve over a future hours or days. A threshold range of the VOC concentration distribution is defined for determining whether the concentration is abnormal.
And carrying out threshold judgment on the regional VOC concentration prediction data, and identifying abnormal concentration data exceeding or falling below a preset threshold value, wherein the abnormal concentration data represent the region and the time point of the abnormal VOC concentration. And positioning a corresponding concentration region scene graph according to the region and time information in the abnormal concentration data.
Region image segments corresponding to the abnormal concentration data are extracted from the original concentration region scene graph, and represent the region scene graph with abnormal VOC concentration.
In this embodiment, step S5 includes the following steps:
Step S51: performing lithium battery pack position matching on the abnormal concentration area scene graph based on the sensor node spatial position to obtain the abnormal concentration position of the lithium battery pack;
Step S52: performing abnormal heat area positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration position of the lithium battery pack so as to obtain an abnormal heat area model;
step S53: performing thermal runaway simulation on the abnormal thermal zone model to obtain abnormal zone thermal runaway simulation data;
step S54: and carrying out battery assembly response analysis on the abnormal region thermal runaway simulation data to generate battery assembly thermal runaway response data.
In this embodiment, the position information of the sensor node is matched with the scene graph of the abnormal concentration area, the area in the scene graph of the abnormal concentration area is associated with the corresponding sensor node position by a spatial position matching algorithm, such as a nearest neighbor algorithm or a matching method based on a distance threshold, the position of the lithium battery pack corresponding to the abnormal concentration area is determined according to the matching result, the position information of the lithium battery pack in the space is provided, the three-dimensional point cloud model of the lithium battery pack is obtained based on the abnormal concentration position information of the lithium battery pack, the geometric shape and the spatial information of the lithium battery pack are obtained by a laser scanning or three-dimensional reconstruction technology, the abnormal thermal area is located and identified according to the abnormal concentration position information in the three-dimensional point cloud model of the lithium battery pack, the clustering algorithm, the abnormal detection algorithm or other image processing methods are used for extracting the abnormal thermal area, integrating the identified abnormal thermal region into an abnormal thermal region model to represent a region with abnormal heat in the lithium battery pack, performing thermal runaway simulation based on the abnormal thermal region model, simulating temperature distribution and thermal coupling effect of the abnormal thermal region by using a thermal conduction model, a thermodynamic model or other related models, in the simulation process, considering physical characteristics, thermal characteristics and thermal runaway mechanisms of the lithium battery pack, simulating temperature change and thermal runaway conditions of the abnormal region, generating abnormal region thermal runaway simulation data including information such as temperature distribution, temperature change trend and the like of the abnormal region, analyzing thermal runaway response of the lithium battery pack according to the abnormal region thermal runaway simulation data, evaluating the thermal runaway degree of the battery pack by analyzing the temperature distribution and the change conditions inside the abnormal region, using the thermal characteristics and performance parameters of the battery pack, the temperature change of the abnormal region is analyzed and evaluated, which provides thermal runaway response data of the battery assembly in the abnormal region, such as a temperature rise rate, a temperature gradient, and the like.
In this embodiment, step S6 includes the following steps:
Step S61: analyzing the thermal runaway fault type of the battery assembly according to the thermal runaway response data of the battery assembly to generate thermal runaway fault type data;
Step S62: performing risk quantification assessment on the thermal runaway fault type data to generate thermal runaway risk assessment data;
Step S63: performing thermal runaway decision analysis on the thermal runaway response data of the battery assembly based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy;
step S64: and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
In this embodiment, the thermal runaway response data of the battery assembly is used to analyze the type of fault in each abnormal region, determine the type of thermal runaway fault that exists according to the temperature change, temperature gradient, temperature rise rate, etc. of the abnormal region, record the type of thermal runaway fault for each abnormal region, generate thermal runaway fault type data for each abnormal region, correlate the type of thermal runaway fault with the corresponding position information of each abnormal region, evaluate the risk of different fault types based on the thermal runaway fault type data, assign a corresponding risk level or score for each thermal runaway fault type in consideration of factors such as severity, occurrence frequency, consequences, etc. of the fault type, perform risk evaluation by a qualitative or quantitative method, for example, using a risk matrix, probability distribution and other methods to generate thermal runaway risk assessment data, associating each thermal runaway fault type with a corresponding risk level or score, carrying out decision analysis on the thermal runaway response data of the battery assembly based on the thermal runaway risk assessment data, taking factors such as risk levels, accumulated risks and consequences of different fault types into consideration, making a thermal runaway decision strategy comprising processing measures, preventive measures and emergency response measures for different fault types, determining a corresponding decision scheme according to the height of the risk levels, associating the thermal runaway decision strategy with the thermal runaway response data of the battery assembly to form a decision strategy library, making a thermal runaway monitoring operation plan of the lithium battery assembly according to the thermal runaway decision strategy library, determining a monitoring frequency, a monitoring parameter and a monitoring method according to the monitoring requirements and the decision strategy, and executing thermal runaway monitoring operation, including periodically or in real time monitoring parameters such as temperature, temperature distribution, temperature change and the like of the battery pack, monitoring by using a sensor, thermal imaging equipment or other monitoring tools, performing real-time analysis and judgment according to monitoring data and a thermal runaway decision strategy, taking corresponding decisions according to monitoring results, including measures such as early warning, alarming, emergency shutdown, maintenance and the like, continuously optimizing the thermal runaway decision strategy according to the result and feedback of the monitoring operation, and updating a decision strategy library to improve the efficiency and accuracy of monitoring and processing.
In the present embodiment, there is provided a VOC-based lithium battery pack thermal runaway monitoring system for performing the VOC-based lithium battery pack thermal runaway monitoring method as described above, including:
the concentration distribution module is used for acquiring the VOC parameters in the real-time lithium battery pack based on the distributed sensor nodes; carrying out spatial concentration distribution analysis on VOC parameters in the real-time lithium battery pack to generate VOC concentration distribution fluctuation trend data;
the concentration distribution scene module is used for constructing a three-dimensional coordinate system based on the distributed sensor nodes; performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to obtain a smooth optimized concentration distribution network; performing global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph;
The three-dimensional model module is used for acquiring a multi-azimuth image of the lithium battery pack; carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes; carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes to construct a three-dimensional point cloud model of the lithium battery pack;
The concentration situation evolution module is used for carrying out concentration situation evolution on the VOC concentration distribution scene graph to generate regional concentration situation evolution data; abnormal concentration identification is carried out on the regional concentration situation evolution data based on a preset VOC concentration distribution threshold value so as to obtain VOC abnormal concentration data; extracting an abnormal concentration region scene graph based on the VOC abnormal concentration data;
The thermal runaway response module is used for carrying out abnormal thermal area positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration area scene graph so as to obtain an abnormal thermal area model; performing thermal runaway response analysis on the abnormal thermal region model to generate thermal runaway response data of the battery assembly;
The thermal runaway decision module is used for performing risk quantification assessment on the thermal runaway response data of the battery assembly so as to generate thermal runaway risk assessment data; performing thermal runaway decision analysis based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy; and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
The invention acquires the VOC parameters in the lithium battery pack through the distributed sensor nodes, monitors the concentration of Volatile Organic Compounds (VOC) in the battery pack in real time, is favorable for knowing the chemical environment in the battery pack, identifying potential volatile matters leakage or abnormal conditions, generating VOC concentration distribution fluctuation trend data, providing detailed analysis on VOC concentration change, facilitating the discovery of abnormal conditions and taking corresponding measures, obtaining a smooth optimized concentration distribution network of VOC concentration in the lithium battery pack by constructing a three-dimensional coordinate system and carrying out space mapping on the VOC concentration fluctuation trend data, providing more accurate and comprehensive VOC concentration information, facilitating the understanding and analysis of the VOC concentration in the internal space of the lithium battery pack, and further improving the quality of a concentration distribution scene graph through global deep convolution extraction, providing more accurate VOC concentration distribution information, determining the positions of assembly nodes of a battery pack by acquiring multi-azimuth images of the lithium battery pack and carrying out key node identification and marking on the images, carrying out three-dimensional point cloud reconstruction on the assembly nodes of the battery pack, constructing a three-dimensional point cloud model of the lithium battery pack, providing three-dimensional representation of the internal structure of the battery pack, facilitating subsequent thermal runaway response analysis and abnormal hot area positioning identification, carrying out concentration situation evolution analysis on a VOC concentration distribution scene graph, generating regional concentration situation evolution data, providing trend and evolution situation of VOC concentration change in a region, helping to find the mode and trend of concentration change, carrying out abnormal concentration identification on the regional concentration situation evolution data based on a preset VOC concentration distribution threshold value, further obtaining VOC abnormal concentration data, the method is favorable for timely finding and positioning abnormal VOC concentration areas, carrying out abnormal thermal area positioning identification on a three-dimensional point cloud model of the lithium battery pack based on an abnormal concentration area scene graph, determining the position of the abnormal thermal area, carrying out thermal runaway response analysis on the abnormal thermal area model, generating thermal runaway response data of the battery pack, providing characteristics and attributes of the abnormal thermal area, facilitating understanding and evaluating the degree and influence of a thermal runaway event, carrying out risk quantification evaluation on the thermal runaway response data of the battery pack, carrying out quantitative evaluation on the thermal runaway risk, generating thermal runaway risk evaluation data, providing information on the risk level and the potential influence of the thermal runaway event, carrying out thermal runaway decision analysis based on the thermal runaway risk evaluation data, and making corresponding decision strategies, such as taking emergency maintenance measures or taking safety isolation measures to reduce the thermal runaway risk, wherein the decision strategies are favorable for timely coping with potential thermal runaway problems and guaranteeing safe operation of the lithium battery pack.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
The foregoing is merely a specific embodiment of the invention to enable a person skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. The lithium battery pack thermal runaway monitoring method based on the VOC is characterized by comprising the following steps of:
Step S1: acquiring the internal VOC parameters of the real-time lithium battery pack based on the distributed sensor nodes; carrying out spatial concentration distribution analysis on VOC parameters in the real-time lithium battery pack to generate VOC concentration distribution fluctuation trend data;
Step S2: constructing a three-dimensional coordinate system based on the distributed sensor nodes; performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to obtain a smooth optimized concentration distribution network; performing global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph;
step S3: acquiring a multi-azimuth image of the lithium battery pack; carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes; carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes to construct a three-dimensional point cloud model of the lithium battery pack;
Step S4: carrying out concentration situation evolution on the VOC concentration distribution scene graph to generate regional concentration situation evolution data; abnormal concentration identification is carried out on the regional concentration situation evolution data based on a preset VOC concentration distribution threshold value so as to obtain VOC abnormal concentration data; extracting an abnormal concentration region scene graph based on the VOC abnormal concentration data;
step S5: carrying out abnormal heat region positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration region scene graph so as to obtain an abnormal heat region model; performing thermal runaway response analysis on the abnormal thermal region model to generate thermal runaway response data of the battery assembly;
Step S6: performing risk quantification assessment on the battery assembly thermal runaway response data to generate thermal runaway risk assessment data; performing thermal runaway decision analysis based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy; and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
2. The VOC-based lithium battery pack thermal runaway monitoring method of claim 1, wherein the specific steps of step S1 are:
Step S11: monitoring the real-time working state of the lithium battery pack based on the distributed sensor nodes, and acquiring the internal VOC parameters of the real-time lithium battery pack;
Step S12: performing time sequence concentration characteristic analysis on the VOC parameters in the real-time lithium battery pack to generate VOC time sequence concentration characteristic data;
step S13: performing spatial concentration distribution analysis on the VOC time sequence concentration characteristic data to obtain VOC time sequence concentration distribution data;
Step S14: and quantifying fluctuation trend of the VOC time sequence concentration distribution data to generate VOC concentration distribution fluctuation trend data.
3. The VOC-based lithium battery pack thermal runaway monitoring method of claim 1, wherein the specific steps of step S2 are:
step S21: carrying out spatial position identification on the distributed sensor nodes to obtain the spatial positions of the sensor nodes;
step S22: constructing a three-dimensional coordinate system based on the spatial positions of the sensor nodes;
Step S23: performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to generate a VOC three-dimensional concentration distribution network;
step S24: performing iterative interpolation smooth optimization on the VOC three-dimensional concentration distribution network to obtain a smooth optimized concentration distribution network;
Step S25: and carrying out global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph.
4. The VOC-based lithium battery pack thermal runaway monitoring method of claim 3, wherein the specific steps of step S25 are:
performing discrete voxel decomposition on the smooth optimized concentration distribution network to obtain a plurality of three-dimensional concentration voxel units;
Carrying out local convolution filtering treatment on a plurality of three-dimensional concentration voxel units, and extracting the parameter combination characteristics of the voxel units;
performing multi-scale deep convolution feature mining on the voxel unit parameter combination features to generate multi-scale global concentration structure information;
And carrying out global feature activation processing on the smooth optimized concentration distribution network by utilizing the multi-scale global concentration structure information so as to obtain a VOC concentration distribution scene graph.
5. The VOC-based lithium battery pack thermal runaway monitoring method of claim 1, wherein the specific steps of step S3 are:
Step S31: acquiring a multi-azimuth image of the lithium battery pack;
Step S32: carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes;
step S33: extracting image edges of multi-azimuth images of the lithium battery pack to obtain multi-azimuth image edge contour lines;
step S34: performing feature point overlapping matching on the battery pack assembly nodes based on the multi-azimuth image edge contour lines to generate an image overlapping region;
Step S35: and carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes according to the image overlapping region, and constructing a three-dimensional point cloud model of the lithium battery pack.
6. The VOC-based lithium battery pack thermal runaway monitoring method of claim 5, wherein the specific steps of step S35 are:
performing three-line intersection method calculation on the battery pack assembly nodes according to the image overlapping areas to obtain initial three-dimensional node coordinates;
Extracting adjacent characteristic points of the battery pack assembly nodes according to the image overlapping areas to obtain characteristic points around the nodes;
performing three-dimensional back projection on the initial three-dimensional node coordinates based on the feature point degrees around the nodes to generate component node three-dimensional space position coordinates;
Carrying out fine correction processing on the three-dimensional space position coordinates of the component nodes to obtain corrected three-dimensional space position coordinates of the nodes;
carrying out three-dimensional structure analysis on the three-dimensional space position coordinates of the correction nodes to obtain three-dimensional structure data of the lithium battery pack;
And carrying out three-dimensional point cloud reconstruction on the three-dimensional structure data of the lithium battery pack, and constructing a three-dimensional point cloud model of the lithium battery pack.
7. The VOC-based lithium battery pack thermal runaway monitoring method of claim 1, wherein the specific steps of step S4 are:
Step S41: performing region segmentation processing on the VOC concentration distribution scene graph to obtain a plurality of concentration region scene graphs;
step S42: carrying out concentration situation evolution on the plurality of concentration area scene graphs to generate area concentration situation evolution data;
step S43: future concentration prediction is carried out on the regional concentration situation evolution data to obtain regional VOC concentration prediction data;
step S44: performing abnormal concentration identification on the regional VOC concentration prediction data based on a preset VOC concentration distribution threshold value to obtain VOC abnormal concentration data;
Step S45: and carrying out abnormal region positioning on the concentration region scene graph based on the VOC abnormal concentration data so as to extract the abnormal concentration region scene graph.
8. The VOC-based lithium battery pack thermal runaway monitoring method of claim 1, wherein the specific steps of step S5 are:
Step S51: performing lithium battery pack position matching on the abnormal concentration area scene graph based on the sensor node spatial position to obtain the abnormal concentration position of the lithium battery pack;
Step S52: performing abnormal heat area positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration position of the lithium battery pack so as to obtain an abnormal heat area model;
step S53: performing thermal runaway simulation on the abnormal thermal zone model to obtain abnormal zone thermal runaway simulation data;
step S54: and carrying out battery assembly response analysis on the abnormal region thermal runaway simulation data to generate battery assembly thermal runaway response data.
9. The VOC-based lithium battery pack thermal runaway monitoring method of claim 1, wherein the specific steps of step S6 are:
Step S61: analyzing the thermal runaway fault type of the battery assembly according to the thermal runaway response data of the battery assembly to generate thermal runaway fault type data;
Step S62: performing risk quantification assessment on the thermal runaway fault type data to generate thermal runaway risk assessment data;
Step S63: performing thermal runaway decision analysis on the thermal runaway response data of the battery assembly based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy;
step S64: and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
10. A VOC-based lithium battery thermal runaway monitoring system for performing the VOC-based lithium battery thermal runaway monitoring method of claim 1, comprising:
the concentration distribution module is used for acquiring the VOC parameters in the real-time lithium battery pack based on the distributed sensor nodes; carrying out spatial concentration distribution analysis on VOC parameters in the real-time lithium battery pack to generate VOC concentration distribution fluctuation trend data;
the concentration distribution scene module is used for constructing a three-dimensional coordinate system based on the distributed sensor nodes; performing space mapping on the three-dimensional coordinate system based on the VOC concentration distribution fluctuation trend data to obtain a smooth optimized concentration distribution network; performing global deep convolution extraction on the smooth optimized concentration distribution network to obtain a VOC concentration distribution scene graph;
The three-dimensional model module is used for acquiring a multi-azimuth image of the lithium battery pack; carrying out key node identification on the multi-azimuth image of the lithium battery pack, and marking the battery pack assembly nodes; carrying out three-dimensional point cloud reconstruction on the battery pack assembly nodes to construct a three-dimensional point cloud model of the lithium battery pack;
The concentration situation evolution module is used for carrying out concentration situation evolution on the VOC concentration distribution scene graph to generate regional concentration situation evolution data; abnormal concentration identification is carried out on the regional concentration situation evolution data based on a preset VOC concentration distribution threshold value so as to obtain VOC abnormal concentration data; extracting an abnormal concentration region scene graph based on the VOC abnormal concentration data;
The thermal runaway response module is used for carrying out abnormal thermal area positioning identification on the three-dimensional point cloud model of the lithium battery pack based on the abnormal concentration area scene graph so as to obtain an abnormal thermal area model; performing thermal runaway response analysis on the abnormal thermal region model to generate thermal runaway response data of the battery assembly;
The thermal runaway decision module is used for performing risk quantification assessment on the thermal runaway response data of the battery assembly so as to generate thermal runaway risk assessment data; performing thermal runaway decision analysis based on the thermal runaway risk assessment data, and constructing a thermal runaway decision strategy; and performing thermal runaway monitoring operation of the lithium battery pack based on the thermal runaway decision strategy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410902615.7A CN118483596B (en) | 2024-07-08 | 2024-07-08 | VOC-based lithium battery pack thermal runaway monitoring method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410902615.7A CN118483596B (en) | 2024-07-08 | 2024-07-08 | VOC-based lithium battery pack thermal runaway monitoring method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118483596A true CN118483596A (en) | 2024-08-13 |
| CN118483596B CN118483596B (en) | 2024-09-17 |
Family
ID=92198142
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410902615.7A Active CN118483596B (en) | 2024-07-08 | 2024-07-08 | VOC-based lithium battery pack thermal runaway monitoring method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118483596B (en) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019174653A2 (en) * | 2019-06-17 | 2019-09-19 | 广东恒翼能科技有限公司 | Lithium battery thermal runaway early warning protection system and method |
| CN113671385A (en) * | 2021-03-29 | 2021-11-19 | 国网江苏省电力有限公司南京供电分公司 | Evaluation method and system for abnormal state of lithium ion battery based on VOC gas |
| CN114035053A (en) * | 2021-11-11 | 2022-02-11 | 合肥工业大学 | Lithium battery thermal runaway early warning method and system thereof |
| CN114559815A (en) * | 2020-11-27 | 2022-05-31 | 李玉钦 | New energy automobile battery thermal runaway judging and handling method |
| CN114695986A (en) * | 2022-03-29 | 2022-07-01 | 湘潭大学 | High-capacity LFP battery thermal runaway early warning method based on gas evolution characteristics |
| WO2023061039A1 (en) * | 2021-10-13 | 2023-04-20 | 中通服和信科技有限公司 | Tailing pond risk monitoring and early-warning system based on internet of things |
| US20230130896A1 (en) * | 2021-10-15 | 2023-04-27 | Samsung Sdi Co., Ltd. | Battery charging method and battery pack |
| CN116184209A (en) * | 2022-12-06 | 2023-05-30 | 浙江凌骁能源科技有限公司 | Early warning method and device suitable for thermal runaway of power battery |
| CN116973763A (en) * | 2023-07-14 | 2023-10-31 | 吉林建筑大学 | Method and device for early warning out of control of lithium battery Bao Re |
| US20230375492A1 (en) * | 2022-02-09 | 2023-11-23 | Purdue Research Foundation | Methods and apparatus for sensing volatile organic compounds and gases released from electrochemical cells |
| CN117761559A (en) * | 2023-12-21 | 2024-03-26 | 一汽解放汽车有限公司 | Method, device, equipment and medium for constructing fuel cell thermal runaway prediction model |
| CN117969760A (en) * | 2024-02-01 | 2024-05-03 | 广州邦禾检测技术有限公司 | Method for detecting thermal runaway gas of battery |
| CN118151017A (en) * | 2024-03-28 | 2024-06-07 | 上海电力大学 | Lithium battery thermal runaway early warning method and medium |
| CN118144571A (en) * | 2024-05-11 | 2024-06-07 | 太原科技大学 | Multi-model fusion power battery thermal runaway early warning system and method |
-
2024
- 2024-07-08 CN CN202410902615.7A patent/CN118483596B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019174653A2 (en) * | 2019-06-17 | 2019-09-19 | 广东恒翼能科技有限公司 | Lithium battery thermal runaway early warning protection system and method |
| CN114559815A (en) * | 2020-11-27 | 2022-05-31 | 李玉钦 | New energy automobile battery thermal runaway judging and handling method |
| CN113671385A (en) * | 2021-03-29 | 2021-11-19 | 国网江苏省电力有限公司南京供电分公司 | Evaluation method and system for abnormal state of lithium ion battery based on VOC gas |
| WO2023061039A1 (en) * | 2021-10-13 | 2023-04-20 | 中通服和信科技有限公司 | Tailing pond risk monitoring and early-warning system based on internet of things |
| US20230130896A1 (en) * | 2021-10-15 | 2023-04-27 | Samsung Sdi Co., Ltd. | Battery charging method and battery pack |
| CN114035053A (en) * | 2021-11-11 | 2022-02-11 | 合肥工业大学 | Lithium battery thermal runaway early warning method and system thereof |
| US20230375492A1 (en) * | 2022-02-09 | 2023-11-23 | Purdue Research Foundation | Methods and apparatus for sensing volatile organic compounds and gases released from electrochemical cells |
| CN114695986A (en) * | 2022-03-29 | 2022-07-01 | 湘潭大学 | High-capacity LFP battery thermal runaway early warning method based on gas evolution characteristics |
| CN116184209A (en) * | 2022-12-06 | 2023-05-30 | 浙江凌骁能源科技有限公司 | Early warning method and device suitable for thermal runaway of power battery |
| CN116973763A (en) * | 2023-07-14 | 2023-10-31 | 吉林建筑大学 | Method and device for early warning out of control of lithium battery Bao Re |
| CN117761559A (en) * | 2023-12-21 | 2024-03-26 | 一汽解放汽车有限公司 | Method, device, equipment and medium for constructing fuel cell thermal runaway prediction model |
| CN117969760A (en) * | 2024-02-01 | 2024-05-03 | 广州邦禾检测技术有限公司 | Method for detecting thermal runaway gas of battery |
| CN118151017A (en) * | 2024-03-28 | 2024-06-07 | 上海电力大学 | Lithium battery thermal runaway early warning method and medium |
| CN118144571A (en) * | 2024-05-11 | 2024-06-07 | 太原科技大学 | Multi-model fusion power battery thermal runaway early warning system and method |
Non-Patent Citations (1)
| Title |
|---|
| 钟睿;雷印杰;周新志;谭凯;林健;: "基于神经网络参数优化的热失控状态温度预测研究", 计算机测量与控制, no. 12, 25 December 2015 (2015-12-25) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118483596B (en) | 2024-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116129366B (en) | Digital twinning-based park monitoring method and related device | |
| CN109870617B (en) | Intelligent power plant electrical equipment fault diagnosis method based on width learning and infrared image space-time characteristics | |
| CN113569956B (en) | Mountain fire disaster investigation and identification method based on AI algorithm | |
| CN116704733B (en) | Aging early warning method and system for aluminum alloy cable | |
| CN118094438A (en) | Distribution cable operation analysis method and system based on big data | |
| CN118470550B (en) | Natural resource asset data acquisition method and platform | |
| CN117711131A (en) | Constructor warning protection system based on gesture danger analysis | |
| CN115130369A (en) | Transformer substation equipment comprehensive state evaluation and active early warning method based on digital twinning | |
| CN116308185A (en) | Visual progress input and early warning method for constructional engineering | |
| CN117422938B (en) | Dam slope concrete structure anomaly analysis method based on three-dimensional analysis platform | |
| CN117828309A (en) | Substation safety early warning method for multi-source data fusion ranging | |
| CN118483596B (en) | VOC-based lithium battery pack thermal runaway monitoring method and system | |
| CN117272082A (en) | Data processing method and system based on industrial Internet of things | |
| Lowin et al. | From physical to virtual: leveraging drone imagery to automate photovoltaic system maintenance | |
| CN118134044B (en) | Intelligent biological laboratory instrument sharing management method and system | |
| CN118334561B (en) | Smart panoramic inspection monitoring method and system for high-voltage cable | |
| CN118508438B (en) | Photovoltaic module power loss prediction method, system, equipment and storage medium | |
| CN117830032B (en) | Method and system for monitoring snapshot and risk assessment of power transmission line network | |
| CN118655475A (en) | Performance test method and system for energy storage power supply | |
| CN118397000B (en) | Dynamic data processing method and system based on industrial equipment | |
| CN118839175A (en) | Pollution source online monitoring method and system | |
| CN118521264A (en) | Forestry resource dynamic monitoring and asset assessment integrated technology | |
| CN118570211A (en) | Water quality abnormality detection analysis method and system based on deep learning | |
| CN118508438A (en) | Photovoltaic module power loss prediction method, system, equipment and storage medium | |
| WO2024035640A2 (en) | Probability of detection of lifecycle phases of corrosion under insulation using artificial intelligence and temporal thermography |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |