CN-121633857-B - Power battery safety detection and assessment system based on multi-mode visualization

CN121633857BCN 121633857 BCN121633857 BCN 121633857BCN-121633857-B

Abstract

The invention relates to the technical field of battery safety detection and discloses a power battery safety detection and evaluation system based on multi-mode visualization, which comprises a detection device and an evaluation device, wherein the detection device is connected with the evaluation device; the detection device comprises a closed test cavity, a safety triggering and loading module, a multi-mode parameter acquisition module and a data processing and visualizing module, wherein an analysis module constructs a coupling characteristic vector based on temperature characteristics, smoke characteristics and deformation characteristics, determines a safety evaluation value of a battery to be detected according to the coupling characteristic vector, judges whether the safety evaluation value is optimized according to electric power data, determines an optimization coefficient of the safety evaluation value based on air pressure and electric power data in a bin if the safety evaluation value is optimized, obtains an optimized safety evaluation value, and determines the safety level of the battery to be detected according to the optimized safety evaluation value. The invention can comprehensively, accurately and intuitively detect and evaluate the safety condition of the power battery.

Inventors

MA TIANYI
WANG XIAO
WANG FANG
HAN CE
SUN ZHIPENG
LIU LEI
LI YUPENG
HAN LIQIONG
Liu Daifan
HAO WEIJIAN

Assignees

中汽研新能源汽车检验中心(天津)有限公司
中国汽车技术研究中心有限公司

Dates

Publication Date: 20260508
Application Date: 20260130

Claims (9)

1. A multi-modality visualization-based power battery safety detection and assessment system, comprising: the device comprises a detection device and an evaluation device, wherein the detection device is connected with the evaluation device, the detection device comprises a closed test cavity, a safety triggering and loading module, a multi-mode parameter acquisition module and a data processing and visualization module, the evaluation device comprises an analysis module, a judgment optimization module and a judgment module, A battery to be detected is placed in the closed test cavity; the safety triggering and loading module is used for applying triggering loading to the battery to be detected under the controlled condition and obtaining response data of the battery to be detected under the triggering loading condition; the multi-mode parameter acquisition module is used for acquiring response data and preprocessing the response data; the data processing and visualizing module is used for drawing a stackable thermal field diagram, a smoke equivalent surface and a deformation curve of the battery to be detected according to the preprocessed response data; The analysis module is used for extracting temperature characteristics of the stackable thermal field diagram, smoke characteristics of the smoke equivalent surface and deformation characteristics of the deformation curve, constructing a coupling characteristic vector based on the temperature characteristics, the smoke characteristics and the deformation characteristics, and determining a safety evaluation value of the battery to be detected according to the coupling characteristic vector; The judging and optimizing module is used for collecting the power data of the battery to be detected, judging whether to optimize the safety evaluation value according to the power data, if so, collecting the air pressure in the bin of the closed test cavity, determining the optimizing coefficient of the safety evaluation value based on the air pressure in the bin and the power data, and obtaining an optimized safety evaluation value; the judging module is used for determining the safety level of the battery to be detected according to the optimized safety evaluation value; constructing a coupling feature vector based on the temperature feature, the smoke feature and the deformation feature, and determining a safety evaluation value of the battery to be detected according to the coupling feature vector, wherein the method comprises the following steps: Respectively carrying out normalization treatment on the temperature characteristic, the smoke characteristic and the deformation characteristic to obtain normalization characteristics; feature enhancement is carried out on the normalized features to obtain enhanced features; Performing time alignment and spatial alignment on the enhanced features to form an aligned multi-modal feature set; constructing coupling items among modes according to the multi-mode feature set, and combining the coupling items with the multi-mode feature set to form a coupling feature vector; Comparing the coupling characteristic vector with the historical coupling characteristic vector group, and determining a safety evaluation value of the battery to be detected according to the comparison result; The coupling items comprise a coupling item of temperature change rate and smoke diffusion gradient, a coupling item of smoke concentration peak value and deformation accumulation rate and a coupling item of temperature space mutation and deformation inflection point time sequence difference.
2. The multi-modality visualization-based power battery safety detection and assessment system of claim 1, wherein the controlled conditions include any one or more combination of triggering of overcharging/overdischarging, heating, squeezing, needling of the battery to be detected.
3. The multi-modal visualization-based power battery safety detection and assessment system according to claim 2, wherein when drawing a stackable thermal field map, a smoke iso-surface and a deformation curve of a battery to be detected according to the preprocessed response data, the system comprises: The response data comprises temperature data, smoke concentration data and deformation data; Processing the preprocessed temperature data, mapping temperature values at different moments and different positions to a two-dimensional plane by using a thermal imaging algorithm, and representing the temperature by the color depth to draw a stackable thermal field map; Processing the preprocessed flue gas concentration data, determining a distribution boundary of the flue gas in space by adopting an isosurface generating algorithm, and drawing a flue gas isosurface; And processing the preprocessed deformation data, and drawing a deformation curve by using time as a horizontal axis and deformation as a vertical axis and a fitting curve.
4. The multi-modality visualization-based power battery safety detection and assessment system according to claim 1, wherein when determining the safety assessment value of the battery to be detected according to the comparison result, the system comprises: when the historical coupling feature vector group has the same historical coupling feature vector as the coupling feature vector, taking a historical security evaluation value corresponding to the historical coupling feature vector as a security evaluation value; when the historical coupling feature vector which is the same as the coupling feature vector does not exist in the historical coupling feature vector group, calculating the state offset rate of the battery to be detected according to the coupling feature vector, and determining the safety evaluation value of the battery to be detected according to the state offset rate.
5. The multi-modality visualization-based power battery safety detection and assessment system of claim 4, wherein when calculating the state shift rate of the battery to be detected from the coupling feature vector, comprising: Acquiring a time sequence of the coupling feature vectors, and calculating Euclidean distances between the coupling feature vectors at adjacent moments based on the time sequence of the coupling feature vectors to acquire coupling variation; weighting the coupling variable quantity, wherein the weight consists of a temperature characteristic weight, a smoke characteristic weight and a deformation characteristic weight so as to obtain a weighted coupling variable quantity; and obtaining the change rate of the weighted coupling change quantity in a preset time window to obtain the state offset rate.
6. The multi-modality visualization-based power battery safety detection and assessment system of claim 5, wherein when determining the safety assessment value of the battery to be detected based on the state shift rate, comprising: Comparing the state offset rate with a first state offset rate and a second state offset rate, and determining a safety evaluation value according to the comparison result, wherein the first state offset rate is smaller than the second state offset rate; when the state offset rate is less than or equal to the first state offset rate, determining the security evaluation value as a first security evaluation value; When the state offset rate is greater than the first state offset rate and less than or equal to the second state offset rate, determining the security evaluation value as a second security evaluation value; and when the state offset rate is greater than the second state offset rate, determining the safety evaluation value as a third safety evaluation value.
7. The multi-modality visualization-based power battery safety detection and assessment system of claim 6, wherein determining whether to optimize the safety assessment value based on the power data comprises: Analyzing the electric power data, and carrying out denoising and filtering treatment on the electric power data; acquiring power mutation indexes based on the filtered power data, wherein the power mutation indexes comprise amplitude mutation rate, relative change rate and duration; comparing the power mutation index with a preset threshold value, and judging that the power data has a power mutation value only when the power mutation index meets the preset amplitude threshold value and continuously exceeds the preset times within a preset time window; When the power data is judged to have the power mutation value, the safety evaluation value is judged to be required to be optimized; When it is determined that the power data does not have the power abrupt value, it is determined that the safety evaluation value does not need to be optimized.
8. The multi-modality visualization-based power battery safety detection and assessment system of claim 7, wherein when determining the optimization factor of the safety assessment value based on the in-bin air pressure and power data and obtaining the optimized safety assessment value, comprising: analyzing the air pressure in the bin to obtain the air pressure change rate; Analyzing the power data to obtain a real-time current value; Comparing the air pressure change rate with an air pressure change threshold, comparing the real-time current value with the current threshold, and determining an optimization coefficient according to the comparison result; When the air pressure change rate is greater than or equal to the air pressure change threshold value and the real-time current value is greater than or equal to the current threshold value, determining the optimization coefficient as a first optimization coefficient; when the air pressure change rate is greater than or equal to the air pressure change threshold value and the real-time current value is smaller than the current threshold value, determining the optimization coefficient as a second optimization coefficient; When the air pressure change rate is smaller than the air pressure change threshold value and the real-time current value is larger than or equal to the current threshold value, determining the optimization coefficient as a third optimization coefficient; When the air pressure change rate is smaller than the air pressure change threshold value and the real-time current value is smaller than the current threshold value, determining the optimization coefficient as a fourth optimization coefficient; and taking the product value of the optimization coefficient and the safety evaluation value as an optimized safety evaluation value.
9. The multi-modality visualization-based power battery safety detection and assessment system of claim 8, wherein when determining the safety level of the battery to be detected based on the optimized safety assessment value, comprising: And comparing the optimized safety evaluation value with a preset safety grade mapping table, and determining the safety grade of the battery to be detected according to the comparison result.

Description

Power battery safety detection and assessment system based on multi-mode visualization Technical Field The invention relates to the technical field of battery safety detection, in particular to a power battery safety detection and evaluation system based on multi-mode visualization. Background With the rapid improvement of the permeability of the new energy automobile, how the safety problem of the power battery changes from safety or not to safety is visualized and quantitatively evaluated. The existing laboratory detection device is mostly unfolded around a single working condition or parameter, is used for collecting and displaying link fracture, is difficult to restore real evolution of a thermal, electric, force and gas coupling process, is difficult to migrate to different battery systems and complex scenes due to the fact that a simple threshold value or linear weighted score is adopted in data application, and results are insufficient in interpretation, contrast and reproducibility. Meanwhile, the battery object is expanded from a traditional lithium ion system to an all-solid-state battery, and when the solid-state system is abnormal, the corrosive or toxic gas and high-temperature solid/gas two-phase coupling behavior is accompanied, so that higher requirements are put on test cavity materials and the like. The conventional device is mainly a general metal cavity and a conventional gas sensor scheme, so that the performances of temperature resistance, corrosion resistance and the like are difficult to consider, and the conventional system only stays on a simple monitoring level for visualization, so that linkage analysis of thermal field-smoke-mechanical deformation cannot be performed, and interpretation of battery safety is affected. On the test execution level, the traditional platform has various problems that the cavity volume and the temperature/pressure resistant allowance are limited, the high-capacity monomer exothermic and outgassing peak value is difficult to cover, the efficiency and consistency of the opening and closing and sealing links are insufficient, the triggering/loading units are distributed, the parameter resolution is limited, the programmable working condition path scanning is difficult to complete, and the data acquisition lacks a unified model, so that the subsequent analysis and comparison cost is high. These factors limit the systematic awareness of battery safety, especially cross-system, cross-scenario consistency assessment. Therefore, there is a need to design a multi-modal visualization-based power battery safety detection and assessment system to solve the problems in the prior art. Disclosure of Invention In view of the above, the present invention provides a system for detecting and evaluating safety of a power battery based on multi-modal visualization, which aims to solve the above-mentioned problems. The invention provides a power battery safety detection and evaluation system based on multi-mode visualization, which comprises the following steps: the device comprises a detection device and an evaluation device, wherein the detection device is connected with the evaluation device, the detection device comprises a closed test cavity, a safety triggering and loading module, a multi-mode parameter acquisition module and a data processing and visualization module, the evaluation device comprises an analysis module, a judgment optimization module and a judgment module, A battery to be detected is placed in the closed test cavity; the safety triggering and loading module is used for applying triggering loading to the battery to be detected under the controlled condition and obtaining response data of the battery to be detected under the triggering loading condition; the multi-mode parameter acquisition module is used for acquiring response data and preprocessing the response data; the data processing and visualizing module is used for drawing a stackable thermal field diagram, a smoke equivalent surface and a deformation curve of the battery to be detected according to the preprocessed response data; The analysis module is used for extracting temperature characteristics of the stackable thermal field diagram, smoke characteristics of the smoke equivalent surface and deformation characteristics of the deformation curve, constructing a coupling characteristic vector based on the temperature characteristics, the smoke characteristics and the deformation characteristics, and determining a safety evaluation value of the battery to be detected according to the coupling characteristic vector; The judging and optimizing module is used for collecting the power data of the battery to be detected, judging whether to optimize the safety evaluation value according to the power data, if so, collecting the air pressure in the bin of the closed test cavity, determining the optimizing coefficient of the safety evaluation value based on the air pressure in the bin and the power dat