CN-122017593-A - Fault prediction method, system, equipment, medium and product based on multi-physical field coupling

CN122017593ACN 122017593 ACN122017593 ACN 122017593ACN-122017593-A

Abstract

The invention relates to the technical field of electric power systems, and discloses a fault prediction method, a system, equipment, a medium and a product based on multi-physical field coupling, wherein the method determines the current multi-physical field distribution of a battery module by inputting the current operation data of the battery module into a multi-physical field coupling dynamics model, thereby revealing a complex coupling evolution mechanism in the battery from the multi-physical field coupling view angle, estimating the current estimation state of the battery module by adopting extended Kalman filtering, the method comprises the steps of correcting the current state of the battery module, predicting the estimated state of the battery module at the future preset moment through a risk trend prediction model based on LSTM, and predicting the fault state of the battery module through the estimated state of the battery module, so that the fault state of the battery module is predicted through multidimensional physical quantity, the battery fault can be comprehensively and timely accurately predicted, and the false alarm rate and the missing report rate of the battery fault are reduced.

Inventors

WANG QINGBIN
YAN HANGANG
WANG YUXI
YANG YUN
ZHAO XIANZHONG
YI BIN
YANG JIANFENG
Huang Gancai

Assignees

广东电网有限责任公司云浮供电局

Dates

Publication Date: 20260512
Application Date: 20260213

Claims (10)

1. A fault prediction method based on multi-physical field coupling, comprising: Determining current multi-physical-field distribution of the battery module by combining current operation data of the battery module according to a pre-constructed multi-physical-field coupling dynamics model of the battery module, wherein the multi-physical-field coupling dynamics model is obtained by coupling a plurality of physical fields including electric heat, stress, structural acoustics and aerodynamic force; Estimating the current estimation state of the battery module by adopting extended Kalman filtering according to the current multi-physical-field distribution; Based on a pre-trained risk trend prediction model of the LSTM, predicting an estimated state of the battery module at a future preset moment by combining the current estimated state of the battery module; and predicting the fault state of the battery module according to the predicted estimated state of the battery module.
2. The multi-physical field coupling based fault prediction method of claim 1, further comprising: Constructing a geometric model of the battery module according to the geometric structure of the battery module; respectively constructing an electrothermal field simulation model, a stress field simulation model, a structural acoustic field simulation model and a aerodynamic acoustic field simulation model according to the geometric model of the battery module; And coupling the electric heating field simulation model, the stress field simulation model, the structural acoustic field simulation model and the aerodynamic acoustic field simulation model to obtain a coupling dynamics model of the multiple physical fields.
3. The multi-physical field coupling based fault prediction method of claim 1, wherein the current multi-physical field profile comprises a stress field profile, a temperature field profile, an acoustic energy field profile, and a gas concentration field profile; Estimating the current estimation state of the battery module by adopting extended Kalman filtering according to the current multi-physical-field distribution, wherein the method comprises the following steps: Determining a state feature vector of the battery module according to the current multi-physical-field distribution and the multi-physical-field actual measurement data of the battery module, which are measured by the sensor; Based on an extended Kalman filtering algorithm, the temperature state, the stress state, the sound energy density state and the gas concentration of the battery module are determined to be used as the current estimation state by combining the state feature vector of the battery module.
4. The fault prediction method based on multiple physical field coupling according to claim 3, wherein the determining the state feature vector of the battery module according to the current multiple physical field distribution and the multiple physical field measured data of the battery module measured by the sensor comprises: Extracting physical field characteristics from the stress field distribution, the temperature field distribution, the acoustic energy field distribution and the aerodynamic field distribution respectively, wherein the physical field characteristics comprise stress gradient, temperature rise rate, acoustic energy density and gas escape concentration rate; acquiring multi-physical-field measured data measured by a sensor on the battery module, and determining multi-physical-field deviation according to residual errors between the multi-physical-field measured data and the current multi-physical-field distribution; and determining a state characteristic vector of the battery module according to the stress gradient, the temperature rise rate, the acoustic energy density, the gas escape concentration rate and the multi-physical-field deviation.
5. The multi-physical field coupling based fault prediction method according to claim 1, wherein the predicting the fault state of the battery module according to the predicted estimated state of the battery module comprises: Carrying out dimensionless transformation on the predicted estimated state of the battery module to obtain a plurality of dimensionless transformed estimated state characteristics; Weighting the plurality of dimensionless transformed estimated state features to obtain a fault risk index of the battery module; comparing the magnitude relation of the fault risk index and the fault risk index interval threshold values corresponding to the plurality of preset early warning levels respectively, and determining the preset early warning level corresponding to the fault risk index according to the comparison result; And determining the fault state of the battery module according to the preset early warning level corresponding to the fault risk index.
6. The multi-physical field coupling based fault prediction method as claimed in claim 3, further comprising: Determining the temperature change rate and the gas concentration change rate of the battery module according to the temperature field distribution and the gas dynamic field distribution; Determining a thermal runaway risk index of the battery module according to the temperature change rate and the gas concentration change rate of the battery module; And under the condition that the thermal runaway risk index continuously exceeds a preset thermal runaway risk threshold value in a preset period, judging that the thermal runaway risk occurs in the battery module.
7. A multi-physical field coupling based fault prediction system, comprising: The system comprises a physical field determining module, a battery module and a control module, wherein the physical field determining module is used for determining the current multi-physical-field distribution of the battery module according to a pre-constructed multi-physical-field coupling dynamics model of the battery module and the current operation data of the battery module, wherein the multi-physical-field coupling dynamics model is obtained by coupling a plurality of physical fields including electric heat, stress, structural acoustics and aerodynamic force; the state estimation module is used for estimating the current estimation state of the battery module by adopting extended Kalman filtering according to the current multi-physical-field distribution; the state prediction module is used for predicting the estimated state of the battery module at a future preset moment by combining the current estimated state of the battery module based on a pre-trained risk trend prediction model of the LSTM; and the fault state prediction module is used for predicting the fault state of the battery module according to the predicted estimated state of the battery module.
8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the multi-physical field coupling based fault prediction method of any one of claims 1-6.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed, implements the steps of the multi-physical field coupling based fault prediction method according to any of claims 1-6.
10. A computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions, wherein the program instructions, when executed by a computer, cause the computer to perform the steps of the multi-physical field coupling based fault prediction method as claimed in any one of claims 1-6.

Description

Fault prediction method, system, equipment, medium and product based on multi-physical field coupling Technical Field The invention relates to the technical field of power systems, in particular to a fault prediction method, a system, equipment, a medium and a product based on multi-physical field coupling. Background With the rapid development of energy storage systems, the safety and reliability of lithium ion batteries are a key challenge in the industry. Under complex working conditions, the battery may have faults such as thermal runaway, internal short circuit, structural fatigue, gas leakage and the like, and the evolution process of the battery involves the coupling action of multiple physical fields such as heat, electricity, force, chemistry, acoustics and the like. For example, mechanical abuse (e.g., impact extrusion) or electrical abuse (e.g., overcharging) may trigger internal shorts, which in turn initiate a chain exothermic reaction, ultimately leading to thermal runaway and even fire explosion. Currently, each standard puts stringent requirements on the thermal safety and the overall package safety of a battery system, and development of high-precision fault prediction and health management technologies is urgently needed. The traditional battery fault prediction method mostly focuses on fault signals with single dimension, and is difficult to predict battery faults comprehensively and timely, and the false alarm rate and the missing report rate are high. Disclosure of Invention In view of the above, the present invention provides a fault prediction method, system, device, medium and product based on multiple physical field coupling. The first aspect of the invention provides a fault prediction method based on multi-physical field coupling, which comprises the following steps: Determining current multi-physical-field distribution of the battery module by combining current operation data of the battery module according to a pre-constructed multi-physical-field coupling dynamics model of the battery module, wherein the multi-physical-field coupling dynamics model is obtained by coupling a plurality of physical fields including electric heat, stress, structural acoustics and aerodynamic force; Estimating the current estimation state of the battery module by adopting extended Kalman filtering according to the current multi-physical-field distribution; Based on a pre-trained risk trend prediction model of the LSTM, predicting an estimated state of the battery module at a future preset moment by combining the current estimated state of the battery module; and predicting the fault state of the battery module according to the predicted estimated state of the battery module. Preferably, the method further comprises: Constructing a geometric model of the battery module according to the geometric structure of the battery module; respectively constructing an electrothermal field simulation model, a stress field simulation model, a structural acoustic field simulation model and a aerodynamic acoustic field simulation model according to the geometric model of the battery module; And coupling the electric heating field simulation model, the stress field simulation model, the structural acoustic field simulation model and the aerodynamic acoustic field simulation model to obtain a coupling dynamics model of the multiple physical fields. Preferably, the current multi-physical field profile comprises a stress field profile, a temperature field profile, an acoustic energy field profile, and a gas concentration field profile; Estimating the current estimation state of the battery module by adopting extended Kalman filtering according to the current multi-physical-field distribution, wherein the method comprises the following steps: Determining a state feature vector of the battery module according to the current multi-physical-field distribution and the multi-physical-field actual measurement data of the battery module, which are measured by the sensor; Based on an extended Kalman filtering algorithm, the temperature state, the stress state, the sound energy density state and the gas concentration of the battery module are determined to be used as the current estimation state by combining the state feature vector of the battery module. Preferably, the determining the state feature vector of the battery module according to the current multi-physical-field distribution and the measured multi-physical-field data of the battery module measured by the sensor includes: Extracting physical field characteristics from the stress field distribution, the temperature field distribution, the acoustic energy field distribution and the aerodynamic field distribution respectively, wherein the physical field characteristics comprise stress gradient, temperature rise rate, acoustic energy density and gas escape concentration rate; acquiring multi-physical-field measured data measured by a sensor on the battery module, and deter