Battery Micro-Short Fault Diagnosis Using Dynamic Voltage Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for diagnosing micro-short circuit faults in batteries are complex, computationally intensive, and lack accuracy, especially in embedded devices due to data timeliness and computational requirements, making them ineffective for timely detection and prevention of battery faults.
Innovation Solution
A method utilizing variational modal decomposition for noise elimination, dynamic reference voltage calculation, eigenvalue and correlation coefficient extraction, and improved Frechet distance analysis to detect micro-short circuits in batteries, reducing complexity and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If model-based method is used for fault diagnosis, then fault detection capability is improved, but calculation complexity and computational amount increase significantly
Solution Approach 1:
The patent extracts only the essential voltage difference information and correlation coefficients from complex battery data, eliminating unnecessary computational elements while retaining fault detection capability. This extraction approach reduces calculation complexity by focusing only on the most relevant features for micro-short circuit detection.
Solution Approach 2:
The patent replaces complex model-based computational systems with a simplified statistical analysis approach using correlation coefficients and voltage differences. This substitution eliminates the need for building and solving complex nonlinear battery models, significantly reducing computational burden while maintaining diagnostic effectiveness.
2Device complexity
If signal processing-based method is used for fault diagnosis, then computational requirements are reduced, but fault diagnosis accuracy decreases due to low nonlinear fitting degree
Solution Approach 1:
The patent transforms the analysis by changing from examining absolute voltage values to analyzing voltage differences and correlation coefficients. This parameter transformation enables the system to capture nonlinear fault patterns while maintaining computational simplicity, thereby improving accuracy without increasing computational requirements.
Solution Approach 2:
The patent introduces a new dimension of analysis by calculating correlation coefficients between different battery voltages, transforming the problem from one-dimensional voltage monitoring to multi-dimensional relational analysis. This dimensional expansion enhances fault detection accuracy by revealing hidden patterns without requiring complex computational models.
3Measurement precision
If machine learning-based method is used for fault diagnosis, then fault detection accuracy is improved, but timeliness decreases and computational requirements increase making it unsuitable for embedded devices
Solution Approach 1:
The patent employs lightweight computational algorithms that can be executed quickly on embedded devices without requiring powerful hardware or extensive training data. This approach prioritizes rapid, timely detection over the high accuracy of complex machine learning models, making the system suitable for real-time embedded deployment where speed is critical.
Solution Approach 2:
The patent performs preliminary data processing and feature extraction using simple statistical methods that prepare the data for rapid analysis. By pre-computing voltage differences and correlation coefficients, the system enables quick fault detection without the need for time-consuming machine learning inference, thus maintaining timeliness while achieving adequate accuracy.
4Ease of operation
If traditional fault diagnosis methods are used, then implementation is simple, but sensitivity to micro-short circuit faults at initial stage is insufficient
Solution Approach 1:
The patent introduces correlation coefficients as an intermediary metric that mediates between simple voltage monitoring and complex fault analysis. This intermediary approach enhances sensitivity to subtle micro-short circuit faults by revealing relational patterns between batteries that are not apparent from individual voltage readings alone, while keeping the implementation relatively simple.
Solution Approach 2:
The patent combines multiple simple indicators (voltage differences, correlation coefficients, and threshold comparisons) into a composite diagnostic approach. This combination enhances detection sensitivity by leveraging the complementary strengths of each indicator, achieving better fault detection performance than any single simple method alone while maintaining implementation simplicity.
Data Source
AI summary
The present invention discloses a method for diagnosing micro-short circuit fault of a battery, comprising: acquiring real-time voltage data of each battery in battery module of electric vehicle, obtaining voltage matrix after noise elimination; calculating dynamic reference voltage sequence of the each battery in the voltage matrix within sliding window length, extracting eigenvalues of the each battery at different times; calculating correlation coefficients between voltages of the each battery at different times and reference voltage within the sliding window; forming feature point matrix of the battery module based on the eigenvalues and the correlation coefficients; calculating dynamic reference feature point sequence based on the feature point matrix, and calculating improved Frechet distances between feature point sequences of the each battery and the dynamic reference feature point sequence, using Frechet distance values as anomaly scores of the each battery; thus, judging whether a short-circuit fault occurs in the each battery.