Battery Cell Voltage Diagnosis Using Statistical Error Masking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery diagnosis methods fail to accurately diagnose abnormal behavior in individual battery cells within a battery pack, necessitating a more precise and efficient method to detect defects such as short-circuit or negative electrode exposure to prevent fires.
Innovation Solution
A battery diagnosis apparatus and method that utilizes time-series data analysis, calculating moving averages and standard deviations, normalizing voltage values, and applying error functions to determine error rates for each cell, masking error rates during high current changes, and comparing cell behaviors to identify abnormal cells.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If battery diagnosis is performed at battery pack level using conventional voltage monitoring, then diagnosis coverage is limited to aggregate behavior, but individual cell abnormalities cannot be detected
Solution Approach 1:
The patent divides the battery pack into individual cell units for separate analysis. By calculating moving averages and standard deviations for each cell independently, the system can detect abnormalities at the cell level rather than only at the pack level, thereby improving diagnosis precision without requiring complex additional hardware
Solution Approach 2:
The patent introduces statistical parameters (moving average and standard deviation) as intermediary tools to bridge the gap between simple voltage monitoring and complex cell-level diagnosis. These intermediaries enable individual cell analysis using existing voltage data, improving detection capability without increasing system complexity
2Reliability
If error rate calculation is performed without masking during high current changes, then all voltage deviations are flagged as potential defects, but false positives increase during normal high current operations
Solution Approach 1:
The patent applies masking as a preliminary filtering step before final defect determination. By identifying high current change periods in advance and masking error rates during these periods, the system prevents false positives from being generated during normal high current operations, thereby improving diagnosis reliability
Solution Approach 2:
The patent applies different processing rules to different operational conditions. Error rates are masked during high current changes but not during normal operations, allowing the system to maintain high sensitivity when needed while reducing false positives during specific problematic conditions
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A battery diagnosis apparatus according to an embodiment disclosed herein includes an information obtaining unit configured to obtain time-series data regarding a voltage of each of a plurality of battery cells and a controller configured to calculate a moving average and a standard deviation for a reference unit from the time-series data of each battery cell and diagnose whether each battery cell is abnormal, based on the moving average and the standard deviation.