Battery Anomaly Detection Using Multi-Feature Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional battery management systems fail to detect internal faults in rechargeable batteries and connected loads/components, relying solely on state information and not considering multiple features like voltage and current, which can lead to thermal runaway and financial losses.
Innovation Solution
A method and apparatus that identify anomalies in rechargeable batteries and connected loads/components using current and voltage measurements during regular operation, determining features such as internal resistance and outputting severity levels, with suggestions for user action.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional battery state estimation methods are used, then the battery model validity can be determined, but internal battery faults cannot be detected
Solution Approach 1:
The patent segments the battery monitoring system into multiple independent feature extraction modules, each analyzing specific aspects (voltage, current, temperature, impedance) separately, then combines them for comprehensive fault detection. This allows detailed analysis of individual parameters while maintaining overall system reliability.
Solution Approach 2:
The patent transitions from conventional single-dimension state estimation to multi-dimensional analysis by extracting and analyzing multiple features simultaneously (voltage features, current features, temperature features, impedance features). This dimensional expansion enables detection of internal faults that single-parameter methods miss.
2Measurement precision
If multiple features related to voltage and current are analyzed, then internal battery faults can be detected, but the device complexity increases
Solution Approach 1:
The patent implements a universal feature extraction framework that handles multiple battery parameters (voltage, current, temperature, impedance) using similar processing algorithms. This multi-functional approach reduces overall system complexity compared to having separate dedicated systems for each parameter.
Solution Approach 2:
The system performs self-diagnosis by automatically extracting features from its own operating data and comparing them against threshold values or historical patterns. This self-service capability eliminates the need for external complex diagnostic equipment, reducing overall device complexity while maintaining high detection precision.
3Device complexity
If conventional estimation methods are used, then the system operates simply, but anomalies in connected loads/components cannot be identified
Solution Approach 1:
The patent merges the battery monitoring function with connected component anomaly detection into a single integrated system. By combining feature extraction from both battery and connected loads into one unified analysis framework, the system achieves comprehensive monitoring without proportionally increasing complexity.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Embodiments herein disclose an apparatus and methods for identifying an anomaly in at least one of a re-chargeable battery and at least one component(s) connected to the re-chargeable battery. Embodiments herein relates to the field of battery management systems, and more particularly to apparatus and methods for identification of anomalies in batteries and loads/component(s) connected to the re-chargeable battery. The embodiments herein includes outputting a severity level of the anomaly in the at least one of the re-chargeable battery and at least one component connected to the re-chargeable battery, based on the analyzed plurality of the threshold values of the determined plurality of the characteristic data corresponding to the voltage data and current data, wherein the severity level of anomaly comprise at least one of, a negligent level, a medium level and a critical level.