Battery Cell Voltage Detection Through Outlier Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery systems face challenges in effectively detecting battery cell voltage outliers, which can lead to abnormal capacity and potential short circuits due to significant voltage differences among battery cells.
Innovation Solution
A method involving data acquisition, clustering algorithms, and threshold-based determination to identify suspected and actual voltage outlier clusters, ensuring accurate detection of battery cell voltage outliers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional voltage detection methods are used, then the detection process is simple, but the detection precision is insufficient to identify battery cell voltage outliers
Solution Approach 1:
The patent segments the voltage detection process into multiple stages: data acquisition from battery cells, clustering analysis to identify outlier patterns, and hierarchical classification to distinguish between temporary and persistent voltage outliers. This segmentation enables precise detection while managing complexity through structured processing steps.
Solution Approach 2:
The patent performs preliminary clustering analysis on voltage data to identify potential outlier patterns before final determination. By pre-processing the data through clustering algorithms, the system prepares candidate outlier sets that are then validated against threshold criteria, improving detection precision without requiring complex real-time analysis.
2Reliability
If comprehensive voltage monitoring of all battery cells is implemented, then detection reliability is improved, but the loss of time and computational resources increases
Solution Approach 1:
The patent applies partial monitoring by focusing computational resources on identifying and analyzing only the voltage outliers detected through clustering, rather than uniformly processing all battery cell data with equal depth. This approach maintains reliability by thoroughly investigating suspicious cases while reducing overall processing time.
Solution Approach 2:
The patent introduces clustering analysis as an intermediary step between raw voltage data collection and final outlier determination. This intermediary process efficiently filters and groups data, enabling the system to maintain high reliability through comprehensive monitoring while reducing processing time by organizing data into manageable clusters before detailed analysis.
3Measurement precision
If clustering algorithms are used to identify voltage outliers, then measurement precision is improved, but the difficulty of detecting and measuring increases due to algorithm complexity
Solution Approach 1:
The patent employs self-service mechanisms where the clustering algorithm automatically adapts to the specific voltage patterns in the battery data. The system uses the data itself to determine clustering parameters and thresholds, reducing the need for manual configuration and expert knowledge while maintaining high identification accuracy.
Solution Approach 2:
The patent utilizes parameter changes in the clustering algorithm to balance precision and implementation difficulty. By adjusting clustering parameters such as distance thresholds and cluster formation criteria, the system achieves accurate outlier identification while keeping the algorithm computationally manageable and easier to implement.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of the present disclosure disclose a voltage detection method, a device, a vehicle, and a storage medium. The method comprises: a data set of a target object in at least one time interval is acquired, where the data set comprises voltages of battery cores included in at least one battery cell of the target object; a voltage of a random battery cell in a random time interval is obtained based on voltages of battery cores included in the random battery cell included in a data set in the random time interval; a voltage of the at least one battery cell in the random time interval is clustered based on a clustering algorithm to obtain multiple battery cell voltage clusters in the random time interval; suspected voltage outlier clusters and voltage outlier clusters are determined from multiple battery cell voltage clusters in the at least one time interval; and whether the target object has battery cell voltage outliers is determined based on a number of the suspected voltage outlier clusters and a number of the voltage outlier clusters. The embodiments of the present disclosure can effectively detect a battery cell voltage and ensure efficient and safe operation of a battery system.