Distributed Battery Abnormality Detection via Localized Distribution Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distributed equipment abnormality detection systems face challenges in managing large numbers of battery cells, as they require significant communication and calculation resources, leading to high costs and potential concealment of abnormalities due to individual differences and variations in temperature and charging states.
Innovation Solution
A distributed equipment abnormality detection system that divides equipment into groups, calculates distribution information for each group, and compares it with integrated distribution information from a management server to determine abnormality, reducing communication and calculation loads while accurately identifying abnormal cells.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If internal resistance values of all battery cells are collected to detect abnormalities, then abnormality detection accuracy is improved, but communication amount and calculation amount increase significantly
Solution Approach 1:
The patent segments the battery management system into multiple management apparatuses, each responsible for a specific battery cell or group of cells. Each management apparatus independently calculates distribution information for its assigned cell, comparing its internal resistance distribution against a reference distribution. This segmentation eliminates the need to collect and process all internal resistance values centrally, significantly reducing communication amount while maintaining detection accuracy through localized statistical comparison.
2Measurement precision
If internal resistance values of all battery cells are collected to detect abnormalities, then abnormality detection accuracy is improved, but device cost increases due to expensive collection and processing apparatus
Solution Approach 1:
The patent implements self-service by enabling each battery cell's management apparatus to autonomously perform abnormality detection using locally available data. Each management apparatus calculates its own cell's internal resistance distribution and compares it against a reference distribution stored locally or obtained through minimal communication. This eliminates the need for expensive centralized collection and processing apparatus, as each unit independently performs the complete detection function using its own resources.
3Quantity of substance
If battery system is divided into multiple subsystems to reduce communication load, then communication amount is reduced, but abnormality detection accuracy decreases due to concealed individual differences
Solution Approach 1:
The patent changes the parameter used for abnormality detection from absolute internal resistance values to distribution information derived from statistical analysis. By transforming raw measurement data into distribution characteristics (such as mean, variance, or histogram-based representations), the system can effectively compare cells while accounting for individual differences and variations. This parameter transformation enables accurate detection within segmented subsystems without requiring centralized data collection, as each subsystem can perform meaningful comparisons using its locally calculated distribution information.
Data Source
AI summary
A distributed equipment abnormality detection system is provided for monitoring physical amounts of equipments of identical type and detecting an abnormality of each equipment. The distributed equipment abnormality detection system includes equipment management apparatuses that manage the equipments; and a management server apparatus for communicating the equipment management apparatuses. Each equipment management apparatus includes a communication unit that communicates with the management server apparatus; a measurement unit that repeatedly measures the physical amount of the equipment; a distribution information generation unit that calculates distribution information from the measured physical amount; a distribution comparison unit that calculates a difference between the distribution information generated by the distribution information generation unit, and integrated distribution information on the entire equipments delivered from the management server apparatus; and an abnormality determination unit that determines whether or not the equipment is abnormal based on the calculated difference.