Sensor Change-Point Grouping for Multi-Abnormality Event Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing system analysis methods struggle to accurately separate and output information for individual events when multiple abnormalities occur in a system, leading to confusion for operators in understanding the system status.
Innovation Solution
A system analysis method that acquires historical sensor data to identify abnormalities, estimates change points, calculates relevance levels between these points, and clusters them into groups to generate output information specific to each abnormality group, allowing for clear separation and identification of individual events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If multiple abnormalities are detected and output together, then the system can provide comprehensive abnormality information, but the operator cannot appropriately grasp the system status due to mixed output
Solution Approach 1:
The patent segments multiple detected abnormalities into separate groups based on their temporal relationships and relevance levels. Change points are classified into different groups, with each group representing a distinct abnormality event. This segmentation allows the system to maintain complete abnormality information while organizing it in a way that is easy for operators to understand and analyze individually.
2Reliability
If all sensor data is analyzed without grouping, then all abnormalities are detected, but the complexity of analyzing multiple mixed events increases
Solution Approach 1:
The patent divides the complex analysis of all sensor data into manageable segments by grouping change points based on their relevance levels and temporal relationships. This segmentation reduces analysis complexity while maintaining detection accuracy by allowing focused analysis of each abnormality group rather than treating all data as a single complex mixture.
Solution Approach 2:
The patent introduces a new dimension for organizing abnormality data by classifying change points into multiple groups based on relevance levels. This dimensional organization transforms the complex multi-dimensional sensor data into structured groups that are easier to analyze, reducing processing complexity while preserving all abnormality information.
3Ease of operation
If change points are classified into multiple groups based on relevance levels, then individual events can be separated, but additional processing steps are required
Solution Approach 1:
The patent performs preliminary classification of change points into groups based on relevance levels during the data processing phase. By organizing the data structure in advance with pre-calculated relevance levels and group assignments, the system enables rapid event separation during operation without requiring additional complex processing steps when abnormalities are detected.
Data Source
AI summary
A system analysis method includes: acquiring history information indicating, based on sensor values outputted by sensors, whether one of sensor values outputted by respective sensors indicates abnormality and/or whether individual relationship between sensor values outputted by different sensors indicates abnormality in time-series manner; estimating a change point group of change points, each indicating a time point system state has changed, based on history information; estimating relevance levels, each indicating relevance to the system state between two arbitrary time points included in the change point group; generating groups of change point groups by classifying the change point group into a plurality of groups based on the history information and the relevance levels; and generating and outputting output information, as information relating abnormality per group of the change point groups.