Abnormality Detection Apparatus for Time-Series Data Correlation Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing system monitoring apparatuses cannot identify the specific factor contributing to failure in monitoring targets, despite calculating correlations between time-series data.
Innovation Solution
A system monitoring apparatus that determines relevance between first and second time-series data, calculates irregularity degrees, and derives the normal or abnormal state of second time-series data using first and second degrees, enabling identification of the factor related to the monitoring target's failure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If correlation analysis is performed between multiple measurement values to detect abnormalities, then abnormality detection capability is improved, but the ability to identify specific failure factors deteriorates
Solution Approach 1:
The patent segments the correlation analysis into two distinct phases: (1) model generation phase using only normal operation data to establish baseline correlations, and (2) abnormality detection phase where deviations from these correlations are analyzed. This segmentation allows the system to maintain correlation-based detection while adding the capability to identify specific failing components by comparing which correlation relationships break down during abnormalities.
Solution Approach 2:
The system performs preliminary action by pre-generating correlation models during normal operation before any abnormality occurs. These models capture the expected relationships between various measurement values. When an abnormality occurs, the system compares actual measurements against these pre-established models to quickly identify which relationships are violated, thereby pinpointing the specific failure factor without losing diagnostic information.
2Measurement precision
If multiple performance indexes are monitored to improve detection accuracy, then detection precision is improved, but system complexity increases
Solution Approach 1:
The patent applies dynamics by making the correlation models adaptive rather than static. The system continuously updates correlation models based on changing operational conditions while maintaining the underlying correlation structures. This allows the system to handle multiple performance indexes dynamically, adjusting to normal variations in system behavior without requiring complex reconfiguration, thereby maintaining detection accuracy across multiple parameters while managing system complexity.
Data Source
AI summary
A system monitoring apparatus has: a determination unit for determining whether the relatedness indicating a relationship that is true for first time-series data of multiple sets measured during a first time period is true for second time-series data of the multiple sets measured during a second time period; an irregularity degree calculation unit for calculating the degree of irregularity indicating the extent to which the second time-series data is irregular; a first extent calculation unit for calculating a first extent indicating the extent to which the irregularity is at a specified value when the second time-series data are normal or irregular; a second extent calculation unit for calculating a second extent indicating the extent to which the second time-series data of the multiple sets are related; and a state calculation unit for finding whether the second time-series data are normal or irregular, based on the first and second extents.