Equipment Abnormality Detection Using Learned Waveform Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional abnormality detection systems erroneously determine equipment abnormalities when event information is not prepared in advance, leading to incorrect assessments during subsequence data analysis.
Innovation Solution
An abnormality detection device that calculates outlier scores and extracts outlier data based on time-series data, collating waveforms with pre-defined conditions to determine equipment normalcy without relying on pre-prepared event information, using a unit for determining waveform types and generating waveform conditions from learning data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the conventional abnormality detection method compares time-series data to detect subsequence data, then abnormality detection capability is improved, but erroneous determination occurs when equipment is operating normally
Solution Approach 1:
The patent introduces waveform conditions as an intermediary between the detected subsequence data and the abnormality determination. Instead of directly determining abnormality from subsequence data, the system compares the waveform against pre-defined waveform conditions (normal operation patterns) to mediate the determination process, thereby reducing erroneous detections while maintaining abnormality detection capability
Solution Approach 2:
The patent performs preliminary action by pre-defining waveform conditions representing normal operation patterns before the actual abnormality detection process. These waveform conditions are established in advance through learning normal operation data, allowing the system to compare detected waveforms against known normal patterns and avoid erroneous determinations
2Measurement precision
If event information is prepared in advance to avoid erroneous determination, then determination accuracy is improved, but device complexity and preparation requirements increase
Solution Approach 1:
The system performs self-service by automatically learning and generating waveform conditions from normal operation data without requiring external event information preparation. The waveform condition generation unit autonomously creates the reference patterns needed for accurate determination, eliminating the need for manual event information management and reducing system complexity
Solution Approach 2:
The patent creates simplified copies of normal operation patterns in the form of waveform conditions. Instead of managing complex event information, the system generates condensed waveform representations that capture essential normal operation characteristics, making the determination process simpler while maintaining accuracy
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
An abnormality detection device is configured so as to include: an outlier score calculating unit (3) for calculating, from abnormality detection time-series data indicating states of equipment which is an abnormality detection target at a plurality of times in time series, a degree of abnormality of the equipment at each of the plurality of times as an abnormality detection outlier score; an outlier data extracting unit (4) for extracting, from among pieces of the abnormality detection time-series data, a piece of abnormality detection time-series data in a time period in which an abnormality may have occurred in the equipment as outlier data on the basis of the abnormality detection outlier score at each of the plurality of times calculated by the outlier score calculating unit (3); and an abnormality determining unit (8) for collating a waveform of the abnormality detection outlier data extracted by the outlier data extracting unit (4) with a waveform condition for determining that a waveform indicating a change in the abnormality detection outlier data is a waveform obtained when the equipment is operating normally, and determining whether or not the equipment is operating abnormally on the basis of a collation result between the waveform condition and the waveform of the abnormality detection outlier data.