Failure Predictor Detection Using Multi-Time Deviation Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional failure predictor detection systems often incorrectly identify environmental changes as equipment failures, leading to false detections, due to deviations in measured values caused by external factors.

Innovation Solution

A failure predictor detection device that compares estimated and measured values at multiple times, using a detection unit to distinguish between environmental changes and actual equipment failures by employing short-term and long-term deviation detection units, thereby reducing false positives.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a learning model trained using only normal operation data is used to detect failure predictors by comparing predicted values with measured values, then failure detection capability is improved, but false detection increases due to environmental changes

Engineering Contradiction:

Improvefailure detection capabilityVSAvoiddetection accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent segments the detection process into multiple time points and introduces intermediate comparison units. Instead of directly comparing predicted values with measured values, the system first compares them at multiple time points, then compares the comparison results to identify trends. This segmentation allows the system to distinguish between temporary environmental fluctuations and actual equipment deterioration, reducing false detections while maintaining failure detection capability.

Inventive Principle:

Principle #1Segmentation

2Ease of operation

If deviation between predicted value and measured value is used as the sole criterion for failure detection, then detection simplicity is improved, but false detection occurs due to environmental changes

Engineering Contradiction:

Improvedetection simplicityVSAvoidfalse detection rate

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent performs preliminary comparisons between predicted values and measured values at multiple time points before making a final failure determination. The comparison unit calculates deviations at each time point, stores these comparison results, and then the determination unit analyzes the trend of these deviations over time. This preliminary action at multiple time points filters out temporary environmental changes before the final detection decision, reducing false detections while maintaining operational simplicity.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentEP4123402B1Failure predictor detection device, failure predictor detection method, failure predictor detection program, learning device, method for generating trained learning model, and trained learning model generation program

Publication Date: 2025.01.01 MITSUBISHI ELECTRIC CORP

Figure 1
Figure 2
Figure 3

AI summary

An object is to obtain a failure predictor detection device that is capable of more adequately detecting a failure predictor. A failure predictor detection device according to the present disclosure includes: an acquisition unit to acquire estimation data and comparison data of targeted equipment for failure predictor detection in a targeted time period for the failure predictor detection; an estimation unit to calculate an estimated value of the comparison data during a normal operation from the estimation data using a learning model; and a detection unit to detect a failure predictor of the equipment on the basis of comparison results at multiple times between the estimated values and measured values shown by the comparison data.