Abnormality Detection Using Reference-Curve Distance

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In target devices using servomotors, steep prediction curves for press load or motor speed during machining operations can lead to false abnormality detections when standard deviation-based methods are used.

Innovation Solution

An abnormality detecting device that acquires first and second index values related to the operation stage and load, respectively, and detects abnormalities based on the distance from a predetermined reference curve on a two-dimensional plane.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If standard deviation-based abnormality detection is used, then abnormality detection capability is improved, but false abnormality detections increase when prediction curves are steep

Engineering Contradiction:

Improveabnormality detection capabilityVSAvoidabnormality determination accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent changes the detection parameter from absolute standard deviation to relative distance from reference curve. Instead of using fixed threshold values based on standard deviation, the system calculates the distance between measured points and the reference curve, then determines abnormalities based on whether this distance exceeds a threshold. This parameter transformation resolves the contradiction by making the detection adaptive to curve steepness.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent implements dynamic threshold adjustment based on local curve characteristics. The abnormality threshold is not fixed but varies according to the steepness of the prediction curve at each point. When the curve is steep, the threshold automatically adjusts to accommodate larger deviations, preventing false positives while maintaining detection sensitivity.

Inventive Principle:

Principle #15Dynamics

2Ease of operation

If fixed threshold abnormality detection is used, then detection simplicity is improved, but adaptability to different operating conditions deteriorates

Engineering Contradiction:

Improvedetection method simplicityVSAvoidadaptability to steep prediction curves

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The patent performs preliminary actions by pre-calculating the reference curve from normal operation data before actual abnormality detection. This reference curve serves as a baseline that captures the expected relationship between variables under normal conditions. By having this reference established in advance, the system can quickly compare actual measurements against it without complex real-time adjustments, maintaining simplicity while improving adaptability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from the reference curve to dynamically adjust detection thresholds. The reference curve provides continuous feedback about expected normal behavior, allowing the system to adapt to different operating conditions including steep prediction curves. This feedback mechanism enables the simple detection method to become adaptable without sacrificing ease of operation.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS12276975B2Abnormality detecting device, abnormality detecting method, and storage medium

Publication Date: 2025.04.15 OMRON CORP

AI summary

An abnormality in a device of interest is detected on the basis of the distance from a predetermined reference curve to a point represented by a first index value and a second index value in a two-dimensional plane having the first index and the second index as axes.