Time-Series Event Waveform Grouping for Equipment Abnormality Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing time-series data processing technologies face challenges in detecting equipment abnormalities without event information, as they rely on pre-identified events which may not be available or easily processable, especially when event data is not electronically recorded.

Innovation Solution

A time-series data processing device comprising an event waveform extracting unit, a co-occurrence rate calculating unit, and an event information generating unit that identifies events by analyzing co-occurrence rates and grouping time-series data to generate event information, even when such information is not provided.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If event information is provided to identify events in subject equipment, then abnormality detection accuracy is improved, but the system requires additional information infrastructure and increases complexity

Engineering Contradiction:

Improveabnormality detection accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system extracts event information automatically from time-series data itself through waveform extraction and co-occurrence rate calculation, making the data self-sufficient for event identification without requiring external event information infrastructure

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

Event waveforms are extracted from time-series data by identifying portions where values change by a predetermined amount or more, separating event-related patterns from the overall data for further analysis

Inventive Principle:

Principle #2Taking out (Extraction)

2Device complexity

If event information is not provided, then system complexity is reduced, but the ability to detect equipment abnormalities accurately deteriorates

Engineering Contradiction:

Improvesystem complexityVSAvoidabnormality detection accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

Co-occurrence rate calculation serves as an intermediary mechanism that bridges the gap between raw time-series data and event identification, enabling event detection without direct event information by analyzing temporal relationships in the data

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary waveform extraction and co-occurrence rate calculation on time-series data to prepare event information in advance, enabling subsequent abnormality detection to proceed accurately even without external event information

Inventive Principle:

Principle #10Preliminary action

3Ease of manufacture

If event information is recorded on print medium, then information storage is simplified, but electronic processing capability is lost

Engineering Contradiction:

Improveinformation storage simplicityVSAvoidelectronic processing capability

Core Design Contradiction:

Ease of manufactureVSEase of operation

Solution Approach 1:

The patent replaces mechanical print medium storage with electronic time-series data storage, and substitutes manual information extraction with automated waveform extraction and co-occurrence rate calculation algorithms that process data electronically

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11531688B2Time-series data processing device, time-series data processing system, and time-series data processing method

Publication Date: 2022.12.20 MITSUBISHI ELECTRIC CORP

AI summary

An event waveform extracting unit (3) extracts an event waveform from time-series data. A co-occurrence rate calculating unit (4) calculates co-occurrence rates of event waveforms among the time-series data. A grouping unit (5) classifies the time-series data into groups depending the co-occurrence rates of the event waveforms. An event information generating unit (6) determines the time at which the periods during which event waveforms occur overlap with each other among the time-series data included in each group, and generates event information identifying an event related to the event waveforms on the basis of the determined time.