Acoustic Fault Localization by Segment and Frequency Band Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing acoustic diagnosis devices struggle to accurately specify abnormal parts in equipment with multiple components due to varying sound generation timings and frequencies across different components.

Innovation Solution

An acoustic diagnosis device and method that collects operational sound, extracts segments, calculates sound pressure levels in specific frequency bands, and specifies abnormal parts based on deviations from normal operation patterns, utilizing a database for accurate identification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If operational sound is analyzed as a whole without segmentation, then analysis simplicity is maintained, but abnormal part specification accuracy deteriorates

Engineering Contradiction:
Improveanalysis simplicityVSAvoidabnormal part specification accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The operational sound is segmented into multiple operational segments based on operation timing. Each segment corresponds to a specific operational phase, allowing precise identification of which component generates abnormal sounds at different times. This segmentation enables accurate abnormal part specification while maintaining manageable analysis through systematic organization.

Inventive Principle:
Principle #1Segmentation

2Reliability

If sound pressure level is analyzed across all frequency bands simultaneously, then analysis comprehensiveness is improved, but abnormal part specification accuracy deteriorates

Engineering Contradiction:
Improveanalysis comprehensivenessVSAvoidabnormal part specification accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The frequency spectrum is segmented into multiple frequency bands, and sound pressure levels are calculated separately for each band. By analyzing which specific frequency band shows abnormal sound pressure levels, the system can identify the type of component abnormality (e.g., bearing defects, gear issues) while maintaining comprehensive monitoring across all frequency ranges.

Inventive Principle:
Principle #1Segmentation

3Speed

If deviation calculation is performed without comparing to normal operation baseline, then real-time monitoring speed is improved, but abnormal detection accuracy deteriorates

Engineering Contradiction:
Improvereal-time monitoring speedVSAvoidabnormal detection accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

Sound pressure level data during normal operation is collected and stored as baseline information before abnormal conditions occur. This preliminary data is used to establish reference values for each operational segment and frequency band. When monitoring, the system quickly compares current measurements against these pre-established baselines, enabling both rapid detection and accurate identification of abnormalities.

Inventive Principle:
Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables precise identification of abnormal equipment parts, facilitating timely maintenance and improving product quality and yield by accurately detecting deviations in sound pressure levels and patterns.

Implementation Method 1

an operational sound collecting unit that collects operational sound of the equipment

Methodology Applied
Scientific EffectAcoustic detection: Acoustics

Data Source

PatentEP4700345A1Acoustic diagnostic device and acoustic diagnostic method
Publication Date: 2026.02.25 JFE STEEL CORP
  • EP4700345A1 patent drawingFigure 1
  • EP4700345A1 patent drawingFigure 2
  • EP4700345A1 patent drawingFigure 3(a)~3(c)

AI summary

An acoustic diagnosis device according to the present invention includes: an operational sound collecting unit that collects operational sound of the equipment; an in-operational-segment operational sound extracting unit that extracts operational sound for each of predetermined operational segments of the equipment from the operational sound collected by the operational sound collecting unit; a sound pressure level acquiring unit that acquires a sound pressure level for each of predetermined frequency bands for each of the operational segments with respect to operational sound for each of the operational segments extracted by the in-operational-segment operational sound extracting unit; a degree-of-deviation calculating unit that calculates a degree of deviation of the sound pressure level for each of the operational segments and each of the frequency bands acquired by the sound pressure level acquiring unit from a sound pressure level of the same equipment, a same operational segment, and a same frequency band during normal operation; and an abnormal part specifying unit that specifies an abnormal part of the equipment on a basis of the degree of deviation calculated by the degree-of-deviation calculating unit.