Wind Turbine Blade State Detection Through Acoustic Spectrograms

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for diagnosing abnormalities in wind turbine blades, such as scratches or cracks, are limited in detection accuracy and cannot effectively identify various states beyond changes in frequency components over time.

Innovation Solution

A state detection system that includes a state detection apparatus and a sound collection apparatus, utilizing acoustic information analysis to detect patterns in sound waves generated by the blades, allowing for improved detection of abnormalities through pattern recognition and classification, including the use of a database for normal patterns and noise reduction techniques.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If Doppler shift component detection method is used to detect blade abnormalities, then detection capability is provided, but detection accuracy of various states is insufficient

Engineering Contradiction:
Improvedetection accuracyVSAvoidstate detection capability
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent transforms acoustic information from time-domain signals into frequency-domain spectrogram images, adding a visual dimension to the analysis. This allows the system to detect not only Doppler shift components but also various other patterns (impact sounds, vibration patterns, etc.) that represent different blade states, thereby improving both detection accuracy and versatility across multiple state types.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If acoustic information analysis is performed to detect blade states, then detection capability is provided, but background noise reduces detection accuracy

Engineering Contradiction:
Improvedetection accuracyVSAvoidbackground noise influence
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent extracts only the relevant frequency components from the acoustic information by generating spectrograms that display frequency distribution over time. This extraction process separates the blade-generated sounds from background noise, allowing the pattern recognition unit to focus on characteristic patterns while filtering out irrelevant noise components.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The spectrogram serves as an intermediary representation between the raw acoustic information and the pattern recognition process. By transforming acoustic signals into visual frequency-time patterns, the system creates an intermediate format that enhances the distinguishability of blade states while suppressing the impact of background noise.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If pattern recognition is used to detect blade states, then detection capability is provided, but clear criteria for determining states are lacking

Engineering Contradiction:
Improvedetection accuracyVSAvoiddetermination criteria complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent pre-registers multiple patterns in the pattern recognition unit, each corresponding to a specific blade state (normal, impact, vibration, etc.). This preliminary preparation of reference patterns establishes clear determination criteria before actual detection occurs, allowing the system to compare incoming acoustic patterns against known state patterns and make accurate determinations without complex real-time analysis.

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

Enhances the detection accuracy of blade abnormalities by identifying various states, facilitating timely inspections and repairs, reducing the influence of background noise, and providing clear criteria for determining blade conditions.

Implementation Method 1

a sound collection apparatus configured to measure sound generated at a detection target and output a result of measurement to the state detection apparatus as acoustic information

Methodology Applied
Scientific EffectSound detection: Sound

Implementation Method 2

a method for detecting the presence of a Doppler shift component, in which a frequency component with high sharpness changes over time, in the analysis results of acoustic information (wind noise) emitted by the blade

Methodology Applied
Scientific EffectDoppler effect: Doppler Effect

Data Source

PatentEP4621214A1State detection device and state detection system
Publication Date: 2025.09.24 YOKOGAWA ELECTRIC CORP
  • EP4621214A1 patent drawingFigure 1
  • EP4621214A1 patent drawingFigure 2
  • EP4621214A1 patent drawingFigure 3

AI summary

A state detection apparatus (10) includes an analyzer (12) that detects a state of a detection target. The analyzer (12) acquires acoustic information corresponding to sound generated at the detection target, detects the state of the detection target based on a pattern included in an image representing the time variation of a frequency component of the acoustic information, and outputs a result of detecting the state of the detection target.