Wind Turbine Blade Thermal Imaging With ML Frame Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional systems for monitoring wind turbine rotor blades using infrared imaging face challenges in efficiently analyzing large volumes of data, which is time-consuming and costly, hindering timely detection of blade damage.

Innovation Solution

A system and method utilizing infrared imaging combined with machine learning techniques to automatically identify blade sections, subsample critical frames, and detect anomalies, reducing data analysis burden and computational costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If infrared imaging is used to monitor rotor blades, then detection capability is improved, but data processing time and computational cost increase

Engineering Contradiction:

Improveblade defect detection capabilityVSAvoiddata processing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system extracts only the most informative frames from the thermal imaging video sequence by identifying frames that capture complete rotor blade sections. This selective extraction reduces the total number of frames requiring detailed analysis, thereby decreasing processing time while maintaining defect detection capability.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The rotor blade monitoring data is segmented into multiple sections, and the system processes each section independently by identifying representative frames for each section. This segmentation allows parallel processing and reduces the overall computational burden compared to analyzing the entire video sequence uniformly.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If infrared imaging is used to monitor rotor blades, then detection capability is improved, but storage requirements increase

Engineering Contradiction:

Improveblade defect detection capabilityVSAvoiddata storage requirements

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The system extracts only the essential thermal imaging frames that contain complete rotor blade sections, discarding redundant frames. This extraction strategy maintains sufficient data for defect detection while significantly reducing the volume of data requiring storage.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Rather than storing and processing every frame from the thermal imaging sequence, the system applies partial action by selecting only the minimum necessary subset of frames that provide complete coverage of all rotor blade sections. This approach provides sufficient monitoring capability with reduced storage demands.

Inventive Principle:

Principle #16Partial or excessive action

3Measurement precision

If machine learning models analyze all image frames, then detection accuracy is improved, but computational complexity increases

Engineering Contradiction:

Improveanomaly detection accuracyVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary action by pre-processing the thermal imaging video to identify and select key frames that contain complete rotor blade sections before applying machine learning analysis. This pre-selection step reduces the input data volume for the machine learning models, decreasing computational complexity while preserving detection accuracy.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The analysis process is segmented into two stages: first, automatic selection of representative frames containing complete blade sections; second, application of machine learning models only to these selected frames. This segmentation of the analysis process reduces overall computational complexity compared to applying machine learning to every frame.

Inventive Principle:

Principle #1Segmentation

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

The system effectively reduces data processing time and storage requirements while enhancing the ability to detect blade defects, enabling proactive maintenance and minimizing downtime.

Implementation Method 1

receiving, via an imaging analytics module of a controller, thermal imaging data of the rotor assembly

Methodology Applied

Scientific EffectThermal radiation: Thermal Radiation

Data Source

PatentUS12546290B2System and method for monitoring wind turbine rotor blades using infrared imaging and machine learning

Publication Date: 2026.02.10 LM WIND POWER AS

AI summary

A method for monitoring a rotor assembly of a wind turbine includes receiving, via an imaging analytics module of a controller, thermal imaging data of the rotor assembly that includes a plurality of image frames. The method includes automatically identifying, via a first machine learning model of the imaging analytics module, a plurality of sections of a rotor blade of the rotor assembly within the plurality of image frames until all sections of the rotor blade are identified. Further, the method includes selecting, via a function of the imaging analytics module, a subset of image frames from the plurality of image frames, the subset of image frames comprising a minimum number of the plurality of image frames required to represent all sections of the rotor blade. Moreover, the method includes generating, via a visualization module of the controller, an image of the rotor assembly using the subset of image frames.