Semi-Supervised Surface Anomaly Detection of Composite Wind Turbine Blades From Drone Imagery

TL;DR

This paper introduces BladeNet, a semi-supervised deep learning system for automated detection and fault analysis of wind turbine blades from UAV imagery, achieving high accuracy with minimal manual annotation.

Contribution

BladeNet is a novel dual architecture that combines unsupervised detection with semi-supervised fault identification, reducing manual labeling effort in wind turbine blade inspection.

Findings

Achieves an Average Precision of 0.995 on offshore dataset

Achieves an Average Precision of 0.223 on DTU dataset

Obtains an AUC of 0.639 for surface anomaly detection

Abstract

Within commercial wind energy generation, the monitoring and predictive maintenance of wind turbine blades in-situ is a crucial task, for which remote monitoring via aerial survey from an Unmanned Aerial Vehicle (UAV) is commonplace. Turbine blades are susceptible to both operational and weather-based damage over time, reducing the energy efficiency output of turbines. In this study, we address automating the otherwise time-consuming task of both blade detection and extraction, together with fault detection within UAV-captured turbine blade inspection imagery. We propose BladeNet, an application-based, robust dual architecture to perform both unsupervised turbine blade detection and extraction, followed by super-pixel generation using the Simple Linear Iterative Clustering (SLIC) method to produce regional clusters. These clusters are then processed by a suite of semi-supervised detection methods. Our dual architecture detects surface faults of glass fibre composite material blades with high aptitude while requiring minimal prior manual image annotation. BladeNet produces an Average Precision (AP) of 0.995 across our {\O}rsted blade inspection dataset for offshore wind turbines and 0.223 across the Danish Technical University (DTU) NordTank turbine blade inspection dataset. BladeNet also obtains an AUC of 0.639 for surface anomaly detection across the {\O}rsted blade inspection dataset.