Damage Detection in Wind Turbine Blades Based on an Improved Broad Learning System Model

Abstract

The research on damage detection in wind turbine blades plays an important role in reducing the risk of shut down in wind turbines. Rapid and accurate damage identification by using efficient detection models is the focus of the current research on damage detection in wind turbine blades. To solve the problems of the complex structure of the model and high time consumption in deep learning models, an improved broad learning system (BLS) model using the algorithm of chunking based on non-local means (NLMs) was proposed, which was called the CBNLM-BLS. The chunked, in-parallel accelerated integral image approach was used to optimize the NLM to speed up the BLS. Experiment results showed that the proposed model achieved a classification accuracy of 99.716%, taking 28.662 s to detect damage in the wind turbine blades. Compared with deep neural network models, such as ResNet, AlexNet and VGG-19, the proposed CBNLM-BLS had higher classification accuracy, shorter training time and less complex model construction and parameters. Compared with traditional BLSs, the CBNLM-BLS had less time complexity. It is of great significance to identify damage in wind turbine blades more efficiently.

A Bird Vocalization Classification Method Based on Improved Adaptive Wavelet Threshold Denoising and Bidirectional FBank

“…The dataset for this study was derived from Xeno-canto 31 . After removing eight extremely short recordings, it comprised 263 audio recordings from 16 bird species across seven orders, nine families, and 15 genera.…”

Review on the Advancements in Wind Turbine Blade Inspection: Integrating Drone and Deep Learning Technologies for Enhanced Defect Detection

“…Zou et al [127] emphasized the importance of detecting damages in WTBs to prevent accidents and economic losses. A novel model, CBNLM-BLS, is introduced that combines chunking and Non-Local Means (NLMs) [128] to enhance the Broad Learning System (BLS) [129], making it more computationally efficient.…”

“…In comparison tests with deep learning models such as ResNet [82], VGG19, and AlexNet [106], the CBNLM-BLS achieved the highest classification accuracy of 99.71% in detecting defects on WTBs and had a computation time of 28.662 s. This research utilized a dataset of 741 images of WTBs, with 556 used for training and 185 for testing. Variations of the CBNLM-BLS model, adjusting parameters N1, N2, and N3, were also evaluated, achieving accuracy rates as high as 99.81% with computation times ranging from 14.078 to 41.752 s. In their innovative work, Zou et al [127] developed the CBNLM-BLS model to detect damages in WTBs. The model's efficacy is visually demonstrated through pre-and post-processing images.…”