/wtbd

Wind turbine blade diagnosis

Primary LanguagePureBasicMIT LicenseMIT

wtbd (Wind Turbine Blade Diagnosis)

This is an TensorFlow demo of an acoustic-based damage detection method for wind turbine blade surfaces using convolutional neural networks (CNNs). Dataset is not included.

Relevant thesis and paper

Tsai et al 2022 Acoustic-based method for identifying surface damage to wind turbine blades by using a convolutional neural network Meas. Sci. Technol.

Tung-Chen Tsai 2021 Acoustical Damage Detection for Wind Turbine Blade Surfaces using Convolutional Neural Networks MSc Thesis National Taiwan University

Usage (via Google Colab)

You can evaluate previous data here or infer from previous data here. For new data, you can evaluate it here or infer here.