Wind energy infrastructure: convolutional neural networks, uncertainty quantification, and digital twins (https://engineering.unt.edu/seminars/2021/wind-energy-infrastructure-convolutional-neural-networks-uncertainty-quantification)

The improvements in wind energy infrastructure have been a constant process throughout many decades. There are new advancements in technology that can further contribute towards prognostics and health management in this industry. These advancements are driven by the need to fully explore the impact of convolutional neural networks, uncertainty quantification, and digital twins. For convolutional neural networks, the burning question has always been the requirement of the amount of data and quality of data. The talk aims to address the question: can data fusion techniques used along with data augmentation improve the damage diagnostics by using the convolutional neural network? Experimental data sets are used to create a framework based on a deep neural network for efficient fault diagnostics. This talk will demonstrate the improvement of structural health diagnosis of composites using a convolutional neural network. Uncertainty quantification for fatigue life of wind energy infrastructure is a topic that has recently gained a lot of attention. The talk aims to address the question: can the structural reliability of an offshore wind turbine under fatigue loading conditions be predicted more consistently? Digital twins have many benefits and have wide applications in the area of manufacturing, aviation, and healthcare. There is interest to explore the real-time function and offline functions of digital twins for wind energy infrastructure. There is also a need to exploit the digital twin environment in regard to the fusion of homogenous and heterogeneous data, Bayesian updating , and multi-fidelity uncertainty quantification. The talk aims to address the question: what is the role of the digital twins in diagnostics and prognostics of wind energy infrastructure?