ABSTRACT
The structures of wind turbine blades (WTBs) are very susceptible to damage from various sources controlled by the geometry, strength of composite material, strength of adhesive, stress distribution, etc. Hence, sustainable materials are important for their design and work. The chapter presents the development of a machine learning (ML) model for predicting the failure of a WTB based on the design parameters, properties of material, and environmental conditions inputted. After the modeling of the WTB, the computational fluid analysis is carried out to obtain the pressure distribution on the blades developed due to the incoming wind. This is followed by the finite element analysis to estimate the deformation produced and stresses developed in the blades due to the combination of these pressure loads and the centrifugal force. At the same time, various ML models are trained based on the available data to infer that the neural network (accuracy of 95%) is the best model for fault detection in the turbine system. As a result, the threshold wind speed for the failure is found to lie in the range of 49.2−55.25 m/s, which is close to the results obtained from the computational study.
