Health Monitoring of Wind Turbine Blades through Vibration Signal Using Advanced Signal Processing Techniques

Wind turbines are used for the transformation of wind energy into electrical energy with the help of rotating blades connected to the generator. The blades of the wind turbine play an important role since these are used for the conversion of kinetic energy into electrical energy. The blades of the wind turbine experience severe vibrations due to adverse environmental conditions, huge size, variation in wind speeds, and continuous operation throughout. These vibrations lead to serious damage to the turbine resulting in the reduction of its productivity and may lead to failure in the future. The faults must be recognized at the early stage so that energy conversion is not affected. Therefore, effective health monitoring and fault diagnosis are important for evading severe damage to the wind turbines. The identification of faults in any system requires physical knowledge of the parts which are not readily available every time. Hence data-driven models are used for classification and diagnosis of faults. The present work aims to diagnose the faults in a wind turbine blade using vibration signals as the measured signal which is acquired from the hardware setup and classify the faults using different machine learning techniques and then the performance of the classifiers are compared. The experimental work is carried out using a wind turbine set-up by introducing the different conditions of the blades i.e. healthy and defective blades. To find the suitability of the proposed method, the signal acquisition is done at three different speeds using a suitable instrumentation system. The required statistical parameters are extracted from the measured vibration signals. Then three different machine learning classifiers are applied for the classification of faults. The performances of the classifiers are evaluated in terms of the percentage of accuracy from the confusion matrix. The proposed work shows that the machine learning technique is a good approach for wind turbine fault classification and the vibration signal is a good choice as a measuring signal for the detection and diagnosis of the faults in wind turbine blades.