Wind Turbine Blade Fault Detection Method Based on TROA-SVM

Wind turbines are predominantly situated in remote, high-altitude regions, where they face a myriad of harsh environmental conditions. Factors such as high humidity, strong gusts, lightning strikes, and heavy snowfall significantly increase the vulnerability of turbine blades to fatigue damage. This susceptibility poses serious risks to the normal operation and longevity of the turbines, necessitating effective monitoring and maintenance strategies. In response to these challenges, this paper proposes a novel fault detection method specifically designed for analyzing wind turbine blade noise signals. This method integrates the Tyrannosaurus Optimization Algorithm (TROA) with a support vector machine (SVM), aiming to enhance the accuracy and reliability of fault detection. The process begins with the careful preprocessing of raw noise signals collected from wind turbines during actual operational conditions. The method extracts vital features from three key perspectives: the time domain, frequency domain, and cepstral domain. By constructing a comprehensive feature matrix that encapsulates multi-dimensional characteristics, the approach ensures that all relevant information is captured. Rigorous analysis and feature selection are subsequently conducted to eliminate redundant data, thereby focusing on retaining the most significant features for classification. A TROA-SVM classification model is then developed to effectively identify the faults of the turbine blades. The performance of this method is validated through extensive experiments, which indicate that the recognition accuracy rate is 98.7%. This accuracy is higher than that of the traditional methods, such as SVM, K-Nearest Neighbors (KNN), and random forest, demonstrating the proposed method's superiority and effectiveness.