Research Survey on Detecting Faults in Wind Turbine Blade Systems

Research Survey on Detecting Faults in Wind Turbine Blade Systems

Aditya Pol1, Sahil Bhagyawant 2, Rutvik Gadade3, Yash Dhayagude4, Yogita Shinkar5

Department of Computer Engineering, (SVPM COE Malegaon)Baramati, Maharashtra, India

Abstract: The importance of wind turbines in the production of renewable energy is well recognized, but the blades of these turbines determine how functional and how long these turbines will last. Factors like cracks, erosion, and other structural failures can directly affect operational efficiency, operational and maintenance expenses, and in the worst-case scenario, catastrophic failures. However, the conventional inspection methods have some disadvantages like reliance on manual examination, and ground-level observations, which require a substantial amount of time and money, and can even be dangerous at times. This research looks into technology-based solutions for wind turbine blades fault detection and integration of NDT and artificial intelligence into one versatile system for wind turbine blades structural health assessment. Data from the blades is collected and analyzed using signal processing and classification techniques such as CNNs after employing vibration sensors, acoustic emission monitoring, and thermographic imaging.

Keywords: Wind Turbine, Convolutional Neural Network, Pre-processing, Machine Learning, Deep Learning