In the search for a sustainable and clean alternative to fossil fuel, renewable energy sources, specifically, wind energy is seen to possess immense potential with a lot of scope for improvement. Wind energy systems, using rotating blades, convert the kinetic energy of winds into other forms of usable energy like electrical or mechanical energy. The most essential part of these systems is the wind turbine. The dependability of these wind turbines directly correlates to obtaining the maximum quantity of energy from the available wind source, resulting in an increase in the overall efficiency of energy generation. However due to varying environmental conditions, defects and damages are unavoidable creating the urgency for new and innovative maintenance strategies to ensure system’s safety, profitability and cost-effectiveness. Techniques such as structural health monitoring (SHM) and fault diagnosis system (FDS) have proven to be imperative in the pre-emptive detection of faults in the wind turbine. These techniques facilitated a proactive feedback mechanism, reduced downtime of the systems thereby increasing its productivity. This study reviewed various strategies, methodologies and machine learning algorithms developed for monitoring of wind turbine performance and early fault detections in these turbines. Additionally, a novel, state of-the-art technique using SHM and FDS for diagnosis on wind turbines is also presented. This study is an extended work of Márquez et al (2012). © TJPRC Pvt. Ltd.