Textbook in PDF format This book thoroughly explores the realm of data-driven blade-icing detection for wind turbines, focusing on multivariate time series classification to enhance the reliability and efficiency of wind energy utilization. The widespread prevalence of sensor technology in wind turbines, coupled with substantial data collection, has paved the way for advanced data-driven methodologies, which do not require extensive domain knowledge or additional mechanical tools. The interdisciplinary appeal of this study has drawn attention from experts in fields like computer science, mechanical engineering, and renewable energy systems. Adopting a comprehensive approach, the book lays down a foundational framework for blade-icing detection, stressing the critical role of sensor data integration and the profound impact of machine learning techniques in refining the detection processes. The book is designed for undergraduate and graduate students keen on renewable energy technologies, researchers delving into machine learning applications in energy systems, and engineers focusing on sustainable solutions for enhancing wind turbine performance