Edge-Enabled Predictive Maintenance of Wind Turbine Blades: Intelligent Embedded Approach

Abstract

This work presents an integrated framework for intelligent monitoring and predictive modeling of wind turbine blade deformation, combining numerical simulations, artificial neural networks (ANNs), and embedded sensing technologies. Recognizing the critical role of blade integrity in wind turbine performance, the proposed approach leverages contactless strain sensors embedded in composite structures, with data acquired wirelessly to facilitate real-time, scalable deployment across wind farms. The framework incorporates not only deformation data but also environmental parameters such as temperature, humidity and ultraviolet radiation, which significantly influence blade stress behavior and can affect also sensor accuracy. A robust architecture is outlined for collecting, transmitting, and analyzing data using edge computing and cloud integration. Priority is given to measurement accuracy, with secure wireless communication ensuring data integrity and reliability for informed decision-making. Numerical simulations using ANSYS Workbench inform optimal sensor placement, while ANN models trained on synthetic and real-world data predict stress evolution under varying conditions. A MATLAB Simulink-based system enables early fault detection through cumulative stress estimation, supporting proactive maintenance strategies. The proposed framework establishes a scalable and secure architecture for intelligent condition monitoring of wind turbine blades, integrating edge computing, wireless sensor networks, and predictive modeling techniques.