An international collaboration between EPFL and the University of Glasgow has led to the development of an advanced machine-learning algorithm that can effectively detect hidden manufacturing defects in wind turbine composite blades, significantly boosting quality assurance processes in the wind energy sector.
Detecting faulty wind turbine blades before they are put into service is crucial to avoid substantial costs for operating companies. Traditionally, quality inspections have been limited to surface checks of specific areas as these composite structures are manufactured. However, the new approach co-created by researchers at EPFL and the University of Glasgow introduces a patented radar technology combined with an AI assistant. This combination allows inspection engineers to identify possible anomalies beneath the surface, offering a non-destructive, non-contact, agile, and low-power solution for data acquisition and analysis.
The research has been published in Elsevier Mechanical Systems and Signal Processing (MSSP) and merges signal processing and AI techniques developed by Olga Fink from EPFL and David Flynn from the University of Glasgow. Fink, who heads the Intelligent Maintenance and Operations Systems Laboratory (IMOS) at EPFL, highlights the increasing complexity of wind turbine designs and materials, which raises the likelihood of defects during manufacturing.
Flynn’s team at the University of Glasgow utilized a Frequency Modulated Continuous Wave radar with a robotic arm to inspect wind turbine blade samples at various distances to isolate features and potential failures. The IMOS team then enhanced the raw data information content through AI algorithms, enabling the identification of anomalies within the turbine parts.
The collaboration plans to further validate their findings with additional data and intends to test the method on existing turbines using robotic arms or drones. Spotting manufacturing defects in turbines before deployment and inspecting them during operation can ensure defect-free operation for up to 20 years. This innovative approach represents a significant advancement in wind turbine quality assurance, signaling a promising future for the industry.
