US wind turbine blade producer TPI Composites has collaborated with a US research centre to design a composite manufacturing process based on digital twin technology.
The approach is the result from working with WindSTAR Center, which has used machine learning (ML) to serve as the digital twin of the blade manufacturing process.
The ML framework provides real-time feedback during fabrication, results in reduced defects, and enables more efficient production of wind blades versus the high computational costs of physics-based models.
Stephen Nolet, Senior Director of Innovation & Technology for TPI, worked alongside student researchers and faculty from the University of Texas at Dallas, as well as technical experts from Olin Epoxy and Westlake Epoxy to develop a framework for the digital twin of the vacuum assisted resin infusion moulding (VARIM) process.
By applying an ML approach, the team achieved predictive accuracy of more than 95% with 100-times faster computation than the physics-based simulations.
Nolet said: “The primary value of utilising a ML framework is leveraging historical results and data to inform current manufacturing at a pace that significantly reduces defects from occurring in a real-time production environment.”
In the coming year, the WindSTAR research team plans to focus on scaling the technology to larger components with greater manufacturing complexity.
The work will apply tools taken from artificial intelligence (AI) to find patterns in historical data and predict outcomes on full-scale wind blade components including blade shells.
The WindSTAR Center is operated jointly by the University of Massachusetts Lowell and the University of Texas Dallas and supported by 18 Industry members including TPI.