Generalizable Temperature Nowcasting with Physics-Constrained RNNs for Predictive Maintenance of Wind Turbine Components

arXiv:2404.04126v1 Announce Type: new
Abstract: Machine learning plays an important role in the operation of current wind energy production systems. One central application is predictive maintenance to increase efficiency and lower electricity costs by reducing downtimes. Integrating physics-based knowledge in neural networks to enforce their physical plausibilty is a promising method to improve current approaches, but incomplete system information often impedes their application in real world scenarios. We describe a simple and efficient way for physics-constrained deep learning-based predictive maintenance …

abstract application arxiv components costs cs.lg cs.sy current eess.sy efficiency electricity energy knowledge machine machine learning maintenance networks neural networks nowcasting physics predictive predictive maintenance production role systems type wind