Digital twin for condition monitoring of actuators

Abstract

Digital Twin is an individual model of the physical system with the support of enabling technologies, including multi-physics simulation, machine learning, etc. We developed Digital Twin for electromechanical actuators (EMA) containing three essential parts: real physical system (real EMA), digital model (virtual EMA), and information interconnection (real-time data from sensors and control), which allows capturing the current state of the physical system in a digital model. The virtual EMA provides the near real-time representation of the physical EMA. It offers the potential to optimize operations and maintenance and further accelerates the new product development process. The virtual EMA enables degradations detection, What-If simulations, RUL estimations, KPIs trending, etc. The degradation tracking performance of the digital twin was evaluated using a test stand with real EMA and seeded faults (partial winding short). The results indicate that this approach can be used for the monitoring of EMA degradations.