Skip to main content
APEC 2025 Main banner
Back

Technical Lecture

AC, DC, BLDC Motor Drives

T16.4 - Neural Network based Digital Twin Health Monitoring of BLDC Motor Drives for Robots

Wednesday, March 19, 2025
9:30 AM – 9:50 AM ET
Location: Level Three, A314
Robots have shown promise in numerous applications, such as space exploration and disaster rescue. Due to the harsh environmental conditions, robotic arm joint motor-drive systems are vulnerable to hardware failures. Digital twin (DT) models, i.e., a virtual replica of the physical system, of the utilized motor drives are paramount for robotic applications since DT models can be used for high-accuracy online health monitoring in safety-critical applications. Thus, this paper presents a novel DT model of a two-level three-phase inverter to drive a robotic arm brushless DC (BLDC) motor based on a dynamic neural network (NN). Various NN architectures have been tested to determine the best trade-off between the model accuracy and computational burden, ensuring that the proposed model can be used in real-time applications. Finally, the efficacy of the proposed NN-based DT approach is validated using experimental data from a lab-scale prototype.