Prediction of Current-Dependent Motor Torque Characteristics Using Deep Learning for Topology Optimization
Taiga Aoyagi, Yoshitsugu Otomo, Hajime Igarashi, Hidenori Sasaki, Yuki Hidaka, Hideaki Arita
Abstract
In this study, we propose a fast topology optimization (TO) method based on a deep neural network (DNN) that predicts the current-dependent motor torque characteristics using its cross-sectional image. The trained DNN is shown to provide the current condition that provides the maximum torque under the assumed motor control method. The proposed method helps perform TO with a reduced number of field computations while maintaining a high search capability.
Topics & Concepts
TorqueComputer scienceTopology optimizationTopology (electrical circuits)Artificial neural networkNetwork topologyCurrent (fluid)ComputationDirect torque controlField (mathematics)Artificial intelligenceControl theory (sociology)Induction motorControl (management)PhysicsAlgorithmVoltageFinite element methodMathematicsElectrical engineeringEngineeringPure mathematicsThermodynamicsOperating systemTopology Optimization in EngineeringPiezoelectric Actuators and ControlStructural Health Monitoring Techniques