An EV-Traction Inverter Data-Driven Modelling for Digital Twin Development

Digital Twin (DT) technology has achieved a significant breakthrough in the development of numerous applications, including electric vehicles (EVs). This technique involves the creation of a virtual version of a real object, utilizing simulation and optimization tools with real data to optimize its functioning. The traction inverter is considered the brain of the EV drive system, as it controls the power flow between the battery and the electric motor. It converts the DC voltage from the battery into AC voltage and adjusts the frequency and amplitude to control the speed and torque of the motor. This research work presents the DT modelling procedures of an EV traction inverter. The basic principles of DT are discussed, and the model-driven DT modelling technique is compared to the data-driven. The motor-drive physical model (test bench) is provided. A data-driven model of the inverter was generated using a nonlinear regression-based artificial neural network ANN. Historical data from experimental tests were used as a modeling reference. The obtained results from the data-driven model achieved high prediction accuracy.