Abstract:Digital twins are emerging in power electronics as a tool for optimized design, fast prototyping, and smart maintenance, given the increasing demand in terms of high power density, reliability, and low cost. A power electronics digital twin can be regarded as a virtual representation of a power electronics system, embodied by a high-fidelity model running in real time, of which the states and model parameters are updated based on the feedback from the sensors or their data fusions. Digital twin models are developed in multiple domains and run in real time or even faster than real time. The data collected from the physical entities are transferred with controlled latencies and quality, and then processed to iterate the models to represent physical power electronics from application-oriented views. However, due to the nonlinearities and fast switching behaviors of power electronics, one significant challenge is to ensure high fidelity and fast computation simultaneously, which stimulates the real-time simulation (RTS) of power electronics with an ultralow time step. Moreover, the electrical, thermal, and electrothermal aspects shall all be considered to improve the model’s accuracy. To consider the uncertainties, the parameters of the digital twin models are either identified by advanced optimization algorithms or represented by probabilistic expressions. To make it possible, real-time communication should be established between the physical and digital twins. Empowered by modern Internet of Things (IoT) technology, many sensor signals can be transferred in a timely fashion to ensure real-time capability of the digital twin. The digital twin gives insightful information about its physical counterpart based on which smart decisions can be made to improve the design, prototyping, control, and lifecycle management of power electronics converters and systems. In this article, we review the digital twin of modern power electronics, and get an insight into its enabling technologies and possible applications.

Digital Twins as Electric Motor Soft-Sensors in the Automotive Industry

Application of Digital Twin in Intelligent Vehicles Research

Digital twins for internal combustion engines: A brief review

Building a Digital Twin Powered Intelligent Predictive Maintenance System for Industrial AC Machines

Development of digital twin model for mechanical application

An Interpretable Digital Twin for Self-Aware Industrial Machines

Health monitoring and prognosis of electric vehicle motor using intelligent‐digital twin

Digital Twins for Modern Power Electronics: an Investigation into the Enabling Technologies and Applications

Research on Electric Vehicle Powertrain Systems Based on Digital Twin Technology

Driving sustainability - The role of digital twin in enhancing battery performance for electric vehicles

The Digital Twin Landscape at the Crossroads of Predictive Maintenance, Machine Learning and Physics Based Modeling

Digital Twin-Enabled Internet of Vehicles Applications

Digital Twin Model of Electric Drives Empowered by EKF

Digital Twin for Flexible Manufacturing Systems and Optimization Through Simulation: A Case Study

Enhancing Digital Twin Model for Connected Vehicles by Powertrain and Longitudinal Dynamics.

Hierarchical modeling for an industrial implementation of a Digital Twin for electrical drives

Low Fidelity Digital Twin for Automated Driving Systems: Use Cases and Automatic Generation

The digital twin in Industry 4.0: A wide‐angle perspective

Overview of Digital Twin Platforms for EV Applications

A Machine Learning-based Digital Twin for Electric Vehicle Battery Modeling

Digital twins in process engineering: An overview on computational and numerical methods