Electric Vehicle Battery Management using Digital Twin

Abstract

In the transportation business, battery-powered electric vehicles (EVs) are regarded the immediate solution to internal combustion engines in light of the growth in environmental pollution. While expanding the use of electric vehicles, battery-related difficulties such as range anxiety, safety concerns, cost, and the availability of charging stations are important concerns. A precise online estimate of the battery’s State of Health (SoH) has the ability to resolve some of these issues. However, computing the SoH on EVs is computationally intensive, necessitating expensive onboard integrated electronics and rapidly draining the EV battery. In addition, the SoH estimating algorithms currently available do not utilise incremental battery usage data. This research presents a digital twin of the EV battery as a solution to the difficulty of onboard computation for incremental SoH prediction. It enables intensive computing and analytics to be performed in the cloud instead of a vehicle’s battery management system (BMS). It calculates the SoH of the battery using an incremental learning method with a mean square error of 0.023.