An AI-based digital twin of the Electric Vehicle

For commercial vehicles, reliability is key since the vehicle is typically linked to the daily earnings of the owner. To ensure continuous vehicle operation, early issue diagnostics and proactive maintenance are important. However, an electric vehicle is a complex and dynamic system consisting of numerous components interacting with each other and with external environments such as road conditions, traffic, weather, and driving behavior. Thus, vehicle operation and performance are highly contextual and for identifying an abnormal operation (diagnostics) the solution must take into account the conditions under which it is driven. To address this, we have built an AI-based digital twin of electric vehicles. The model is an AI-based generative transformer network that generates near-ideal vehicle behavior. The present focus has been on motor, controller and battery. The model is trained using 200 vehicles first 1500 km driven data. To ensure the digital twin model learns near-ideal vehicle behavior, the vehicles used for training are the ones that did not report any issues during the initial and subsequent three months. Results show that on the test set comprising good vehicle behavior, the digital twin model data resulted in an R2 score of 0.97. Results show that 25% of vehicles with reported progressive issues in motor, controller and battery, show higher error (RMSE 50%) show lower health scores at least two weeks in advance, thus enabling the service team to do proactive maintenance for these key components. The vehicle digital twin model can be used to generate a health score for each vehicle by comparing the generated ideal behavior to actual vehicle output. The score can aid the service team in proactively identifying vehicles with issues and addressing them before vehicle breakdown. The digital twin can also help in estimating the remaining useful life and resale value of a vehicle.