Data Driven Energy Economy Prediction for Electric City Buses Using Dynamic Network based Machine Learning

It looks like using electricity to power transportation systems, especially city buses, has a lot of potential. For fleet management and vehicle design to work well, they need to fully understand real-time driving data. Alternative power trains need to be carefully looked at from many technical points of view before they can work at their best. As the need for energy can change quickly, a careful design is used to keep costs low and energy waste to a minimum. Because the problem is so complicated and has many sides, businesses and researchers have not tried to find analytical solutions. If you do operational optimization right, accurate predictions of energy demand will save you a lot of money. In the energy sector of battery electric buses (BEBs), the goal of this study is to make things more open and clear. New machine learning methods were used to get a better understanding of speed profiles by using different sets of variables that could provide different explanations. Five different methods were used to do a full evaluation of the predictions' overall usefulness, reliability, and accuracy. The models used made predictions that were more than 94% accurate, which is great and in line with a careful process for choosing features. The suggested method gives manufacturers, fleet operators, and cities environmentally friendly public transportation options. It will also make people's travel habits very different.