A review of machine learning approaches for electric vehicle energy consumption modelling in urban transportation

Abstract

Global warming and carbon emissions have drawn attention to the need to decarbonize transport. Promoting electric vehicles (EVs) has become an important strategy towards this goal. Although EVs have significant advantages in emission reduction and energy saving, their wider adoption is limited by factors such as range and charging infrastructure. Accurate prediction and modeling of EV energy consumption is the key to solving these challenges. This review first summarizes traditional energy consumption estimation models and, thereafter explores the application of interpretable machine learning, emphasizing its importance in improving model transparency and practicality. The potential of neural networks to enhance prediction accuracy through feature extraction and pattern recognition is also discussed. This paper aims to provide a systematic review of the latest advances in EV energy consumption forecasting. It serves as a reference for researchers and practitioners to optimize and upgrade urban transport systems for sustainable development.