Evaluating the road gradient effects on electric vehicles’ energy consumption and regeneration using digital surface model

The demand for vehicles, coupled with the need to analyse emissions and efficiency on slopes, has prompted the need to assess the impact of terrain on the engine efficiency of electric vehicles (EVs) and their operations. This study aims to assess the operational efficiency of EVs in varied settings or on uneven terrain. The real-world performance of these vehicles involves many conditions in which they are expected to traverse inclined terrain, affecting their energy expenditure and operational efficiency. This study aims to analyse the impact of road inclination on EVs' energy expenditure and the energy used in regenerative braking. This study, based on Terrain’s Digital Surface Models, Geographic Information Systems (GIS), Machine Learning (ML) technologies, and the Analysis of Moment Structures (AMOS) software, assesses the impact of terrain, vehicle characteristics, and climatological factors on the energy performance of EVs. The study analyses energy usage patterns and recovery behaviour in Amman city streets through an analysis of road slopes alongside speed and vehicle weight effects under different traffic and weather conditions, and air conditioning (A/C) and lighting impacts on energy usage. This study also claims that climbing significantly increases energy expenditure, and descending allows for energy recovery through the brakes. EVs can regain 40% of the electricity used when driven downhill. Factors such as the vehicle's weight, vehicle aerodynamic design, and road slope affect energy consumption. Although at high speeds, aerodynamic design has a minor effect. The efficiency of the recovery systems will also decline as external weather conditions increase energy consumption and enhance rolling resistance. Traffic delays also increase energy consumption because of the repetitive acceleration and deceleration. The boosting technique allowed us to develop models to predict energy consumption and regeneration with high precision. The generated boosting models for energy consumption prediction surpass the R-squared regression model of 0.86, as the regression boosting model achieves 0.95. Gradient boosting also demonstrates a good capacity to analyse the nonlinear interactions of diverse variables. For energy vehicle designs, the route research slope and environmental vehicle system designs will provide energy efficiency solutions. The research work provides a starting point to revolutionize energy vehicle technologies and transport systems within a larger scope.