Energy Consumption Prediction of Electric Vehicle Air Conditioning System Using Artificial Intelligence

Electric vehicles (EV) are considered as a solution to the problem of pollution in the transport sector. One of the issues that prevent expanding of EV in use is the difficulty of accurate range prediction. Range prediction is a complicated problem because it is undergoing the influence of many factors: vehicle design, environment, and driver behavior. We apply machine learning techniques to investigate the extent to which these factors affect the range of EVs and estimate the weight of these factors. We focus on the energy consumption associated with Heating Ventilation and Air Conditioning (HVAC) system operation. We considered significant features to use for model training as follows: temperature inside and outside, temperature setting selected by a driver, energy consumption in previous periods of time, and even road inclination. We used a weighted sum of Linear Regression and Random Forest algorithms, and mispredicting was only 65 watts for the 30 min time interval.