A multi-mode electric vehicle range estimator based on driving pattern recognition

Limited driving range and availability of charging infrastructures are still among the main barriers adoption electric vehicles (EVs) in market. Combination those limiting factors causes ‘range anxiety’ EV users. While different battery technologies under development, one short-term solution to reduce users’ anxiety is provide user with an accurate estimation. In this study, estimation technique proposed that recognises current pattern then classifies it into predefined clusters (driving modes). The future energy consumption per kilometre tuned according average each cluster. Having updated rate, calculated based on state-of-charge. Different features considered for clustering where ‘average speed’ power’ were identified as best choices application. effectiveness estimator validated using real data gives error 9% ahead.