Driving cycle construction for electric vehicles based on Markov chain and Monte Carlo method: A case study in Beijing

As a simulation of real-world driving data, driving cycle is widely used for the evaluation of vehicles' economy, emission and driving range. However, most of existing driving cycles are constructed based on traditional vehicles and proved not suitable for electric vehicles. In this work, real-world driving data of 40 electric taxis for 6 months in Beijing area are used to construct a driving cycle to appropriate for electric vehicles' evaluation. Road type data are considered to improve the representativeness of constructed cycle using the conventional Markov chain method for real-world driving data. Here, we extract 12 parameters, which describe the characteristics of driving cycle, to indicate the differences among the constructed driving cycle, NEDC and real-world driving data. Results show that the new constructed driving cycle has improved representativeness for real-world driving data in Beijing compared to NEDC.