Charging management of plug‐in electric vehicles in San Francisco applying Monte Carlo Markov chain and stochastic model predictive control and considering renewables and drag force

Abstract

The charging management of plug‐in electric vehicles (PEVs) in San Francisco considering the effect of drag force on the vehicles, the real driving routes of vehicles, the social aspects of drivers' behaviour, the type of PEVs and the PEV penetration level is presented in this study. In this study, the drivers' responsiveness probability, to provide vehicle‐to‐grid service at the parking lot, is modelled with respect to the value of the incentive, drivers' social class and the real driving routes in San Francisco. Herein, the Monte Carlo Markov Chain is applied to estimate the hourly probability distribution function of the state of charge (SOC) of the PEV fleet in the day. The main data set applied in this study includes the real longitude and latitude of driving routes of vehicles in San Francisco, recorded in every four‐minute interval of the day. In this study, a stochastic model predictive control is applied in the optimisation problem to address the variability and uncertainty issues of PEVs' SOC and renewables' power. Herein, quantum‐inspired simulated annealing algorithm is applied as the optimisation technique. It is demonstrated that the type of PEVs, the PEV penetration level and even the social class of drivers can affect the problem results.

Keywords

• probability distribution function
• PEV fleet
• driving routes
• San Francisco
• stochastic model predictive control
• PEV penetration level