Driving Simulator for Electric Vehicles Using the Markov Chain Monte Carlo Method and Evaluation of the Demand Response Effect in Residential Houses

It is essential to secure flexible resources in power systems to increase the proportion of variable renewable energy generation systems. One flexible resource is demand response (DR) of the batteries of electric vehicles (EVs). In this study, we propose an electric vehicle driving simulator using the Markov chain Monte Carlo (MCMC) method and an EV demand response evaluation model. The former is a highly versatile EV driving simulator that can produce a random driving pattern based on actual EV fleet data, taking into account various features. The latter is a residential DR evaluation model that minimizes a household electricity bill based on the simulated fleet data and enables realistic DR operation using three processes: forecasting, planning, and operation. The contribution of this paper is to enable evaluation of realistic EV battery value in various housing and EV utilization combinations. We found that the EV battery control provides an economic benefit of US$62 (5% of the night-charging case cost) with only charge control and US$370 (31%) with charge and discharge control, as the average expected value based on our assumption of evaluation. Because 40-kWh EV batteries have a sufficiently large capacity to store surplus power from a rooftop PV, they can be operated by determining the operation schedule based on a fixed-fee structure without forecasting or planning.