EV-assisted battery load hiding: A Markov decision process approach

Household load hiding is a customer-oriented approach to avoid possible privacy leakage from electricity-consumption data reported by smart meters. The basic idea is that customers can actively distort their consumption profile using energy storage devices. In this paper, we propose a practical design framework for an electric vehicle (EV)-assisted battery load hiding method to improve privacy as well as reduce customer electricity cost. We formulate a Markov decision process to address the uncertainties that lie in the EV arrival and departure events as well as the future household energy demand requests. A model-free learning algorithm is designed to automatically control the charging process of the EV and the battery in response to the household consumption and pricing information. Simulation results demonstrate that the effectiveness of the proposed learning framework for EV-assisted load hiding through comparison with different benchmark methods.