Optimal real-time pricing and electricity package by retail electric providers based on social learning

Abstract

As the global retail electricity market liberalizes, retail electric providers (REPs) are optimizing pricing schemes and electricity packages (EPs) to meet consumers' individual and diverse electricity plans while reducing volatility risk and ensuring profitability. However, consumer social learning (SL) helps frequently adjust electricity plans as social media evolves, preventing REPs from accurately understanding consumer electricity demand. Therefore, this paper constructs a dynamic game model of the retail electricity market considering social learning. REPs announce price and electricity package information to consumers based on market information and historical experience, and consumers first shared electricity usage plans based on their willingness to learn and share information among themselves, and then respond to REPs based on current information and experience. The main findings of this paper are as follows: (i) social learning increases consumer utility, promotes competition, depresses REPs prices, shrinks price differentials, and induces concessions by REPs, but raises volatility risk. (ii) electricity packages combined with social learning can save electricity consumption and significantly reduce volatility risk. Finally, this paper uses Smart Meter Energy Consumption data in London households to demonstrate the impact of social learning and to provide decision support for REP's electricity sales strategies.