An integrated optimization + learning approach to optimal dynamic pricing for the retailer with multi-type customers in smart grids

摘要: Abstract In this paper, we consider a realistic and meaningful scenario in the context of smart grids where an electricity retailer serves three different types customers, i.e., customers with optimal home energy management system embedded their meters (C-HEMS), only (C-SM), without (C-NONE). The main objective paper is to support make day-ahead dynamic pricing decisions such mixed customer pool. To end, propose two-level decision-making framework acting as upper-level agent firstly announces its prices next 24 h lower-level agents subsequently schedule usages accordingly. For lower level problem, model price responsiveness according unique characteristics. upper optimize for maximize profit subject market constraints. above tackled by genetic algorithms (GA) based distributed optimization methods while feasibility effectiveness are confirmed via simulation results.