Residential aggregator risk constrained profit maximization under demand response program

Abstract

This thesis proposes a Mixed Integer Non-Linear Programming (MINLP) stochastic energy model for an energy aggregator operating in the US distribution systems energy markets. Day- ahead, real-time and spot markets are considered as trading market options for the aggregator. When trading in real-time and spot markets; the aggregator faces multiple risks coming from load variability and uncertain market price. Deciding the selling price to be offered to the aggregator’s customers is a challenge for the aggregator. Uncertainties are modeled via stochastic programming and quantified via Conditional Value at Risk (CVaR). The aggregator’s optimal day-ahead selling prices to be offered to customers under real-time and spot prices uncertainty are determined by solving the proposed stochastic model. Changing the hourly prices offered to customers will change their hourly consumption resulting in a load redistribution during the day. Savings for the aggregator and customers will be gained by shifting the customers load to a lower price periods during the day. A case study is implemented to show the validity of the proposed model and influence of the aggregator in the distribution systems energy market.