Optimal design of demand response model considering uncertainty

In order to solve the challenges brought by new energy access to the power grid, this paper explores how to prompt the user to respond to demand with the lowest scheduling cost. In order to calculate the scheduling cost of demand response, it is first necessary to calculate the baseline load of demand response, and then predict the actual load of the area. This paper uses the BP neural network to predict the actual load after the demand response occurs, so as to obtain the load reduction in the demand response. According to the difference in demand response, this paper establishes two models: an incentive-based demand response model and a price-based demand response model, and introduces uncertain variables and correction factors with uncertain variables to obtain uncertainty Model. Finally, genetic algorithm is used to optimize the above two models, and the optimization results of these two models are compared.