Day-Ahead Intelligent Energy Management Strategy for Manufacturing Load Participating in Demand Response

Abstract

Flexible resources such as adjustable load widely participate in interaction with power grid, which can effectively promote renewable energy consumption. In previous studies, researchers generally focused on industrial and household users, but usually ignored the manufacturing load. Therefore, in this paper, an day-ahead intelligent energy management strategy for manufacturing load is proposed. Firstly, we analyze the power demand behavior of manufacturing load in detail, and describe the energy flow and material flow of manufacturing load through state task network (STN) method and mixed integer linear programming model. Then, the conditional deep convolution generative adversarial networks (C-DCGAN) algorithm is used to describe the uncertainty of new energy and construct a set of scheduling scenarios. Finally, case study shows that the proposed method can effectively improve the regional renewable energy consumption level and economic benefits of enterprises.

Keywords

• Manufacturing load
• state task network (STN) method
• conditional deep convolution generative adversarial networks (C-DCGAN) algorithm
• typical scenario screening method