Low-carbon economic dispatching of microgrid considering generalized integrated demand response and nonlinear conditions

Abstract

With the depletion of fossil fuels and the deterioration of the ecological environment, it is crucial to realize the low-carbon economic dispatching of microgrids (MG). In previous studies, the Demand Response (DR) can reduce microgrid operating costs and CO2 emissions by optimizing the load curve, under this background, this paper considers the coupling relationship between electricity, heat, and gas. And by the introduction of Power-to-Gas (P2G) and Organic Rankine Cycles (ORC), the flexible conversion of electricity-gas and heat-electricity is realized. In addition, based on the characteristics of the time-shifting load, interruptible load, and alternative load, the Generalized Integrated Demand Response (GIDR) model is constructed. Secondly, to restore the non-linear status of energy conversion equipment efficiency under different partial loads, this paper adopts data-driven modeling and uses Multilayer Perceptron (MLP) to fit the curves of key parameters, with the help of the Soft Actor–Critic (SAC) algorithm, the solution of the model is realized. In the case study section, this paper verifies the rationality of the proposed method by using real data from Northwest China. The results show that GIDR can reduce the cost by 16.40% and reduce the CO2 emissions by 32.26%, and compared with not considering nonlinear conditions, the method in this paper can reduce the operating cost by 23.16% and carbon emissions by 33.00%.

Keywords

• Nonlinear conditions
• Multi-energy microgrids
• Deep reinforcement learning
• Integrated demand response