International Journal of Electrical Power & Energy Systems (Aug 2024)

A multilayer voltage intelligent control strategy for distribution networks with V2G and power energy Production-Consumption units

  • Peixiao Fan,
  • Jun Yang,
  • Song Ke,
  • Yuxin Wen,
  • Xuecheng Liu,
  • Leyan Ding,
  • Tahmeed Ullah

Journal volume & issue
Vol. 159
p. 110055

Abstract

Read online

The large-scale application of distributed power generation and the prospects of vehicle to grid (V2G) technology lead to unstable operating voltage of the distribution network, but also new possible approaches to the regulation of the power system. Therefore, a multilayer voltage intelligent control strategy is proposed for a distribution network with V2G and power energy production-consumption units (PECUs). First, a model of the PECU includes facilities such as electric vehicle (EV) charging stations (CSs), urban loads, and distributed energy. The model includes a basic energy unit that can accept distribution network scheduling, provide flexible resources for the power system, and enhance system regulation capabilities. Second, based on the impact of V2G process on the demand of EV users, a CS frequency modulation power controllable boundary evaluation model and a sum power response model were obtained, which enable the CSs to participate in power regulation. Furthermore, to address complex engineering tasks with hidden rewards such as comprehensive control, an evolutionary deep reinforcement learning (EDRL) algorithm is applied and improved based on novelty search operations. This algorithm further enhances the scalability of intelligent agent learning to design a multilayer voltage control structure for the distribution network. Finally, the simulation results show that the applied models can fully utilize the power regulation capabilities while satisfying the internal power supply and demand balance in the PECUs. The proposed strategy has the best coordinated control ability, compared to particle swarm optimization (PSO), network losses are reduced by 23.93 %, and voltage deviations are significantly lowered by 71.95 %, it also ensures the charging demands of EV users, reducing the discharge process of EVs by 6 times.

Keywords