Optimization of a Power Distribution Network Based on the Regulation of Energy Storage and the Siting and Sizing of Renewable Energy

Abstract

This study proposes an optimization model for energy storage regulation and control using a deep reinforcement learning algorithm. The model is trained to optimize the control of energy storage systems, considering the regional wind power output and electricity consumption load. Simulation results demonstrate that the algorithm effectively reduces peak load values, achieves a balanced relationship between renewable energy and electricity demand, and mitigates the issue of trend reversal. Consequently, it stabilizes the distribution of tidal flow at high penetration rates. Subsequently, to stabilize the distribution trend, we introduce a decision model for determining the siting and sizing of renewable energy. This model is based on a multi-objective genetic algorithm, with line loss and voltage volatility as the loss function. We evaluate the model using the IEEE14-bus system and find that it successfully identifies the optimal access siting and sizing for renewable energy. Ultimately, the goal of stabilizing the voltage volatility of the nodal network and reducing line losses was successfully achieved using this model

Keywords: Energy storage optimization, Power distribution network, Siting and sizing of renewable energy, Line loss, Stabilized voltage magnitude