Research on Multi-level and Multi-Objective Optimization of Microgrid Based on Large-scale System Control Theory

Abstract:
In order to implement the goal of carbon neutralization, China has vigorously developed new energy in recent years. With the development of new energy, how to improve comprehensive economics of microgrids composed of new energy and conventional power supply became an urgent and realistic problem. This paper proposes a multi-layer hierarchical control structure based on the large-scale system theory to solve the coupling problem of various units in microgrid dispatching. This paper takes wind power generation systems, photovoltaic power generation systems, solar thermal power generation systems, and diesel generators as microgrid's research objects. It takes the microgrid environmental protection index and renewable energy utilization index as the optimization objective. The coupling between the systems is cut off using interaction balance, model simplification, and decentralized predictive control. The joint power generation system is divided into independent subsystems. Local decision-making units are set up for each subsystem to improve the quantum particle swarm algorithm for each subsystem to optimize. The joint system relies on a multi-level hierarchical structure to repeatedly iterate to obtain the optimal solution. Each subsystem uses a model predictive control algorithm to control each unit group's output to achieve the optimal operating state of the joint power generation system and achieve the comprehensive economy of the system Optimal. Finally, a simulation comparison between the decomposition and coordination scheduling based on the large-scale system theory and the complete scheduling based on the improved quantum particle swarm optimization shows that this method can effectively reduce the total power generation cost of the microgrid by about 2,322$, and the calculation time is reduced from 44.2 seconds to 12.5 seconds.