| With the economic development, energy supply became very tight, how to make use of clean energy more and more becomes a concern to people. Large-scale use of Distributed Generations (DG) can improve the energy and environmental issues. So the research of DG technology and coordinated operation with distribution network has an important significance. Since DG connected to distribution network, distribution network from the original radiation single power supply become multi-mode power supply coexist, it brought the opportunity for diversification of Distribution energy, also brought new challenges for Distribution safe and stable operation.This paper analyzes the Power principle, feature and network characteristics of DG, And study the theory of Multi-Agent Control. On the base of Learning in the traditional Multi-Agent Reinforcement, In order to speed up the progress of learning, I proposed a quantum computing Multi-Agent System Reinforcement Learning Algorithm. Find new ways that it is different from Traditional way of behavior choices, making Use of quantum superposition and entanglement theory fully, which improve the parallelism of Multi-Agent Reinforcement Learning. The experiment results show that the new algorithm is effective. And the algorithm is applied to the coordination control of distribution network with DG. According to the different users load and DG characteristics, establish the coordination and control framework, knowledge base and control strategies. Finally, design the hardware and software of intelligent distribution network coordinated control system.The experiment results show that the new algorithm is effective. It achieve optimal control of distribution network and DG, and can be quickly and effectively to restore power supply when distribution network have fault, Even when extreme events are presence, it can effect to ensure the distribution network critical load continuity and reliability operation. |
PDF Full Text Request