The Study On Self-Healingcontrol Forsmart Distribution Grid Based On Multi-Agent Syetem

With the rapid development of the smart grid, the smart distribution network also begans to continuous development, so that the distribution network continues to increase its capacity, and the levels of its voltage continue to increase. The renewable energy mainly by the wind power, photovoltaic power and micro-grid accesses to distribution network, so that the control of distribution network becomes more and more complex, and the important feature of smart distribution network self-healing control gets more and more attention of scholars at home and abroad.Firstly, the features of the self-healing framework and its advantages and shortcomings was analyzed in this paper, the "2-3-10" self-healing framework was proposed, including two control loops, three-tier structure and ten control module. In the issue of distributed power and micro-grid accessing to smart distribution network, the micro-grid control module adds to the self-healing control framework, and the real-time prediction module also dds to the self-healing control framework. These two modules not only can achieve effective control of distribution networks and micro-networks, but also can predict the load and voltage changings according to the current operation of the power grid, make the distribution network ready to cope with the changes in time, guarantee the operation of the distribution networksafely and stably.Secondly, fuzzy clustering method is selected by the distribution network to partitioned, so that the distribution network can do quickly controlof local coordination with the global slow-control, for example, IEEE 10-generator 39-node, the system is divided into six districts, completing distribution network partition.Finally, the IEEE 10-machine 39-node system was used for self-healing control simulation based on Multi-Agent system in this paper. In normal operation state, distribution network uses global optimization control of the minimum network loss as the goal, making the distribution network operation in minimum energy loss, the most environmentally friendly economic status; In the non-normal state, the distribution network adopts prevention and control, compensate control, emergency control, restorative control, robust control, island control,etc, for different operating status:alert status, abnormal state,emergency state, recovery state.Taking appropriate control methods to make the operational status of the distribution network continue to develop to a good direction, and finally the distribution network to run the economy state.The distribution network can maintain the state of the economy through the prediction of the power output and load changes, can continue to run with security and stabilityand can provide good quality of power.