Application Of Immune Algorithm And Fuzzy Evaluation For Distribution Network Self-healing Restoration

As the self-healing problem for faults of distribution network is complex and nonlinear, it is a mixed optimization problem with multi-objectives and constrains. However, since it has significant impact on grid reliability and realization of self-healing, the study of the self-healing problem for faults of distribution network has important theoretical and practical significances.A strategy based on immune algorithm and fuzzy evaluation is proposed to solve this problem in the event of a large-scale blackout. It means a combination approach of immune algorithm and fuzzy evaluation is used to solve the mathematical model of the self-healing problem for faults. First, the method of power supply capability assessment for a single node is used to obtain the tie-switches’ capacity, which are the necessary constraints in immune algorithm. Then, immune algorithm searches feasible schemes for self-healing restoration. When the population is initialized, the constraints of network and tie-switches’ capacity ensure initial solutions to be feasibility and reasonable. So the avoiding of unfeasible solutions greatly reduces the search space, and the impact on immune searching due to the unfeasible initial solutions is also greatly decreased. In the process of searching for the best solution, antigens were changed dynamically in order to get the best solution as quickly as possible. Fuzzy evaluation is used to evaluate the schemes from four aspects, which include switch operation times, net loss, level of voltage drop and tie-switches’ capacity margin. Weighted sum of four indices provides a good guideline for choosing the optimal scheme for self-healing operation. The weight of the four indicators in fuzzy evaluation can be varied by the different purpose of self-healing problem, which can also appropriately reflect the grid’s situation after restoration.The calculation results of33-bus distribution system show that the proposed algorithm has a highly computational efficiency and the final scheme can effectively ensure the system run safely and economically after self-healing restoration.