Study Of Distribution Network Optimization With Distributed Generation Based On NSGA-? Algorithm

As the problems of the shortage of fossil energy and the environmental pollution are increasingly serious, the distributed generation has grown rapidly because of its advantages of clean and efficiency. But with the interconnection of a large number of DGs, the network will become active and interconnected network between power and user from passive network, so the interconnection of DG will affect the operation of distribution network. In order to improve the performance of the distribution network with DG, this paper prepares to study the multi-objective optimization of distribution network with DG using distribution network reconfiguration and reactive power optimization.By analyzing the character of all kinds of DGs, this paper makes them equivalent to PQ?PQ(U)?PI and PV type nodes respectively. Because of the forward-backward sweep method can't deal with PV node directly, this paper uses the reactive correction method to improve it. The simulation results show that the improved algorithm can deal with the power flow calculation of network with all kinds of DGs, and the algorithm is fast and stable.This paper establishes the multi-objective reconfiguration model of distribution network with DGs, in which the optimization functions are minimum the network loss and the load balance. In order to reduce the number of the infeasible solution, this paper proposes the operation of coding?crossover?mutation based on loop and the method of judging infeasible solution. Because of the frequent change of the network topology, the load flow calculation is very difficult. In order to solve of the problem, this paper proposes the method of node layering. We applied NSGA-? algorithm that adopted above operation to the distribution network with DG and no DG, the simulation result shows the optimization effect is remarkableConsidering the reactive power of DG can be adjusted independently, this paper combines DG with traditional reactive power optimization means to optimize the distribution network. And the paper establishes the multi-objective reactive power optimization model of distribution network, in which the optimization functions are minimum the network loss and the voltage deviation. Dealing with the continuous variables and discrete variables by right encoding and decoding, we apply NSGA-? algorithm to the solution of the model. The simulation result shows that the reasonable utilization of the DG's reactive power can improve the performance of the distribution network better.In addition, the NSGA-? algorithm can get a set of optimal solutions and the policymaker can choose from them, so the algorithm avoids the blindness of weight method. The result verifies the feasibility of applying the algorithm to multi-objective distribution network reconfiguration and reactive power optimization.