Reactive Power Optimization In Distribution Network With Distributed Generation

In order to deal with the problem of energy shortage and environmental pollution, distributed power supply (DG) has been widely introduced into the distribution network, because it has many advantages such as environmental protection, reliable power supply and flexible power generation. The interconnection of DG provides supplement and effective powerful support to distribution network, but it also changes the single power supply mode of traditional distribution network , which will has important effect on the network loss and voltage etc. Therefore, it is necessary to control the distributed generation and optimize the allocation of reactive power compensation equipment, which can make the system stable and economical. So it is of important theoretical and practical significance to make a study on the reactive optimization of distribution network with distributed generation . The content of this paper is mainly divided into the following three parts :(1)The typical distributed power generation technology is summarized. And the node processing method of different distributed power supply is analyzed in detail. The Newton-Raphson method is used to calculate the power flow. The influence of grid-connected DG on system loss and voltage quality is analyzed.(2)Aiming at the problem of reactive power optimization of distribution network with DG, an improved multi-objective teaching and learning algorithm is proposed. The self-learning mutation stage is added to the teaching and learning algorithm, and the self-adaptive linear weight is introduced to improve the convergence accuracy of the algorithm. The roulette method of the niche technology is employed to select the "teacher"to maintain the diversity of particle swarm, and the Pareto solution is reduced by the non-dominated sorting method and the phasing-out strategy,which improves the diversity and uniformity of the Pareto solution.(3)The multi-objective optimization strategy of reactive power in distribution network with DG is studied, and multi-objective reactive power optimization model with DG including minimum power loss,minimum voltage deviation and lowest reactive power cost is established.By using the IEEE 33-node power distribution system as a simulation example, an improved multi-objective teaching and learning algorithm is proposed to solve the reactive power optimization problem, and a set of diverse solutions are obtained, which provides a flexible alternative about reactive power optimization for decision makers.