Study On The Application Of Reactive Power Optimization Algorithm To Mianyang Urban Distribution Network

Reactive power optimization of distribution system can effectively reduce network losses, improve the quality of voltage, and make the system security, economic and high-quality operation. In order to effectively reduce network losses and reasonablely carry on reactive power compensation, how to choose the number, the location and the capacity of the reactive power compensation buses of 10 kV distribution networks are mainly researched in this paper. The main research results are as follows:According to reactive power optimization problems in distribution network, a reactive optimization model that make network active power loss to be minimumed as the objective function, including the constraints of the power balance of buses, the reactive power of electric source bus, the bus voltage magnitudes, reactive power compensation quantity of the compensation buses and the economy is presented in the paper. The optimal network loss differental rate rule is introduced in the model to assure the number and the location of the compensation buses automatically chosen that is reasonable in economica in the process of optimization. The overall sensitivity methods and local sensitivity methods to chosse the compensation buses in reactive power optimization of the distribution network, and the fundamental principle and the calculating steps of the forward/back sweep in the distribution network are discussed.In terms of the characteristics of the above model and the two sensitivity methods, the ant colony optimization algorithm (ACOA) of the reactive power optimization in distribution network based on overall sensitivity method and local sensitivity method are proposed. On the basis of analyzing the basic principles of ACOA and TSP problem solved by ACOA, the sensitivity of methods and the optimal network loss differental rate rule are combined with ACOA to construct a new algorithm. A new state transition probability rule that put the sensitivity of the candidate buses as the visibility factor is brought forward and the ant colony optimization search strategy is improved. Due to the ACOAs not only considering active power loss of the entire network, but also taking into account the impact of the sensitivity factor on the process of optimization, they can find compensation buses that have a greater impact on network losses and improve the search speed.A Matlab program of ACOA based on overall sensitivity methods and local sensitivity methods is implemented. The typical systems of IEEE 33-bus and IEEE 69-bus were calculated and analyzed by the program. The results show the feasibility, accuracy and effectiveness of the method and the program in this paper. One of Mianyang urban radial distribution networks is calculated by the program and the optimal compensation scheme can be provided a basis for planning and operation decision-making.Simulation studies show that compared with the pure sensitivity methods, ACOA based on overall sensitivity methods and local sensitivity methods perented in the paper have some merits in reactive power optimization of the distribution network, such as reduction of the human interference, shrinkage of the search space, automatic selection to the reasonable compensation buses, choice of the less number of buses and simpler calculation; and compared with genetic algorithm have the characteristics of faster search speed and simpler programming.