Studies On Intelligent Technology Of Reactive Power Optimization In Rural Distribution Power Network

Rural power grid which is the important part of the electric system and is also the basic condition of the rural economy, agricultural production and rural social development. Constructing smart country grid is the major theme of constructing uniform Strong Smart Grid. The Optimal Reactive Power of the country grid is very meaningful for improving the reliability of the country grid, reducing the power loss, increasing the voltage passing rate and providing high quality electric energy to users. Consequently, it is very important to complete the Optimal Reactive power of the country grid which is the important part of constructing smart country grid.The paper puts forward a whole scientific reasonable scheme of the optimal reactive power from distribution line optimal reactive power planning to operation optimization and to specific implementation, which is based on system analysis the present state of country grid, makes use of present data resource and automatic level.The distribution network is the basic of the reactive power optimization. According to the characteristics of load flow calculation in rural distribution network, the paper put forward a new node numbering optimization based on forward and backward substitution method. This method could compactly and effectively record the network topology, and quickly create the raw data storage table of the Forward and backward substitution method needed. The algorithm is Fast calculation and reliable convergence.This article divided the issue of distribution line reactive planning into two sub-issues: (1) The selection of compensation node; (2) The best configuration of compensation capacity. The selection of compensation node has a crucial role in the distribution line reactive planning. Base on the principle of reactive power balance in each area, this paper defined a new moment "linear moment of impedance power", and presented a new method of searching compensation node by that moment. Firstly, the whole feeder is divided into several areas by using that moment, searching compensation node in each area. The method is base on the principle of reactive power balance in each area, can effectively avoid compensation node non-uniformity and range overlap. In the case of compensation location determining, the paper established a multi load level model for optimal reactive power planning to solve the best compensation capacity configuration, use improved genetic algorithm to solve the model. Initial population will directly affect the quality of optimization algorithms and speed, focusing on improving the initial population generation, modified the initial population generated strategy, which can orderly and dynamically determine the max compensation capacitor for each compensation node, and achieved a reasonable configuration for each compensation capacitor in different load levels. Reactive power optimization intelligence system of distribution lines is centralized implement of optimal planning and running. The system is mainly consists of upper computer system in scheduling room, reactive power compensation controller and communication. The system use technology of network and the GPRS to achieve data exchange between computer in scheduling room and each compensation controller (lower computer) in distribution lines. Lower computer collected operational parameters and transmitted it to the host computer in scheduling room, SCADA system also transmitted data of substation export to the host computer, these data are applied to calculate with the proposed optimal reactive power running strategy as raw data input, The host computer transmitted the switching instructions to lower computer according to the optimal results, achieved optimal reactive power running control of distribution lines. Software system, achieved flexible topology maintenance, real-time updating the displaying of compensation devices running, the query of historical data and the drawing curve, statistic of switching capacity and reactive power and so on.The paper proposed system control strategy of optimal reactive power running in distribution line based on improved tabu search algorithm. According to reactive power compensation’s influence on power flow and load characteristics, and characteristics of rural distribution line parallel capacitor optimal switching problem, improved some aspects of basic tabu search algorithm. Including binary code optimization, "Moving" starting point selection, the initial solution generation method, the selection of starting point for cycle, increased computing speed to meet the online operation, solved the real-time optimal switching of compensation capacitor in distribution lines. The paper compared the effects of compensation with the local control strategy through an example. The results show that system control strategy the paper proposed reduce network losses, improve voltage quality, and make reactive power optimized running.In the case of classification compensation of transformer and distribution line make the power factor of 10kV bus reached corresponding targets, substation compensation capacity should be the sum of main transformer power loss and difference considering standard of main transformer primary side. Put forward the automatic control mode in the secondary side of reactive power dynamic balance, based on the main transformer reactive power demand in the perspective of local reactive power balancing.In view of the proposed approaches in this paper, corresponding algorithms and programs are made, and those methods are tested on various example grids.The results shows the feasibility of the proposed methods.