Research On Heuristic Hybrid Intelligent Algorithm Of Reactive Power Optimization In Transmission Network

Reactive power optimization which plays an important role in realizing system's safe and economic operation has not been solved completely because of the complexity of mixed-integer nonlinear programming with many variables and many constraints. Voltage/reactive power adjustment and static reactive power optimization are fully studied based on the basic principles of voltage/reactive power control in actual power systems and immune evolutionary programming algorithm. The contents presented in the paper are as follows:Firstly, an effective heuristic adjustment algorithm using hierarchical and regional method for voltage and reactive power in the transmission network is proposed. Based on the principles of hierarchical and regional balance and local compensation of reactive power as well as inverse regulation of pilot node voltage, new conceptions of layer-area, load rate, reactive power adjustment ability and reactive power unbalance level are defined to ensure the ideal aim of reactive power balance and voltage inverse regulation, and evaluate the reactive power balance level of the network. Then the strategy of hierarchical and regional adjustment of voltage and reactive power is presented, it includes 3 stages:â‘ global adjustment of voltage and reactive power;â‘¡hierarchical and regional adjustment of reactive power;â‘¢partial adjustment of voltage and reactive power. After analyzing the character of power flow in each stage, the reasonable heuristic regulations of voltage/reactive power adjustment are made concerning the major problems, aiming to make voltage qualified and reactive power balanced in hierarchy and region as much as possible.Secondly, a new heuristic hybrid intelligent algorithm of reactive power optimization in the transmission network is proposed through the voltage/reactive power heuristic adjustment regulations and the immune evolutionary programming algorithm. The initial group can be made using the proposed heuristic method, which makes the majority of individuals feasible, and compresses the search space of the algorithm effectively. Then, the control variables'mutation directions of infeasible individuals in the immune evolutionary programming can be controlled using the proposed heuristic method, after analyzing the global or partial reasons of infeasible individuals, the control variables need to be adjusted are determined and their mutation values are made, which can improve the mutation efficiency of infeasible individuals, avoid lots of invalid searches, conquer the phenomena of inefficient search and early maturity in the immune evolutionary programming. The algorithm has the advantages of higher research efficiency, better stability and more perfect global optimization ability compared to the common immune evolutionary programming.Finally, the validity of the two proposed algorithms in the paper is verified by the test on a 43-bus simulation system and Chongqing high voltage network.