Research On Reactive Power Optimization Of Power System Based On Improved Genetic Algorithm

With the rapid economic growth of China and the development of industry, the powerquality requirements of various sectors of the country have become more strict than ever. Inthe power system, reactive power plays a special role. On one hand, reactive power createsthe necessary conditions in exchange, transportation, conversion for electrical energy, on theother hand, if the distribution of the reactive power and the reactive loads is unreasonable, itwill affect the economy and stability of the power system and lower the power quality.Therefore, in order to reduce network losses caused by unreasonable reactive powerdistribution and improve voltage quality, studying the power system reactive poweroptimization is of great significance.The reactive power optimization of the power system is essentially an optimizationproblem, it’s kind of multi-variable, often more than one objective, mathematical models ofcomplex, large-scale processing and the real-time requirements of the algorithm is also high.Especially in recent years, the scale of the power system is much larger and more complex.Because of its inherent limitations, the traditional algorithm has not well adapted to thecurrent large power system. In recent years, intelligent algorithms began to be used in thefield of reactive power optimization, where genetic algorithms applied more widely to otheralgorithms. In this thesis, an improved genetic algorithm on the basis of previous studies isproposed in order to further improve the speed and accuracy of their solvings.Firstly, the background and significance of reactive power optimization are described,the content and features are analyzed in order to propose a model for reactive poweroptimization. The model took into account the minimum net loss and the maintaining powersystem stability. Then, in order to improve the global convergence and speed of the normalgenetic algorithm the improved catastrophic genetic algorithm (ICGA) is proposed. Thealgorithm introduces catastrophic operator and has a dynamic range of disaster control tosolve the problem that conventional genetic algorithm easily falling into local optimization,but also greatly inhance the convergence speed. Finally, the test function verifies the validityof the improved algorithm.In this thesis, the improved genetic algorithm combined with power flow calculation isapplied to reactive power optimization and simulations of two standard nodes systemsrecommended by the Institude of Electrical and Electronics Engineers (IEEE) to verify theeffectiveness of the algorithm. The results show that ICGA has good performance inmaintaining population diversity and improve search efficiency and other aspects in reactivepower optimization of power system.