Reasearch Of Improved Genetic Algorithm In Reactive Power Optimization Of Power System

Reactive power optimization in power systems is a mixed nonlinear optimization problem with a large number of variables and constrains, the operating variables include continuous and discrete variables, so the optimization becomes very complex. Recent researches and developments of reactive power optimizations are introduced. And then a mathematic model of reactive power optimization, regarding the minimization of active power losses as objective function, is established.The simple genetic algorithm is constrained by its poor converging performance,readily leads to local optimization, its computing speed is slow , in view of the deficiency of simple genetic algorithm, this paper make some improvement ,in order to reactive power optimization in power system to be better applied to obtain a better optimization of performance, to reduce the active power loss and improve voltage quality . Main is a continuous variable binary coding, discrete variables used in the encoding of real numbers to form hybrid coding used in this way, this method not only can overcome the binary-coded discrete variables when dealing with errors, but also can be overcome Binary code when dealing with lack of discrete variables, and resolve to understand the accuracy no longer rely on the length of encoded strings, retaining the real-coded solution, high precision, easy to search the advantages of a larger space. The fitness function value of the ratio method and the Combination of the best preserved of the hybrid selection strategy, this can prevent the group's best individual has been eliminated and the prevention of the worst individual is selected. In this paper, improved crossover and mutation are used in adaptive genetic crossover and mutation operators, which can speed up the convergence rate of genetic algorithms to effectively improve the ability of genetic algorithm optimization This article also uses the best individual at least to retain the greatest genetic algebra and algebra, a combination of criteria for the termination of evolution, so that in the global scope can be guaranteed to find the optimal solution, to avoid falling into local optimum. Improve the genetic algorithm applied to the power system reactive power optimization algorithm not only can improve the convergence rate, but also can improve the algorithm accuracy. he improved algorithm in this paper is applied to the IEEE-30 bus system, and compares the result of simple genetic algorithm, the results verify that the proposed algorithm advance the speed of calculate and improve the astringency of algorithm.