Application in Improved Genetic Algorithms for Optimization of Reactive Power

The genetic algorithm is a self-adapting probabilistic iterated search method, which is based on a principle of the natural choice and the natural genetic mechanisms. And it can simulate the development law of biological evolution in the natural world and can be used in the complex nonlinear optimization problems in continuous variables and discrete variables mixed.This paper uses the genetic algorithm in the reactive power optimization and improves the basic genetic algorithm. The linear scale transformation method is used in the fitness function. The championship method is used in the choice. The cross-rate and mutation rate with index changes are used in the operation and second variation is used in it after evolutionary to some algebraic. The genetic algorithms can jump out of the local optimal solution with the above methods in the optimization process, enhances the global optimization capability, improves the accuracy and retains the advantages of the basic genetic algorithm.