Prcatical Research Of The Reactive Power Optimization In Changxing Power System

In this thesis, the significance of reactive power compensation, role and compensation are introduced and its importance in modern power system is illustrated. And to Changxing Power system of the reactive power optimization case study object, according to Changxing Voltage-Reactive Power Control of system performance, considering the regional balance of power, real-time reactive security constraints and voltage requirements of the system objective function of minimizing loss and constraints on the state variable of the way by using penalty function treatment, establishment of mathematical model for optimal operation of power, ultimately the regional grid of dynamic reactive power optimization control.In this thesis, the basic principles of genetic algorithms, implementation method are described. Learning the characteristics of artificial immune system, it will be introduced into the genetic algorithm, the immune genetic algorithm optimization applied in reactive power optimization calculation. In accordance with Changxing Power Grid, the genetic algorithm and immune genetic algorithm have been analyzed in the thesis.Reactive power optimization problem is a typical nonlinear programming problems. It is nonlinear, discontinuous, uncertainties more and so on. Genetic algorithm uses probabilistic search technique, the objective function is not continuously differentiable demand. Applied to reactive power optimization problem has its distinctive advantages. Immune genetic algorithm reserves the characteristics of stochastic global parallel search from genetic algorithm, enhanced local search capabilities. It has good results in the node voltage control, reducing loss and obtaining the maximum overall efficiency, and more fast calculating speed.Reactive power optimization is a very important means in the power system security and economic operation. This thesis proposes genetic algorithm in reactive power optimization as a new priority net, that solves the problem of nonlinear programming and linear programming method can not deal with discrete variable problem. It opens up broad prospects for the practical-application of reactive power optimization. The Changxing power system is used to evaluate the proposed optimization algorithm.From the comparison immune genetic algorithm with genetic algorithm,the results show that the immune genetic algorithm has better capacity of the global convergence. It can break away from local optimal solution effectively, search the best solution and have better stability. Meanwhile this method also has faster convergence speed.