Power System Supply And Distribution Energy-saving-Study On Reactive Power Optimization

Power system reactive power optimization is an important part of study in powersaving and energy reducing, a way to improve power quality, basis of power systemoperation, analysis and control, solve the problem of reactive power optimization willbring significant improvements in power system operation. This thesis is based on thesaving of energy, make a special research on reactive power optimization, using thegenetic algorithm, and make changes according to the characteristics of power systemand the actual situation, then make simulation on Matlab platform through the IEEEstandard bus system to improve the superiority of genetic algorithm.Research on practical application on the genetic algorithm in reactive poweroptimization problem, especially for the problem in simple genetic algorithm such asconverging slowly, stop early, local optimization, and combined with thecharacteristics of reactive power optimization’s mathematical model, make somechanges on the genetic algorithm. Including the use of real integer hybrid encodingmode, handling discrete variables precisely, and compared to the single binaryencoding mode, which can improve the speed of computation effectively; refers to thefitness of the tension, not only can effectively avoid the premature local optimum, butalso improves the speed of the evolutionary; the selection operator use the tournamentmethod and optimal preservation combining, laid a better foundation for the evolution,application of improved adaptive genetic algorithm make a better performance ofgenetic algorithm; in the termination criterion, I set two fixed value which are themaximum genetic algebra and retain at least algebra, ensures that the algorithm hasbetter accuracy.Reactive power optimization is a multi variable, multi-objective, nonlinear complexsystem optimization process, this paper selects the minimum active power loss andapplication of penalty function of voltage limits as target function, which is the basisof this calculation; power flow calculation is the most basic analysis and calculationof power system reactive power optimization problem, power flow calculationmainly calculates the active power loss and if the relevant state variable meets therequirements of power optimization, in order to get the final objective function tomeet the constraints of the value; after compare the advantages and disadvantages ofthree kinds of classic power flow calculation method, I select the P-Q decompositionmethod as the power flow calculation method, ensure power optimization operation agood foundation, can also make genetic algorithm have better applications. In order to verify the effect of the improved genetic algorithm to solve the reactivepower optimization, using the Matlab software and the IEEE standard test system tomake simulation to compare the simple genetic algorithm and improved geneticalgorithm, this paper selects the IEEE-30node system, the results showed that: theconvergence performance and effectiveness of the improved genetic algorithm hasobvious advantages, is more suitable for solving the problem of reactive poweroptimization in power system, and also has certain practical value.