| The benefit evaluation of distribution network planning is based on the whole power grid,considering the reliability and equipment cost recovery problems,which is of great significance for improving the efficiency of grid asset management and obtaining long-term economic benefits.Providing good technical support for the development of electric power enterprises is conducive to scientific decision-making,ensuring the best allocation of resources in the limited grid construction and achieving the maximization of social benefits and business benefits.This thesis studies two data processing methods based on support vector machine(PSO)particle swarm optimization for distribution network load forecasting and variable population size genetic algorithm.Some basic theoretical probl ems of adaptive genetic algorithm which are suitable for distribution network are studied.The theoretical research and improvement of the algorithm are mainly studied on the time complexity of the algorithm and the convergence speed of the algorithm.An improved genetic algorithm is presented.The experimental results show that compared with the basic genetic algorithm,the improved genetic algorithm presented in this thesis has a smoother population convergence rate and a more reasonable initial population selection.The individuals with good gene fragments in the population are very good,The premature convergence of the algorithm is effectively avoided.When the population with high fitness in the population gets into difficulties in many iterations,the population can continue to evolve.After the algorithm is applied to the distribution network case,the voltage quality of the distribution network is obviously improved and the network loss rate is effectively reduced,which shows that the improved genetic algorithm can be applied in the distribution network planning.The algorithm has good evolvement ability,Benefit assessment is of great significance.This thesis constructs the index system of distribution planning benefit evaluation.Gives the basic principles of constructing the comprehensive benefit evaluation index system of distribution network planning.This thesis studies subjective weight hierarchy analysis method and objective weight entropy method weight assignment method,and uses subjective and objective combination weighting method to calculate the weighting index system of planning benefit evaluation of distribution network.Studied the matter element theory,cloud element theory two evaluation methods,using cloud element method to assess the distribution network planning benefits.It provides a feasible reference for the benefit evaluation of distribution network planning.This thesis builds a distribution network planning benefit evaluation experiment platform.Data cleaning was conducted on the constructed platform to test the efficiency of different types of data on the same algorithm.The differences between the two algorithms were analyzed.The same data was analyzed using different platforms to study the differences in processing speed.The comprehensive evaluation model of power grid planning benefit evaluation and the evaluation index system of power distribution network established in this thesis are of great significance to properly grasp the investment scale and timing,promote the scientific development of power grid and maximize the benefit of power grid. |
