Research On Optimization Of Substation Resources

China is in a period of rapid development,and the load of power consumption has increased rapidly.The original power system is not able to meet the power supply requirements.The power grid needs a planning,adding or expanding some substations.The selection of the distribution of substation sites,the lay out of power supply area and the distribution of substation outlet intervals are of important influence and significance for future reliability,economy and safety of the grid.Site selection of the substation is closely related to the actual situation,and there are many factors involved.Considering the economical factor of the station can not meet the requirements.In this paper,a site selection method based on improved genetic algorithm is proposed in consideration of the economic and geographical environment of the substation.First based on the economical factor,this method applies an improved genetic algorith:m to select the initial location of the substation.Adaptive adjustment parameters are introduced into the improved genetic algorithm,which can adjust the crossover rate and mutation rate parameters consciously,and increase the diversity of search.It makes up for the slow convergence rate and early maturity of genetic algorithm,and has better convergence performance.After determining the candidate of location,this method considers the geographical environment around substation,uses AHP to re-locate the substation.Finally,the economical factor and the geographical environment factor are unified and optimized.The calculation results show that the proposed method can meet the requirements of substation planning.A reasonable lay-out of power supply area is an important guarantee for the safety,reliability and economy of power supply.The division of substation power supply area is very similar to the composition of V diagram.The characteristics of eff-ective range,hollow circle,linear characterization and local dynamic indicate that V graph model can be used to solve the optimization problem of substation power supply area division.The characteristics of weighted V graph can be applied to real problems of substation planning,which is,the use of different weight values to reflect the influence of uneven load distribution,different capacity of substations and different load rates on the substation power supply range division.The general model of weighted V graph cannot consider the load rate and power supply radius constraint problems of substation in the calculation at the same time.So the derivation according to the physical meaning of the weighted V graph derived a new formula to calculate the weight;and then put forward the definition of variable weights and calculation formula is deduced,finally combined with the improved weighted V graph and alternative location algorithm to local search the power supply region.According to the characteristics of dynamic adjustment of variable weight,the algorithm of weighted V graph which is improved by cellular automata is proposed to improve the calculation speed.The optimization problem of substation outgoing interval is a multi-objective,multi constraint and nonlinear optimization problem.Considering the multiple evaluation indexes such as load rate,alternate interval number and non public interval number,and considering the future load change of power supply area,a mathematical model of substation outlet interval optimization is established.SPSS clustering analysis is used to prioritize,and Hopfield neural network is constructed based on Lagrange multiplier.By introducing the global memory of PSO,it can avoid falling into the local minimum solution in the process of solving.