Research On Multi Objective Group Decision Making For Load Forecasting And Distribution Network Planning In ZW District

Urban distribution network power system as an important part of,for the rational integration of various resources,power quality guarantee and promotion of an for the economic development of the region has a pivotal role,so rational,science,improve the distribution network planning is particularly important.With the rapid development of economy,the construction of the original distribution network can not keep up with the pace of development,to a certain extent even restricts the development of power system,thereby affecting the economic prosperity and development,in order to meet the demand of power load to the ZW city economic development and resource integration,the District Planning and the reasonable transformation of the distribution network,the network planning scheme for scientific and effective decision-making is necessary,to verify the effectiveness of the program on the one hand,on the other hand can be found through the decision of distribution network planning project,provide important reference basis for other distribution network planning project.Power load forecasting is the foundation and core of power supply planning,and the power load forecasting is the basic work of urban power network planning,which plays a key role in the quality of urban power network planning.Because the power load is influenced by random factors of uncertainty,there are characteristics of non-stationary,nonlinear and time-varying data series in power load,this paper changed the traditional power load forecasting method of spatial load,the intelligent algorithm is applied to the actual power grid planning and load forecasting,through the collection the empirical mode decomposition(Ensemble Empirical Mode Decomposition,EEMD)data of power load is decomposed into a series of independent intrinsic mode functions and a residual function,and the genetic algorithm(Genetic Algorithm GA),a series of function decomposition after certain weights were given according to the sequence features of each function respectively.By using the method of least squares support vector machine(Least Square Support Vector Machine,LSSVM)and non parametric generalized autoregressive conditional NPGARCH heteroskedasticity(non-parametric generalized resorting to the conditional heteroskedasticity)was predicted.The predicted results on cumulative,target of the electric power load forecasting results.The example shows that the method can effectively improve the accuracy of electric power load forecasting,and provide a reliable foundation for the scientific and security of distribution network planning.Secondly,about ZW city distribution network profiles were analyzed,from the distribution network planning objectives,principles and thefocus of the studies the problems existing in the voltage level of distribution network in the city,according to the problems,analysis in distribution network planning relates to the interests of the main body of interests and needs,so as to construct a multi-objective group decision making evaluation index,based on entropy weight theory of distribution network planning multi-objective group decision making model is proposed.The model is applied to ZW city distribution network planning,the results show that the city's distribution network planning program to meet the overall interests of the parties,so the plan is more reasonable and scientific.