Research On Analysis And Forecasting Method Of Medium And Long Term Power Load Under The New Normal Of Economy

Analysis and prediction of medium and long term power load is a basic part of power market analysis or power grid planning.With China's economic development entering the new normal,analysizing and forecasting the medium and long term power load accuratly,and on the basis of this,balance the power and electricity is of great significance for optimizing the allocation of power resources and promoting the transformation and upgrading of the power industry.Although the traditional load analysizing and forecasting methods have been widely used,but there are still some insufficiencies.On the one hand,the national electricity consumption growth has been slowed down,showing obvious characteristics of S-type and uncertainty,which makes the methods applying to steady growth phase of load more limited;on the other hand,the historical data which the load forecasting in the new normal phase of economy can rely on,is mainly the data of the "12th Five-Year" period,and the number of samples is small,which brings great difficulties for modeling.Therefore,the medium and long term load analysis and prediction methods adapting to the new economic normal is studied and discussed in this paper.First of all,starting from the analysis of the connotation of new normal,the impact of the three macro performance of new economic normal including velocity change,structure optimization and driving force conversion to the power load is analyzed,then the four aspects of new characteristics of the development of China's power load under the new normal is summed up.Secondly,8 micro indicators are selected to reflect the impact of the macro performance of new economic normal to electric power load,and a hierarchical index system of power load influencing factors combining the macroscopic view with microscopic view is established.As traditional grey correlation analysis model cannot well consider historical data differences and future electricity development,a quantitative analysis model of power load influencing factors based on double weighting grey relational degree is proposed.On the one hand,according to the principle that the weight is big when near and small when far,the correlation coefficient of each period is fuzzy weighted;on the other hand,AHP is used to give the importance weight for each index,quantifying the judge of experts and operators for the future situation.In addition,a medium and long term load forecasting model based on partial least-square regression and scenario analysis method is proposed.The model of partial least squares regression which can be modeled under the condition of small sample size and multi-ple correlation of variables is used to obtain the relation equation between power load and its influencing factors under the new normal;and according to the characteristics of economy and power industry under the new normal,the scenario analysis method is used to set up several scenarios and corresponding parameters,and then get the different load forecasting results under different scenarios,in order to reduce the forecasting risk.Finally,the proposed models are applied to the analysis and prediction of Zhejiang power grid,and good results are achieved,which verifies the validity of the models.