| The load forecasting of power system has been an important link for the safety andeconomical operation of power system, especially in the short-term load forecasting. Itsprecision will influence the economic and secure operation of power system and quality ofpower supply. The features of short-term load forecasting can be generalized asfollowings: many data need to be forecasted, factors which influence forecast arecomplicated and random,and high precision of forecast is demanded.At present China is in the stage of economic development. The load forecasting ofPower demand faces many difficulties with great uncertainties due to various complexfactors. The traditional load forecasting model based on fixed structure and parameters canno longer meet new load forecast requirements. So load forecast studies are graduallyfocused onto how to consider various uncertainties and achieve intelligent forecasting. Thesuperiority of Intelligent Calculation in the current short-term load forecasting applicationhas been very good performance.This paper conducted a theoretical study of power system load forecasting andanalysed various types of factors affecting the load and their internal relations. Based onthe theoretical analysis of traditional single forecasting models, this paper proposed to useintelligent calculation method and establish a comprehensive optimal forecasting model.On the basis of theoretical studying, this paper made a variety of applications forintelligent calculation in the process of load forecasting. And it managed load of Suzhoucity, using single forecasting models and the intelligent combination forecasting model formaking daily load forecasting. By comparing with the results of forecasting, this paperproved the superiority of intelligent calculation combination forecasting model. The resultof data analysis indicate that the prediction accuracy, precision and overall stability havebeen improved. |