Research on Saturated Spatial Power Load Forecasting Based on Land Utility

Abstract

With the increasing demand for electricity, accurate prediction of power load is of great significance for improving the quality of power grid planning and construction. Traditional saturated load forecasting method is greatly affected by historical information. This paper proposed an improved spatial load forecasting(SLF) method based on error model transformation. Considering the nature and development time of urban land, each district is divided into different blocks, the blocks are classified into two categories: homogeneous blocks and simultaneous blocks. An iterative load forecasting system architecture is proposed, which transforms blocks load forecasting problem into two sub-problems: multi-level gridding parameter training and model integrative prediction. Land utility and historical data are both investigated during the load forecasting procedure. Simulation result indicates that accuracy of calculated prediction result is higher and less affected by the noise.