A data mining approach for spatial modeling in small area loadforecast

摘要

In a competitive power market, locations of future load growthnhave to be described with sufficient geographic precision to permitnvalid marketing strategy and siting of future T&D equipment. Smallnarea load forecast which provides information of future electric demandnthat includes spatial and temporal characteristics, is useful fornT&D and market planning. Domain experts for spatial load forecastnrequire long term practicing and are difficult to find. In order toncapture the meaningful associations between spatial data and the loadnchanges, and to provide a useful tool for spatial load forecast, a datanmining. technique based on a "Knowledge Discovery in Database (KDD)"nprocedure is proposed to determine automatically the preferentialn"scores" of land use changes. The proposed spatial modeling approach isnan exploratory data analysis, trying to discover useful patterns innspatial data that are not obvious to the data user and are useful in thenspatial load forecast