An approach of electric power demand forecasting using data-mining method: a case study of application of data-mining technique to improve decision making

In this paper, the monthly electric demand prediction approach with dynamic and adaptive mechanism connected to the business environment is proposed with the aim of building the management plan of stable electric power supply and accomodation. The proposed prediction adopts the Kalman-Filter as the basic prediction scheme and possesses two characteristics stated below. One is the state-space built with the principal component time-series integrated with time-series PCA (Principal Component Method) from multi business indices related to the targeted time-series. The other is the self-organised auto-updating of the state-space by structured neural networks. The proposed scheme shows considerably more accurate prediction than any other models with single variable time-series and the obvious effect appears to be the high accuracy achieved by adopting time-series PCA as a Data-Mining technique. Given these results, the proposed prediction scheme might be considered to improve stable electric power supply and accommodation. This prediction scheme can be applied to various management areas, and so it might be considered to be an effective method for decision-making support.