Demand power forecasting with data mining method in smart grid

Nowadays increasing electricity demand is a key issue. As the demand is increasing day by day, obtaining energy efficiency is also getting important. Hence developing accurate demand forecasting methods is crucial for ensuring energy efficiency through efficient system operation. In this paper, we suggested a demand forecasting method with data mining techniques. We proposed a hybrid method which combined K-means clustering, Bayesian classification and ARIMA. Most of the previous research tried to solve this issue from supply side management but here in this paper the proposed forecasting model works on consumer side. Case study has been carried out with actual load profile from Jeju island, South Korea. The minimum error rate is 0.1853 from proposed Hybrid Model. The performance of the proposed model was also compared with the Neural Network based forecasting. The comparison shows better performance of proposed model compared to Neural Network.