Smart Grid Demand Forecasting Based on Data Mining
With the increasing demand for electricity,obtaining energy efficiency becomes more and more important.Therefore,the development of accurate demand forecasting methods is very important to ensure energy efficiency through effective system operation.A demand forecasting method based on data mining technology is proposed.This method combines k-means cluste-ring,Bayesian classification and ARIMA.Most previous studies have tried to solve this problem from the perspective of supply side management,but the prediction models proposed are suitable for the consumer side.In the experiment,the actual load profile of a city is studied.The results show that the minimum error rate of the proposed method is nearly 0.3,which can meet the actual needs.
万方数据
维普
