Short-term electricity consumption forecasting and correcting method for residential users based on fuzzy clustering classification and Elman neural network algorithm
Electricity consumption forecasting is an important issue in smart grid construction,being accurate means great reference value to power grid planning and economic sector management decision-making.In this paper, the short-term electricity consumption prediction and modification method based on fuzzy clustering and Elman neural network algorithm has been proposed, residential users’data collection can be accomplished by automatic metering system every 15 minutes.First, the approach classified the users according to the usage behavior by fuzzy clustering; next calculated the weight of each type of influencing factors with the path coefficient; then utilized the weighted factors and historical electricity consumption as the training samples of the Elman neural network; and finally, the modified algorithm is applied to get the optimized result.The analysis proved that after classification it’s effective and feasible with the obviously increased accuracy compared with the overall prediction, in addition, the modified algorithm further promoted the predictive accuracy.