In this paper

an improved prediction method based on time segmentation is proposed to increase the electricity accuracy of prediction. The 12 months of the year are broken down into known and unknown periods of time.on the basis of which the electricity quantity of the latter is predicted

and the predicted results are revised by combining the known electricity quantity. To verify the feasibility

a practical case is studied using the proposed method and other common methods. The variance analysis of the prediction results indicate that the time segmentation based prediction method significantly improve the rationality and the reliability of the prediction result.