A prediction method of residential load transformer demand factor based on time series. (September 2020)

• Record Type: Journal Article
• Authors: Sun, Ming; Jiang, Hang; Lu, Yangchao; Liu, Qing; Yang, Yong; Yu, Bing
• Is Part Of: Journal of physics. Volume 1633(2020)
• Publication Date: 2020-09
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1633/1/012161
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: The use of residential transformer in distribution network directly affects the assessment result of the feeder installed capacity. With the development of power system reform and the improvement of power supply economy, it is more and more important to evaluate the demand factor of these transformers scientifically. In this paper, a method based on time series is proposed to predict the demand factor of residential load transformer in distribution network. Firstly, the actual current data of distribution transformer is collected. After data pre-processing, the maximum value of monthly demand factor is calculated, and the historical data based on time series can be obtained. Then, the general development trend of demand factor is predicted through gray model. Finally, more accurate demand factor prediction value can be obtained by using neural network model training. The data of a distribution transformer in a capital city of China is selected to illustrate the practicability and effectiveness of the proposed algorithm. The proposed algorithm can fully consider the dynamic characteristics of the demand factor changing in time, and will greatly improve the accuracy of the feeders installed capacity assessment.