Study on the Identification of the Differential Lasso in the Low-voltage Transformer Area. (June 2020)

• Record Type: Journal Article
• Authors: Wang, Yi; Liu, Zhitong; Li, Yuanyi; Pang, Bo; Zhao, Nan; Liu, Jia; Shi, Diancai; Hou, Gang; Wang, Qi; Zhao, Chuanzong; Wang, Qinghao
• Is Part Of: Journal of physics. Volume 1549:Number 5(2020)
• Publication Date: 2020-06
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1549/5/052063
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: The validity of the transformer area's file in the Power information acquisition system is very important, the man-made way is time-consuming, and the cost of relying on the equipment is high. In order to provide efficient, convenient and economical identification technology of the transformer area archives, this paper proposes a low voltage topology recognition method based on differential Lasso, which can complete automatic topology recognition only by using the power consumption information of the transformer area. First of all, the statistical model of Lasso is established for the data of electricity consumption. According to the characteristics of the meter in the meter box belonging to the same area, the differential Lasso model is further constructed. The regression coefficient based on the differential lasso can be calculated for each transformer area in turn by polling, and select the meter according to the threshold. The method of this paper is used to analyze the field collected electricity data. The results of topology recognition show the effectiveness of the proposed method.