Research on cross-region interconnection detection method of abnormal data in power monitoring system. Issue 1 (1st June 2022)

• Record Type: Journal Article
• Authors: Bin, Dongmei; Ling, Ying; Yu, Tong; Yang, Chunyan; Liu, Muxian; Chen, Haiguang
• Is Part Of: Journal of physics. Volume 2290:Issue 1(2022)
• Publication Date: 2022-06-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/2290/1/012028
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: When conventional methods collect cross region interconnection detection data of power monitoring system, one data has multiple signal strengths, resulting in low detection rate and high false alarm rate of cross region interconnection abnormal data. Therefore, a cross region interconnection detection method of abnormal data in power monitoring system is proposed. Divide the cross region interconnection scene types of power monitoring system, make the detection end send data packets to the security partition corresponding to different scenes, and Gaussian filter the signals received in the security partition, and only retain the signals with probability distribution density function of [0.5, 1]; Convert multiple signal strength indications of MAC address into one, so that the security partition only responds to the MAC address, obtains the MAC address detection data, and inputs it into the deep confidence network after preprocessing. If the MAC address matches successfully, it is judged that there is cross region interconnection in the power monitoring system, and the cross region interconnection detection of abnormal data is completed. The cross region interconnection alarm data provided by a power Co., Ltd. is selected as the experimental data, and the comparative experiment is set. The results show that the design method improves the abnormal data detection rate, reduces the false alarm rate, and ensures the detection accuracy of large-scale cross region interconnection data.