Smart Grid Monitoring by Wireless Sensors Using Binary Logistic Regression

Energies

Ramya Kuppusamy¹  HamidReza Baghaee²  Hariprasath Manoharan³  Yuvaraja Teekaraman⁴  Irina Kirpichnikova⁴  Srete Nikolovski⁵ 

[1] Department of Electrical & Electronics Engineering, Sri Sairam College of Engineering, Bangalore 562106, India;Department of Electrical Engineering, Amirkabir University of Technology, Tehran 15875–4413, Iran;Department of Electronics and Communication Engineering, Audisankara College of Engineering and Technology, Gudur 524 101, India;Faculty of Energy and Power Engineering, South Ural State University, Chelyabinsk 454 080, Russia;Power Engineering Department, Faculty of Electrical Engineering, Computer Science and Information Technology, University of Osijek, 31000 Osijek, Croatia;

关键词: smart grids (intelligent networks);    phasor machine learning;    binary logistic regression;    wireless network;    Sensors;   

DOI  :  10.3390/en13153974

来源: DOAJ

【 摘 要 】

This article focuses on addressing the data aggregation faults caused by the Phasor Measuring Unit (PMU) by installing Wireless Sensor Networks (WSN) in the grid. All data that is monitored by PMU should be sent to the base station for further action. But the data that is sent from PMU does not reach the main server properly in many situations. To avoid this situation, a sensor-based technology has been introduced in the proposed method for sensing the values that are monitored by PMU. Also, the basic parameters that are necessary for determining optimal solutions like energy consumption, distance and cost have been calculated for wireless sensors, whereas, for PMU optimal placements with cost analysis have been restrained. For analyzing and improving the accuracy of the proposed method, an effective Binary Logistic Regression (BLR) algorithm has been integrated with an objective function. The sensor will report all measured PMU values to an Online Monitoring System (OMS). To examine the effectiveness of the proposed method, the examined values are visualized in MATLAB and results prove that the proposed method using BLR is more effective than existing methods in terms of all parametric values and the much improved results have been obtained at a rate of 81.2%.

【 授权许可】

Unknown