USING SVM ALGORITHM TO CLASSIFY ELECTRICAL LOADS

Abstract

With the development of Industry 4.0, almost all areas of life in general are applying smart techniques, IoT systems, and artificial intelligence with the goal of increasing efficiency and productivity at work. manage, monitor, increase economic efficiency, etc. With the electricity industry also focusing on the smart use and transmission of electricity, one of the smart ones is the self-categorization of loads used in daily life. In addition to load classification, there are information customers, information users, manufacturers, and suppliers. Power supply can rely on load classification of possible areas to predict service users. From these projects, manufacturers have more optimal solutions for suitable phase balance in the process. power supply for each area. In this paper, the author presents the results of performing deep learning dependent classification for electrical categories, specifically using the Support Vector Machine (SVM) method to perform subcategory loading. The results obtained are higher than the method distribution system. The accuracy ratio of the method is greater than 0.93.