Smart Grid Security Framework for Data Transmissions with Adaptive Practices Using Machine Learning Algorithm

Abstract

This research presents an analysis of smart grid units to enhance connected units’ security during data transmissions. The major advantage of the proposed method is that the system model encompasses multiple aspects such as network flow monitoring, data expansion, control association, throughput, and losses. In addition, all the above-mentioned aspects are carried out with neural networks and adaptive optimizations to enhance the operation of smart grid networks. Moreover, the quantitative analysis of the optimization algorithm is discussed concerning two case studies, thereby achieving early convergence at reduced complexities. The suggested method ensures that each communication unit has its own distinct channels, maximizing the possibility of accurate measurements. This results in the provision of only the original data values, hence enhancing security. Both power and line values are individually observed to establish control in smart grid-connected channels, even in the presence of adaptive settings. A comparison analysis is conducted to showcase the results, using simulation studies involving four scenarios and two case studies. The proposed method exhibits reduced complexity, resulting in a throughput gain of over 90%.