Automation industry in cyber physical systems based on 4G IoT networks for operation controlling and monitoring management

Abstract

Cyber-Physical System (CPS) is an emerging technique that focuses on the integration of computation applications as a network of communicating physical and cyber components which controls real time physical substructures of the industrial automation. Since implementation, operation and design of CPS and management of automation substructures are plays major significance in the various industrial applications. This study presents a Deep learning (DL) based intrusion controlling and monitoring management system (DL-ICMMS) for CPS in the automation industry for the detection of the occurrence of intrusions. To transform the input data in a compatible manner, data normalization is carried out and the Adam optimizer is applied to handle hyperparameter values of Restricted Boltzmann Machine (RBM) method. Experimental results analysis stated that the DL-ICMMS method increased accuy of 98.80% whereas the Multilayer Perceptron (MLP), K Nearest Neighbor (KNN), Deep Belief Network (DBN), Convolutional Neural Network (CNN), and Long short-term memory (LSTM) approaches have shown reduced accuy of 93.10%, 93.62%, 91.60%, 90.88%, and 91.75%, correspondingly.