A Federated CNN Integrated Cyber Secure Smart Grid for Energy Efficient Robotic Machines

Abstract

In order to enhance the cyber-secure configuration of smart grids in industrial settings, this research offers a Federated Convolutional Neural Network (CNN)-integrated approach, which focuses at the junction of the latest technological advances. The main goal is to examine how to integrate a smart grid with strong cyber security features into an industrial distribution network. This entails the creation of a brand-new generation of robotic devices that are powered by DC and are positioned strategically to overcome the energy constraints of existing industrial machinery. The focus is on utilizing energy recovery, storage, and exchange at the industrial level to support environmentally friendly production methods. The network's real-time cyber-attack detection and mitigation capabilities are further enhanced with the incorporation of Federated CNN. A key element in increasing the smart grid's overall intelligence is the Federated CNN, a distributed deep learning model. This approach provides for successful threat detection without jeopardizing sensitive data by spreading the learning process among multiple nodes while preserving data privacy. The integration of the model ensures a flexible and dynamic reaction to new cyber threats, improving the industrial network's resilience. In the age of smart manufacturing, the results should have real-world applications for sectors looking at implementing energy-saving measures while protecting vital infrastructure from online attacks. With an accuracy of around 98.12 %, the recommended Fed-CNN—which has been implemented in Python software is 3.37 % more precise than other current approaches like Fed-ANN, Fed-LSTM, and Fed-GAN.