Author : Prof. Kiruthika K, B Jnanesh, Vikram J R, Rakshita Hanamant Mathad, Shaik Dilshad
Date of Publication : July 2026
Abstract: Smart grids use modern communication, control, and computing technologies to make power systems more efficient, reliable, and sustainable. However, the increased use of connected devices and communication networks also makes smart grids vulnerable to different cyberattacks, including Denial of Service (DoS), data injection, scanning, rogue node, switching, connection loss, and double command attacks. Such attacks can affect grid operation, cause power interruptions, and create financial and safety concerns. This work presents a machine learning-based system for detecting cyberattacks in smart grids using MATLAB and Raspberry Pi 3B+. A publicly available smart grid cybersecurity dataset containing different attack categories is used to train a Bi-lateral Convolutional Neural Network (CNN)-based Artificial Neural Network (ANN) model in MATLAB. The trained model is then deployed on the Raspberry Pi 3B+ for real-time attack detection. A MATLAB App Designer-based GUI is used to provide input data and display the detected attack, while a 16×2 LCD provides a local indication. The proposed system offers a simple, low-cost, portable, and practical approach for real-time cybersecurity monitoring in smart grid environments.
Reference :