A Machine-Learning-Based Cyber Attack Detection Model for Wireless Sensor Networks in Microgrids

Abstract

In this article, an accurate secured framework to detect and stop data integrity attacks in wireless sensor networks in microgrids is proposed. An intelligent anomaly detection method based on prediction intervals (PIs) is introduced to distinguish malicious attacks with different severities during a secured operation. The proposed anomaly detection method is constructed based on the lower and upper bound estimation method to provide optimal feasible PIs over the smart meter readings at electric consumers. It also makes use of the combinatorial concept of PIs to solve the instability issues arising from the neural networks. Due to the high complexity and oscillatory nature of the electric consumers' data, a new modified optimization algorithm based on symbiotic organisms search is developed to adjust the NN parameters. The high accuracy and satisfying performance of the proposed model are assessed on the practical data of a residential microgrid.