Deep-Block Network for Cyberattack Mitigation and Assessment in Smart Grid Power System with Resilience Indices

Abstract

Distributed renewable energy sources, with wide communication components in a microgrid infrastructure, make cyber security assessment and mitigation a developing cyber-physical system study. Extensive cybersecurity threads are prevailing in modernized smart grid. Hence, to detect and mitigate cyber threads an advanced cost-effective resilience cyber risk assessment and mitigation mechanism is needed. To enhance cyber-physical security in smart grids, a secured deep learning algorithm with blockchain technology (BlockDeepNet) is proposed. Distributed secured data analysis is carried by using deep learning approach, while blockchain helps in the implementation of secured decentralized resilient control. To validate, real-time cosimulation on IEEE 15 bus system was conducted. Also, for evaluating cyber security breach, four types of cyberattacks were introduced to validate the effectiveness of proposed security assessment and resilience operation. We obtained normalized resilience index of 2.36 for grid communication failure, 0.91 for replay attack, 1.34 for false data injection, and 1.74 for DoS attack. The obtained results on simulation case study by real-time hardware in the loop implementation showed that the proposed BlockDeepNet accurately reduce load loss for various cyberattack and provide robust resiliency. Overall, this research provides a platform for cybersecurity assessment and enhanced resilience operation of cyber-physical power energy system.