Cascade Failure Management in Distributed Smart Grid using Multi-agent Control

The interdependent network of centralized power systems makes the system more vulnerable to failures. The existing cyber-physical architecture of power systems possesses single-point contingency i.e. by compromising the master node, the slave nodes will not be capable to operate. Prior research studies primarily focused on the network design for the detection and mitigation of cascade failure. Nevertheless, in practical scenarios, detection and restoration are not considered through the same innovative approach. In this paper, we proposed a distributed cyber-physical smart grid for the detection and restoration of cascade failures. We considered a multi-agent control model for physical and cyber faults. We address distributed coordination control among generation and distribution agents for the management of cascade failure. The cascade failure in distributed smart grid is managed in two stages: detection and restoration. In the first stage, the cyber-physical agent triggered an alert upon detection of fault and an emergency state is initiated. In the second stage, the faulty agent gets isolated from the network and an alternate agent is integrated for the restoration of the power network. The simulations are performed on IEEE 9 bus network in MATLAB and JADE for the cascade failure management. The effectiveness of multi-agent control fault management is validated analytically and experimentally for distributed control smart grids.