Enhanced Frequency Response Based on Multiagent Distributed Power Agreement

In this paper, we propose a multiagent based distributed algorithm for enhanced frequency response in large-scale multiarea power systems. Control nodes associated with these areas preemptively agree on their relative contributions in supporting the frequency response in a completely distributed manner. Consequently, in case of system contingencies, each area knows the total change in generation required to facilitate faster frequency restoration during primary frequency control. Within an area, local controllers associated with generators execute a distributed power agreement via sparse information exchange to achieve this required change in generation. As a result of these coordinated control actions, frequency deviation from the nominal remains minimal, thereby reducing the dependency on secondary frequency control. Inspired from graph theoretic techniques in networked controls, the proposed architecture requires nodes with minimalistic computation and communication capabilities, thereby increasing its applicability. Simulations performed on a large test case demonstrate the efficacy of the proposed algorithm in improving the transient performance of the frequency response.