Regression-based Self-Triggered Control over Wireless Channel for Distributed Economic Dispatch Problem

Abstract

<p>An essential building block of the smart grid of today is the integration of various renewable power generators in the system. Renewable distributed generators (DG) provide environmentally friendly energy at a low cost. A distributed economic dispatch (ED) system for renewable DGs has been formulated in this work to enhance scalability and reduce single points of failure as compared to the centralized ED system. A self-triggered consensus control mechanism is devised over a wireless channel to reduce the communication cost due to communication overhead in distributed systems. Communication losses due to wireless communication between local DGs have been addressed in this work, and a regression-based estimation has been proposed to account for lost or corrupted packets. The proposed approach is compared with the existing centralized ED system and also with periodic communication. The proposed self-triggered communication approach achieves the same level of performance as periodic communication while using half the resources. </p>