Data-Centric Decentralised Controller for Effective P-F and Q-V Control in AC Microgrids

Abstract

—Accurate sharing of active and reactive powers in AC microgrids is one of the challenging problems. Implemen tation of Artificial Intelligence (AI) techniques seems to be a promising solution to enhance the control and operation of microgrids. This paper presents a novel method of data-centric AI-based decentralized frequency (f) and voltage (V ) controller while sharing the proportional active and reactive power among the distributed generation (DG) units in the microgrids. In the proposed decentralized controller a Multi-Output Regressor based AI model is used for faster control action with accuracy. Once the controller model is trained, validated and tested on the microgrid data set for benchmark accuracy, it can be effectively used in real-time operation. The effectiveness of the proposed controller has been demonstrated on a Voltage Source Converter (VSC) based microgrid and compared with the traditional droop controller under various loading conditions.