Active Power Coordination for a Large Population of Grid-Forming and Grid-Following Inverters Based on Mean Field Games Theory

Abstract

In a grid with a high proportion of renewable energy, the coordinated control between grid-forming (GFM) and grid-following (GFL) inverters becomes crucial for maintaining aspects such as frequency and voltage regulation. This paper presents a decentralized competitive power coordination scheme, designed for a large-scale networked inverter system based on the Mean Field Games (MFG) concept. This scheme effectively harmonizes GFM and GFL inverters to achieve both frequency recovery and active power distribution within the distribution network. Each inverter devises its optimal active power control strategy by maximizing a frequency-constrained revenue function. Through the market price, each inverter's control strategy influences others' strategies. Importantly, this scheme simplifies the interaction among inverters by averaging the impacts generated by each inverter's control strategies. For the MFG control scheme, a decentralized policy iteration algorithm is proposed, eliminating the need for information exchange between inverters and significantly decreasing the communication burden. The effectiveness of the proposed control framework is corroborated using the IEEE 33 test system.