Adaptive Fuzzy-Neural-Network Power Decoupling Strategy for Virtual Synchronous Generator in Micro-Grid
Yu Wang, Rong‐Jong Wai
Abstract
This article proposes an online-trained adaptive fuzzy-neural-network power decoupling (AFNNPD) strategy for a virtual synchronous generator (VSG) control in a micro-grid (MG). First, the mechanism of the power coupling for the VSG control in an MG is analyzed, and the system dynamic model for the proposed power decoupling method is derived. Then, a total sliding-mode control (TSMC) is designed for the power decoupling with the characteristics of the strong robustness and fast dynamic response. Moreover, an adaptive fuzzy-neural-network (AFNN) control is designed to mimic the TSMC law for relaxing the requirement of the detail system information in the TSMC. In addition, adaptive tuning laws for network parameters are derived according to the projection algorithm and the Lyapunov stability theorem for guaranteeing the network convergence as well as the totally power decoupling performance. Furthermore, experimental results are provided to verify the superiority of the proposed AFNNPD method in comparison with conventional methods.