Electric Power ›› 2025, Vol. 58 ›› Issue (9): 33-43.DOI: 10.11930/j.issn.1004-9649.202410050

• Key Technologies for Enhancing the Grid Connection Safety Capability of New Energy and New Grid-Connected Entities • Previous Articles     Next Articles

Virtual Inertia Flexible Control of Photovoltaic Storage VSG Based on AHFS and RBF

XIAO Xiangqi1,2(), DENG Hanjun1,2, ZOU Sheng1,2, XIAO Jianhong1,2, LI Kai1,2, MA Bin1,2   

  1. 1. State Grid Hunan Electric Power Company Limited Power Supply Service Center, Changsha 410004, China
    2. Hunan Key Laboratory of Intelligent Electrical Measurement & Application Technology, Changsha 410004, China
  • Received:2024-10-15 Online:2025-09-26 Published:2025-09-28
  • Supported by:
    This work is supported by Science and Technology Project of State Grid Hunan Electric Power Company Limited (No.5216AG22000V).

Abstract:

To address the issue of active power overshoot and the challenge of system dynamic oscillations in grid integration of traditional photovoltaic and energy storage virtual synchronous generators (VSGs), This article proposes a virtual inertia flexible control strategy for photovoltaic storage VSG based on active high-frequency feedback suppression (AHFS) and radial basis function (RBF). Firstly, a first-order filtering link is introduced to enhance the system's high-frequency suppression capability, and an active power differential term is added to the power feedback loop to modify the dynamic performance of the system. Secondly, an objective function considering both active power overshoot and rise time is constructed, and the optimal active power differential coefficient is determined via particle swarm optimization (PSO) algorithm. Finally, an adaptive control strategy for virtual inertia is designed using RBF neural network, which can flexibly adjust the virtual inertia in real-time based on the system's angular velocity and its rate of change. The coordinated control method effectively mitigates the issues of power overshoot and frequency overshooting in traditional VSGs. Simulation results show that the proposed control strategy can effectively suppress frequency deviation and active power overshoot, thereby enhancing the system's transient stability.

Key words: photovoltaic storage VSG, grid disturbance, frequency deviation, active power oscillation, RBF neural network