高压电器2025，Vol.61Issue(11)：98-108,11.DOI:10.13296/j.1001-1609.hva.2025.11.009
基于神经网络的GFVSG多参数协同自适应优化控制策略
Neural Network-based Multi-parameter Collaborative Adaptive Optimization Control Strategy for GFVSG
摘要
Abstract
Grid-forming virtual synchronous generator(GFVSG)can provide active support to the power grid,yet it has suh problems as slow active-power response and weak frequency interference resistance ability.In this paper,with the advantage of flexible and adjustable GFVSG paramater,a multi-parameter cooperative adaptive optimiza-tion control strategy based on neural networks is proposed.First,a mathematical model is set up to analyze the dynam-ic influence mechanism of rotational inertia and damping coefficient on system.Then,by integrating the synchronous generator power-angle characteristics with the frequency oscillation trajectory,dynamic tuning rules and real-time pa-rameter requirements are derived.Finally,an RBF neural network with strong nonlinear mapping capability is intro-duced to construct a self-tuning framework,design composite error fucntion,introduce dynamic dynamic coupling co-efficients and achieve cooperative self adaptive optimizatino control of multiple parameters.The simulation results show that,in case of sudden change of power command,the proposed strategy can effectively suppresses overshoot and oscillation in both output power and frequency,have response speed and,in case of fault at the grid side,can provide strnger inertia response support.关键词
构网型虚拟同步发电机/转动惯量/阻尼系数/神经网络/自适应Key words
grid-forming virtual synchronous generator/rotational inertia/damping coefficient/neural network/adaptive引用本文复制引用
王旭,贾彦,郭强,郭名洋..基于神经网络的GFVSG多参数协同自适应优化控制策略[J].高压电器,2025,61(11):98-108,11.基金项目
国家重点研发计划"储能与智能电网技术"(2024YFB2408400) （2024YFB2408400）
内蒙古自然科学基金(2023ZD20). Project Supported by National Key Research and Development Plan"Energy Storage and Smart Grid Technology"Project(2024YFB2408400),Inner Mongolia Natural Science Fund(2023ZD20). （2023ZD20）
