并网变换器虚拟同步无模型参数鲁棒增强控制策略

随着新型电力系统的发展，并网变换器已成为能量传输的关键设备。模型预测控制的虚拟同步机参数鲁棒性较低，当并网参数失配时，输出电流纹波增大，同步机功率支撑下降。针对此问题，提出一种改进的虚拟同步无模型参数鲁棒预测控制方法。首先，该方法采用四阶Runge-kutta优化超局部无模型得到虚拟同步机参数增强鲁棒模型。然后，采用Lagrange插值法求解模型中K参数，通过前4个时刻的采样值预测下一时刻的输出。同时设计虚拟惯量无模型自适应预测算法，实现了惯量动态需求响应。最后，经过价值函数寻优比较得到虚拟同步机的优化电压矢量，实现参数鲁棒增强控制。实验结果表明，所提出的控制策略在参数失配情况下功率支撑能力稳定，频率波动时虚拟惯量能够动态响应，具有良好的稳态和动态响应性能。

Robust enhanced control strategy of a virtual synchronous grid-connected inverter with model-free parameters

With the development of new power systems, grid-connected converters have become key pieces of equipment for energy transmission. The virtual synchronous machine parameters of model predictive control have low robustness, and when the parameters are mismatched, the output current ripple increases, and the synchronous machine power support decreases. To address this issue, an improved virtual synchronous model-free parameter robust predictive control method is proposed. This method first uses a fourth-order Runge-Kutta optimization super-local model-free method to obtain a robust model that enhances the virtual synchronous machine parameters. Then, the Lagrange interpolation method is used to solve the K parameter in the model, and the output for the next moment is predicted using the sampling values from the previous four moments. Also, a virtual inertia model-free adaptive prediction algorithm is designed to achieve dynamic response to inertia dynamic demand. Finally, the optimal voltage vector for virtual synchronous machines is obtained by optimizing the value function, achieving robust parameter-enhanced control. Experimental results show that the proposed control strategy has stable power support capability under parameter mismatch, and the virtual inertia can dynamically respond to frequency fluctuations, with good steady-state and dynamic response performance.