三相四桥臂并网逆变器模型电流预测控制方法

Study on Model Current Predictive Control Method for Three-phase Four-leg Grid-connected Inverters

分布式电源通过三相四桥臂变流器接入低压配电网,传统的三相四桥臂逆变器控制方法中,SAPWM法、三维SVM法等控制原理复杂,难以实现工程化应用,且系统稳定性较差。为此,将模型预测电流控制方法与三相四桥臂并网型逆变器数学模型相结合,提出一种易于实现的三相四桥臂并网型逆变器模型预测电流控制方法。该控制方法摒弃了传统控制方法中大量的非线性控制环节和PWM调制环节,建立了逆变器的离散化数学模型,对逆变器输出电流进行在线评估。在PSCAD/EMTDC仿真平台上搭建光伏发电系统并网模型,并对所提出的控制算法进行验证,结果表明:与传统的控制方法相比,模型预测电流法具有良好的静、动态特性和鲁棒性,控制方法简单,易于实现数字化。

The distributed generators are connected to low voltage distribution network through three-phase four-leg inverters. In the traditional control methods for three-phase four-leg inverters, including saddle pulse width modulation （SAPWM） method and three-dimensional space vector modulation （SVM） method, the control principles are complicated and the system has poor stability, which makes it difficult to achieve engineering application. By combining the model predictive current control method and the mathematical model of three-phase four-leg grid-connected inverter, this paper presents a method of model predictive current control for three-phase four-leg grid-connected inverters which is easy to implement. This method builds the discrete mathematical model of the inverters, avoiding plentiful nonlinear control links and PWM modulation parts in the traditional methods, and the output current of the inverters are evaluated online. The model for photovoltaic power generation system is established in the PSCAD/EMTDC simulation platform in this paper, and the proposed control algorithm＇ is verified. The results suggest that compared with the traditional methods, model predictive current method possesses good steady stability, dynamic stability and robustness, and is simple, making it easy to realize digitization.