DocumentCode
736696
Title
A three-phase voltage source PWM rectifier with strong DC immunity based on model predictive control
Author
Shulin, Cai ; Hui, Cao ; Danxi, Liang ; Ruigen, Niu ; Lixin, Jia ; Ahmad, Waseem
Author_Institution
School of Electrical Engineering, Xi´an Jiaotong University, Xi´an 710049, P.R. China
fYear
2015
fDate
28-30 July 2015
Firstpage
7871
Lastpage
7876
Abstract
A Pulse Width Modulation (PWM) rectifier with strong DC disturbance rejection is presented based on Model Predictive Control (MPC). The design of outer voltage square loop and inner current loop using model predictive control is presented, where voltage square loop indicates the square of output DC voltage calculated to be linear with active current. A discrete-time model of the rectifier is constructed to predict the future current for all voltage vectors. Then, one voltage vector minimizing the current errors is selected by solving the cost function. Two weighting matrices, which affect the performance of the rectifier greatly, are used in the cost function to weight prediction error and control actions. This paper indicates the function of the two different weighting coefficients and the way to choose proper weighting factors on a typical three phase PWM rectifier under different conditions. The static, dynamic, pre-charging performance and anti-disturbance ability are discussed upon the proposed rectifier model. The results are proved by simulation that the proposed method can achieve strong DC immunity and a fast dynamic performance with pre-charging strategy compared to the traditional method.
Keywords
Cost function; Mathematical model; Predictive control; Predictive models; Pulse width modulation; Switches; Voltage control; Current model predictive control; Finite set of switch state; Outer voltage square loop; Pre-charging; Strong DC Immunity;
fLanguage
English
Publisher
ieee
Conference_Titel
Control Conference (CCC), 2015 34th Chinese
Conference_Location
Hangzhou, China
Type
conf
DOI
10.1109/ChiCC.2015.7260890
Filename
7260890
Link To Document