Open AccessModel predictive direct power control scheme for Vienna rectifier with constant switching frequency
Author(s) -
Dang Chaoliang,
Wang Fei,
Liu Ding,
Tong Xiangqian,
Song Weizhang,
Pat Wheeler
Publication year - 2021
Publication title -
iet power electronics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.637
H-Index - 77
eISSN - 1755-4543
pISSN - 1755-4535
DOI - 10.1049/pel2.12218
Subject(s) - control theory (sociology) , ripple , rectifier (neural networks) , robustness (evolution) , model predictive control , pwm rectifier , computer science , power control , voltage , power (physics) , power factor , engineering , physics , control (management) , biochemistry , chemistry , stochastic neural network , quantum mechanics , artificial intelligence , machine learning , recurrent neural network , artificial neural network , electrical engineering , gene
Abstract Model predictive direct power control (MPDPC) has strong robustness and fast dynamic response, so it is widely applied in the control system of grid‐connected converter. However, the traditional FCS‐MPC bears with variable switching state, which will reduce the current tracking accuracy, accidentally produce current ripple and electromagnetic noise. Here, a double closed‐loop control strategy with constant switching frequency based on optimal switching sequence synthesis of model predictive direct power control (MPDPC‐CF) in the inner loop and SMC control in the outer loop is proposed for the non‐linear system of Vienna rectifier. Firstly, the Vienna rectifier is modelled to obtain the predicted values of input power. Then, the given values of active power and reactive power are obtained through the outer loop. Three adjacent voltage vectors with the lowest cost function in the sector are selected to synthesize the optimal voltage vector, and the pulse time of the corresponding vector is calculated according to the cost function of the voltage vector. To verify the correctness of the theoretical analysis, the Vienna rectifier is taken as the research object, and the comparison with the traditional MPDPC shows that the proposed constant frequency model predictive control has good steadystate and dynamic performance.