Open AccessGA–PSO approach for optimising space‐vector PWM control sequence
Author(s) -
Ning Yong,
Dai Yuxing,
Peng Zishun
Publication year - 2019
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/iet-pel.2018.5915
Subject(s) - particle swarm optimization , sequence (biology) , genetic algorithm , control theory (sociology) , matlab , pulse width modulation , total harmonic distortion , inverter , computer science , algorithm , mathematics , mathematical optimization , voltage , engineering , control (management) , artificial intelligence , biology , electrical engineering , genetics , operating system
Space‐vector pulse width modulation (SVPWM) provides several degrees of freedom, which can be optimised to improve the harmonic performance of the three‐phase inverter. Genetic algorithm (GA) and immune algorithm (IA) are the two classical probabilistic optimisation algorithms, which are simple in structure and do not need an accurate mathematical model. However, the optimisation accuracy and reliability are low when they optimise the high‐dimensional non‐linear problem, such as SVPWM control sequence of the three‐phase inverter. To cope with these problems, a genetic algorithm–particle swarm optimisation (GA–PSO) is proposed here, which introduces the mutation of GA into discrete PSO. The global and local optimisation ability of the algorithm is greatly improved by the introduction of mutation operation. The results of MATLAB/SIMULINK simulation show that the weighted total harmonic distortion (WTHD) by the optimal SVPWM control sequence based on GA–PSO is 0.199%, which is much better than that of the PSO, IA, and GA. The average generation number of GA–PSO is only 1/500 of IAs. Further experimental data verify that the WTHD by the optimal SVPWM control sequence based on GA–PSO is lower than that of conventional SVPWM and IA.