Research On Intelligent Control Technique For Voltage Source Inverter

It is well known that the inverter system is working in switching mode, and can be regarded as a nonlinear system, so the mathematical model of the inverter is difficult to acquire. Because the performance of the classic controller is restrained by the precision of the model, designing an excellent classic scheme of the inverter become harder. In this paper, the intelligent control technology is introduced into the inverter system, and fuzzy control technique, genetic algorithm and fuzzy neural network control technique, which widely applied in the control of nonlinear system, are studied.The disturbance of the inverter include periodic and non-periodic disturbance. Repetitive control technique based on the inter model principle is a strategy to compensate the periodic disturbance, however, it is incapable to compensate the non-periodic disturbance. Several hybrid control schemes of intelligent control and repetitive control are studied in the paper. The intelligent controller is used to improve transient performance, while repetitive controller is applied to generate high-quality sinusoidal output voltage in steady state. Based on the mathematical model, the repetitive controller is designed firstly. PD controller is introduced, due to the bad transient response of the repetitive controller, at the same time, the PI controller is designed to insure the precision of output voltage.Considered the nonlinear of the inverter, direct fuzzy control scheme is designed to improve the transient performance of the inverter and the performance of the hybrid controller is analyzed. In order to acquire the good parameters, an optimized design method based on the genetic algorithm is studied in this paper. The results of the simulation and are given to analyze and compare the performance of the PD repetitive controller and direct fuzzy repetitive controller.The parameters of fuzzy controller can not adaptively adjust to acquire an optimum structure online, because it is designed offline. A fuzzy neural network control strategy is presented to improve the adaptability of the inverter under the condition of sudden changes in the load. Simulation and experimental results proved the advantage of intelligent control technique in the real time control of the inverter, as well as,some questions of the application of intelligent control technique are discussed and analyzed. Finally, some suggestions for farther research are presented.