Skip to Main Content
The brushless DC motor has such advantages as simple structure, convenient to control, high reliability, and has been applied in many industrial fields. In order to simplify the converter topology and lower the system cost, four-switch three-phase BLDCM recently becomes research highlight of scholars. Conventional hysteresis controllers suffer from big phase current ripple and inaccuracy current threshold adjusting of the four-switch three-phase BLDCM. To overcome the shortcomings of the hysteresis controller, this paper presents a novel direct current control strategy based on current threshold online identification using intelligent controller for four-switch three-phase BLDCM. A radial basis function neural network is built to identify the relationship of load, current threshold and expected speed online. When the given speed and load is setting, current threshold identifier give the suitable threshold output to the current controller. Also the system use two PID controller based on RBF neural network online regulation to control phase current Ia and Ib separately. Current controller constructs the online reference model, implements self-learning of PID controller parameters by RBF neural network. The intelligent controller individually regulate duty cycle of PWM signals working on the inverter bridge to make phase current fall in the specified threshold quickly and smoothly. Simulated and experimental systems are build to fully prove the performance of the control scheme. Excellent flexibility and adaptability as well as high precision and good robustness are obtained by the proposed strategy.