The Research On Sensorless Control System For Brushless DC Motors Based On DSPACE

Brushless DC motor (BLDCM) is used widely for the reason that it have many advantages of good low-voltage performance, strong torque overload capacity, great starting torque and low starting current. In order to realize the phase-change control of three phase stator windings, traditional BLDCM always has an internal position sensor to detect the six phase-change position. However, its existence increases the volume of BLDCM as well as reduces the reliability of the system. So it is significant and promising to study the sensorless control of BLDCM. This paper focus on several key issues of sensorless control technology based on principle and math model analysis of BLDCM, such as estimation of rotor position, positioning control of the resting rotors and double closed-loop drive. And MATLAB/Simulink simulation and dSPACE experiment were done about them.Back-emf which is famous for the simple principle and reliable method occupies the main position in the field of estimation of rotor position. This paper obtains the Phase-change point by detecting the zero-crossing point of the line back-emf, the method avoids 30°phase-shift as well as reduces dependence on motor parameter.Another difficulty about sensorless control is how to fix the rotor position and start it. Complex locating method is proposed by this paper, firstly, imposed 12 voltage vectors which is produced by two-two conduction and three-three conduction to stators, obtained resting rotor position in 30°precision by detecting stator peak-current to judge the change of inductance. Secondly, drive the motors to precision position by using pre-setting way. Finally, start motor with constant voltage/frequency ratio, once detecting the signals of steady position estimation of rotors, the sensorless control starts working.Sensorless BLDCM double close-loop drive system were set up in MATLAB/Simulink by using S-function.,what's more, simulation analysis were taken on the theory proposed by this paper in order to verify the rationality and feasibility of the theory.Estimation errors of position were caused by parameter change, voltage and current sampling errors, filter phase delay. The emphasis of this paper is put on the influence to the estimation errors of rotor position. Finally, the experimental investigation was conducted for sensorless BLDCM control by using dSPACE hardware-in-the-loop simulation belonged to the lab, and it will provide theoretical support and experimental basis for the BLDCM controller.