Abstract
Since the brushless DC motor (BDCM) drive with position sensorless control possesses higher capability in many industry applications, the major purpose of this dissertation is to establish a DSP-based sensorless BDCM drive, and to develop some key technologies for improving its driving control performance. For facilitating the implementation of the developed sophisticated control rules and the performance test, an experimental motor drive is first constructed. The motor is powered by a voltage source inverter with properly designed isolating drive circuits. The TMS320C240 DSP augmented with a specific ASIC is utilized to build up the digital control computer. This configuration makes the fully digital control for motor drives become possible. Although many BDCM sensorless control methods have been proposed till now, their performances are rather sensitive to the variations of system parameters and operating condition. To improve this, an intelligent sensorless control strategy is developed in this dissertation. The rotor position for making commutation is first roughly estimated by a proposed switching signal generation scheme using motor terminal voltages. Then the commutation instant is finely adjusted to yield better torque generating capability. This is achieved by a simple but stable self-tuning algorithm for seeking the minimization of motor line drawn current. In addition to the proper commutation, the winding current waveform also significantly affects the torque generating performance of a BDCM. As generally recognized, better motor driving control performance can be obtained for any type of BDCMs if the current-mode PWM control is applied. Accordingly, a robust current-mode control scheme is developed to let the BDCM drive possess fast and robust winding current tracking response. Since no commutation signals are available from the sensed motor terminal voltages at standstill, a suitable means is indispensable for starting a sensorless BDCM drive. A simple method is developed to start the BDCM stably and smoothly like a synchronous motor until a steady-state speed is reached. As to the speed monitoring and control, a speed estimation approach is proposed. And based on the estimated speed, the dynamic model is estimated and a speed controller considering the effect of system dead-time is developed. Finally, a switch-mode rectifier is employed as the input stage of the inverter for drawing power with good power quality from utility grid. The robust voltage ripple cancellation and current control approaches are developed such that good tracking control performance in input current waveform is obtained even if the output filtering capacitor with reasonably small value is used. The effectiveness of all the developed control approaches and the driving performance of the established sensorless BDCM drive are demonstrated experimentally.