Improved Fault-Tolerant Control for Brushless Permanent Magnet Motor Drives With Defective Hall Sensors
Brushless permanent magnet motor drives based on Hall sensors have received significant attention in recent years. In this area, the faults of Hall sensors become a new concern and several fault-tolerant control (FTC) methods have been proposed. However, most of the state-of-the-art FTC methods requ...
Saved in:
Published in | IEEE transactions on energy conversion Vol. 31; no. 2; pp. 789 - 799 |
---|---|
Main Authors | , , , |
Format | Journal Article |
Language | English |
Published |
IEEE
01.06.2016
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Brushless permanent magnet motor drives based on Hall sensors have received significant attention in recent years. In this area, the faults of Hall sensors become a new concern and several fault-tolerant control (FTC) methods have been proposed. However, most of the state-of-the-art FTC methods require some time to reconstruct the correct Hall sensor signals, which results in significant transient currents and speed dip during fault diagnostic process (FDP). In this paper, a new and improved FTC scheme based on FDP and vector-tracking observer is proposed. A method to identify the duration of FDP is proposed based on the analysis of acceleration estimation and the fault diagnosis results. During FDP, the method defaults to an open-loop observer control, which removes the undesirable current/torque transient. After that, the close-loop observer is re-enabled and the motor operation is restored. The proposed FTC is demonstrated in detailed simulations and experimentally on 120° brushless dc motor drives and sinusoidal PM motor drives. For both types of drives, a significant improvement is achieved in steady state and transient operation with faults of up to two Hall sensors, which has not been possible with available alternative FTC approaches (unless a sensorless control is used). |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 0885-8969 1558-0059 |
DOI: | 10.1109/TEC.2016.2526621 |