Analysis of Luenberger observers and sliding mode observer for brushless dc motor
Abstract
RELEVANCE of the research lies in the need to identify a more accurate and stable sensorless vector control system for a brushless direct current motor for applications in electric drives where mass and dimensional indicators or the provision of a high degree of housing hermeticity are critically important.
Objective
To perform a comparative analysis of the dynamic characteristics and robustness of two types of observers, namely a full-order Luenberger observer and a sliding mode observer, with a sensorless vector control system for a brushless direct current motor to identify a more accurate and stable control system.
Methods
This work was carried out using mathematical modeling methods in the SimInTech environment. A complete dynamic model of the electric drive control system was developed, observer submodels and commutation tables were implemented, and all features of observer synthesis and speed calculation based on back-EMF force were described in detail. Testing was conducted in two modes: acceleration to the nominal rotation speed followed by load application, and an increase in stator resistance by 30%.
Results
The research established that while both observers ensure the fundamental operability of the system, they possess different sensitivities to parameter variations. The Luenberger observer demonstrated higher accuracy in average speed estimation under parameter drift conditions compared to the sliding mode observer, despite the presence of high-frequency ripples in the signal. The sliding mode observer exhibited lower robustness to changes in stator resistance, which resulted in an increase in the steady-state error and the emergence of a pronounced chattering effect. It was revealed that the application of a blind start for these observers is not applicable and leads to torque spikes during the motor acceleration period.
Conclusion
Recommendations were formulated for selecting control strategies for various operating conditions of electric drives. The research results can be applied in the development of electric drives where mass and dimensional indicators or the provision of a high degree of housing hermeticity are critically important, such as traction motors for unmanned aerial vehicles or high-precision industrial manipulators.