This paper proposes a new method of modelling of inductance spatial harmonic components for a position sensorless control. The conventional methods for the sensorless control based on mathematical model consider only fundamental components of the inductances. This cause several problems including limited controller's bandwidth. The proposed method is based on the fact that the estimated back electromotive force (EMF) by the observer contains harmonic voltages induced by inductance spatial harmonics. By processing and analysing the estimated EMF at every operational region, only an interested order component among harmonic voltages is investigated in order to be modelled mathematically. Additionally, a simple feedforward compensation of the observer's input voltage by proposed mathematical model in sensorless control reduces the estimation errors of position and velocity and therefore the controller's high bandwidth is realized.
SUMMARYThis paper proposes a current controller for signal injection-based control schemes such as those for parameter identification and position sensorless control. In a conventional current control system with voltage signal injection, the separation of the injected frequency components from the current is carried out by using band stop filters. However, the separation is not accurate, especially when the system is in the transient state and/or when the current controller bandwidth is close to the injected signal frequency. This leads to distortion of the transient response and affects the stability of the current control system. The proposed controller estimates the injected frequency components in the current on the basis of a motor model by using the injected signal as the input. Therefore, accurate signal separation is accomplished even in the transient state. The current controller's stability and dynamic performance are improved significantly. Experimental results are presented to validate the proposed method. Factors that play an important role in the implementation of the controller, such as time delay and parameter variation, are discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.