This paper describes the study and experimental verification of sensorless control of permanent magnet synchronous motors with a high precision drive using two novel estimation methods. All the studies of the modified Luenberger observer, reference model, and unscented Kalman filter are presented with algorithm details. The main part determines trials with a full range of reference speeds with a special near-zero speed area taken into account. In order to compare the estimation performances of the observers, both are designed for the same motor and control system and run in the same environment. The experimental results indicate that the presented methods are capable of tracking the actual values of speed and motor position with small deviation, sufficient for precise control.

This paper deals with the problem of rotor speed and position detection in sensorless permanent magnet synchronous motor (PMSM) drives. A concept based on detecting the back EMF induced in stator windings was developed and modified. A general structure of an adaptive observer with the proportional-integral function of a corrector is introduced. The non-stationary character of the observer presented in this paper requires an adaptive change of observer corrector settings. Such observer structure was implemented on a DSP system and verified experimentally. Both simulation and experimental results show good properties of the proposed observer structure.

In the paper a new sensorless control structure for the PMSM drive is presented. Such a structure is especially recommended for speed in the range of single revolutions per second (excluding standstill). The method uses a back EMF observer for position estimation. However, there is no need to estimate the speed. This is a big advantage because of possible irregularity of estimated back EMF in this speed range, which makes the calculation of speed difficult or impossible. The proposed structure is similar to the model following control, but the reference model attempts to track the motor operating point. The main idea is to utilize as a reference a model of the whole drive, including speed and current controllers and motor model. Such a model produces reference voltage for the real inverter. However, an extra unit – called rotator – is needed to provide precise control of direct axis current, which is sensitive to the improper position of the voltage vector. The rotator acts as a kind of compensator for error of position estimation.

Purpose – The aim of the paper is to find a simple structure of speed controller robust against drive parameters variations. Application of artificial neural network (ANN) in the controller of PI type creates proper non-linear characteristics, which ensures controller robustness. Design/methodology/approach – The robustness of the controller is based on its non-linear characteristic introduced by ANN. The paper proposes a novel approach to neural controller synthesis to be performed in two stages. The first stage consists in training the ANN to form the proper shape of the control surface, which represents the non-linear characteristic of the controller. At the second stage, the PI controller settings are adjusted by means of the random weight change (RWC) procedure, which optimises the control quality index formulated in the paper. The synthesis is performed using simulation techniques and subsequently the behaviour of a laboratory speed control system is validated in the experimental set-up. The control algorithms of the system are performed by a microprocessor floating point DSP control system. Findings – The proposed controller structure with proper control surface created by ANN guarantees expected robustness. Originality/value – The original method of robust controller synthesis was proposed and validated by simulation and experimental investigations.