Jakub Kolota

Adaptive control of permanent magnet synchronous motor with constrained reference current exploiting backstepping methodology

The presented paper proposes a novel control method of Permanent Magnet Synchronous Machine. Thanks to a recent technique called adaptive backstepping, adjustment of all motor parameters is obtained considering well known dq model. Furthermore, the introduction of a new trajectory generator gives the possibility to limit the reference current without breaking the adaptation process. Examples are calculated to illustrate the properties of the new control method.

Sensorless position estimator applied to nonlinear IPMC model

This paper addresses the issue of estimating position for an ionic polymer metal composite (IPMC) known as electro active polymer (EAP). The key step is the construction of a sensorless mode considering only current feedback. This work takes into account nonlinearities caused by electrochemical effects in the material. Owing to the recent observer design technique, the authors obtained both Lyapunov function based estimation law as well as sliding mode observer. To accomplish the observer design, the IPMC model was identified through a series of experiments. The research comprises time domain measurements. The identification process was completed by means of geometric scaling of three test samples. In the proposed design, the estimated position accurately tracks the polymer position, which is illustrated by the experiments.

Multiple models input-output adaptive controller applied to Ionic Polymer Metal Composite

Ionic Polymer Metal Composites(IPMCs) belong to the class of wet Electro-active Polymers (EAPs) and is promising candidate actuator for various potential applications mainly due to its flexible, low voltage/power requirements, small and compact design and lack of moving parts. Among the smart materials available, IPMCs seem to fit nicely in the constraints of robotics because they react mechanically when stimulated by an electrical signal. However, being a widely used material in industry, it requires complex control methods due to its strongly nonlinear nature. This paper presents a novel approach to tuning multiple models adaptive controller. By adding additional filters we significantly improve convergence of the estimates to the true values of the IPMC parameters as compared to conventional adaptive controllers. Numerical results and comparison with experimental data are presented. The effectiveness of the proposed multiple models adaptive control strategy is verified in experiments.

Analysis of 3D model of reluctance stepper motor with a novel construction

A method of modeling and numerical simulation of two-phased variable reluctance stepper motor using time stepping finite element method (FEM) is presented. In the proposed model, the nonlinear electromagnetic field equations and the circuit equations are strongly coupled and solved together with the motion equation at each time step. The subject of this paper is a novel form of construction of reluctance stepper motor bases purely on variable-reluctance principle. This approach demonstrates that the dynamics of classic reluctance stepper motor construction can be improved by 3D nonlinear distributed parameters models. The optimization of the motor geometry has been solved numerically and the proposed model may contribute to the improvement of the drives dynamics in the future prototypes.

The proportional derivative indirect adaptive control

Abstract The presented paper proposes novel method of Indirect Model Reference Adaptive Control. Thanks to a recent technique called Immersion & Invariance, the parameter adjustment is run with the proportional and derivative term. The former is a standard in indirect adaptive control. The latter is introduced by authors and it provides an enhancement of the transients. Furthermore, it is shown that the proposed control scheme supports the multi model methodology. Illustrative examples are calculated to show the properties of the new control method.

The FEM Parallel Simulation with Look Up Tables Applied to the Brushless DC Motor Optimization

The Finite Element Method (FEM) is widely used to model and simulate electromechanical devices. It enables us to analyze device’s properties and improve its performance. Although, its relevance for electric machine design has been proved, its drawbacks are time consuming calculations. Therefore, it is proposed a novel parallel solver to the FEM modeling technique. However, the presented simulation algorithm is limited to magnetic linear models without eddy currents. The FEM parallel simulation technique is applied to simulate brushless DC motor. The analysis of the field distribution and movement characteristics are included. Moreover, the torque ripple problem is described and its influence of motor movement is shown. The brushless DC motor optimization is performed based on the simulated annealing algorithm. The objective function variables are a dimension of the stator teeth and a width of the permanent magnets. The optimization algorithm characteristics are presented and analyzed. Additionally, the reasons for and against of the simulated annealing algorithm as multivariable optimization tool for the FEM models are examined.

Parallel Computations of Multi-Phase Voltage Forced Electromagnetic Systems

This research presents a methodology to parallel computations of the finite element models of electromagnetic systems excited from voltage source. In the model an electric circuit equations, field equations and motion equation are sequentially coupled. The circuit equations for each phase are calculated parallel using client – server system.