Design of a Speed Sensor of a High Speed Drive

OData support
Supervisor:
Dr. Stumpf Péter Pál
Department of Automation and Applied Informatics

Over the last few decades, automation has developed into a central technological strategy. Automation technologies augment human life in many different fields . Electric machines play important role in automated systems from robotics through milling machines to traction. In the last decade increasing attention has been given to high speed induction and synchronous machine drives to reduce system size and improve power conversion efficiency. The angular speed of hundreds of thousands or even above million can occur. In modern technology control is an essential part of the systems. For well functioning closed loop motor drive systems, angular velocity measurement is a dispensable feature. Because of the really huge range of angular speed of these machines the measurement poses many challenges. The effective measurement method is different in low and high speed. A proper sensor system in the whole range of the possible speeds should choose the right method independently. The paper investigate the options of angular velocity measurement, especially executed by optical encoder. In the final project I implemented in a Raspberry PI board an algorithm to process the signals from an optical encoder. Depending on the speed of the rotation two claculation method is implemented.

In safety critical applications as a back-up method, (usually for failure detection) estimators can be used because the reduced cost by avoiding the use of more sensors. In this project the options and results of using and estimator is briefly described too. An estimator capable to calculate the mechanical speed of an asychronous machine is introduced in my final project. A complete simulation model is developed in Matlab/Simulink environment. It was demonstrated by simulation that, the estimator works properly and the relative error between the real and the estimated speed is less than 0.5%.

Downloads

Please sign in to download the files of this thesis.