Development of a measurement unit for asynchronous machine

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

Nowadays, the use of asynchronous motors or also known as induction motors is the

most widespread in the field of factory automation and building service engineering, on the

grounds of their high reliability and low maintenance needs. Over the past decades, the

development of power electronics and decreasing prices of power electronic devices have

created new opportunities for controlling induction motors. For example, high quality drive can

be achieved by direct torque or field oriented control. The advancement of electric vehicles is

opening new, uncharted territories for the application of such motors, since loss mitigation can

be achieved through controlled magnetic flow at high rotational speed resulting in low working


Knowledge of the rotational speed, and the angle closed by the fixed and rotating

magnetic field, due to the slip, is essential for field oriented controlling. Such data can be

obtained by using expensive rotation speed measuring sensors, necessary because of the large

scale figures. To decrease the expenses, a functional method for estimating the rotational speed

can be applied based on the mathematic machine model. This method is one of the most

researched topics in the field of induction motors.

In this piece of work, a rotational speed estimation method has been tested, selected

after an in-depth review of corresponding academic literature and algorithms from various case

studies. Firstly, the chosen method has been tested in a simulation environment with and

without load. For the simulation, MATLAB Simulink software environment and the previously

measured electric parameters of a real induction motor have been used. The necessary devices

and parts have been selected in view of the measurement range, processing time and precision

requirement. The next step has been to design the printed circuit of the measuring device.

Having tested the designed device, the estimating algorithm has been implemented in a

STM32F4 microcontroller. After that, the testing and tuning of the implemented algorithm have

been carried out. The display of the measured values has been achieved by a computer sided,

MATLAB based user interface.

The simulation results of the chosen algorithm clearly show that in ideal circumstances

the rotational speed can be estimated with minimal errors.

The algorithm precision can be increased by constant rotor flux supply, due to the field

oriented supply. As a consequence, in real circumstances numerous factors can have an impacton the estimates, such as the fluctuating rotor flux, or changes in the motor parameters in

operation, which is proportional to the temperature variations.

The estimation can give rather precise results in real circumstances too, with the

compensation of the above mentioned errors and the online identification of the parameters.


Please sign in to download the files of this thesis.