Modelling of natural field orientation control for induction motor drivers with artificial neural network

ABSTRACT Master Thesis MODELLING OF NATURAL FIELD ORD2NTATION CONTROL FOR INDUCTION MOTOR DRIVERS WITH ARTD7ICIAL NEURAL NETWORK Gamze YDLDIZ Firat University Graduate School of Natural and Applied Sciences Department of Electric and Electronics Engineering 2005, Page:53 Efficient four-quadrant operation and good transient and steady state performance are the main desired features of a modern drive system. In this study, induction motor is preferred because of the direct current motor has several drawbacks such as big size, difficult maintenance due to brushes and commutators, being expensive, not working at explosion and dusty environment. Control techniques of induction motor are generally given. This method is used because of several advantages of sensorless vector control compared with other vector controls. These advantages are cost, accuracy and stability of systems. Natural field orientation is a kind of sensorless vector control. The performance of this control technique in low speed is high and their equations are simple. Thus this method has advantages compared to other methods. In this study, speed and torque controls of natural field orientation control techniques modeled by Matlab/Simulink software, are done. The results showed that the performance of natural field orientation control in low and high speed is very well. Flux angles are plotted using Xartificial neural network models. Error rates are decreased to 0,0001 in figures which are plotted in different speed and number of neural cells. Keywords: Induction Motor, Natural Field Orientation Control, Artificial Neural Network Model XI