الفهرس 
CHAPTER 1 : INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 124 |  |  |
| | | from | 124 | | |
Abstract
The induction motor exhibits highly coupled, nonlinear, multivariable structure, which can be considered as a good example for a nonlinear system. Nowadays, high performance induction motor drives are very important in industrial applications such as robotics, rolling mills and machine tools. Such applications reqmre ~ast, well damped and precise torque response, parameter-insensitive control characteristics and rapid recovery from speed DROP caused by impact loads.
Recently, many adaptive control techniques have been devoted to cope with the variations of the induction motor parameters. In fact, parameter variations, which essentially depen~ on temperature variation, skin effects, and saturation can heavily affect the performance of the induction motor drives. Thus, without the exact knowledge of the machine parameters, optimum decoupling and machine linearization for high performance applications can not be achieved. Moreover, the adoption of modern control te~hniques becomes useless. For this purpose, the efforts of researchers are focused on the challenging problem of the on-line estimation of the induction motor electrical parameters.
On the other hand, motor drive systems without speed sensors have gained increasing popularity in industrial applications. The main advantages coming from such types of drives are quick response, high reliability and low cost. In this way, a sensorless control of the induction motor enables to keep the main advantages of the induction motor like ruggedness, mechanical simplicity, low maintenance and acquisition costs. Several s,chemes have been described in the literature related to speed or position sensorless operation of induction motors. These sensorless schemes are often based on parameters of machine model. Therefore, the accurate estimation of the motor parameters and motor speed
becomes vital and necessary to• realize high performance applications of induction motor drives.
In this thesis, a modified gain rotor flux observer of the induction motor drive with speed and parameters adaptation scheme has been presented. The flux observer has been employed to identify the stator current and rotor flux accurately. The parameter adaptive scheme has been used in combination with the flux observer for estimating the stator and rotor resistances and motor speed. The adaptive scheme for identifying the stator and rotor resistances and motor speed have been presented based on the modified gain rotor flux observer. The control -laws of these adaptive schemes \. have been designed based on Lyapunov’s theorem.
Two new approaches have been considered for modifying the observer gain. In the first approach, the observer gain has been determined by solving the observer model m the stationary reference frame. A simple closed form equation for the observer gain has been derived. This gain is found to be a function of the stator frequency in addition to the estimated stator parameters. The stator and rotor resistances and motor speed are all estimated when the motor is’ fed from• a variable voltage and frequency source. Simulation results show significant improvements in the transieJ:.lt responses of the estimated parameters using the proposed observer compared with the conventional one. The rotor flux and stator current component errors between the actual and their estimated values decay faster to steady-state zero value with the proposed observer.
In the second approach, a modified gain rotor flux observer has been derived based on the principle of field-orientation in the d-q synchronously rotating reference frame. The optimal value of the observer gain has been proved in this case by minimizing the error of stator current between the estimated and its actual one and it is found to be a function of all electrical machine parameters. The observer with a modified gain obtained from this approach has been applied to direct and indirect vector control schemes of the induction motor drives with and without speed sensors. All simulation results are comp8red with those obtained from the corresponding schemes using the conventional observer. It has been proved that the vector controlled induction motor drives (direct or indirect) with the proposed observer is robust against stator and rotor resistance variations. Moreover, The transient responses of the estimated parameters and motor speed are improved significantly and converge quickly to their actual values. In addition, it has the capability to operate under load disturbance and wide range of speed variations. In other words, the vector control schemes with and without speed sensor using the proposed observer can exhibit good dynamic and steady-state performances in comparison with those obtained using the conventional one. The thesis falls in six chapters