Vector Control of Induction Machine with Estimates and Adaptation of Rotor Time Constant

The variation of the rotor time constant exerts a dominant effect in the performance degradation of the orientation of the rotor flux method of induction machine and the online estimation of the time constant ( r T ) is necessary in order to use it in the flux orientation and keeping the performance of the vector control. We presented our work vector control of induction machine by estimating and online adaptation of the rotor time constant of our machine only by measurement of voltage, current and speed, in line on the machine, to provide information to the orientation block on the variation of the parameter studied r T .


Introduction
The asynchronous machine constitutes more and more today a high percentage of the motors used for high-performance drives by advantageously replacing the DC motor .but the difficulty for controlling asynchronous machines is the fact that there is a strong coupling between the flux and torque and any action on one affects the other and more the transitional regime of the machine depends on several variables related to each other in a nonlinear way this fact makes its control more complex than direct current machine because it is very difficult to obtain effective decoupling of the two control parameters which are magnetic flux and mechanical torque. To overcome this limitation of the control of induction machines, several techniques use the method of flux-oriented (vector control), allowing decoupling the speed and the flux of the machine. However, these performances degrade at high speed since the de-fluxing turns so necessary and also when the variation of internal parameters and outside the system studied. The interest raised by the indirect method of flux orientation is motivated by no need to measure the flux, unlike the direct method. We start with the presentation of the model of the asynchronous machine for the vector control. Next, we show improved indirect control by the introduction of the algorithm estimator of the time constant. Finally, the results obtained with the proposed algorithm of the r T estimator with adaptation, and the simulation.

Modeling the Machine
The model of the machine used for the control is given in the ( q d, ) by the following Equations (1):

Adaptation of the Rotor Time Constant
From measurable quantities of the machine (voltage, current, speed) we estimate r T using the equation (10). Once determined, we must ensure to maintain the time constant at the orientation block equal to the estimated value of which corresponds to the constant real time of the machine.  We will simulate the closed loop system without adaptation with a variation of the resistance in the first stage, then with adaptation r T . We tested the system behavior when the rotor resistance varies. The instants 1s and 2s the system is subjected to an application then removing a torque disturbance Nm C r 10 = Note in Figure (3) the rotor flux increases with rotor resistance, so decoupling between flux and torque lost at times 1s and 2s and the estimated flux does not follow the actual flux of the machine the same for the estimated time constant r T does not follow real r T .

Characteristic of MAS
As there is no adaptation of the rotor time constantly, one can observe the effect of variation r T on the dynamic response of flux and torque at the machine.
It is clearly visible from Figure (4) that the curves that the adaptation constant function of the rotor time is accomplished despite the variation applied to the rotor resistance of the machine.

Conclusions
The results of this work on the training adjusted by conventional methods of indirect rotor flux orientation are obtained with improved performance of the control system studied is realized by the development of a blueprint indirect stronger whatever the variation of parameters with online correction of the variation of the time constant of rotor to maintain the decoupling block in perfect agreement with the actual conditions of engine operation.