چكيده به لاتين
The control of high voltage electrical motors by using multilevel inverters has become one of the interesting fields in our country’s industrial applications. These electrical drives are very common for motors with a power rating between 1 MW up to 4 MW and the voltage rating between 3.3 KV and 6.6 KV. These systems are very expensive and are usually installed in critical places. Therefore, the manufacturers and the users of these devices pay close attention to their reliability and maintenance.
There are many research works with the focus on the reliability enhancement of the multilevel-cascaded H-bridge inverters in which the efforts have been made to balance the output voltages of a faulty inverter based on various mathematical algorithms. However, the modeling of these inverters under faulty condition has not been considered in the literatures. In this thesis, in order to show the importance of the inverter nonlinearity modeling, it will be shown how neglecting the inverter nonlinearity can lead to the induction motor instability while driving light loads. In order to achieve the model of multi-level inverters systematically, at first, a numerical method is presented to improve the performance of two-level inverters that are very common in low-voltage industrial applications. Then, a general method for modeling the nonlinear behavior of voltage-source inverters is proposed to be used in induction motor parameter estimation process (auto-tuning). One of the unresolved issues of the multilevel-cascaded H-bridges is an unwanted dc voltage increase of some cells when the inverter is operating with some H-bridge modules being bypassed. This occurs at special motor power factor operating conditions. Many research articles have tried to prevent the shut down of these inverters at these load power factors by changing the reference voltages. In this thesis, it has been shown that the reliability of the converter can be improved by vector control of the induction motor power factor and it prevents the increase of the dc voltages of power cells. In addition, a method has been proposed to improve the estimation of the stator inductance. The validity of this method is verified experimentally. At the end, the effect of the modeling of voltage source inverter nonlinearity on sensorless vector control of an induction motor at low speeds is presented.
Keywords:
Modeling the nonlinear curve of voltage source inverter, multilevel-cascaded H-bridge, induction motor parameters estimation, induction motor power factor control.
