Study Of Main Flux Saturation Influence In Speed Estimating Of Induction Machine

Sensorless vector control of induction motor is not only an important aspect but also a hotspot of investigation in modern AC drive. The key problem is how we can get the speed of motor accurately. The way of speed estimation is based on measurement of stator voltages and currents which are easy to measure. Induction machine is a controlled object that is non-linear, multi variables, strong coupled. Whereas its complication of controlling rotor flux, states observer is used in the machine vector control system widely, but the accuracy of speed observer lies on the machine parameters and flux adjust depends on parameters too. Speed estimation utilized in this paper is based on the way of full order flux observer. The observer makes use of motor parameters to estimate speed, but these parameters (stator and rotor resistance, magnetizing inductance) will vary because of the change of environmental temperature, main flux saturation and iron loss. So parameter variation will lead to the error of estimation. In conventional drive system, it is often keep magnetizing current as a constant, then change torque current to control torque of the machine, at this time the machine is considered operating at the point of flux saturated where all parameters of the machine are constants, including magnetizing inductance and rotor resistance. But, among the various applications, the induction motors used in the hoist, cranes and the electric vehicles are often driven above their base speed, and the magnetizing current can be changed in order to get flexible ability of controlling torque, then the degree of main flux saturation will be changed. The magnetizing inductance is also varied nonlinearly according to the flux saturation. When the motor operates in field weakening region, the main flux saturation will obviously influence the precision of speed observer, take a great error in veracity field-oriented and bring great error of speed estimation without saturation compensation in speed observer system. So, it must be considered to compensate in sensorless vector control. According to analysis of these reasons, this paper used voltage model to identify magnetizing inductance, then rotor speed is estimated by using the identified magnetizing inductance taking account in flux saturation. According to the simulation, it can be proved that the structure and principle of the compensated observer is simple and easy to realize. Usually, the establishment of dynamic mathematical model neglected saturation and iron loss, but saturation and iron loss exist truly. After analyzing saturated compensation, the motor model with saturation and iron loss is established in this paper. Then the compensated controller is analyzed.