Search

In a vector-controlled induction machine drive, accurate knowledge of the machine electrical parameters is required to ensure correct alignment of the stator current vector relative to the rotor flux vector, to decouple the flux- and torque-producing currents and to tune the current control loops. This paper presents a new method for on-line identification of the induction machine parameters required to tune a rotor-flux-oriented (RFO) vector control scheme. Accuracy of the slip frequency estimation required for RFO vector control is achieved by utilizing the parameter independent 'flux pulse' rotor time constant estimation scheme, which utilizes short-duration pulses injected into the flux-producing current. The parameters required to tune the synchronous frame current control loops with a decoupling circuit are estimated using a recursive estimation scheme derived from the synchronous frame voltage equations. As the 'flux pulse' scheme requires signal injection into the flux-producing current a new rotor time constant estimation scheme is presented, based on the sensitivity analysis of the recursive parameter estimation scheme. Simulation and experimental results are presented which demonstrate the effectiveness of the online parameter identification and control loop tuning technique. Index Terms--Current control, induction machine, parameter estimation.

For high performance induction machine control, the technique of indirect rotor flux oriented vector control is commonly utilized. The torque performance of this technique is significantly affected by the performance of the current control loops. In this paper a new adaptive high bandwidth current controller for induction machines is presented. The technique is derived and validated through simulation and experimental results. The new architecture is shown to achieve a dead-beat response, with a rise time of one sample period and no overshoot, when the inverter voltage limit is not exceeded. When the voltage limit is exceeded the current response is achieved in a minimum time with no overshoot. It is shown that the new control technique achieves similar dynamic response to the conventional dead-beat control scheme while eliminating parameter sensitivity issues. It is demonstrated that the new algorithm offers higher bandwidth than the commonly used synchronous frame proportional plus integral control technique while maintaining good steady state performance. Index Terms--Adaptive control, current control, current control loops, dead-beat control, dynamic response, indirect RFO vector control, induction machine, synchronous frame, vector control.

This paper proposes a novel rotor resistance identification method for an indirect rotor flux-orientated controlled induction machine drive. The method is effectively integrated with the intermediate current control loop of the system. A decoupled synchronous voltage control scheme is used to achieve a fast, accurate current control response and indicates the relative thermal change of the rotor resistance. A model reference adaptive control scheme is then used to track the variation of the rotor resistance. Other issues, such as the nonideal characteristics of the power devices, stator resistance variation and comparison with two other parameter identification methods, are included. This method is less complex and more effective than others, and this is supported by theoretical analysis, and verified by simulation and experimental results. Index Terms--Induction machines, parameter estimation, rotor resistance, vector control.

This paper shows that it is possible to use currently available commercial software to model and simulate a vector-controlled induction machine system. The components of a typical vector control system are introduced and methods given for incorporating these in the MATLAB/SIMULINK software package. The identification of rotor resistance is important in vector control, if high-performance torque control is needed, and modeling of the extended Kalman filter (EKF) algorithm for parameter identification is discussed. It is certainly advisable, when feasible, to precede implementation of new algorithms, whether for control or identification purposes, with an extensive simulation phase. Additionally, a technique for generating pulse-width modulation (PWM) phase commands to extend machine operation to higher speeds before field weakening occurs is simulated in a vector-controlled induction machine, driven by a PWM inverter. This demonstrates the versatility of the vector-controlled induction machine system model. Index Terms - Induction machines, modeling, parameter estimation, simulation, vector control.

The estimation of rotor time constant, or rotor resistance, in a vector-controlled induction machine is necessary to achieve high-performance torque control. A new method of estimating the rotor resistance on-line, for use in a vector-controlled induction machine, is presented. It uses short duration pulses added to the constant flux reference current [Mathematical Expression Omitted] and based on the resultant torque command current produced by a proportional-integral controller [Mathematical Expression Omitted] adjusts the rotor resistance estimate. This method of self-tuning the vector controller to the rotor time constant, when operating in a closed-loop speed control loop, does not produce torque pulsations when correctly tuned. In comparison to other on-line methods such as the extended Kalman filter and the extended Luenberger observer, this method does not require voltage sensors and is computationally simpler. The rotor resistance estimation technique is illustrated through simulation and practical implementation of a vector-controlled induction machine. Index Terms - Induction machines, parameter estimation, vector control.