A Unified-Model-Based Analysis of MRAS for Online Rotor Time Constant Estimation in an Induction Motor Drive

Depending on the functional candidates, such as rotor flux, stator voltage, electromagnetic torque, and reactive power, various schemes based on a model reference adaptive system (MRAS) have been proposed to update rotor time constant. In this paper, a unified model for rotor time constant identification based on MRAS in an induction motor drive with indirect field-oriented control is developed. It is shown that kinds of MRAS-based models are of unity in essence and share a unified mathematical expression despite their different implementations, providing a new viewpoint to analyze the estimation models and supplying theoretical basis for evolutions and comparisons among them. Furthermore, to expand the effective range of the estimation model, stator frequency or q-axis stator current must be incorporated into the corresponding adaptive mechanism. From the closed-loop transfer function of the system, established in this paper, the properties of high nonlinearity and strong coupling behind the seemingly simple structure of these MRAS-based identification methods are exposed, paving a basis to improve the performance further. The validity of the proposed theory is verified by simulations and representative experimental results.