Speed-Sensorless Induction Machine Control in the Field-Weakening Region Using Discrete Speed-Adaptive Full-Order Observer

This paper presents a novel field-weakening strategy combined with a modified discrete speed-adaptive full-order observer to improve the performance of speed-sensorless induction machine drives in the field-weakening region. The robust field-weakening controller regulates the field current and the clamp of torque current according to the machine operating mode. Compared with the conventional inverse proportion to speed method and optimal flux level calculation method, the proposed controller shows the superiorities over maximum torque selection and parameter robustness. Moreover, a discrete speed-adaptive full-order observer is designed to estimate the rotor speed. However, when the machine operates at high speed, two poles of the conventional forward Euler approximation-based observer are outside the unit circle, which leads to system instability. To solve the conflict between stability, accuracy, and computational burden, a modified Euler approximation is used to discretize the observer. A comparative study of different approximations substantiates the presented approximation. The effectiveness of the proposed solution is demonstrated experimentally on an industrial induction machine drive.

I. Introduction

In electrical vehicles (EVs) and tool spindle drives application areas, cage induction machines (IMs) are extensively used due to high ruggedness, low manufacturing cost, and good performance in the extended speed range [1], [2]. To extend the machine upper speed with the limited dc-link voltage, the field-weakening strategy based on rotor field-oriented control (RFOC) is considered effective [3], [4]. It is desirable to achieve maximum torque capacity over the whole speed range, which puts forward a high request to the current distribution strategy. The recent researches can be divided into two main categories.