Dissertation in the field of Electrical Engineering, Zengcai Qu

The title of thesis is Control aspects for energy-efficient and sensorless AC motor drives

This research proposes control methods for improving the energy efficiency and stability of
sensorless AC motor drives. The study focuses on induction motors (IMs) and synchronous
reluctance motors (SyRMs). Loss-minimizing methods are developed for both IM and SyRM
drives. The loss-minimizing control applies dynamic space-vector motor models which take
into account hysteresis losses and eddy-current losses as well as the magnetic saturation. The minimum points of the loss function are numerically searched in order to calculate the
efficiency-optimal control variable. Magnetic saturation effects can affect the energy
optimization more significantly than core-loss parameters. Additionally, flux-angle and rotorangle estimation methods in sensorless drives are also sensitive to inductance parameters. A saturation model was proposed for SyRMs using explicit power functions. The proposed model takes into account cross saturation and fulfills the reciprocity condition. In order to improve the stability of the sensorless IM drives, especially at low speeds, a gain scheduling method was proposed for a full-order flux observer. The observer gains are selected as functions of the rotor speed estimate in order to improve the damping and robustness of the closed-loop system. The observer is augmented with a stator-resistance adaptation scheme in the low-speed region. In high-speed applications with limited sampling frequency, dynamic performance of the discrete-time approximation of a continuous-time controller can dramatically decrease, and can, in the worst case, even become unstable. A discrete-time current controller was proposed for SyRMs. The current controller is designed based on the exact discrete-time motor model that includes the effects of the zero-order hold and delays. The dynamic performance and robustness are improved, especially at low sampling to fundamental frequency ratios.

Opponent: professor Giacomo Scelba, University of Catania, Italy

Supervisor: Professor Marko Hinkkanen Aalto University School of Electrical Engineering, Department of Electrical Engineering and Automation