Sizing and control of a brushless doubly fed reluctance motor drive system for variable speed applications

Since their early development in the 19th century, alternating current (AC) electric motors (EMs) have evolved to become one of the founding technologies of modern society. It is estimated that, worldwide, electric motors consume half of the total produced electric energy, mostly driving systems such as compressors, pumps, and fans. Climate change and cost of energy have pushed industrial actors to reduce their energy consumption and increase the energy efficiency of their plants and products. One of the answers to this problem lies in driving EMs in at variable speed to cope with varying input and output requirements of the driven system, increasing the energy efficiency by up to 35 %. However, this comes at the expense of adding a variable speed drive (VSD), an electric system responsible for supplying EMs with a variable voltage and frequency to control and adapt the speed following the requirements of the driven system. Despite the evolution of VSDs and their fundamental components, the power semiconductor devices, the relatively high cost of VSDs, especially at high power ratings, still represents an important barrier towards its further market penetration.
The brushless doubly fed reluctance motor (BDFRM) is an interesting alternative to conventional EMs, offering the possibility of reducing the cost of the VSD thanks to its doubly fed nature. Indeed, the BDFRM is an EM that could merge the electric power of two independent AC voltage supplies into a mechanical power to drive mechanical systems at variable speed. This opens the path to supplying the BDFRM with mains supply on the one hand, and a downscaled VSD on the other hand, thereby reducing the cost of the drive system. Even though very promising, the BDFRM is currently addressed mainly on academic level and is still facing an important gap towards its adoption in an industrial context.
The goal of this dissertation is to address this gap by investigating the relevance of the BDFRM as a motor for industrial variable speed applications. To reach that goal, first, the working principles of the BDFRM are investigated on the level of the magnetics and the field distribution inside the air gap. Based on this, the electromechanical model of the machine is derived and analyzed to clarify the active and reactive power exchange with the two independent AC supplies. Then, a model-based control system design is proposed, befitting an industrial application, which proved to be effective for variable speed operation and load rejection in real-time on a low power BDFRM prototype. Finally, the acquired knowledge allowed for the development of a scalable sizing methodology for all the elements of the drive system, including the machine itself, the VSD and capacitor banks to ensure the power quality at the mains supply. The proposed methodology allows for engineers to determine the volume of the machine and ratings of different components, which could be extended further towards the cost assessment of the whole drive system.
The main findings of this dissertation show that the BDFRM could be used as a variable speed drive but that it is mainly interesting for the speed range 500 RPM – 1000 RPM. The BDFRM could also realize a significant downsizing of the VSD, but this comes at the expense of a larger machine volume due to the low torque density of the BDFRM, and at the expense of capacitor banks at mains supply for power factor compensation.