Control of a Brushless Doubly-Fed Reluctance Machine

Scriptie/masterproef: Master's Thesis


The industry is on a never-ending pursuit for more cost effective and efficient electric drives for their applications, including the compressor industry. For compressor-like applications, variable speed drives are required and with the prospect of a cost effective drive, a possible alternative candidate is the brushless doubly-fed reluctance machine. Thanks to the two available windings and its doubly-fed nature, an extra degree of freedom becomes accessible for optimal use. Furthermore, only one of the two available windings necessitates the use of a converter, while both windings can support the required power flow. This results in a potential downscaled converter, consequently reducing the costs of the complete drive.
While this machine has its roots in the begin period of the 20th century, it is only in the last few decades a lot of interesting improvements are discovered concerning this machine type, supported by the renewed interest of the researching community for the unusual and complex operating principle of the BDFRM.
While assisting ongoing research regarding the viability of the BDFRM for a compressor- like application, the goal of this thesis is to provide insight in this complex machine and propose an appropriate control law for the intended application. This thesis will therefore start with an explanation of the complex operating principle of this machine, followed by the conception of a simulation model of the BDFRM. Then, different existing control laws and state observers are studied and based on this acquired knowledge, an appropriate, however unusual control law is proposed. Namely, a cascaded output and state feedback controller, based on the Linear Quadratic Regulator, in combination with an extended Kalman filter as state observer. An unusual combination for this machine type, yet to be found in the literature. This is followed by deriving the models of the BDFRM for control purposes, implementing them in simulation and eventually validating the control law in simulation, as a first proof of concept.
Finally and most importantly, to validate the simulations in reality, different tests are carried out on a test-setup comprising of a BDFRM prototype, but first, to be able to design the control algorithm, the machine electrical and mechanical parameters are identified. The tests mimic, on one hand, normal conditions in which the BDFRM would drive a compressor, and, on the other hand, a more extreme condition to test the robustness of the controller. The control law performances show satisfactory results. Eventually the outcome of this thesis is once more concluded and possible further developments of this control law are discussed.
Datum Prijs25 jun 2021

Citeer dit