Please use this identifier to cite or link to this item:
Title: Studies on the control of high-gain DC-DC boost converters
Authors: Jiang, Wentao
Keywords: DRNTU::Engineering::Electrical and electronic engineering::Power electronics
Issue Date: 2018
Source: Jiang, W. (2018). Studies on the control of high-gain DC-DC boost converters. Doctoral thesis, Nanyang Technological University, Singapore.
Abstract: The traditional dc-dc boost converter has been widely applied in industrial applications. However, due to the parasitic resistance of the inductor and the serious reverse-recovery problem, the voltage gain of this converter is very limited. To solve this problem, many high voltage gain dc-dc boost converters have been proposed in the last decade. The high voltage gain dc-dc boost converters can be roughly separated as isolated converters and non-isolated converters. Since most of the isolated converters suffer from large amount of power losses caused by the leakage current problem, the non-isolated converters are more preferred in applications where electrical isolation is not necessary. Since the non-isolated high voltage gain dc-dc boost converters are generally high-order non-minimum phase systems, it is more difficult to regulate these converters as compared to regulating their traditional counterpart. In this thesis, some studies on the control aspects of such non-isolated high voltage gain dc-dc boost converters are presented. Firstly, the study on how to select the most suitable state variables to design the current-mode controller for the high voltage gain dc-dc converter is presented. For the current-mode control technique, the measurement of the inductor current for feedback purpose is necessary. However, some of the high voltage gain converters, such as the hybrid-type dc-dc boost converter, contain two or more inductors. As such, the issues such as which inductor current is more suitable for the design of the controller should be answered. To address this, a comparative study of the adaptive current-mode controllers for the hybrid-type high-order dc-dc boost converter was carried out. The Routh-Hurwitz stability criterion was used to determine the most suitable inductor current for the controller design. Some simulation as well as experimental results are also presented to verify the theoretical conclusions. Next, the problem of regulation of high-order dc-dc converters using least number of state variables for feedback purposes is addressed. To this end, three output feedback control laws for various non-isolated high voltage gain dc-dc boost-type converters are proposed. In these control laws, only the converter output voltage is required for the feedback purposes. This feature results in that, these control laws are very suitable for the applications where there is a cost limit or power density constraint to accommodate the current sensor. Initially, a voltage-mode controller for a dc-dc multilevel boost converter is presented. Unlike some of the existing voltage-mode controllers for the high-order dc-dc converters, the selection of the controller gains of the proposed controller does not rely on a trial and error approach. Since the proposed controller uses the new structure, the frequency domain method could be used to select the appropriate values for the controller gains to ensure robust stability. As such, it is easier to achieve the desired robust control performance. Next, the development of a novel output feedback control strategy for the positive output super-lift re-lift Luo (POSRL) converter is presented. The main feature of this controller is that, despite the non-minimum phase obstacle presented by the converter, the output voltage is regulated directly. Apart from this, the structure of the proposed controller is such that there is no risk of saturation in the control law due to division by zero, and the “remaining dynamics” for the controlled converter has only one equilibrium point which is always stable. Finally, an improved voltage-mode controller for the quadratic boost dc-dc converter is presented. A new structure for the integral action is adopted in this controller. Since the adopted integrand is bounded by a user-defined constant, the extreme changes in the control signal can be avoided. As such, the proposed controller provides better control performance as compared to its counterparts which use the traditional integral action.
DOI: 10.32657/10220/46278
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:EEE Theses

Files in This Item:
File Description SizeFormat 
ThesisJWT.pdfMain Article3.31 MBAdobe PDFThumbnail

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.