Back to results list
Please use this identifier to cite or link to this item:
|Title:||Design of robust and fast controller for high-performance DC-DC converters||Authors:||Ali, Majid||Advisors:||Loo, K. H. (EIE)
Lai, Y. M. (EIE)
|Keywords:||Feedback control systems
|Issue Date:||2019||Publisher:||The Hong Kong Polytechnic University||Abstract:||To provide a well-regulated supply voltage for their operation, dc-dc power converters are widely used in computers, cell phones, telecommunication system, grid-connected energy storage systems, distributed generators, photovoltaic systems, and many others. Fast dynamic response performance is an essential requirement for these dc-dc power converters, and this is achievable with a well-designed feedback controller. Traditional methods for designing feedback controller are based on an accurate small-signal model of the power converter. However, model uncertainties and power converter's parameter variations due to, for example temperature or aging effects, will adversely affect the performance of the designed feedback controller. Another major disadvantage of the conventional feedback control approach is that the feedback controller is required to meet both stability and transient response performance requirements, and these two requirements usually cannot be met with one set of controller parameters. Hence, there is always a tradeoff between stability and dynamic response performance. These factors make the design of high-performance feedback controller a challenging task. To address these issues various adaptive and nonlinear control methods have been proposed. However, these methods are relatively complex and require high-performance microcontroller for there implementation. The principal focus of this work is to develop and implement a parameter insensitive, hardware-efficient and high-performance feedback controller for dc-dc power converters. For illustration, the feedback controller design for non-isolated dc-dc buck converter commonly used for low power applications and the isolated dc-dc dual-active-bridge (DAB) converter commonly used for medium to high power level has been selected for this study. A novel non-intrusive method for online estimation of power-stage parameters and autotuning of the feedback controller for buck converter is presented in the first part of this research work. The autotuning controller presented has been specially developed to handle wide-range variations in the resonant frequency of the LoCo output filter, equivalent series resistance (ESR) zero of the output capacitor, and input voltage. The proposed controller accurately estimates the resonant frequency ωo and the ESR zero frequency ωESR by examining the converter's startup transient and by online measurement of output voltage ripple. As a result, the proposed method is non-intrusive and does not disturb the converter's normal operation or output voltage regulation. Based on these results, the digital compensator is automatically tuned based on some user-defined phase margin and crossover frequency to provide the desired transient response and output voltage regulation over wide-range variations of power-stage parameters and operating points. It has been demonstrated by experimental results that the presented controller can maintain absolute stability and consistent transient performance over a wide-range of power-stage parameters and operating points of the converter.
In the second part of this research work, in order to enhance the transient response of the front-end DAB converter and reduce the amplitude of the second-order ripple component in the dc-bus voltage of a cascaded converter system comprising a DAB dc-dc converter followed by a single-phase inverter, a disturbance observer based dc-bus voltage control has been proposed. In the proposed method, a disturbance observer is designed based on minimal plant information and is combined with a proportional controller to form a decoupled composite controller for dc-bus voltage control. The input to the disturbance observer is the sensed dc-bus voltage and the control signal. Hence, the proposed solution does not require additional voltage or current sensor. Using this information, the disturbance observer can accurately estimate and compensate for the lumped disturbance which includes external disturbances, model uncertainties, and circuit parameter variations due to temperature and aging effects. With these disturbances accounted for and their effects compensated, superior control performance can always be achieved. In the final part of this research work, in order to overcome the bandwidth limitation of conventional feedback control methods, an uncertainty and disturbance estimator (UDE) assisted sensorless load current feedforward control scheme for two-stage single-phase inverter system has been proposed. The proposed control method consists of a feedforward path, an UDE, and a voltage feedback loop. The load current is estimated from the estimated average current of the DAB's secondary bridge, and lossless sensing of the dc-link capacitor current is achieved using a digital filter, and the estimated load current is fed forward to achieve fast dynamic response. However, the estimation accuracy of the load current and calculation of the optimum feedforward gain depends on the values of circuit parameters which may be not known with high precision. Hence, an UDE is used to compensate for model uncertainties and parameters variations, and the voltage feedback loop is designed to ensure good converter's stability. As compared to traditional single-loop voltage-mode control or voltage-mode control with load current feedforward, the proposed UDE assisted current sensorless load current feedforward control scheme results in an improved dynamic response performance and reduced dc-bus voltage ripple.
|Description:||xxv, 138 pages : color illustrations
PolyU Library Call No.: [THS] LG51 .H577P EIE 2019 Ali
|URI:||http://hdl.handle.net/10397/81469||Rights:||All rights reserved.|
|Appears in Collections:||Thesis|
Show full item record
Files in This Item:
|991022270854803411_link.htm||For PolyU Users||168 B||HTML||View/Open|
|991022270854803411_pira.pdf||For All Users (Non-printable)||4.11 MB||Adobe PDF||View/Open|
Citations as of Dec 4, 2019
Citations as of Dec 4, 2019
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.