Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.


Hybrid modeling and optimal control of a step-up DC-DC converter


Michèle J. Arnold

Semester/Bachelor Thesis, SS 05

1.1 Problem Formulation
Switch-mode dc-dc step-up converters are intrinsically hybrid and therefore hard to control using classic control methods. The control objective is the regulation of the output voltage despite changes in the input voltage and the output load. The difficulties in controlling dc-dc step-up converters arise from their hybrid nature and the present constraints. The basic approach is averaging of the switched dynamics and therefore bypass the hybrid nature of the system. The currently used controllers are linear controllers, such as PI-type controllers, that are designed based on a linearization around an operating point. A new approach is to derive a model that captures the hybrid characteristic of the system. A control scheme which incorporates the hybrid system and the contstrains needs to be investigated.

1.2 Methods
To capture the hybrid characteristic, an exact discrete-time model is implemented.Model Predictive Control is used to regulate the output voltage by implementing an online optimization scheme. Using Model Predictive Control, the hybrid characteristics can be incorporated and constraints are handled explicitly. The control moves are obtained by minimizing an objective function which includes the control objectives. The physics of the dc-dc step-up converter implicate two specific characteristics which need to be incorporated into the objective function. To minimize the objective function, a descent method is implemented. Two different algorithms to compute the search direction are applied.

1.3 Results
The non linear model based on the exact discrete-time model is a very accurate way to capture the hybrid nature of the dc-dc step-up converter. As the main control objective is the regulation of the output voltage in the presence of input voltage and output load variations, the online optimization scheme is appropriate to achieve the control objective. The online optimization scheme allows for online adjustment of the model parameters. Two final objective functions are obtained, which incorporate the specific characteristics. Simulation results show the balancing to occurring changes in the input voltage and the output load.

1.4 Outlook
The mathematical parameterization leads to expressions too large to handle. A different approach needs to be investigated to parameterize the infliction point line and to adjust the current limiting accordingly. The derived approaches assume the output load to be known. This is never the case in reality. Therefore, an estimation scheme needs to be implemented, by applying a standard Kalman Filter. To implement the optimization scheme in a real plant, a very efficient and tailored optimization routine needs to be developed.

Supervisors: T. Geyer, Dr. G. Papafotiou, Prof. M. Morari


Type of Publication:

(13)Semester/Bachelor Thesis

M. Morari

No Files for download available.
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2005:IFA_2794
Permanent link