MODELLING, ANALYSIS & CONTROL OF SINGLE-INDUCTOR DUAL-OUTPUT DC-DC BUCK CONVERTER SYSTEM

Abstract

The thesis contains a detailed study on the modelling, analysis and control of SIDO DC-DC buck converter system. The system is described in detail by means of state-space and small-signal modelling, leading to a multi-input multi-output transfer function matrix which represents the coupled system dynamics. These indexes, namely, the Relative Gain Array (RGA), the Hankel Interaction Index Array (HIIA) and the H2 norm, are used to systematically quantify the interaction between outputs and demonstrate that the coupling is parameter dependent and frequency dependent, and that the coupling is very sensitive to variations in the load. To overcome the limitations of conventional control approaches, two control strategies are studied: a conventional PID controller based on equivalent Single Input Single-Output (SISO) modelling, and a structured H-infinity based PID controller designed using the full multi-variable system representation. Many simulation studies have been carried out to evaluate the performance of these controllers such as transient response, frequency domain analysis, stability assessment and load and line regulation. From the results obtained, the classical PID controller is easy to implement but it has some drawbacks when it comes to interaction and disturbance rejection and hence a higher overshoot and cross regulation effects is obtained. The structured H-infinity based PID controller, on the contrary gives better damping, settling time, robustness and cross coupling suppression of the outputs. In conclusion, the paper highlights the importance of interaction-aware modelling and strong control design in multi-output DC–DC converters, and it shows how the structured H-infinity based approach is a suitable and effective solution for achieving high performance applications.