Analysis and Design of Quasi Resonant Buck Converter

Quasi Resonant Buck Converter is analysed using Generalized State Space Averaging method to obtain transfer function. A PI control algorithm is designed to obtain the closed-loop performance of 54V, 2.916kW Converter with respect to line and load regulation. Evaluation of output is performed by MATLAB software at 200 kHz and results are depicted for five operating conditions where peak overshoot and settling time are used to measure the dynamic performance of converter. It is verified that PI controller optimized for a small-signal transient at an operating condition offers good dynamic performance; however, for large-signal disturbances, it is observed that Fuzzy control yields good rejection ability for supply and load disturbances. However, standard method doesn't exist for transformation of expert's knowledge into rule and database of a fuzzy system and necessity arises to use an algorithm on Neuro-fuzzy integrated system. It is verified that Neuro controller is superior to PI or Fuzzy controller resulting in good transient response under various operating conditions. Experimental results for PI control algorithm are in agreement with simulation.

Autorentext

A.Rameshkumar received his B.E Degree in Electrical Engineering in April 1988 and M.E Degree in Applied Electronics in 1996 from Bharathiar University, Coimbatore. He obtained his Doctoral degreein Electrical Engineering. He is in teaching since 1988 and is presently the Principal at Surendra Institute Of Engineering and Management, Siliguri.