Framework for calculating Root-Mean-Squares of state variables in power electronic converter models

This thesis introduces a new framework to enhance the simulation performance of switch converters, motivated by the pursuit of more efficient energy conversion. This thesis offers improvements in the cyclic-mode model simulation of converters. Initially, a comprehensive review from the classification and operating principles of switch converters to their modelling methods sets the stage for subsequent analysis. Additionally, the thesis analyses and introduces the design methodologies for different types of switch converters. Subsequently, a cyclic-mode model for a class EF2 inverter is constructed, and the accuracy of this simulation method is verified. A new application of Newton's method to accurately determine the multimode steady-state operating conditions of a switching converter by combining it with cyclic-mode model is then demonstrated, thus extending the applicability of the cyclic-mode model. Furthermore, a novel technique based on cyclic-mode modelling is proposed, which allows obtaining the root-means-square values of switch converter signals without generating waveforms, this greatly reduces the difficulty of finding the exact RMS value of a switching power supply. Finally, this technique, combined with the Newton method, forms a complete simulation method for power supplies. This method accurately and swiftly analyses DC-input power supplies in multiple operating modes, precisely obtaining their signal waveforms, periodic averages, and root-mean-square values.