Abordagem de conversores ressonantes classe e quadrático com aplicação de PSO

Abstract

This dissertation presents the application of the particle swarm optimization algorithm (PSO) to obtain design parameters for the inverting step of a quadratic Class E converter. This converter can be designed by modeling the rectifier step as an equivalent impedance load. As for the inverting step, the decomposition into resonant parameters is performed, as it allows for the normalization of the system. The differential equations that represent the circuit depend on a set of variables that, according to methodologies in the literature, depend on values estimated from figures of merit, for example. As a contribution of this work, a PSO algorithm is used here to select these values, removing the need to arbitrate values for variables and also reducing the human-machine interaction time in the design of circuit variables. Based on the data obtained, a simulation was carried out and subsequent implementation to validate the theoretical results. The practical results were satisfactory and close to the developed model.