Algorithms for Power Converter Design Optimization

A practical optimization approach for power converters is established which allows conception of a design to meet all powercircuit performance requirements and concurrently to optimize a defined quantity such as weight or losses. In addition to facilitating a cost effective design, the computer-aided approach provides a means to readily assess a) the weight-efficiency tradeoff, b) impacts of converter requirements and component characteristics on a given design, and c) optimum power-system configurations. The following two popular algorithms for nonlinearly constrained optimization are utilized to design the power converter: 1) the sequential unconstrained minimization technique, SUMT Version 4; 2) the ALAG penalty function technique, ALAG5. These algorithms are compared, and suggestions are made for improving the efficiency of the optimization algorithms for power converter design.     

Design Optimization for a Half-Bridge DC-DC Converter

A nonlinear programming technique using the penalty function method is employed for the minimum weight design of a halfbridge dc-dc buck power converter. The cost-effective computeraided approach provides a minimum weight design down to the details of the component level which concurrently meets all power circuit performance requirements. Important design insights and guidelines are obtained to readily assess 1) the tradeoffs between weight and loss as the switching frequency is increased, 2) the optimum operating frequency range over which the total weight/loss is minimized, 3) the impact of device characteristics on the total weight and loss profile.