Orthogonal Optimization Algorithm in Application to Circuit Design - Swarm Intelligence Algorithm Based on the Analysis of Variance Ratio

摘要

This paper put forward a novel intelligent algorithm of orthogonal design for circuit parameters. The intelligent orthogonal optimization differs from conventional orthogonal design. According to variance and variance ratio analysis of the orthogonal design, the next searching direction and range of each variable are determined, which is able to be circulating in the optimization of searching. The intelligent orthogonal optimization is performed until error value of the variance ratio for each variable is zero, namely the calculation is terminated when the variance ratio of each variable is approximately equal, which is the optimal solution. Correspondingly, the variable range of the orthogonal array in preceding step is the tolerance of the optimal solution. The characteristic of this method is no need for special tolerance design. The tolerance of optimal solution is obtained when the parameter design is completed. In contrast with general orthogonal design method and other optimization methods of Genetic algorithm (GA), particle swarm optimization (PSO), and so on, the intelligent orthogonal optimization has the advantage of less calculation time, higher speed, no exiting of prematurity of local circulation and slow convergence of global search. A circuit of the regulated power supply serves as optimization design paradigm in this paper. The calculation process and result show that this method is simple and practical, and also it will have a useful application prospect in the field of circuit design.