Automatic loop shaping in QFT using hybrid optimization and constraint propagation techniques

摘要

An efficient method for automatic loop shaping in Quantitative Feedback Theory (QFT) is proposed. The proposed method uses hybrid optimization and constraint propagation techniques. The hybrid optimization part combines interval global optimization and nonlinear local optimization methods. To supplement the optimization tools, a quick solution approach is developed based on ideas of constraint propagation. The quick solution is found to be quite effective in discarding infeasible controller parameter regions in the optimization search. The proposed overall method is demonstrated on an example taken from the QFT literature where it automatically synthesizes a QFT controller of given structure in quick time. Copyright (c) 2006 John Wiley & Sons, Ltd.