中文摘要：

简要介绍了在水文模型参数优选中有代表性的3个算法，即SCE—UA（Shuffled Complex Evolution）、遗传算法和改进的单纯形法，采用基于MATLAB的多次独立的复杂函数全局优化测验，并将3种方法应用于三水源新安江模型参数自动优选，对各优化算法的应用效果进行分析比较，并讨论各算法的收敛速度、计算量以及稳定性和有效性．结果表明SCE—UA综合了遗传算法和单纯形法优点，能够全局一致、快速地收敛到全局最优解，取得了较为满意的结果．

英文摘要：

Three representative optimization methods in hydrologic parameters optimization are briefly introduced; namely SCE-UA （shuffled complex evolution-University of Arizona）, genetic algorithm and revised simplex method. Three optimization methods are tested by optimizing some representative complicated functions based on MATLAB and applied to the automatic parameter calibration of three-component Xinanjiang Model. The application effectiveness of three methods is analyzed and compared. The convergence rate, computational effort, stability as well as efficiency of each algorithm are discussed. The results show that SCE-UA integrating the advantages of genetic algorithm and simplex method can globally, coherently and rapidly converge to optimal solutions and achieve satisfactory results.