Comparison of parameter optimization algorithms for environmental model
环境模型参数优化方法的比较

Parameters identification is achieved through the minimization of objective function based on model outputs and the observed data. Because of ever increasing complexity of environmental models, there are significant difficulty for conventional optimal methods to present a global optimization. On the contrast, however, direct optimization algorithms are widely developed in recent years due to increasing computer efficiency and show promising applications. Four direct optimal algorithms, i.e. CRS algorithm, SCE _ UA algorithm, SA algorithm and Annealing Simplex algorithm, were thus selected in this paper to compare their performances via case studies.