基于GLUE法的多指标水质模型参数率定方法

A new calibration approach for multi-output water quality model based on GLUE method

目前对水质模型进行参数率定通常利用计算机算法来实现,但由于水质模型日趋复杂的非线性结构往往会导致"异参同效"现象,无法通过单个似然度判断参数的取值是否能够取得真值.为避免这一情况,本文提出了一套基于GLUE法的多目标模型参数率定方法,并以WASP水质模型的应用为例,通过Sobol方法确定模型的敏感参数,并利用DO、CBOD、氨氮、硝态氮4项指标的似然函数对参数同时进行率定.结果表明,该方法既可以有效地避免因追求"过拟合"而造成模型参数取值不当,也可以减小模型参数的不确定性,为具有"异参同效"现象的复杂模型的参数率定工作提供了一个更为可靠的方法.

To avoid error values generated from the parameter calibration of a complex water quality model,a new approach is proposed based on a multiobjective GLUE method. By applying a water quality model WASP,the four indicators of DO,CBOD,ammonia and NO-3are calibrated at the same time to find the true values of the sensitive parameters selected by Sobol method. The result shows that this new approach can effectively avoid error values from＂over fitting＂in single-objective parameter optimization,and reduce the uncertainty of parameters. It is a more accurate approach for avoiding＂equifinality＂in complex model calibration.