随机因素对河流水质模型参数最优估值影响

 用多宾斯－BOD－DO（Dobbins－BOD－DO）水质模型，生成真实的无干扰和有干扰的仿真水质数据。在此基础上，用绘制目标函数响应面的方法，分析水质监测数据中的随机噪声对水质模型参数最优估值的影响。分析结果表明，随机噪声干扰使最优估计参数＂漂离＂系统真实参数，并增加最优估计参数的不唯一性问题。

Effects of random data errors on the optimized parameter values for a water quality model

he Dobbins-BOD-DO water quality model was used to generate synthetic error-free water quality data; and random errors were then introduced into them. Based on these data the random errors' effects on the optimized parameters in the water quality model were analysed with response surfaces for objective function. The results show that the random errors"drift"the optimized parameters away from their true values and increase the parameters' non-unity problem.