河流突发性污染事故反演模型建模及其应用

Inverse modeling for river pollution incident and its application

以污染源识别问题为例,采用时空全域径向基函数配点法建立河流突发性污染事故反演模型,通过污染事故发生地下游断面监测数据来反演污染源的排放历史。数值实验表明,反演模型不仅能准确识别一维和二维河流突发性事故中的污染源历史,以及污染物浓度的时空分布,而且能控制反演解的误差水平与输入测量数据的误差水平保持一致。反演模型应用于2005年11月13日松花江重大水污染事故,根据下游巴彦港断面硝基苯实测浓度准确反演了上游苏家屯断面硝基苯浓度随时间变化过程,反演浓度峰值误差为15.8%。

The inverse model for river pollution incident based on the global space-time MQ collocation method was proposed to identify pollution source,in which the release history at the source location is recovered from the measured concentration at the downstream monitoring sites.The one-and two-dimensional numerical results show that the method not only can recover the history of pollution source,and simulate the concentration distribution,but is also capable of controlling the calculation error of the inverse solution at the same level with the input measurement noise.The model was used to solve Songhuajiang River pollution incident which occurred on November 13,2005,and the result shows that a reasonable estimation of the concentration history of nitrobenzene at the upstream observation site,Sujiatun,can be obtained from the measured data at the downstream observation site,Bayangang.The error of the maximum concentration is about 15.8%.