适用于移动平台的河流突发污染溯源算法研究

Algorithm for identifying the point source contaminant in the rivers chemically polluted for the moving water quality monitoring platform

提出了一种适用于小型移动水质检测平台的一维均一地表水单点源污染物溯源新算法,即通过计算污染物在时间域和空间域的质量浓度导数来反推污染物初始排放点的位置、排放时间及污染物总量。首先,通过算例仿真研究表明,此方法具有较高的溯源精度(与理论距离偏差在10%以内)。其次,结合目前移动检测的实际限制(平台长度、取样时间等),对算法中涉及的采样时间、距离等因素的选择进行分析,对移动平台的溯源操作给出了优化建议,以确保有较高的溯源精度。最后,通过实验室水槽缩尺模型试验验证了算法有较高溯源精度(最大偏差在30%以内),从而在原理上证明了该算法适用于特定流域下的移动水质污染溯源。

In this article,we would like to present a novel inversion algorithm for calculating the point source contaminants in the rivers with chemical spills.The said algorithm has been designed and developed to be applied for the moving water quality monitoring platform in case of the need to test the point source pollution.As in the actual conditions,it is often to identify the location of the contaminant＇s source along with the release time and the real quantity of the contaminants involved based on a platform which can be made to float in the river,to detect the pollutant or the contaminant concentration and transferring the results found for the online center to monitor.To be exact,the algorithm has to be used to study an inverse problem based on a classic equation which can be used to work out the pollutant concentration with the different changing time and location in the one-dimension river.And,then,we have arranged to conduct a test over the partial derivatives in time-space domain to identify the source information（such as the location,the quantity and the pollutant release time） directly.In addition,we have managed to apply the said algorithm to the entire actual process to test its availability for the actual targets in hoping to make an evaluation of the method.A series of numerical experiments has been done based on the hypothesis prove that this method performs well in identifying and evaluation of the contaminants,with the relative errors in paneling the pollution source location at a rate no more than 10%.And,then,since the target of this article is to confine our platform in the range of monitoring the mobile water quality,it is necessary to take into consideration the limited size of the platform and sampling time so as to reduce the relative errors based on the numerical experimental results.And,finally,the results of the experiments we have done indicate that the actual relative error of the pollution source location monitoring has been within the 30%,which proves that the method we have proposed is valuable and convenient for its application.