突发性水污染事件溯源方法

On source identification method for sudden water pollution accidents

为快速准确地求解突发性水污染溯源问题,在微分进化与蒙特卡罗基础上提出了一种新的溯源方法。该方法将溯源问题视为贝叶斯估计问题,推导出污染源强度、位置和排放时刻等未知参数的后验概率密度函数;结合微分进化和蒙特卡罗模拟方法对后验概率分布进行采样,进而估计出这些未知参数,确定污染源项。通过算例与贝叶斯-蒙特卡罗方法进行对比,结果表明:该方法可使迭代次数有效缩减3/4,污染源强度、位置和排放时刻的平均相对误差分别减少1.23%、2.23%和4.15%,均值误差分别降低0.39%、0.83%和1.49%,其稳定性和可靠性明显高于贝叶斯-蒙特卡罗方法,能较好地识别突发性水污染源,为解决突发水污染事件中的追踪溯源难点问题提供了新的思路和方法。

In order to solve the source identification problem of sudden water pollution accident accurately and quick- ly, a method based on the Differential Evolution and Markov Chain Monte Carlo （MCMC） is presented. First, the problem is considered as a Bayesian estimation problem, and the posterior probability distribution of the unknown pa- rameters that include source＇s position, intensity and events＇ initial time are deduced with Bayesian inference. Second, these unknown parameters are estimated by sampling the posterior probability distribution using the Differential Evolu- tion algorithm and Markov Chain Monte Carlo simulation, and the sources are further identified. To test the effective- ness and accuracy of the proposed method, numerical experiments are carried out, and the model result is compared to that of the Bayesian-MCMC method. The conclusions are as following： three fourth of the iterations can be reduced, the average relative error of the source＇s position, intensity and events initial time are reduced 1.23%, 2.23% and 4. 15%, their mean errors are decreased 0.39% ,0.83% and 1.49% by using the proposed method. The latter is thus more stable and robust than the Bayesian-MCMC method, and is able to identify the sudden water pollution accidents＇ source effectively. Therefore, this study provides a new approach and method to solve the difficult traceability problem of sudden water pollution accidents.