College of Earth and Environmental Sciences, Lanzhou University, Chengguan District, Lanzhou, Gansu Province, China.
J Environ Manage. 2013 Jun 15;122:130-6. doi: 10.1016/j.jenvman.2013.03.015. Epub 2013 Apr 10.
There is always uncertainty in any water quality risk assessment. A Monte Carlo simulation (MCS) is regarded as a flexible, efficient method for characterizing such uncertainties. However, the required computational effort for MCS-based risk assessment is great, particularly when the number of random variables is large and the complicated water quality models have to be calculated by a computationally expensive numerical method, such as the finite element method (FEM). To address this issue, this paper presents an improved method that incorporates an artificial neural network (ANN) into the MCS to enhance the computational efficiency of conventional risk assessment. The conventional risk assessment uses the FEM to create multiple water quality models, which can be time consuming or cumbersome. In this paper, an ANN model was used as a substitute for the iterative FEM runs, and thus, the number of water quality models that must be calculated can be dramatically reduced. A case study on the chemical oxygen demand (COD) pollution risks in the Lanzhou section of the Yellow River in China was taken as a reference. Compared with the conventional risk assessment method, the ANN-MCS-based method can save much computational effort without a loss of accuracy. The results show that the proposed method in this paper is more applicable to assess water quality risks. Because the characteristics of this ANN-MCS-based technique are quite general, it is hoped that the technique can also be applied to other MCS-based uncertainty analysis in the environmental field.
在任何水质风险评估中都存在不确定性。蒙特卡罗模拟（MCS）被认为是一种灵活、高效的方法，可以描述这种不确定性。然而，基于 MCS 的风险评估所需的计算工作量很大，特别是当随机变量的数量很大并且必须使用计算成本高的数值方法（例如有限元法（FEM））来计算复杂的水质模型时。为了解决这个问题，本文提出了一种改进的方法，将人工神经网络（ANN）纳入 MCS 中，以提高传统风险评估的计算效率。传统的风险评估使用 FEM 来创建多个水质模型，这可能会很耗时或繁琐。在本文中，ANN 模型被用作迭代 FEM 运行的替代品，因此，必须计算的水质模型数量可以大大减少。以中国黄河兰州段化学需氧量（COD）污染风险为例。与传统的风险评估方法相比，基于 ANN-MCS 的方法可以在不损失准确性的情况下节省大量计算工作量。结果表明，本文提出的方法更适用于评估水质风险。由于这种基于 ANN-MCS 的技术的特点非常通用，因此希望该技术也可以应用于环境领域的其他基于 MCS 的不确定性分析。
