Optimal Water Pollution Control in River Streams

不同的两个模拟/优化模型有用的河流水质保护已经开发，可以提供最佳的战略控制污染物负荷从点源沿着流。一种是确定性优化模型，可以被视为线性规划问题，而另一种是随机优化模型（鲁棒优化模型），考虑了问题的数学公式中涉及的参数的不确定性。根据模型的特点，将水流动力学和水质动力学作为等式约束条件进行优化，这两个模型都能够在水质标准或目标水质（BOD和DO）的约束下，确定河流允许的最大污染物负荷及其沿沿着的空间分布。为了增强模型的通用性，采用有限元方法对流动和输运进行了模拟（BOD和DO）方程转化为代数形式的等式约束。该确定性模型适用于具有结点的复杂环状和/或辫状网络流，并且能够确定从多个点源加载的污染物的允许最大量以及其对这些源中的每一个的最优分配。模型能将河流流量、水深、断面面积、水温、风速、盐度等参数的不确定性反映到优化解中，从而提供了一些可行的解决方案。因此，它是有用的，制定一些替代战略的河流水质管理，决策者可以选择一个有效的。

Different two simulation/optimization models useful for conservation of water quality in river streams have been developed that can provide optimal strategies for controlling pollutant loads from point sources along the stream. One is a deterministic optimization model that can be regarded as a linear programming problem, while the other is a stochastic optimization model (robust optimization model) that considers uncertainties of the parameters involved in the mathematical formulation of the problem. The results are summarized in terms of the characteristics of the models as follows.Taking into account the dynamics of water motion and water quality as equality constraints in the optimization process, both of the models developed are capable of determining allowable maximum pollutant loads and their spatial allocation along the streams under the constraints of the water quality standards or targeted water qualities given for BOD and DO.In order to enhance the versatility of the models, the finite element method is employed to cast the flow and transport (BOD and DO) equations of partial differential form into the equality constraints of algebraic form.The deterministic model is applicable to the complex looped and/or braided network streams with junctions, and enables determination of the allowable maximum amount of pollutants loaded from multiple point sources as well as its optimal allocation to each of these sources.The stochastic (robust) model can reflect uncertainties of the parameters such as river discharge, water depth, cross-sectional area, water temperature, wind speed and salinity to the solution of optimization, thus providing some feasible solutions. It is therefore useful for making up some alternative strategies for stream water quality management from which decision makers can select an efficacious one.