某氨氮污染地下水体抽出 -处理系统优化模拟研究

抽出 -处理系统设计多侧重于考虑修复初期的效率，在修复后期通常效率低下，产生拖尾现象，其优化的关键在于布设的井群系统能否高效抽出受污染的地下水体。利用溶质运移数值模拟可为井群布设和抽水方案优化提供依据。本研究旨在优化我国北方某化肥厂高浓度氨氮污染的地下水体的抽出 -处理修复系统，节约时间和成本。在水文地质调查及氨氮浓度监测的基础上，综合考虑井数、抽水天数和总抽水量三个变量，采用中轴线法与三角形法结合的布井方法，利用GMS软件反复试算，筛选出三种较优抽水方案并进一步模拟优化，最终从中选出最优抽水方案。结果，相比最初方案（方案1），最优方案（方案3）将修复周期缩短了23个月，抽水总量减少了约31.9×104 m3，而抽水井数量仅增加了1口。该模型进行了稳定流水位拟合验证和4期非稳定流实测溶质浓度验证，较符合实际。结果表明，针对抽水井数量不足引起的拖尾问题，关键因素在于合理的井位布设与分阶段的抽水模式。在修复过程中，及时对地下水中污染物进行监测，并随着污染羽变化过程及时调整抽水方案，保证高浓度区一直有抽水井进行较大流量抽水，可有效提高修复效率并缩短修复周期。

Simulation and optimization of a pumping and treating system for the remediation of ammonia polluted groundwater

Many pumping and treating systems tend to consider more about the efficiency of earlier remediation period, and this causes the low efficiency and the towing tail phenomenon at the end of remediation period. Efficiently capturing the flow of contaminated groundwater is the key technology in the pumping and treating remediation system optimization. Solute transport simulation can offer guidance for the design of pumping wells and schemes. Many previous studies focused on the coupling of the model and optimization algorithm, such as the Modular Groundwater Optimizer (MGO), but MGO can only apply to a steady flow model. In this study we choose to conduct a lot of artificial trial calculation by GMS instead of MGO because the transient flow model is more realistic than the steady flow. In addition, we verify the model not only by groundwater level data, but also by solute concentration data at 4 times to make sure the more accurate model. This study is aim to optimize the pumping and treating system, and reduce the period and the cost for the remediation of ammonia contaminated groundwater in a chemical fertilizer plant site in northern China. Based on the hydrogeological survey and ammonia nitrogen concentration monitoring, we arrange the pumping wells in the central axis and on both sides of the contamination plume combining the central axis method and the triangle method. After a lot of trial calculation by GMS, three pumping -and -treating schemes are simulated and evaluated, considering the number of wells, remediation period and the whole volume of sewage. Finally, scheme 3 is thought to be the most optimal scheme. Its remediation period is 1 year and 11 months shorter than that of scheme 1, and its sewage volume is 319 million liter less than that of scheme 1. The simulation results show that reasonable arrangement of wells and pumping mode by stages is the key factor to make sure the contaminant plume to be remediated completely and efficiently. Besides, in a groundwater remediation project, the contaminant concentration in groundwater should be monitored regularly and the simulation should be adjusted accordingly. The remediation scheme can be revised in time to ensure that the high concentration zone always has pumping wells with large flow rate.