Application Of Optimization Algorithm In Simulating Ammonia Contaminated Groundwater On P&T

Water is the fundamental to human life.Groundwater,as a part of water resources,is widely used in agriculture,industry and domestic due to its stable water supply and less pollution,which has become closely related to human society.Groundwater is an important or major source of water supply in many parts of China.In recent years,ammonia nitrogen pollution is serious,it has a certain toxic effect on human body and aquatic organisms,and it is also a highly oxygen-consuming substance.Once the concentration of ammonia nitrogen reaches a high value,it will directly lead to the deterioration of groundwater.Therefore,it is urgent to solve the problem of ammonia nitrogen pollution of groundwater.In this paper,an ammonia-contaminated groundwater site in Hebei Province was taken as the research object.In order to obtain the geological and hydrogeological conditions of the site,exploration and pumping tests were carried out to obtain the relevant hydrogeological parameters.The experimental data were fitted by numerical model to determine the permeability coefficient of the study area was 7.18m/d,the maximum pumping capacity of a single well was 115m~3/d,and the impact radius of the pumping well was 22m.According to the drilling data and the water level survey data,the boundary conditions,the structure of aquifer,source and sink terms of the study area were generalized,and a hydrogeological conceptual model was established.The Visual MODFLOW was used to establish the groundwater flow model,and the MT3DMS module was used to establish the pollutant solute transport model.The model was adjusted,identified and verified to determine that the established water flow model can accurately reflect the groundwater flow characteristics and pollutant migration in the simulated area.The sensitivity analysis of model parameters shows that the model is not susceptible to the permeability coefficient of sensitive factors and has good stability.After the model parameters were adjusted,the water model and the solute transport model conform to the actual situation of the water flow and pollutant transport in the study area,using the MGO module which in Visual MODFLOW to optimize the arrangement of pumping wells in the simulated area by genetic algorithm.The minimum pumping amount was set as the objective function,and the number of pumping wells and the target treatment concentration were set as the constraints.By comparing the two schemes for optimized design of the pumping wells in six cases,it is concluded that the five wellsB1,B2,B4,B5 and B6 are the optimal pumping scheme with a constant pumping rate of 103.9 m~3/d,74.19m~3/d,22.26m~3/d,55.35 m~3/d,3.71m~3/d pumping 700d respectively,and the optimized results are verified.