Surrogate models of multi-phase flow simulation model for DNAPL-contaminated aquifer remediation
HOU Ze-yu1, WANG Yu2, LU Wen-xi2
1. College of Construction Engineering, Jilin University, Changchun 130000, China;
2. College of New Energy and Environment, Jilin University, Changchun 130021, China
Kriging, support vector regression (SVR), kernel extreme learning machine (KELM) were applied to building the surrogate models of multi-phase flow simulation model, and set pair analysis (SPA) was applied to building ensemble surrogate models. The applicability of different surrogate models was analyzed via a comparison study. kriging model was with the highest accuracy, followed by KELM model and SVR model. Compared with Kriging model, set pair weighted ensemble surrogate model significantly improved the approximation accuracy. The mean of residuals and the mean of relative errors between ensemble model outputs and simulation model outputs were only 0.4009% and 0.5373%, respectively. Furthermore, it only took 1.5s to run the set pair weighted ensemble surrogate model. Replacing the simulation model with an ensemble surrogate model considerably reduced the computational burden of the simulation-optimization process and maintained high computation accuracy for optimizing the DNAPL-contaminated aquifer remediation strategy.
Fernández-Garcia D, Bolster D, Sanchez-Vila X, et al. A Bayesian approach to integrate temporal data into probabilistic risk analysis of monitored NAPL remediation[J]. Advances in Water Resources, 2012,36:108-120.
[2]
Fetter C W. Contaminant hydrogeology (second edition)[D]. New York:Macmillan Publishing Company, 1999.
[3]
王锐.非水相流体在土壤中运移规律研究[D]. 西安:西北农林科技大学, 2004. Wang R. Experimental study on nonaqueous phase liquids transport in soils[D]. Xi-an:Northwest Sci-Tech University of Agriculture and Forestry, 2004.
[4]
Falta R W, Suresh Rao P, Basu N. Assessing the impacts of partial mass depletion in DNAPL source zones:Ⅰ. Analytical modeling of source strength functions and plume response[J]. Journal of Contaminant Hydrology, 2005,78(4):259-280.
[5]
Qin X S, Huang G H, Chakma A, et al. Simulation-based process optimization for surfactant-enhanced aquifer remediation at heterogeneous DNAPL-contaminated sites[J]. Science of the Total Environment, 2007,381(1-3):17-37.
[6]
Delshad M, Pope G A, Sepehrnoori K. A compositional simulator for modeling surfactant enhanced aquifer remediation, 1. formulation[J]. Journal of Contaminant Hydrology, 1996,23(4):303-327.
[7]
秦梓钧,刘保君,张雪,等. COMSOL Multiphysics有限元软件数值模拟气液两相流的可行性研究[J]. 当代化工, 2016,45(5):916-919. Qin Z J, Liu B J, Zhang X, et al. Study on the feasibility of using COMSOL Multiphysics software in numerical simulation of gas-liquid two-phase flow[J]. Contemporary Chemical Industry, 2016, 45(5):916-919.
[8]
张蔚.渗透率空间变异性对重非水相液体运移影响的数值模拟[D]. 南京:南京大学, 2013. Zhang W. Impact of the spatial variation of permeability on the transport fate of dense non-aqueous phase liquids[D]. Nanjing:Nanjing University, 2013.
[9]
郑菲.典型DNAPLs在饱和多孔介质中的运移及修复研究[D]. 南京:南京大学, 2015. Zheng F. Investigation on the migration and remediation of typical DNAPLs in saturated porous media[D]. Nanjing:Nanjing University, 2015.
[10]
Biagi J, Agarwal R, Zhang Z M. Simulation and optimization of enhanced gas recovery utilizing CO2[J]. Energy, 2016,94:78-86.
[11]
Lu W X, Chu H B. Optimization of denser nonaqueous phase liquids-contaminated groundwater remediation based on Kriging surrogate model[J]. Water Practice & Technology, 2013,8(2):304-313.
[12]
Asher M J, Croke B F W, Jakeman A J, et al. A review of surrogate models and their application to groundwater modeling[J]. Water Resources Research, 2015,51(8):5957-5973.
[13]
侯泽宇.地下水DNAPLs污染源反演识别及不确定性分析研究[D]. 长春:吉林大学, 2018. Hou Z Y. Source identification and uncertainty analysis for DNAPLs-contaminated groudwater[D]. Changchun:Jilin University, 2018.
[14]
Huang Y F, Li J B, Huang G H, et al. Integrated simulation optimization approach for real-time dynamic modeling and process control of surfactant-enhanced remediation at petroleum-contaminated sites[J]. Practice Periodical of Hazardous, Toxic, and Radioactive Waste Management, 2003,7(2):95-105.
[15]
He L, Huang G H, Zeng G M, et al. An Integrated simulation, inference, and optimization method for identifying groundwater remediation strategies at petroleum-contaminated aquifers in western Canada[J]. Water Research, 2008,42(10/11):2629-2639.
[16]
Rao S V N. A computationally efficient technique for source identification problems in Three-Dimensional Aquifer Systems using neural networks and simulated annealing[J]. Environmental Forensics, 2006,7(3):233-240.
[17]
Mirghani B Y, Zechman E M, Ranjithan R S. Enhanced simulation-optimization approach using surrogate modeling for solving inverse problems[J]. Environmental Forensics, 2012,13(4):348-363.
[18]
Luo J N, Lu W X, Xin X, et al. Surrogate model application to the identification of an optimal surfactant-enhanced aquifer remediation strategy for DNAPL-contaminated sites[J]. Journal of Earth Science, 2013,24(6):1023-1032.
[19]
Srivastava D, Singh R M. Breakthrough curves characterization and identification of an unknown pollution source in groundwater system using an artificial neural network (ANN)[J]. Environmental Forensics, 2014,15(2):175-189.
[20]
Srivastava D, Singh R M. Groundwater system modeling for simultaneous identification of pollution sources and parameters with uncertainty characterization[J]. Water Resources Management, 2015, 29:4607-4627.
[21]
Zhao Y N, Lu W X, Xiao C N. A Kriging surrogate model coupled in simulation-optimization approach for identifying release history of groundwater sources[J]. Journal of Contaminant Hydrology, 2016,185:51-60.
[22]
宋健,吴剑锋,杨蕴,等.基于含水层DNAPL污染修复替代模型的多目标优化研究[J]. 中国环境科学, 2016,36(11):3390-3396. Song J, Wu J F, Yang Y, et al. A Kriging-based surrogate model for multi-objective optimization of DNAPL-contaminated aquifer remediation[J]. China Environmental Science, 2016,36(11):3390-3396.
[23]
鞠磊.基于多源数据同化的含水层异质性刻画[D]. 杭州:浙江大学, 2018. Ju L. Characterization of aquifer heterogeneity based on multi-source data assimilation[D]. Hangzhou:Zhejiang University, 2018.
[24]
Zhang Y S, Kimberg D Y, Coslett H B, et al. Multivariate lesion-symptom mapping using support vector regression[J]. Human Brain Mapping, 2014,35(12):5861-5876.
[25]
Smola A J, Scholkopf B. A tutorial on support vector regression[J]. Statistics and Computing, 2004,14(3):199-222.
[26]
Hou Z Y, Lu W X, Xue H B, et al. A comparative research of different ensemble surrogate models based on set pair analysis for the DNAPL-contaminated aquifer remediation strategy optimization[J]. Journal of Contaminant Hydrology, 2017,203:28-37.
[27]
Hou Z Y, Lu W X, Chu H B, et al. Selecting parameter-optimized surrogate models in DNAPL-contaminated aquifer remediation strategies[J]. Environmental Engineering Science, 2015,32(12):1016-1026.
[28]
王涵,卢文喜,李久辉,等.地下水DNAPLs污染多相流的随机模拟及其不确定性分析[J]. 中国环境科学, 2018,38(7):2572-2579. Wang H, Lu W X, Li J H, et al. Stochastic simulation and uncertainty analysis of multi-phase flow of groundwater polluted by DNAPLs[J]. China Environmental Science, 2018,38(7):2572-2579.
[29]
范越,卢文喜,欧阳琦,等.基于Kriging替代模型的地下水污染监测井网优化设计[J]. 中国环境科学, 2017,37(10):3800-3806. Fan Y, Lu W X, Ouyang Q, et al. Optimum design of groundwater pollution monitoring well network based on Kriging surrogate model[J]. China Environmental Science, 2017,37(10):3800-3806.
[30]
姜雪. DNAPLs污染含水层修复方案优选及不确定性分析[D]. 长春:吉林大学, 2016. Jiang X. Optimization and uncertainty analysis of DNAPLs contaminated aquifer repair scheme[D]. Changchun:Jilin University, 2016.
[31]
Hou Z Y, Lu W X. Comparative study of surrogate models for groundwater contamination source identification at DNAPL-contaminated sites[J]. Hydrogeology Journal, 2018,26:923-932.
[32]
Müller J, Shoemaker C A. Influence of ensemble surrogate models and sampling strategy on the solution quality of algorithms for computationally expensive black-box global optimization problems[J]. Journal of Global Optimization, 2014,60(2):123-144.
[33]
王文圣,向红莲,李跃清,等.基于集对分析的年径流丰枯分类新方法[J]. 四川大学学报(工程科技版), 2008,40(5):1-6. Wang W S, Xiang H L, Li Y Q, et al. A new approach to annual runoff classification based on set pair analysis[J]. Journal of Sichuan University (Engineering Science Edition), 2008,40(5):1-6.
[34]
Cao Q K, Li L J, Yu B. Application of dynamic set-pair analysis in coal and gas outburst prediction[J]. Journal of Coal Science and Engineering (China), 2008,14(1):77-80.
[35]
Wang W S, Jin J L, Ding J, et al. A new approach to water resources system assessment-set pair analysis method[J]. Science in China Series E:Technological Sciences, 2009,52(10):3017-3023.
[36]
Luo J N, Lu W X. Sobol' sensitivity analysis of NAPL-contaminated aquifer remediation process based on multiple surrogates[J]. Computers & Geosciences, 2014,67:110-116.
[37]
Chang Chih-Chung, Lin Chih-Jen. LIBSVM:a library for support vector machines[EB/Z]. Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm. Accessed on December 22, 2016.
[38]
罗建男.基于替代模型的DNAPLs污染含水层修复方案优选[D]. 长春:吉林大学, 2014. Luo J N. Optimization of remediation strategy of DNAPLs-contaminated aquifer based on surrogate[D]. Changchun:Jilin University, 2014.
