不同建议分布MCMC算法在地下水污染源反演识别中的对比研究

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (1200 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract In order to obtain the optimal proposal distribution by exploring the influences of normal distribution, uniform distribution and random walk of the MCMC algorithms on the inversion identification results of groundwater pollution source, this paper established a simulation model of pollutant transport and a surrogate model using Kriging method with a hypothetical case, and the MCMC algorithm based on normal distribution, uniform distribution and random walk was developed to identify the release history of groundwater pollution sources. The results show that the inversion algorithm with uniform distribution as the proposal distribution has the advantages of high inversion accuracy, good stability and fast convergence speed, which proves to be the most suitable proposal distribution for groundwater pollution sources inversion.

Key words： groundwater pollution source identification Bayesian theory MCMC algorithm proposal distribution

Received: 10 August 2022

PACS:

X523

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	LI Xue-li
	LUO Jian-nan
	LIU Yong

Cite this article:

LI Xue-li,LUO Jian-nan,LIU Yong. Comparative study of different proposal distribution MCMC algorithms in groundwater pollution source identification[J]. CHINA ENVIRONMENTAL SCIENCECE, 2023, 43(4): 1646-1654.

URL:

http://www.zghjkx.com.cn/EN/ OR http://www.zghjkx.com.cn/EN/Y2023/V43/I4/1646

[1]	张宇,吕军,卢文喜,等.基于随机统计方法的地下水污染源反演识别[J]. 科学技术与工程, 2020,20(2):535-543. Zhang Y, Lü J, Lu W X, et al. Groundwater pollution source inverse identification based on random statistical methods[J]. Science Technology and Engineering, 2020,20(2):535-543.
[2]	张江江.地下水污染源解析的贝叶斯监测设计与参数反演方法[D]. 杭州:浙江大学, 2017. Zhang J J. Bayesian monitoring design and parameter inversion for groundwater contaminant source identification[D]. Hangzhou:Zhejiang University, 2017.
[3]	Moghaddam M B, Mazaheri M, Samani J M V. Inverse modeling of contaminant transport for pollution source identification in surface and groundwaters:A review[J]. Groundwater for Sustainable Development, 2021,15:100651.
[4]	张宇.基于随机统计方法的地下水污染源反演识别研究[D]. 长春:吉林大学, 2016. Zhang Y. Groundwater pollution source identification based on random statistical methods[D]. Changchun:Jilin University, 2016.
[5]	王景瑞,胡立堂.地下水污染源识别的数学方法研究进展[J]. 水科学进展, 2017,28(6):943-952. Wang J R, Hu L T. Advances in mathematical methods of groundwater pollution source identification[J]. Advances in Water Science, 2017, 28(6):943-952.
[6]	张双圣,刘汉湖,强静,等.基于贝叶斯公式的地下水污染源及含水层参数同步反演[J]. 中国环境科学, 2019,39(7):2902-2912. Zhang S S, Liu H H, Qiang J, et al. Synchronous inversion of groundwater pollution source and aquifer parameters based on Bayesian formula.[J]. China Environmental Science, 2019,39(7):2902-2912.
[7]	Zhang J J, Zhang Q, Chen D J, et al. Surrogate-based Bayesian Inverse modeling of the hydrological system:An adaptive approach considering surrogate approximation error[J]. Water Resources Research, 2020,56(1):e2019WR025721.
[8]	An Y K, Yan X M, Lu W X, et al. An improved Bayesian approach linked to a surrogate model for identifying groundwater pollution sources[J]. Hydrogeology Journal, 2022,30(2):601-616.
[9]	Bai Y K, Lu W X, Li J H, et al. Groundwater contamination source identification using improved differential evolution Markov chain algorithm[J]. Environmental science and pollution research international, 2021, 29(13):19679-19692.
[10]	Chib S, Greenberg E. Understanding the Metropolis-Hastings Algorithm, The American Statistician, 1995,49(4):327-335.
[11]	Wang H, Jin X. Characterization of groundwater contaminant source using Bayesian method[J]. Stochastic Environmental Research and Risk Assessment, 2013,27:867-876.
[12]	陈平,徐若曦.Metropolis-Hastings自适应算法及其应用[J]. 系统工程理论与实践, 2008,(1):100-108. Cheng P, Xu R X. Metropolis-hastings adaptive algorithm and its application[J]. Systems Engineering-Theory & Practice, 2008,(1):100-108.
[13]	朱嵩,刘国华,王立忠,等.水动力-水质耦合模型污染源识别的贝叶斯方法[J]. 四川大学学报(工程科学版), 2009,41(5):30-35. Zhu S, Liu G H, Wang L Z, et al. A Bayesian approach for the identification of pollution source in water quality model coupled with hydrodynamics[J]. Advanced Engineering Sciences, 2009,41(5):30-35.
[14]	顾文龙,卢文喜,张宇,等.基于贝叶斯推理与改进的MCMC方法反演地下水污染源释放历史[J]. 水利学报, 2016,47(6):772-779. Gu W L, Lu W X, Zhang Y, et al. Reconstructing the release history of groundwater contamination sources based on the Bayesian inference and improved MCMC method[J]. Journal of Hydraulic Engineering, 2016,47(6):772-779.
[15]	闫雪嫚.基于贝叶斯理论的地下水DNAPLs污染源反演识别研究[D]. 长春:吉林大学, 2021. Yan X M. Identification of DNAPLs-contaminated groundwater pollution sources based on Bayes Theory[D]. Changchun:Jilin University, 2021.
[16]	孙策.基于CA-MCMC方法的河流突发水污染溯源研究[D]. 青岛:山东大学, 2020. Sun C. Source identification of river sudden water pollution based on CA-MCMC method[D]. Qingdao:Shandong University, 2020.
[17]	朱嵩,毛根海,程伟平,等.基于贝叶斯推理的水环境系统参数识别[J]. 江苏大学学报(自然科学版), 2007,(3):237-240. Zhu S, Mao G H, Cheng W P, et al. Parameters identification for water environmental system based on Bayesian inference[J]. Journal of Jiangsu University (Nature Science Edition), 2007,(3):237-240.
[18]	Kaipio J, Somersalo E. Statistical and computational inverse problems[J]. Springer-Verlag, New York, 2005,:xvi,339.
[19]	赵永胜.统计反问题的贝叶斯理论与方法研究[D]. 青岛:中国石油大学(华东), 2011. Zhao Y S. Bayesian theory and methods of statistical inverse problems[D]. Qingdao:China University of Petroleum (East China), 2011.
[20]	刘艳杰.参数反演的贝叶斯方法及其应用研究[D]. 青岛:山东理工大学, 2020. Liu Y J. Bayesian method of parameter inversion and its application[D]. Qingdao:Shandong University of Technology, 2020.
[21]	朱嵩.基于贝叶斯推理的环境水力学反问题研究[D]. 杭州:浙江大学, 2008. Zhu S. Research on inverse problems of environmental hydraulics based on Bayesian inference[D]. Hangzhou:Zhejiang University, 2008.
[22]	Sacks J, Welch W J, Mitchell T J, et al. Design and analysis of computer experiments[J]. Statistical Science, 1989,4(4):409-435.
[23]	Jin Y. A comprehensive survey of fitness approximation in evolutionary computation[J]. Soft Computing:A Fusion of Foundations, Methodologies and Applications, 2005,9(1):3-12.
[24]	Luo J N, Lu W X. Comparison of surrogate models with different methods in groundwater remediation process[J]. Journal of Earth System Science, 2014,123(7):1579-1589.
[25]	Ayvaz M T. A linked simulation-optimization model for solving the unknown groundwater pollution source identification problems[J]. Journal of Contaminant Hydrology, 2010,117(1-4):46-59.
[26]	Zhao Y, Lu W, Xiao C. A Kriging surrogate model coupled in simulation-optimization approach for identifying release history of groundwater sources[J]. Journal of Contaminant Hydrology, 2016, 185-186:51-60.
[27]	邢贞相,曲睿卓,赵莹,等.不同替代模型在地下水污染源释放历史反演中适用性研究[J]. 东北农业大学学报, 2018,49(12):59-68. Xing Z X, Qu R Z, Zhao Y, et al. Applicability study of different surrogate models in retrieving release history of groundwater contaminant sources[J]. Journal of Northeast Agricultural University, 2018,49(12):59-68.
[28]	刘金山,夏强.基于MCMC算法的贝叶斯统计方法[M].北京:科学出版社, 2016. Liu J S, Xia Q. The Bayesian statistics methods based on MCMC algorithm[M]. Beijing:Science Press, 2016.