利用贝叶斯推理估计二维含源对流扩散方程参数

朱嵩; 刘国华; 毛根海

您当前的位置：

首页 >

文章列表页 >

利用贝叶斯推理估计二维含源对流扩散方程参数

水利与土木工程 | 更新时间：2023-11-28

- 利用贝叶斯推理估计二维含源对流扩散方程参数
- Application of Bayesian Inference to Estimate the Parameters in 2D Convection diffusion Equation with Source
- 工程科学与技术 2008年40卷第2期页码：38-43
- 作者机构：
  
  浙江大学水利与海洋工程学系浙江,杭州,310027
- 作者简介：
- 基金信息：
  
  国家自然科学基金资助项目(50609024);973课题资助项目(2005CB724202)
- DOI：
  中图分类号：
- 纸质出版日期：2008，
扫描看全文
朱嵩,刘国华,毛根海.利用贝叶斯推理估计二维含源对流扩散方程参数[J].工程科学与技术,2008,40(2):38-43.

Application of Bayesian Inference to Estimate the Parameters in 2D Convection diffusion Equation with Source[J]. Advanced Engineering Sciences, 2008,40(2):38-43.
朱嵩,刘国华,毛根海.利用贝叶斯推理估计二维含源对流扩散方程参数[J].工程科学与技术,2008,40(2):38-43. DOI：

Application of Bayesian Inference to Estimate the Parameters in 2D Convection diffusion Equation with Source[J]. Advanced Engineering Sciences, 2008,40(2):38-43. DOI：

摘要

中文摘要: 为了克服观测数据的不确定性给参数反演带来的困难，利用贝叶斯推理建立了二维含源对流扩散方程参数估计的数学模型。通过贝叶斯定理，获得了模型参数的后验分布，从而获得反问题的解。对于多参数反演问题，基于数值解计算得到的参数后验分布很难直观地表现出来，采用马尔科夫链蒙特卡罗方法对参数的后验分布进行采样，获得了扩散系数和降解系数的估计值。研究了观测点位置对计算结果的影响；同时研究了似然函数的形式对估计结果的影响，结果表明在异常值可能出现时采用Laplace分布型的似然函数可以获得稳健估计。对不同观测点数目下的估计值进行了对比，认为对于二维稳态对流扩散方程的双参数估计问题，至少需要两个观测点才有可能得到合理的解。

Abstract

Abstract:To overcome the parameter inversion’s difficulty from the measurement uncertainty

mathematical model based on Bayesian inference was set up to estimate the coefficients in 2D convection diffusion equation with source. By Bayes’ theorem

the posterior distribution of model parameters was obtained

which is the solution of the inverse problem. While for multi parameter inversion problem

the posterior distribution obtained by numerical computation is hard to express easily

Markov chain Monte Carlo method is used to explore the posterior distribution to get the estimation. The effect of the observation position on inversion results was investigated and the form of the likelihood function’s effect was also studied. It indicates that the Laplacian distribution can lead to a robust estimation. Comparing the estimates under different number of measurement points

it can be concluded that at least two points are needed to ensure reasonable solution for the two parameters estimation problem of 2D steady convection diffusion equation.

关键词

贝叶斯推理对流扩散方程马尔科夫链蒙特卡罗模拟环境水力学稳健估计

Keywords

Bayesian inferenceconvection diffusion equationMarkov chain Monte Carlo simulationenvironmental hydraulicsrobust estimation

references

Cao Shuyou.Status quo and tasks of basic research in hydraulics and fluvial dynamics[J].Journal of Sichuan University:Engineering Science Edition,2002,34(1):1-5.[曹叔尤.水力学及河流动力学基本问题研究的现状与任务[J].四川大学学报:工程科学版,2002,34(1):1-5.]

Dai Huichao,Wang Lingling.Study and application of the inverse problem on hydraulics[J].Journal of Sichuan University:Engineering Science Edition,2006,38(1):15-19.[戴会超,王玲玲.工程水力学反问题研究及应用[J].四川大学学报:工程科学版,2006,38(1):15-19.]

Ming Tao,Zhou Xiaode,Feng Mingquan.Inverse problem of multiple parameters identification for water quality model using evolutionary algorithm[J].Journal of Hydraulic Engineering,2003(10):119-123.[闵涛,周孝德,冯民权.河流水质多参数识别反问题的演化算法[J].水利学报,2003(10):119-123.]

Xu Min,Zeng Guangming,Xie Gengxin,et al.Parameter estimation of river water quality model based on real-coded genetic algorithm[J].Journal of Hunan University:Natural Sciences,2004,31(5):41-45.[徐敏,曾光明,谢更新,等.基于实码遗传算法的河流水质模型的参数估计[J].湖南大学学报:自然科学版,2004,31(5):41-45.]

Wang Wei,Zeng Guangming,He Li.Estimation of water quality model parameters with simulated annealing algorithm[J].Shuili Xuebao,2004(6):61-67[王薇,曾光明,何理.用模拟退火算法估计水质模型参数[J].水利学报,2004(6):61-67.]

Huang Minghai,Qi Erong,Wu Jian.A genetic-gradient algorithm for parameter optimization of water quality mathematical model[J].Acta Scientiae Circumstantiae,2002,22(3):315-319.[黄明海,齐鄂荣,吴剑.遗传梯度法在水质数学模型参数估值的应用[J].环境科学学报,2002,22(3):315-319.]

Charles L K,Igor Y,Keith N.Solving inverse initial-value,boundary-value problems via genetic algorithm[J].Engineering Applications of Artificial Intelligence,2000(13):625-633.

Jari K,Erkki S.Statistical and computational inverse problem[M].New York:Springer,2005.

Christophe A,Nando D F,Arnaud D,et al.An introduction to MCMC for machine learning[J].Machine Learning,2003(50):5-43.

浏览量

456

下载量

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

水动力-水质耦合模型污染源识别的贝叶斯方法

基于贝叶斯推理的标准k－ε湍流模型参数识别

基于CUDA的并行改良随机抽样一致性算法