This paper proposes a new technique based on inverse Markov chain Monte Carlo algorithm for finding the smallest generalized eigenpair of the large scale matrices. Some numerical examples show that the proposed method...This paper proposes a new technique based on inverse Markov chain Monte Carlo algorithm for finding the smallest generalized eigenpair of the large scale matrices. Some numerical examples show that the proposed method is efficient.展开更多
This paper considers the parameters and reliability characteristics estimation problem of the generalized Rayleigh distribution under progressively Type-Ⅱ censoring with random removals,that is,the number of units re...This paper considers the parameters and reliability characteristics estimation problem of the generalized Rayleigh distribution under progressively Type-Ⅱ censoring with random removals,that is,the number of units removed at each failure time follows the binomial distribution.The maximum likelihood estimation and the Bayesian estimation are derived.In the meanwhile,through a great quantity of Monte Carlo simulation experiments we have studied different hyperparameters as well as symmetric and asymmetric loss functions in the Bayesian estimation procedure.A real industrial case is presented to justify and illustrate the proposed methods.We also investigate the expected experimentation time and discuss the influence of the parameters on the termination point to complete the censoring test.展开更多
Estimation for the parameters of the generalized logistic distribution (GLD) is obtained based on record statistics from a Bayesian and non-Bayesian approach. The Bayes estimators cannot be obtained in explicit forms....Estimation for the parameters of the generalized logistic distribution (GLD) is obtained based on record statistics from a Bayesian and non-Bayesian approach. The Bayes estimators cannot be obtained in explicit forms. So the Markov chain Monte Carlo (MCMC) algorithms are used for computing the Bayes estimates. Point estimation and confidence intervals based on maximum likelihood and the parametric bootstrap methods are proposed for estimating the unknown parameters. A numerical example has been analyzed for illustrative purposes. Comparisons are made between Bayesian and maximum likelihood estimators via Monte Carlo simulation.展开更多
概率密度演化方法(probability density evolution equation,PDEM)为非线性随机结构的动力响应分析提供了新的途径.通过PDEM获得结构响应概率密度函数(probability density function,PDF)的关键步骤是求解广义概率密度演化方程(generali...概率密度演化方法(probability density evolution equation,PDEM)为非线性随机结构的动力响应分析提供了新的途径.通过PDEM获得结构响应概率密度函数(probability density function,PDF)的关键步骤是求解广义概率密度演化方程(generalized probability density evolution equation,GDEE).对于GDEE的求解通常采用有限差分法,然而,由于GDEE是初始条件间断的变系数一阶双曲偏微分方程,通过有限差分法求解GDEE可能会面临网格敏感性问题、数值色散和数值耗散现象.文章从全局逼近的角度出发,基于Chebyshev拟谱法为GDEE构造了全局插值格式,解决了数值色散、数值耗散以及网格敏感性问题.考虑GDEE的系数在每个时间步长均为常数,推导了GDEE在每一个时间步长内时域上的序列矩阵指数解.由于序列矩阵指数解形式上是解析的,从而很好地克服了数值稳定性问题.两个数值算例表明,通过Chebyshev拟谱法结合时域的序列矩阵指数解求解GDEE得到的结果与精确解以及Monte Carlo模拟的结果非常吻合,且数值耗散和数值色散现象几乎可以忽略.此外,拟谱法具有高效的收敛性且序列矩阵指数解不受CFL (Courant-Friedrichs-Lewy)条件的限制,因此该方法具有良好的数值稳定性和计算效率.展开更多
In this paper we propose a general framework for the uncertainty quantification of quantities of interest for high-contrast single-phase flow problems.It is based on the generalized multiscale finite element method(GM...In this paper we propose a general framework for the uncertainty quantification of quantities of interest for high-contrast single-phase flow problems.It is based on the generalized multiscale finite element method(GMsFEM)and multilevel Monte Carlo(MLMC)methods.The former provides a hierarchy of approximations of different resolution,whereas the latter gives an efficient way to estimate quantities of interest using samples on different levels.The number of basis functions in the online GMsFEM stage can be varied to determine the solution resolution and the computational cost,and to efficiently generate samples at different levels.In particular,it is cheap to generate samples on coarse grids but with low resolution,and it is expensive to generate samples on fine grids with high accuracy.By suitably choosing the number of samples at different levels,one can leverage the expensive computation in larger fine-grid spaces toward smaller coarse-grid spaces,while retaining the accuracy of the final Monte Carlo estimate.Further,we describe a multilevel Markov chain Monte Carlo method,which sequentially screens the proposal with different levels of approximations and reduces the number of evaluations required on fine grids,while combining the samples at different levels to arrive at an accurate estimate.The framework seamlessly integrates the multiscale features of the GMsFEM with the multilevel feature of the MLMC methods following the work in[26],and our numerical experiments illustrate its efficiency and accuracy in comparison with standard Monte Carlo estimates.展开更多
文摘This paper proposes a new technique based on inverse Markov chain Monte Carlo algorithm for finding the smallest generalized eigenpair of the large scale matrices. Some numerical examples show that the proposed method is efficient.
基金supported by the National Statistical Science Research Project of China(2019LZ32)
文摘This paper considers the parameters and reliability characteristics estimation problem of the generalized Rayleigh distribution under progressively Type-Ⅱ censoring with random removals,that is,the number of units removed at each failure time follows the binomial distribution.The maximum likelihood estimation and the Bayesian estimation are derived.In the meanwhile,through a great quantity of Monte Carlo simulation experiments we have studied different hyperparameters as well as symmetric and asymmetric loss functions in the Bayesian estimation procedure.A real industrial case is presented to justify and illustrate the proposed methods.We also investigate the expected experimentation time and discuss the influence of the parameters on the termination point to complete the censoring test.
文摘Estimation for the parameters of the generalized logistic distribution (GLD) is obtained based on record statistics from a Bayesian and non-Bayesian approach. The Bayes estimators cannot be obtained in explicit forms. So the Markov chain Monte Carlo (MCMC) algorithms are used for computing the Bayes estimates. Point estimation and confidence intervals based on maximum likelihood and the parametric bootstrap methods are proposed for estimating the unknown parameters. A numerical example has been analyzed for illustrative purposes. Comparisons are made between Bayesian and maximum likelihood estimators via Monte Carlo simulation.
文摘概率密度演化方法(probability density evolution equation,PDEM)为非线性随机结构的动力响应分析提供了新的途径.通过PDEM获得结构响应概率密度函数(probability density function,PDF)的关键步骤是求解广义概率密度演化方程(generalized probability density evolution equation,GDEE).对于GDEE的求解通常采用有限差分法,然而,由于GDEE是初始条件间断的变系数一阶双曲偏微分方程,通过有限差分法求解GDEE可能会面临网格敏感性问题、数值色散和数值耗散现象.文章从全局逼近的角度出发,基于Chebyshev拟谱法为GDEE构造了全局插值格式,解决了数值色散、数值耗散以及网格敏感性问题.考虑GDEE的系数在每个时间步长均为常数,推导了GDEE在每一个时间步长内时域上的序列矩阵指数解.由于序列矩阵指数解形式上是解析的,从而很好地克服了数值稳定性问题.两个数值算例表明,通过Chebyshev拟谱法结合时域的序列矩阵指数解求解GDEE得到的结果与精确解以及Monte Carlo模拟的结果非常吻合,且数值耗散和数值色散现象几乎可以忽略.此外,拟谱法具有高效的收敛性且序列矩阵指数解不受CFL (Courant-Friedrichs-Lewy)条件的限制,因此该方法具有良好的数值稳定性和计算效率.
基金Y.Efendiev’s work is partially supported by the U.S.Department of Energy Office of Science,Office of Advanced Scientific Computing Research,Applied Mathematics program under Award Number DE-FG02-13ER26165 and the DoD Army ARO ProjectThe research of B.Jin is partly supported by NSF Grant DMS-1319052.
文摘In this paper we propose a general framework for the uncertainty quantification of quantities of interest for high-contrast single-phase flow problems.It is based on the generalized multiscale finite element method(GMsFEM)and multilevel Monte Carlo(MLMC)methods.The former provides a hierarchy of approximations of different resolution,whereas the latter gives an efficient way to estimate quantities of interest using samples on different levels.The number of basis functions in the online GMsFEM stage can be varied to determine the solution resolution and the computational cost,and to efficiently generate samples at different levels.In particular,it is cheap to generate samples on coarse grids but with low resolution,and it is expensive to generate samples on fine grids with high accuracy.By suitably choosing the number of samples at different levels,one can leverage the expensive computation in larger fine-grid spaces toward smaller coarse-grid spaces,while retaining the accuracy of the final Monte Carlo estimate.Further,we describe a multilevel Markov chain Monte Carlo method,which sequentially screens the proposal with different levels of approximations and reduces the number of evaluations required on fine grids,while combining the samples at different levels to arrive at an accurate estimate.The framework seamlessly integrates the multiscale features of the GMsFEM with the multilevel feature of the MLMC methods following the work in[26],and our numerical experiments illustrate its efficiency and accuracy in comparison with standard Monte Carlo estimates.
