EFFICIENT SEISMIC PERFORMANCE ESTIMATION METHOD BY SURROGATE MODELING BASED ON ADAPTIVE KRIGING AND MARKOV CHAIN MONTE CARLO

2020 ◽  
Vol 76 (1) ◽  
pp. 75-86
Author(s):  
Masaru KITAHARA ◽  
Matteo BROGGI ◽  
Michael BEER
Keyword(s):  
Monte Carlo ◽  
Markov Chain ◽  
Markov Chain Monte Carlo ◽  
Seismic Performance ◽  
Estimation Method ◽  
Surrogate Modeling ◽  
Performance Estimation ◽  
Adaptive Kriging

Parameter Estimation of an Electrohydraulic Servo System Using a Markov Chain Monte Carlo Method

2012 ◽  
Vol 135 (1) ◽  
Author(s):  
Junhong Liu ◽  
Huapeng Wu ◽  
Heikki Handroos ◽  
Heikki Haario
Keyword(s):  
Monte Carlo ◽  
Parameter Estimation ◽  
Markov Chain ◽  
Markov Chain Monte Carlo ◽  
Measurement Data ◽  
Estimation Method ◽  
Operating Conditions ◽  
Point Estimate ◽  
Best Fitting ◽  
The One

A parameter estimation method is presented by an example of an electrohydraulic position servo. The method is based on the Markov chain Monte Carlo approach. The method allows utilization of noisy measurement data in identification process, making use of original physical data possible without the requirement of a filter. The method seeks for the best fitting point estimate of the unknown model parameter vector, but the solution to the parameter estimation problem is given as a statistical distribution that contains “all” the possible parameter combinations. The robustness of the model developed with the proposed method is further demonstrated by verification in operating conditions that are independent of each other and the one used in the identification step. Results show that the system model with the hybrid leakage formula for the studied valve describes the system dynamics more precisely and matches the real responses better.

scholarly journals Estimasi Parameter Distribusi Mixture Eksponensial dan Weibull dengan Metode Bayesian Markov Chain Monte Carlo

2019 ◽  
Vol 1 (1) ◽  
pp. 34
Author(s):  
Ulfa Destiarina ◽  
Mustika Hadijati ◽  
Desy Komalasari ◽  
Nurul Fitriyani
Keyword(s):  
Monte Carlo ◽  
Parameter Estimation ◽  
Markov Chain ◽  
Markov Chain Monte Carlo ◽  
Estimation Method ◽  
Mixture Distribution ◽  
Integration Process ◽  
Simulation Data ◽  
Mcmc Estimation

In parameter estimation, sometimes there are several problems that require the completion of a mixture distribution. This study aimed to apply the parameter estimation of exponential and Weibull mixture distribution in simulation data using the Bayesian Markov Chain Monte Carlo (MCMC) estimation method. The results obtained indicate that the analytic calculations of parameter estimation were more accurate than the calculations with the help of software, based on the terms of the suitability of the theory and its integration process.

On Markov Chain Monte Carlo Acceleration

1994 ◽  
Author(s):  
Alan E. Gelfand ◽  
Sujit K. Sahu
Keyword(s):  
Monte Carlo ◽  
Markov Chain ◽  
Markov Chain Monte Carlo
Sign in / Sign up

Export Citation Format

Share Document