Bayesian parameter estimation for a jet-milling model using Metropolis–Hastings and Wang–Landau sampling

Abstract Bayesian parameter estimates for a computationally expensive multi-response jet-milling model are computed using the Metropolis–Hastings and Wang–Landau Markov Chain Monte Carlo sampling algorithms. The model is accompanied by data obtained from 74 experiments at different process settings which is used to estimate the model parameters. The experimentally measured quantities are the 10th, 50th and 90th quantiles of the resulting particle size distributions. Parameter estimation is performed on a population balance jet-milling model composed of three subprocesses: jet expansion, milling and classification. The model contains eight parameters requiring estimation and can compute the same quantities that are determined in the experiments. As the model is computationally expensive to solve, the sampling algorithms are applied to a surrogate model to establish algorithm specific parameters and to obtain model parameter estimates. The resulting parameter estimates are given with a discussion of their reliability and the observed behaviour of the two sampling algorithms. Comparison of the autocorrelation function between samples generated by the two algorithms shows that the Wang–Landau algorithm exhibits more rapid decay. Trace plots of the parameter samples from the two algorithms appear to be analogous and encourage the supposition that the Markov Chains have converged to the distribution of interest. One- and two-dimensional density plots indicate a unimodal distribution for all parameters, which suggests that the obtained estimates are unique. The two-dimensional density plots also suggest correlation between at least two of the model parameters. The realised distribution generated by both algorithms produced consistent results and demonstrated similar behaviour. For the application considered in this work, the Wang–Landau algorithm is found to exhibit superior performance with respect to the correlation and equivalent performance in all other respects.

[1]  L. Tierney.  A note on Metropolis-Hastings kernels for general state spaces , 1998 .

[2]  H.J.C. Gommeren,et al.  Modelling and control of a jet mill plant , 2000 .

[3]  David J. Spiegelhalter,et al.  Introducing Markov chain Monte Carlo , 1995 .

[4]  Handoko Adi,et al.  Use of milling and wet sieving to produce narrow particle size distributions of lactose monohydrate in the sub-sieve range , 2007 .

[5]  Markus Kraft,et al.  Parameter estimation in a multidimensional granulation model , 2010 .

[6]  N Midoux,et al.  Micronization of pharmaceutical substances in a spiral jet mill , 1999 .

[7]  G. Casella,et al.  Explaining the Gibbs Sampler , 1992 .

[8]  Sebastian Mosbach,et al.  Real-Time Evaluation of a Detailed Chemistry HCCI Engine Model Using a Tabulation Technique , 2008 .

[9]  Markus Kraft,et al.  The inverse problem in granulation modeling—Two different statistical approaches , 2011 .

[10]  Walter R. Gilks,et al.  Adaptive Direction Sampling , 1994 .

[11]  Kenneth J. Beers.  Numerical Methods for Chemical Engineering: Preface , 2006 .

[12]  ČernýV..  Thermodynamical approach to the traveling salesman problem , 1985 .

[13]  Michael Frenklach,et al.  Optimization and analysis of large chemical kinetic mechanisms using the solution mapping method—combustion of methane , 1992 .

[14]  A. Fernandez-Pello,et al.  SOOTING BEHAVIOR DYNAMICS OF A NON-BUOYANT LAMINAR DIFFUSION FLAME , 2007 .

[15]  C. Geyer,et al.  Annealing Markov chain Monte Carlo with applications to ancestral inference , 1995 .

[16]  Gareth O. Roberts,et al.  Examples of Adaptive MCMC , 2009 .

[17]  Markus Kraft,et al.  Droplets population balance in a rotating disc contactor: An inverse problem approach , 2006 .

[18]  F. Egolfopoulos,et al.  An optimized kinetic model of H2/CO combustion , 2005 .

[19]  Walter R. Gilks,et al.  Strategies for improving MCMC , 1995 .

[20]  Don W. Green,et al.  Perry's chemical engineers' handbook. 7th ed. , 1997 .

[21]  Tomas Jansson,et al.  Optimization of Draw-In for an Automotive Sheet Metal Part An evaluation using surrogate models and response surfaces , 2005 .

[22]  Lamberto Cesari,et al.  Optimization-Theory And Applications , 1983 .

[23]  A. Gelman,et al.  Weak convergence and optimal scaling of random walk Metropolis algorithms , 1997 .

[24]  Christophe Andrieu,et al.  A tutorial on adaptive MCMC , 2008, Stat. Comput..

[25]  Markus Kraft,et al.  Incorporating experimental uncertainties into multivariate granulation modelling , 2010 .

[26]  John A. Dodds,et al.  Approximate calculation of breakage parameters from batch grinding tests , 1996 .

[27]  Markus Kraft,et al.  Statistical Approximation of the Inverse Problem in Multivariate Population Balance Modeling , 2010 .

[28]  Venkat Venkatasubramanian,et al.  High fidelity mathematical model building with experimental data: A Bayesian approach , 2008, Comput. Chem. Eng..

[29]  Jun S. Liu,et al.  The Wang-Landau algorithm in general state spaces: Applications and convergence analysis , 2010 .

[30]  Sebastian Mosbach,et al.  Iterative improvement of Bayesian parameter estimates for an engine model by means of experimental design , 2012 .

[31]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[32]  Terese Løvås,et al.  Spectral uncertainty quantification, propagation and optimization of a detailed kinetic model for ethylene combustion , 2009 .

[33]  T. Allen.  Particle Size Measurement , 1965, Nature.

[34]  Sebastian Mosbach,et al.  The future of computational modelling in reaction engineering , 2010, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[35]  S. Dreyfus,et al.  Thermodynamical Approach to the Traveling Salesman Problem : An Efficient Simulation Algorithm , 2004 .

[36]  D. Landau,et al.  Efficient, multiple-range random walk algorithm to calculate the density of states. , 2000, Physical review letters.

[37]  Markus Kraft,et al.  Resolving conflicting parameter estimates in multivariate population balance models , 2010 .

[38]  Douglas C. Montgomery,et al.  Response Surface Methodology: Process and Product Optimization Using Designed Experiments , 1995 .

[39]  K. Mengersen,et al.  Use in practice of importance sampling for repeated MCMC for Poisson models , 2010 .

[40]  Don W. Green,et al.  Perry's Chemical Engineers' Handbook , 2007 .

[41]  B. Carlin,et al.  Diagnostics: A Comparative Review , 2022 .