Optimal experiment design under parametric uncertainty: A comparison of a sensitivities based approach versus a polynomial chaos based stochastic approach

Abstract In order to estimate parameters accurately in nonlinear dynamic systems, experiments that yield a maximum of information are invaluable. Such experiments can be obtained by exploiting model-based optimal experiment design techniques, which use the current guess for the parameters. This guess can differ from the actual system. Consequently, the experiment can result in a lower information content than expected and constraints are potentially violated. In this paper an efficient approach for stochastic optimal experiment design is exploited based on polynomial chaos expansion. This stochastic approach is compared with a sensitivities based approximate robust approach which aims to exploit (higher order) derivative information. Both approaches aim at a more conservative experiment design with respect to the information content and constraint violation. Based on two simulation case studies, practical guidelines are provided on which approach is best suited for robustness with respect to information content and robustness with respect to state constraints.

[1]  Jan Van Impe,et al.  Robust multi-objective optimal control of uncertain (bio)chemical processes , 2011 .

[2]  Graham C. Goodwin,et al.  Robust optimal experiment design for system identification , 2007, Autom..

[3]  N. Wiener.  The Homogeneous Chaos , 1938 .

[4]  Filip Logist,et al.  Guaranteed robust optimal experiment design for nonlinear dynamic systems , 2013, 2013 European Control Conference (ECC).

[5]  Kishalay Mitra,et al.  Multiobjective optimization of an industrial grinding operation under uncertainty , 2009 .

[6]  Sandro Macchietto,et al.  Model-based design of experiments for parameter precision: State of the art , 2008 .

[7]  Sandro Macchietto,et al.  Designing robust optimal dynamic experiments , 2002 .

[8]  Kristel Bernaerts,et al.  Optimal Fed Batch Experiment Design for Estimation of Monod Kinetics of Azospirillumbrasilense: From Theory to Practice , 2007, Biotechnology progress.

[9]  V. Vassiliadis,et al.  Second-order sensitivities of general dynamic systems with application to optimal control problems , 1999 .

[10]  E. Walter,et al.  Robust experiment design via stochastic approximation , 1985 .

[11]  John F. MacGregor,et al.  Robust multi-variable identification: Optimal experimental design with constraints , 2006 .

[12]  Lorenz T. Biegler,et al.  On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming , 2006, Math. Program..

[13]  Richard D. Braatz,et al.  Stochastic nonlinear model predictive control with probabilistic constraints , 2014, 2014 American Control Conference.

[14]  E. Walter,et al.  Robust experiment design via maximin optimization , 1988 .

[15]  M. Wendt,et al.  Nonlinear Chance-Constrained Process Optimization under Uncertainty , 2002 .

[16]  Michael P. H. Stumpf,et al.  Maximizing the Information Content of Experiments in Systems Biology , 2013, PLoS Comput. Biol..

[17]  Eva Balsa-Canto,et al.  Dynamic optimization of chemical and biochemical processes using restricted second order information , 2001 .

[18]  Richard D. Braatz,et al.  Open-loop and closed-loop robust optimal control of batch processes using distributional and worst-case analysis , 2004 .

[19]  Johannes P. Schlöder,et al.  Numerical methods for optimal control problems in design of robust optimal experiments for nonlinear dynamic processes , 2004, Optim. Methods Softw..

[20]  Filip Logist,et al.  Robustifying optimal experiment design for nonlinear, dynamic (bio)chemical systems , 2014, Comput. Chem. Eng..

[21]  Eric Walter,et al.  Identification of Parametric Models: from Experimental Data , 1997 .

[22]  Kurt E. Häggblom,et al.  A New Optimization-Based Approach to Experiment Design for Dynamic MIMO Identification , 2017 .

[23]  Dongbin Xiu,et al.  The Wiener-Askey Polynomial Chaos for Stochastic Differential Equations , 2002, SIAM J. Sci. Comput..

[24]  J. Banga,et al.  Computational procedures for optimal experimental design in biological systems. , 2008, IET systems biology.

[25]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[26]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[27]  Joel A. Paulson,et al.  Arbitrary Polynomial Chaos for Uncertainty Propagation of Correlated Random Variables in Dynamic Systems , 2017 .

[28]  Robert King,et al.  A new approach for robust model predictive control of biological production processes , 2007 .

[29]  Massimiliano Barolo,et al.  A backoff strategy for model‐based experiment design under parametric uncertainty , 2009 .

[30]  Juergen Hahn,et al.  Integrating parameter selection with experimental design under uncertainty for nonlinear dynamic systems , 2008 .

[31]  Sandro Macchietto,et al.  Statistical tools for optimal dynamic model building , 2000 .

[32]  A Kremling,et al.  Optimal experimental design with the sigma point method. , 2009, IET systems biology.

[33]  Kristel Bernaerts,et al.  Multi-objective experimental design for (13)C-based metabolic flux analysis. , 2015, Mathematical biosciences.

[34]  Stefan Streif,et al.  A Probabilistic Approach to Robust Optimal Experiment Design with Chance Constraints , 2014, ArXiv.

[35]  Dries Telen,et al.  A sampling-based stochastic optimal experiment design formulation with application to the Williams-Otto reactor , 2017 .

[36]  Filip Logist,et al.  Optimal experiment design under process noise using riccati differential equations , 2013 .

[37]  Moritz Diehl,et al.  An Augmented Lagrangian Based Algorithm for Distributed NonConvex Optimization , 2016, SIAM J. Optim..

[38]  Carine Jauberthie,et al.  An optimal input design procedure , 2006, Autom..

[39]  Habib N. Najm,et al.  Numerical Challenges in the Use of Polynomial Chaos Representations for Stochastic Processes , 2005, SIAM J. Sci. Comput..

[40]  Z. Nagy,et al.  Distributional uncertainty analysis using power series and polynomial chaos expansions , 2007 .

[41]  Filip Logist,et al.  Dynamic optimization of biological networks under parametric uncertainty , 2016, BMC Systems Biology.

[42]  J. Kiefer,et al.  Optimum Designs in Regression Problems , 1959 .

[43]  Jan Van Impe,et al.  Robust Optimization of Nonlinear Dynamic Systems with Application to a Jacketed Tubular Reactor , 2012 .

[44]  Filip Logist,et al.  Optimal experiment design for dynamic bioprocesses: A multi-objective approach , 2012 .

[45]  Håkan Hjalmarsson,et al.  System identification of complex and structured systems , 2009, 2009 European Control Conference (ECC).

[46]  Dominique Bonvin,et al.  Dynamic optimization of batch processes: II. Role of measurements in handling uncertainty , 2003, Comput. Chem. Eng..

[47]  Rudibert King,et al.  Derivative-free optimal experimental design , 2008 .

[48]  Sandro Macchietto,et al.  The optimal design of dynamic experiments , 1989 .

[49]  Filip Logist,et al.  Approximate robust optimization of nonlinear systems under parametric uncertainty and process noise , 2015 .

[50]  Sebastian Sager,et al.  Sampling Decisions in Optimum Experimental Design in the Light of Pontryagin's Maximum Principle , 2013, SIAM J. Control. Optim..