Optimal design of dynamic experiments for guaranteed parameter estimation

This paper presents, for the first time in literature, an approach and preliminary results for the design of dynamic experiments in the framework of bounded-error (guaranteed) parameter estimation that determines confidence limits on the identified parameters similarly to a posteriori analysis in standard maximum-likelihood (e.g. least-squares) estimation. As in the classical approaches to the design of experiments, being established for other types of estimation, an essential part of the solution procedure is the approximation of the resulting joint-confidence region. In this contribution, we develop and thoroughly analyze the procedure and different ways of achieving an over-approximation of the set resulting from guaranteed parameter estimation based on the expected values of parameters. Finally we solve the problem of the design of experiments as a bilevel program that is regularized in order to be well-posed. We demonstrate our approach on a case study from the domain of chemical engineering.

[1]  Dominique Bonvin,et al.  Dynamic optimization of batch processes: I. Characterization of the nominal solution , 2003, Comput. Chem. Eng..

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

[3]  Johannes P. Schlöder,et al.  Regularizing Bilevel Nonlinear Programs by Lifting , 2013 .

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

[5]  Johannes P. Schlöder,et al.  Estimating Parameters in Optimal Control Problems , 2012, SIAM J. Sci. Comput..

[6]  Christodoulos A. Floudas,et al.  Global Optimization of Nonlinear Bilevel Programming Problems , 2001, J. Glob. Optim..

[7]  Moritz Diehl,et al.  Handling uncertainty in economic nonlinear model predictive control: A comparative case study , 2014 .

[8]  Sven Leyffer,et al.  Local Convergence of SQP Methods for Mathematical Programs with Equilibrium Constraints , 2006, SIAM J. Optim..

[9]  Benoît Chachuat,et al.  Unified framework for the propagation of continuous-time enclosures for parametric nonlinear ODEs , 2015, J. Glob. Optim..

[10]  Moritz Diehl,et al.  CasADi -- A symbolic package for automatic differentiation and optimal control , 2012 .

[11]  Paul I. Barton,et al.  Global solution of bilevel programs with a nonconvex inner program , 2008, J. Glob. Optim..

[12]  Benoît Chachuat,et al.  Guaranteed parameter estimation of non-linear dynamic systems using high-order bounding techniques with domain and CPU-time reduction strategies , 2016, IMA J. Math. Control. Inf..

[13]  B. Srinivasan,et al.  Optimization of a semi-batch reaction system under safety constraints , 1999, 1999 European Control Conference (ECC).

[14]  Paul I. Barton,et al.  Towards global bilevel dynamic optimization , 2009, J. Glob. Optim..

[15]  Eric Walter,et al.  Set inversion via interval analysis for nonlinear bounded-error estimation , 1993, Autom..

[16]  R. Fletcher,et al.  Numerical experience with solving MPECs as NLPs , 2002 .

[17]  Eric Walter,et al.  Guaranteed estimation of the parameters of nonlinear continuous‐time models: Contributions of interval analysis , 2011 .

[18]  Benoît Chachuat,et al.  Guaranteed parameter estimation in nonlinear dynamic systems using improved bounding techniques , 2013, 2013 European Control Conference (ECC).

[19]  J. I.  The Design of Experiments , 1936, Nature.