A novel adaptive sampling based methodology for feasible region identification of compute intensive models using artificial neural network
Funding information: Consortium Agreement between Janssen Pharmaceutica, University of Ghent and Rutgers University; U.S. Food and Drug Administration, Grant/Award Numbers: 11695471, 1U01FD006487-01
Identification of feasible region of operations in multivariate processes is a problem of interest in several fields. This is particularly challenging when the process model is black-box in nature and/or is computationally expensive, as analytical solutions are not available and the number of possible model evaluations is limited. An efficient methodology is required to identify samples where the model is evaluated for developing a computationally efficient surrogate model. In this work, an artificial neural network based surrogate model is proposed which is integrated with a statistical-based approach (Jack-knifing) to estimate the variance of the surrogate model prediction. This allows implementation of an adaptive sampling approach where new samples are identified close to the feasible region boundary or in regions of high prediction uncertainty. The proposed approach performs better than a previously published kriging based method for different dimensionality case studies.
|aic17095-sup-0001-supinfo.docxWord 2007 document , 327 KB||
Appendix S1: Supporting Information
Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
- 1, . Operability, Resiliency, and Flexibility: Process Design Objectives for a Changing World. Carnegie-Mellon University: Design Research Center; 1983.
- 2, . Optimal process design under uncertainty. AIChE J. 1983; 29(3): 425-433.
- 3, . Active constraint strategy for flexibility analysis in chemical processes. Comput Chem Eng. 1987; 11(6): 675-693.
- 4, . Determination of operability limits using simplicial approximation. AIChE J. 2002; 48(12): 2902-2909.
- 5, . Feasibility evaluation of nonconvex systems using shape reconstruction techniques. Indus Eng Chem Res. 2005; 44(10): 3638-3647.
- 6, , , , , . An optimization-based framework to define the probabilistic design space of pharmaceutical processes with model uncertainty. Processes. 2019; 7(2): 96.
- 7, , . Computationally efficient black-box modeling for feasibility analysis. Comput Chem Eng. 2010; 34(9): 1515-1521.
- 8, , , . Computationally efficient identification of probabilistic design spaces through application of metamodeling and adaptive sampling. Comput Chem Eng. 2020; 132:106608.
- 9, . Feasibility and flexibility analysis of black-box processes part 1: surrogate-based feasibility analysis. Chem Eng Sci. 2015; 137: 986-1004.
- 10, . A novel feasibility analysis method for black-box processes using a radial basis function adaptive sampling approach. Aiche J. 2017; 63(2): 532-550.
- 11, , , . 110th anniversary: surrogate models based on artificial neural networks to simulate and optimize pressure swing adsorption cycles for CO2 capture. Indus Eng Chem Res. 2019; 58(39): 18241-18252.
- 12. Optimization of Fischer-Tropsch synthesis using neural networks. Chem Eng Technol: Indus Chem-Plant Equipment-Process Eng-Biotechnol. 2006; 29(4): 449-453.
- 13, , , . CFD-based optimization of a flooded bed algae bioreactor. Indus Eng Chem Res. 2012; 52(22): 7181-7188.
- 14, . Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research. J Pharm Biomed Anal. 2000; 22(5): 717-727.
- 15, . Process optimization via neural network metamodeling. Int J Production Economics. 2002; 79(2): 93-100.
- 16, , . A comprehensive review for industrial applicability of artificial neural networks. IEEE Transac Indus Electronics. 2003; 50(3): 585-601.
- 17, . Application-driven sequential designs for simulation experiments: kriging metamodelling. J Operational Res Soc. 2004; 55(8): 876-883.
- 18, . Feasibility analysis of black-box processes using an adaptive sampling kriging-based method. Comput Chem Eng. 2012; 36: 358-368.
- 19, . Modular design optimization using machine learning-based flexibility analysis. J Process Control. 2020; 90: 18-34.
- 20, , . Evolution of concepts and models for quantifying resiliency and flexibility of chemical processes. Comput Chem Eng. 2014; 70: 22-34.
- 21, , , . Adaptive sampling for surrogate modelling with artificial neural network and its application in an industrial cracking furnace. Can J Chem Eng. 2016; 94(2): 262-272.
- 22, , . Efficient global optimization of expensive black-box functions. J Global Opt. 1998; 13(4): 455-492.
- 23. Rasmussen CE. A unifying view of sparse approximate Gaussian process regression. J Machine Learning Res. 2005; 6: 1939-1959.
- 24, , . A survey of adaptive sampling for global metamodeling in support of simulation-based complex engineering design. Struct Multidisciplinary Opt. 2018; 57(1): 393-416.
- 25, , , . Design and analysis of computer experiments. Statistical Sci. 1989; 4(4): 409-423.
- 26, , , . A matlab kriging toolbox. Technical University of Denmark, Kongens Lyngby, Technical Report No IMM-TR-2002-12. 2002.
- 27, , . Learning representations by back-propagating errors. Cognitive Modeling. 1988; 5(3): 1.
- 28, , . Multilayer feedforward networks are universal approximators. Neural Netw. 1989; 2(5): 359-366.
- 29, . Bayesian regularization of neural networks. Artificial Neural Networks. Silver Spring, MD, USA: Springer; 2008: 23-42.
- 30, . Adaptive sequential sampling for surrogate model generation with artificial neural networks. Comput Chem Eng. 2014; 68: 220-232.
- 31, , . Exploration of metamodeling sampling criteria for constrained global optimization. Eng Opt. 2002; 34(3): 263-278.
- 32, , , , . Efficient global optimization for high-dimensional constrained problems by using the kriging models combined with the partial least squares method. Eng Opt. 2018; 50(12): 2038-2053.
- 33, , , et al. Dynamic flowsheet model development and sensitivity analysis of a continuous pharmaceutical tablet manufacturing process using the wet granulation route. Processes. 2019; 7(4): 234.
- 34 FDA U. Guidance for Industry. q8 (r2) Pharmaceutical Development. Silver Spring, MD, USA: Department of Human and Human Services; 2009.
- 35 gFormulatedProducts [computer program]. 2019.