LAPSE:2018.0144
Preprint
LAPSE:2018.0144
Global Deterministic Optimization with Artificial Neural Networks Embedded
July 4, 2018
Artificial neural networks (ANNs) are used in various applications for data-driven black-box modeling and subsequent optimization. Herein, we present an efficient method for deterministic global optimization of ANN embedded optimization problems. The proposed method is based on relaxations of algorithms using McCormick relaxations in a reduced-space [\textit{SIOPT}, 20 (2009), pp. 573-601] including the convex and concave envelopes of the nonlinear activation function of ANNs. The optimization problem is solved using our in-house global deterministic solver MAiNGO. The performance of the proposed method is shown in four optimization examples: an illustrative function, a fermentation process, a compressor plant and a chemical process optimization. The results show that computational solution time is favorable compared to the global general-purpose optimization solver BARON.
Title in Alternate Language
Global deterministische Optimierung von Optimierungsproblemen mit k√ľnstlichen neuronalen Netzwerken
Suggested Citation
Schweidtmann AM, Mitsos A. Global Deterministic Optimization with Artificial Neural Networks Embedded. (2018). LAPSE:2018.0144
Author Affiliations
Schweidtmann AM: [ORCID] [Google Scholar]
Mitsos A: [ORCID] [Google Scholar]
[Login] to see author email addresses.
Version Comments
Original Submission
Record Map
Preprint

LAPSE:2018.0144
This Record
arXiv Record

arXiv:1801.07114
Download
Files
[Download 1v1.pdf] (1.5 MB)
Jul 4, 2018
Pre-Print
License
None Specified
 
Meta
Record Statistics
Record Views
90
Version History
[v1] (Original Submission)
Jul 4, 2018
 
Verified by curator on
Jul 4, 2018
This Version Number
v1
Citations
Most Recent
This Version
URL Here
http://psecommunity.org/LAPSE:2018.0144
 
Original Submitter
ArturSchweidtmann
Links to Related Works
Directly Related to This Work