Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Feasibility of computational intelligent techniques for the estimation of spring constant at joint of structural glass plates: a dome-shaped glass panel structure

Feasibility of computational intelligent techniques for the estimation of spring constant at... While serving as a core modelling strategy for explaining physical processes, classical and physics-based modelling is nevertheless associated with weaknesses such as computational burden, time consumption, and the inability to represent chaotic and complicated processes in glass science and engineering. On the other hand, machine learning (ML) models have been known to overcome the preceding limitation, particularly when accurate and reliable estimation is required from experiments, including axial load. The data (2879 instances) obtained from the experiment, including axial load (N) and four different displacements (mm), were pre-processed, and partitioned into 70% calibration and 30% verification. Subsequently, sensitivity analysis was conducted in which model combinations (M1–4) were generated. The model complies with the latest machine learning industry standards and allows low-cost experimental performances on low-powered edge computing devices. Regarding RMSE, GRNN-M2 and NN-M3 emerged as the best models; however, the confidence interval values indicated that they were superior to the best model (95%). The outcomes based on three performance evaluation criteria (NSE, RMSE, and R) depicted that the ML model performed meritoriously and proved reliable for the estimation of K. In comparison, the NN-M4 model outperformed that of SVM-M1, GRNN-M4, and MLR-M4 by 6, 4, and 5%, respectively, with average increments of 5% by all the models. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Glass Structures & Engineering Springer Journals

Feasibility of computational intelligent techniques for the estimation of spring constant at joint of structural glass plates: a dome-shaped glass panel structure

Loading next page...
 
/lp/springer-journals/feasibility-of-computational-intelligent-techniques-for-the-estimation-kzRurRRR7y

References (51)

Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022. corrected publication 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
ISSN
2363-5142
eISSN
2363-5150
DOI
10.1007/s40940-022-00209-6
Publisher site
See Article on Publisher Site

Abstract

While serving as a core modelling strategy for explaining physical processes, classical and physics-based modelling is nevertheless associated with weaknesses such as computational burden, time consumption, and the inability to represent chaotic and complicated processes in glass science and engineering. On the other hand, machine learning (ML) models have been known to overcome the preceding limitation, particularly when accurate and reliable estimation is required from experiments, including axial load. The data (2879 instances) obtained from the experiment, including axial load (N) and four different displacements (mm), were pre-processed, and partitioned into 70% calibration and 30% verification. Subsequently, sensitivity analysis was conducted in which model combinations (M1–4) were generated. The model complies with the latest machine learning industry standards and allows low-cost experimental performances on low-powered edge computing devices. Regarding RMSE, GRNN-M2 and NN-M3 emerged as the best models; however, the confidence interval values indicated that they were superior to the best model (95%). The outcomes based on three performance evaluation criteria (NSE, RMSE, and R) depicted that the ML model performed meritoriously and proved reliable for the estimation of K. In comparison, the NN-M4 model outperformed that of SVM-M1, GRNN-M4, and MLR-M4 by 6, 4, and 5%, respectively, with average increments of 5% by all the models.

Journal

Glass Structures & EngineeringSpringer Journals

Published: Mar 1, 2023

Keywords: Artificial intelligence; Glass science; Spring constant; Axial load; Glass panel structure; Machine learning

There are no references for this article.