Hyperparameter Estimation Based on Gaussian Process and its Application in Injection Molding

Jun Yan Ma; Xiao Ping Liao; Wei Xia; Xue Lian Yan

doi:10.4028/www.scientific.net/AMR.328-330.524

Paper Titles

Formation of Side Electrode on Quartz MEMS-Based Double-Ended Tuning Fork Using Enhanced Lift-Off Process
p.507

Drawing and CNC Software System for Light Gage Steel Structure
p.511

Active and Passive Deformation of Large Thin-Walled Workpiece in Heat Treatment Process
p.515

Direct Rapid Manufacturing Technology with Laser for Metal Parts
p.520

Hyperparameter Estimation Based on Gaussian Process and its Application in Injection Molding
p.524

A Research on the Measurement of Friction Coefficient of Plastic Forming by Floating Die Extrusion
p.530

Research on Logistics Management Information System
p.538

A Computer Aided Testing System for an Aero Turbo-Shaft Engine
p.542

Adsorption Characteristics of Superplasticizer with Waste Plexiglas (WPS) onto Cement Particle Surface
p.547

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 328-330Hyperparameter Estimation Based on Gaussian...

Hyperparameter Estimation Based on Gaussian Process and its Application in Injection Molding

Abstract:

As a powerful modeling tool, Gaussian process (GP) employs a Bayesian statistics approach and adopts a highly nonlinear regression technique for general scientific and engineering tasks. In the first step of constructing Gaussian process model is to estimate the best value of the hyperparameter which turned to be used in the second step where a satisfactory nonlinear model was fitted. In this paper, a modified Wolfe line search approach for hyperparameters estimation by maximizing the marginal likelihood based on conjugate gradient method is proposed. And then we analyze parameter correlation according to the value of hyperparameters to control the warpage which is a main defect for a thin shell structure part in injection molding.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 328-330)

Pages:

524-529

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.328-330.524

Citation:

Cite this paper

Online since:

September 2011

Authors:

Jun Yan Ma, Xiao Ping Liao, Wei Xia, Xue Lian Yan

Keywords:

Gaussian Process (GP), Hyperparameter Estimation, Injection Molding, Modified Wolfe Line Search, Parameter Correlation

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] J. Zhou. and L.S. Turng: Advances in Polymer Technology, Vol. 26(2007)No. 2, p.71.

Google Scholar

[2] C. E. Rasmussen and C. K. I. Williams: Gaussian Processes for Machine Learning(the MIT Press, England 2006).

Google Scholar

[3] Davide Anguita, Sandro Ridella, Fabio Rivieccio and Rodolfo Zunino: Neurocomputing, Vol. 55(2003) , p.109.

DOI: 10.1016/s0925-2312(03)00430-2

Google Scholar

[4] Vadim Strijov and Gerhard Wilhelm Weberb: Computers and Mathematics with Applications, Vol. 60(2010) , p.981.

Google Scholar

[5] Daisuke kaji and Sumio Watanabe: Neurocomputing, in formation on journal homepage: www. elsevier. com /locate/neucom.

Google Scholar

[6] Gaohang Yu, Lutai Guan and Zengxin Wei: Applied Mathematics and Computation, Vol. 215(2009), p.3082.

Google Scholar

[7] Fernando di Sciascio and Adriana N. Amicarelli: Computers & Chemical Engineering, vol. 32(2008), p.3264.

Google Scholar

[8] Xiaoping Liao, Xuelian Yan, Wei Xia and Bin Luo: Third International Workshop on Advanced Computational Intelligence(Suzhou, Jiangsu, China, August 25-27, 2010).

DOI: 10.1109/iwaci.2010.5585160

Google Scholar

[9] Aynur O. Aptula and Nina G. Jeliazkova: QSAR &Combinatorial Science, vol. 24(2005), p.385.

Google Scholar