Surrogate Modelling Using Phantom-Space Design of Experiments and Adaptive Regression

Electrical and Control Technologies 2011: Proceedings of the 6th International Conference on Electrical and Control Technologies (ECT-2011) 2011
Gints Jēkabsons, Andis Lagzdiņš, Jurijs Lavendels

In many industrial design applications, cutting down the cost of natural experiments or complex simulations involves use of so-called surrogate models that mimic the behaviour of the original system as closely as possible while being much cheaper to evaluate. This includes minimizing the necessary number of sample points and utilizing an efficient modelling method. In this paper, a surrogate modelling procedure is proposed incorporating a phantom-space method for designing of experiments and an adaptive regression modelling method based. The paper also demonstrates the efficiency of the proposed procedure in a practical application in the field or robotics.

Atslēgas vārdi
Surrogate modelling, design of experiments, space-filling design, adaptive regression
Hipersaite
http://www.ktu.lt/ect/archive/Proceedings%20ECT-2011.pdf#page=40

Jēkabsons, G., Lagzdiņš, A., Lavendels, J. Surrogate Modelling Using Phantom-Space Design of Experiments and Adaptive Regression. No: Electrical and Control Technologies 2011: Proceedings of the 6th International Conference on Electrical and Control Technologies (ECT-2011), Lietuva, Kaunas, 5.-6. maijs, 2011. Kaunas: Technologija, 2011, 39.-44.lpp. ISSN 1822-5934.

Publikācijas valoda
English (en)