Modelagem da resposta subjetiva a vibrações em aeronaves usando métodos de aprendizado de máquina
DOI:
https://doi.org/10.5944/ribim.18.2.45309Keywords:
Aviation, Comfort, Artificial Intelligence, Attribute Selection.Abstract
The increased competition in the aircraft market has motivated the aircraft industry to develop higher quality services for the customers. In this sense, an inter-institutional project has being developed for modeling passenger subjective responses related to adjectives (comfort, constancy, force, supportability) occasioned by physical stimuli vibrations on aircrafts simulators. In this work, it is used Computational Intelligence methods, specifically Machine Learning, to perform feature selection and to build regression models. In the feature selection task it was used a filter approach. Machine learning algorithms were used to build Artificial Neural
Networks, Multiple Linear Regression and Regression Trees. The experimental method proposed aims to predict passenger subjective responses only considering most important frequency bands of vibratory stimuli according to each adjective. The model evaluation was performed based on predictive quality and complexity. Is is shown that the proposed method allowed a reduction from 86.42% to 98.15% in the amount of used frequency bands to induce models without impairing the quality of predictions. In addition, 25% of the models showed statistically significant improvement. According to experts the built models were promising, both in terms of
complexity and predictive quality.
Downloads
References
[1] Quehl, J., “Comfort studies on aircraft interior sound and vibration”, Tese de Doutorado, Universitat Oldenburg,
Oldenburg, Alemanha (2001)
[2] Schaefer, R.L., Bitencourt, R.F., Teixeira, J.A.M.S., Iturrieta, C.N., Gerges, S.N.Y., “Aircraft vibration perception
in a laboratory situation”, Proceedings of the International Congress of Mechanical Engineering
(COBEM), Gramado, Brasil (2009)
[3] Schaefer, R.L., Bitencourt, R.F., Gerges, S.N.Y., “Interferência da vibração no campo acústico em um simulador
de aeronaves”, Anais do Congresso da Academia Trinacional de Ciências, Foz do Iguacu, Brasil (2009)
[4] Schaefer, R.L., Bitencourt, R.F., Teixeira, J.A.M.S., Gerges, S.N.Y., “Aircraft vibroacoustic perception
threshold in a laboratory situation compared to ISO 2631/1997 Standard”, Proceedings of the International
Congress and Exposition on Noise Control Engineering, Lisboa, Portugal (2010)
[5] Schaefer, R.L., “Estudo da resposta subjetiva a vibrações em aeronaves”, Dissertação de Mestrado, Universidade
Federal de Santa Catarina, Florianópolis, Brasil (2011)
[6] Bitencourt, R.F., “Desempenho de métodos na avaliação do conforto acústico no interior de aeronaves”, Tese
de Doutorado, Universidade Federal de Santa Catarina, Florianópolis, Brasil (2008)
[7] Bitencourt, R.F., Chamon, L.F.O., Futatsug, S., Yanagihara, J.I., Gerges, S.N.Y., “Preliminary results on the
modeling of aircraft vibroacoustic comfort”, Proceedings of the International Congress and Exposition on
Noise Control Engineering, New York, USA (2012)
[8] Rohrmann, B., “Verbal qualifiers for rating scales: Sociolinguistic considerations and psychometric data”,
Project Report, University of Melbourne, Melbourne, Australia (2003)
[9] Witten, I., Frank, E., Machine Learning: Practical Machine Learning Tools and Techniques, Morgan Kaufmann,
São Francisco, USA (2005)
[10] Liu, H., Motoda, H., Computational Methods of Feature Selection, Chapman & Hall/CRC (2007)
[11] Langley, P., “Selection of Relevant Features in Machine Learning”, Proceedings of the AAAI Fall Symposium
on Relevance, 140-144 (1994)
[12] Mitchell, T.M., Machine Learning, McGraw-Hill, New York, USA (1997)
[13] Alpaydin, E., Introduction to Machine Learning, MIT Press (2004)
[14] Haykin, S., Neural Networks: a comprehensive foundation, Pretince-Hall, 2 ed., Upper Saddle River, USA
(1999)
[15] Rezende, S.O., Sistemas Inteligentes: Fundamentos e Aplicações, Manole, Barueri, Brasil (2003)
[16] Munakata, T., Fundamentals of the New Artificial Intelligence: Neural, Evolutionary, Fuzzy, and More,
Springer, 2 ed. (2008)
[17] Kundu, D., Murali, G., “Model selection in linear regression”, Computational Statistics & Data Analysis, 22,
461-469 (1996)
[18] Hastie, T., Tibshirani, R., Friedman, J., The Elements of Statistical Learning, Springer, 2 ed. (2009)
[19] Rousseeow, P.J., “Least Median of Square Regression”, Journal of the American Statistical Association,
79(388), 871-880 (1984)
[20] Hekimoglu, S., Erengoglu, R.C., Kalina, J., “Outlier detection by means of robust regression estimators for
use in engineering science”, Journal of Zhejiang University Science A, 10(6), 909-921 (2008)
[21] Quinlan, R.J., “Learning with Continuous Classes”, Proceedings of the 5th Australian Joint Conference on
Artificial Intelligence, Singapore, 343-348 (1992)
[22] Wang, Y., Witten, I.H., “Induction of model trees for predicting continuous classes”, Proceedings of the 9th
European Conference on Machine Learning, 128-137 (1997)
[23] Doria, U., Introdução à Bioestatística: para simples mortais, Elsevier, São Paulo, Brasil (1999)
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2014 Ricardo Luís Schaefer, Carlos Andrés Ferrero, Samir Nagi Yousri Gerges

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Los autores conservan los derechos de autor y ceden de forma no exclusiva los derechos de explotación de los trabajos aceptados para su publicación en Revista Iberoamericana de Ingeniería Mecánica, garantizan a la revista el derecho de ser la primera publicación del trabajo.
Los autores permiten que la revista distribuya los trabajos publicados bajo la licencia de uso Creative Commons Atribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0 International). Esta licencia permite a otros compartir el trabajo con un reconocimiento de la autoría del trabajo y de la publicación inicial en esta revista. Se puede copiar, usar, difundir, transmitir y exponer públicamente, siempre que: i) se cite la autoría y la fuente original de su publicación (revista, editorial y URL de la obra); ii) no se use para fines comerciales.
Se permite y se anima a los autores a difundir electrónicamente (en un repositorio institucional, etc) la versión publicada del artículo (VOR) siempre con referencia a su publicación en la Revista Iberoamericana de Ingeniería Mecánica.
La publicación en esta revista es gratuita y no impone ningún coste a los autores.