Comparative analysis between SOM Networks and Bayesian Networks applied to structural failure detection
In this paper is carried out a comparative study between Self-Organizing Maps (SOM) and Bayesian Networks, to evaluate their performance in the field of structural health as damage detector of structural failures type III, which detects where the damaged area is and percentage of damage within this the area. The implemented classifiers are trained to detect structural failures in a simply supported beam and a truss of 13 elements, the detection is performed using modal information from different test scenarios, obtained by OpenSees® and MATLAB®. The simulations show a satisfactory performance of Bayesian networks to provide the range of stiffness variation in each element of the studied systems. Meanwhile, SOM networks are useful in predicting the decrease in elastic modulus, which is assumed as specific percentage of damage. Based on this results is proposed a hybrid methodology (BAYSOM) seeking to reduce computational cost and improve performance in diagnosis and detection of fault conditions in structures.
Farrar CR, Doebling SW, Nix DA, Vibration-Based Structural Damage Identification. Philosophical Transactions of the Royal Society: Mathematical, Physical & Engineering Sciences, 2001; 131–149
Vanik MW, Beck JL, Au SK, Bayesian probabilistic approach to structural health monitoring. Journal of Engineering Mechanics; 126 (7): 738-745
Cox RT. The algebra of probable inference, Baltimore: Johns Hopkins University, 1961
Rhim J, Lee SW. A neural network approach for damage detection and identification of structures. Computational Mechanics; 16: 437-443
Marwalla T. Probabilistic fault identification using a committee of neural networks and vibrational data. America Institute of Aeronautics and Astronautics, Journal of Aircraft; 38 (1): 138-146
Yuan S, Wang L, Peng G. Neural Network Method Based on a New Damage Signature for On-Line Structural Health Monitoring. Nanjing: Nanjing University of Aeronautics and Astronautics, 2003
Kolakowski P, Mujica L, Vehi J. Two Approaches to Structural Damage Identification: Model Updating versus Soft Computing. Journal of Intelligent Material Systems and Structures; 17 (1): 63-79. DOI: 10.1177/1045389X06056073
Addin A, Sapuan S, Mahdi E, Osman M, Ismail N. Bayesian network approach to classify damages and f-folds feature subset selection method in laminated composite materials. Serdang: Universiti Putra Malaysia; 3-6
Huiyong G, Zhengliang L. Bayesian Fusion Theory and its Application to the Identification of Structural Damages. Fourth International Conference on Natural Computation; 4: 110-114
Sofge DA. Structural health monitoring using neural network based vibrational system identification. Cambridge: NeuroDyne Inc, 1994; 91-94
Tamminen S, Pirttikangas S, Röning J. Self organizing maps in adaptative health monitoring. Oulu: University of Oulu. Machine Vision and Media Processing Group, 1999; 1-6
Pearl J. Probabilistic Reasoning in Intelligent Systems. San Francisco: Morgan Kaufmann. 2ed, 1988
Ben-Gal I. Bayesian Networks. Encyclopedia of Statistics in Quality and Reliability, John Wiley & Sons. Ruggeri F., Faltin F. & Kenett R. (Eds.), 2007
Ramirez D, Gómez I, Quiroga J. Structural damage detection using PSO optimization technique. Ingeniería y Desarrollo; 28: 89-101
Andrew Ng. CS229 Machine learning: Lecture notes unsupervised learning [online]. 2010. Disponible en internet: http://www.stanford.edu/class/cs229/notes/cs229-notes7a.pdf.
FullBNT-1.0.4 develop by Kervin Murphy at the California University, Berkeley.
SOM Toolbox Version 2.0 beta, programed on MATLAB® and developed by por Esa Alhoniemi, Johan Himberg, Juha Parhankangas y Juha Vesanto
M. Pedroza, E. Mariotte. “ANALISIS COMPARATIVO ENTRE REDES SOM Y REDES BAYESIANAS APLICADO A LA DETECCION DE FALLAS EN ESTRUCTURAS”. B.S. thesis. Dept. Mech. Eng., UIS. Bucaramanga, STD, 2011. [online] avaible: http://tangara.uis.edu.co/biblioweb/
- There are currently no refbacks.
Copyright (c) 2016 TECCIENCIA