Please wait a minute...

Frontiers of Mechanical Engineering

Front Mech Eng    2013, Vol. 8 Issue (3) : 305-310     https://doi.org/10.1007/s11465-013-0259-5
RESEARCH ARTICLE
The research on structural damage identification using rough set and integrated neural network
Juelong LI1, Hairui LI2(), Jianchun XING2, Qiliang YANG2
1. Naval-port Airport Barracks Department of Navy Logistics, Beijing 100841, China; 2. PLA University of Science & Technology, Nanjing 210007, China
Download: PDF(116 KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks
Abstract

A huge amount of information and identification accuracy in large civil engineering structural damage identification has not been addressed yet. To efficiently solve this problem, a new damage identification method based on rough set and integrated neural network is first proposed. In brief, rough set was used to reduce attributes so as to decrease spatial dimensions of data and extract effective features. And then the reduced attributes will be put into the sub-neural network. The sub-neural network can give the preliminary diagnosis from different aspects of damage. The decision fusion network will give the final damage identification results. The identification examples show that this method can simplify the redundant information to reduce the neural network model, making full use of the range of information to effectively improve the accuracy of structural damage identification.

Keywords rough set      integrated neural network      damage identification      decision making fusion     
Corresponding Author(s): LI Hairui,Email:loveme2006lhr@sina.com   
Issue Date: 05 September 2013
 Cite this article:   
Juelong LI,Hairui LI,Jianchun XING, et al. The research on structural damage identification using rough set and integrated neural network[J]. Front Mech Eng, 2013, 8(3): 305-310.
 URL:  
http://journal.hep.com.cn/fme/EN/10.1007/s11465-013-0259-5
http://journal.hep.com.cn/fme/EN/Y2013/V8/I3/305
Fig.1  The system diagram of structural damage identification using rough set and integrated neural network
Fig.2  The model of integrated neural network
Fig.3  The finite element of beam
Element numberThe extent of damageρNF1ρNF2ρNF9DSN1DSN4RF1RF5RF6
15%0.090.0550.0350.00180.01410.01430.0090.0086
110%0.18470.11310.07040.00380.030.02920.01830.0175
125%0.87810.5440.30930.00980.08390.07620.04960.0464
140%0.92570.59050.30010.01660.1560.12770.08570.0784
160%0.72080.50130.20690.02770.29210.20570.14190.1252
25%0.08640.6420.7160.00050.00730.00130.00830.0083
210%0.08590.66260.74230.00110.01530.00280.01720.0169
225%0.10440.72570.83250.00330.04480.00850.04840.0434
240%0.12940.81890.94980.00630.08720.01680.08840.0716
260%0.16640.875210.0130.1860.03650.16140.1115
65%0.85710.52380.0350.129610.01430.0090.0086
610%10.61220.07040.126710.02920.01830.0175
625%0.87810.5440.30930.116410.07620.04960.0464
640%0.78290.49940.30010.106210.12770.08570.0784
660%0.6910.48060.20690.094910.20570.14190.1252
Tab.1  The subset of training sampling of neural network
Number of the sub-networkNeural network structureNetwork input parameterNetwork output parameter
112 inputs/25 the node of hidden layer/6 outputsρNF1,ρNF2,…,ρNF12OUT1,OUT2,…,OUT6
210 inputs/21 the node of hidden layer/7 outputsρNF1, ρNF2,…,ρNF6OUT1, OUT2, …, OUT7
DS1, DS2, …, DS4
310 inputs/21 the node of hidden layer/7 outputsRSN1, RSN2, ..., RSN6OUT1, OUT2, …, OUT7
DS1, DS2, …, DS4
Tab.2  The structure of every sub-network
1 Barai S V, Pandy P C. Vibration signature analysis using artificial neural networks. Journal of Computing in Civil Engineering , 1995, 9(4): 259–265
doi: 10.1061/(ASCE)0887-3801(1995)9:4(259)
2 Weng G Y, Wang S L. Structure damage detection based on BP neural network. Journal of Hua Zhong University of Science and Technology (Urban Science Edition) , 2009, 26(2): 16–22
3 Li Y, Jiang D S, Luo P, Wang L X. Structural damage detection using rough set theory. Journal of Wu Han University of Technology , 2006, 28(2): 90–93
4 Wang J M, Chen L Z. Application of the combined neural network in structural damage detection. Journal of Computational Mechanics , 2005, 22(2): 247–251
5 Zhang D B, Wang Y N, Yi L Z. Rough neural network and its application to intelligent (information) procession. Control and Decision , 2005, 20(2): 121–126
6 Song L H. Matlab simulation design for K-means clustering. Experimental Technology and Management , 2010, 27(10): 101–103
7 Luo Y G, Peng Y H, Wen B H. Study on integrated neural network for damage identification of engineering structures. Journal of Vibration and Shock , 2006, 25(1): 14–17
8 Wang B S, Ni Y Q, Gao Z M. Input parameters for artificial neural networks in frame connection damage identification. Journal of Vibration and Shock , 2000, 13(1): 137–142
9 Jiang S F, Zhang S. Structural damage identification method with data fusion based on fuzzy neural network. Engineering Mechanics , 2008, 25(2): 95–101
Related articles from Frontiers Journals
[1] Guan LU, Dakai LIANG, . Smart optical-fiber structure monitoring based on granular computing[J]. Front. Mech. Eng., 2009, 4(4): 462-465.
[2] LIANG Lin, XU Guang-hua. Reduction of rough set attribute based on immune clone selection[J]. Front. Mech. Eng., 2006, 1(4): 413-417.
Viewed
Full text


Abstract

Cited

  Shared   0
  Discussed