Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 82
Edited by: B.H.V. Topping
Paper 44

Neural-Network Based Models of a Diagnosis System for Concrete Structures using Non-destructive Test Data

B. Cho+, S.-C. Lee* and Y.-S. Cho$

+New Engineering Consultants Inc, Seoul, Korea
*Research Institute of Technology, Samsung Corporation (Eng. & Cons.), Bundang, Korea
$Department of Architectural Engineering, Hanyang University, Ansan, Korea

Full Bibliographic Reference for this paper
B. Cho, S.-C. Lee, Y.-S. Cho, "Neural-Network Based Models of a Diagnosis System for Concrete Structures using Non-destructive Test Data", in B.H.V. Topping, (Editor), "Proceedings of the Eighth International Conference on the Application of Artificial Intelligence to Civil, Structural and Environmental Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 44, 2005. doi:10.4203/ccp.82.44
Keywords: diagnosis system, concrete structure, artificial neural networks, non-destructive test, impact-echo method, SASW method.

The purpose of this paper is to develop the HI-CONS (High-Intelligence-based diagnosis system for CONcrete Structure) that can give the engineering judgement for ensuring the safety and the serviceability of concrete structures. For this purpose, the system is developed with artificial neural networks (ANN) that can learn cylinder test and non-destructive test results using the Impact-echo method [1] and SASW (Spectral Analysis of Surface Wave method) as training patterns.

The system is composed of two modules. One is to predict the concrete strength that can provide in-place strength information of the concrete to facilitate concrete form removal and quality control of concrete structures. The other is to detect the flow of concrete that can predict thickness of concrete slabs and the kinds of embedded entities in concrete slabs to ensure the maintenance and safety evaluation of concrete structures.

The first module, for predicting the concrete strength, is classified as the ANN-I and the ANN-II models as shown in Figure 44.1. The ANN-I model [2] predicts the concrete strength based on basic information, material properties, measurement, temperature and humidity history from pouring day to 28th day after pouring. The ANN-II model predicts concrete strength based on additional non-destructive test data.

The second module, for detecting the concrete flaw, is classified as the ANN-III and ANN-IV models as shown in Figure 44.1. The ANN-III model predicts the concrete strength and the thickness of the concrete slab. The ANN-IV model can predict the kinds of embedded entities in concrete slab. For this, the results of non-destructive test as signals are used as training patterns in the ANN model.

Figure 44.1: Architecture of HI-CONS.

Abraham,O., Leonard, C., Cote, P and Piwakowski, B, "Time Frequency Analysis of Impact-Echo Signals: Numerical Modeling and Experimental Validation", ACI Materials Journal. Vol.97, No.6, pp. 645-657, 2000.
S.C. Lee, "Prediction of Concrete Strength using Artificial Neural Network Model", Engineering Structures, vol.25, pp. 849-857, 2002. doi:10.1016/S0141-0296(03)00004-X

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £80 +P&P)