Keywords: neural networks, back propagation, radial basis function, ANFIS, compressive strength, light weight concrete.

Structural lightweight concrete has its obvious advantages of higher strength to weight ratio, better tensile strain capacity, lower coefficient of thermal expansion, and superior heat and sound insulation characteristics due to air voids in the lightweight aggregate. Also the reduction in the dead weight of the construction materials, by the use of lightweight aggregate in concrete, could result in a decrease in cross section of the concrete structural elements (columns, beams, plates, foundation). It is also possible to reduce steel reinforcement [1,2].

This paper illustrates the application of artificial neural networks (ANNs) for prediction of compressive strength a of structural light weight concrete (SLWC) mixtures. Among a feed-forward back propagation (BP) model, a radial basis function (RBF) model, and an adaptive neural network-based fuzzy interface system (ANFIS) employed for this investigation. To predict the compressive strength of LWC mixtures using neural networks.

ANNs are suited to complex problems, where the relationships between the variables to be modeled are not well understood. This is because they belong to the class of data-driven approaches that have the ability to determine which model inputs are critical, so there is no need for a priori rationalization about relationships between variables [3].

Among a feed-forward back propagation (BP) model, a radial basis function (RBF) model, and an adaptive neural network-based fuzzy interface system (ANFIS) employed for this problem, the four-layer feed-forward BP neural network were found to be superior to ANFIS model and RBF neural networks prediction of 3, 7, 14, 28, and 90 days compressive strength of SLWC mixtures. The results indicated that the four layers BPNN (7-14-10-5) can be predict the compressive strength of mixtures SLWC with adequate accuracy required for practical design purpose.

1

H. Al-Khaiat, M.N. Haque, "Effect of initial curing on early strength and physical properties of lightweight concrete", Cement Concrete Research 28 (6) pp 859-866, 1998. doi:10.1016/S0008-8846(98)00051-9

2

I.B. Topcu, "Semi - lightweight concretes produced by volcanic slags", Cement Concrete Research 27 (1) pp 15-21, 1997. doi:10.1016/S0008-8846(96)00190-1

3

A. Maren, C. Harston, R. Pap, "Handbook of Neural Computing Applications. Academic", San Diego, California, 488 pp. 1990.

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 £105 +P&P)