• The International Concrete Abstracts Portal is an ACI led collaboration with leading technical organizations from within the international concrete industry and offers the most comprehensive collection of published concrete abstracts.

International Concrete Abstracts Portal


Title: Neural Network Model for Preformed-Foam Cellular Concrete

Author(s): M. Nehdi, Y. Djebbar, and A. Khan

Publication: Materials Journal

Volume: 98

Issue: 5

Appears on pages(s): 402-409

Keywords: cellular concrete; compressive strength; density; models

DOI: 10.14359/10730

Date: 9/1/2001

Cellular concrete is a lightweight material consisting of portland cement paste or mortar with a homogeneous void or cell structure created by introducing air or gas in the form of small bubbles (usually 0.1 to 1.0 mm in diameter) during the mixing process. This material has traditionally been used in heat insulation and sound attenuation, nonload bearing walls, roof decks, and is gaining wider acceptance in tunneling and geotechnical applications. A major concern with the production of cellular concrete is achieving product consistency and predictability of performance. Producers of the material have generated extensive experimental data over the years, but the analysis of such data using traditional statistical tools has not produced reliable predictive models. This research investigates the use of artificial neural networks (ANN) to predict the performance of cellular concrete mixtures. The ANN method can capture complex interactions among input/output variables in a system without any prior knowledge of the nature of these interactions and without having to explicitly assume a model form. Indeed, such a model form is generated by the data points themselves. This paper describes the database assembled, the selection and training process of the ANN model, and its validation. Results show that production yield, foamed density, unfoamed density, and compressive strength of cellular concrete mixtures can be predicted much more accurately using the ANN method compared to existing parametric methods.