Prediction of properties of waste AAC aggregate concrete using artificial neural network

Creative Commons License

Topçu İ. B., Saridemir M.

COMPUTATIONAL MATERIALS SCIENCE, vol.41, no.1, pp.117-125, 2007 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 41 Issue: 1
  • Publication Date: 2007
  • Doi Number: 10.1016/j.commatsci.2007.03.010
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.117-125
  • Keywords: autoclaved aerated concrete, unit weight, compressive strength, ultrasound pulse velocity, artificial neural networks, COMPRESSIVE STRENGTH
  • Eskisehir Osmangazi University Affiliated: Yes


In this study, waste crushed autoclaved aerated concrete aggregates are used as crushed stone in concrete production which have two different sizes in the range of (4-16) and (16-31.5) mm in diameter. Unit weight, cylindrical compressive strength and ultrasound pulse velocity of hardened concrete are determined experimentally for waste autoclaved aerated concrete aggregate concrete types and dynamic elasticity modulus of these concrete types are calculated. It is seen that concrete lighter than crushed stone concrete can be produced by using waste autoclaved aerated concrete aggregates and the usage of waste autoclaved aerated concrete aggregates is suitable for concrete production according to the experimental results. A model is constructed by using artificial neural networks and experimental results are compared to the results of the model. It is concluded that the properties of waste autoclaved aerated concrete aggregated concrete can be obtained without any experimental when the testing in artificial neural networks model results are discussed. It is seen that training and testing results are similar to the experimental results. (C) 2007 Elsevier B.V. All rights reserved.