Properties of pumice aggregate concretes at elevated temperatures and comparison with ANN models


TÜRKMEN İ., BİNGÖL A. F., TORTUM A., Demirboga R., Guel R.

FIRE AND MATERIALS, cilt.41, sa.2, ss.142-153, 2017 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 41 Sayı: 2
  • Basım Tarihi: 2017
  • Doi Numarası: 10.1002/fam.2374
  • Dergi Adı: FIRE AND MATERIALS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.142-153
  • Anahtar Kelimeler: concrete, elevated temperatures, pumice aggregate, thermal conductivity, compressive strength, dynamic elasticity modulus and artificial neural network, COMPRESSIVE STRENGTH, THERMAL-CONDUCTIVITY, NEURAL-NETWORKS, FLY-ASH
  • İnönü Üniversitesi Adresli: Evet

Özet

The mechanical properties and thermal conductivity of concretes including pumice aggregate (PA) exposed to elevated temperature were analyzed by thermal conductivity, compressive strength, flexure strength, dynamic elasticity modulus (DEM) and dry unit weight tests. PA concrete specimens were cast by replacing a varying part of the normal aggregate (0-2 mm) with the PA. All concrete samples were prepared and cured at 23 +/- 10C lime saturated water for 28 days. Compressive strength of concretes including PA decreased that reductions were 14, 19, 25 and 34% for 25, 50, 75 and 100% PA, respectively. The maximum thermal conductivity of 1.9382W/mK was observed with the control samples containing normal aggregate. The tests were carried out by subjecting the samples to a temperature of 0, 100, 200, 300, 400 500, 600 and 700 degrees C for 3 h, then cooling by air cooling or in water method. The results indicated that all concretes exposed to a temperature of 500 and 700 degrees C occurred a significant decrease in thermal conductivity, compressive strength, flexure strength and DEM. An artificial neural network (ANN) approach was used to model the thermal and mechanical properties of PA concretes. The predicted values of the ANN were in accordance with the experimental data. The results indicate that the model can predict the concrete properties after elevated temperatures with adequate accuracy. Copyright (C) 2016 John Wiley & Sons, Ltd.