Prediction of Compressive Strength of Self-Compacting Concrete with ANFIS


Creative Commons License

TOPÇU İ. B. , Olgun M. O. , GÜLBANDILAR E.

JOURNAL OF POLYTECHNIC-POLITEKNIK DERGISI, vol.22, no.3, pp.641-647, 2019 (Journal Indexed in ESCI) identifier

  • Publication Type: Article / Article
  • Volume: 22 Issue: 3
  • Publication Date: 2019
  • Doi Number: 10.2339/politeknik.537209
  • Title of Journal : JOURNAL OF POLYTECHNIC-POLITEKNIK DERGISI
  • Page Numbers: pp.641-647

Abstract

It can be said that the self-compacting concrete (SCC) has been the biggest development in concrete technology in recent years. Self-compacting concretes are very heavily reinforced without vibrators and provide high strength in complex cross-section structures. Both the speed of production and the quality of concrete can be increased by using self-compacting concrete. In the content of the KYB; Superplasticizer use, viscosity additive use, water/binder ratio, sand/total aggregate ratio and total coarse aggregate amount. This study is designed to investigate the changes in the compressive strength and cement/water ratio of the concrete with the proportion of superplasticizer admixtures in self-compacting concrete (SCC) using the predictive model of Adaptive Neural Fuzzy Inference Systems (ANFIS). In the design of ANFIS model were selected cement dosage (300, 400 and 450), SCC ratio and cement ratio (cement/water) as input variables, and compressive strength as output variables. Training data set which 53 randomly selected from 63 data sets were used for the training of the designed model. For the testing of the trained model were, it was used by the 10 data sets remainder of the total set. When comparing correlation between the results obtained from the test parameters and the experiment alone, the designed model showed successfully in predicting the 28-day compressive strengths (R-2 = 0.933, p < 0.01, c = 0.966).