SUBAŞI, Serkan Düzce Üniversitesi - Teknik Eğitim Fakültesi - Yapı Eğitimi Bölümü, Turkey , ŞAHİN, İbrahim Düzce Üniversitesi - Teknik Eğitim Fakültesi - Yapı Eğitimi Bölümü, Turkey , ÇOMAK, Bekir Düzce Üniversitesi - Teknik Eğitim Fakültesi - Yapı Eğitimi Bölümü, Turkey

PREDICTION OF COMPRESSIVE STRENGTH OF CONCRETES CONTAININIG FLY ASH USING NON-DESTRUCTIVE TEST RESULTS BY ANFIS

35734

The purpose of this research work is to develop a mathematical model, for prediction of compressive strength for fly ash added concrete using Adaptive Neuro-Fuzzy Inference Systems (ANFIS). Mixing F class fly ash obtained from Orhaneli Thermoelectric Power Plant into concrete in the amounts of 0%, 5%, 10%, 20% and 30% of cement amounts, five different concrete mixtures were prepared and 12 samples from each mixture were taken for experimental study. Non-destructive test were applied to 6 samples from each mixture and destructive tests were applies to the other 6 samples. On 7th and 28th days, pull-off, pulse velocity, compressive strength properties were measured. Three of the total 60 measurements were eliminated because of vibration errors in the sample. The remaining 57 measurements were used forming ANFIS model. Concrete age, pull-off and pulse velocity were used as the inputs and compressive strength was used as the output of the model. Newly developed ANFIS model’s prediction ability is tested by comparing the predicted compressive strength values obtained from the ANFIS model and the real compressive strength values. As a result, it has been concluded that using the newly developed ANSIF model, which was trained with non-destructive test results, compressive strength values of different concrete can be predicted with an ignorable error rate.

9

Concrete , fly ash , non , destructive test methods , compressive strength , ANFIS

Sdu International Technologic Science

16