Adaptive Neuro-Fuzzy Inference System for Predicting Strength of High-Performance Concrete

Abstract

This study examines the performance of Adaptive Neuro-Fuzzy Inference System (ANFIS) for estimation of compressive strength of High-Performance Concrete (HPC) from given mix proportion. An ANFIS model merges advantages of both ANN and Fuzzy Logic. A total of 54 experimental datasets were used, where 36 datasets were used in training and 18 datasets were used for validating the model. Six input parameters include water binder ratio, age of testing, silica fumes, coarse and fine aggregate and superplasticizer, whereas compressive strength is the single output parameter. The experimental and obtained results were compared. The result illustrates that ANFIS model can be used as an alternative method to predict the compressive strength of high-performance concrete.