2015 CSCE Annual Conference Regina - Building on our Growth Opportunities

2015 CSCE Annual Conference Regina - Building on our Growth Opportunities Conference


Title
Prediction of Strength Properties of Engineered Cementitious Composites using Artificial Neural Network

Return to Session

Author(s)
Ms. Shirin GhatrehSamani (Presenter)
Mr. Ali Ehsani Yeganeh
Dr. Khandaker Hossain, Ryerson University
Abstract

This paper describes the development of Artificial Neural Network (ANN) models for the prediction of compressive strength of the Engineered Cementitious Composite (ECC) based on mix design parameters. A database consisting of large number of ECC mix designs from previous and current research studies are used for training and validation of ANN models. The influence of mix design parameters on the strength properties are evaluated to determine the appropriate input variables for the ANN models with optimized network architecture. The performance of ANN models is found to be good based on various statistical indicators. The ANN models can be used confidently for the optimization of ECC mix design parameters to obtain targeted strength properties.

Keywords: Artificial Neural Network, Engineered Cementitious Composite, Strength properties, Mix design, Training, Validation