New Frontiers in Communication and Intelligent Systems
Strength Assessment of HVFA Concrete using Soft Computing Techniques
Authors: Rajeshwari Ramachandra and Sukomal Mandal
Publishing Date: 08-10-2022
ISBN: 978-81-95502-00-4
Abstract
High volumes of fly ash (HVFA) for cement in concrete, the future material of the construction industry is extensively explored for sustainable developments on a large scale. More ingredients in concrete make its nature complicated and determining whose properties become tedious and uneconomical using traditional prediction methods. From the literature, it is evident that soft computing techniques (SCT) have proven their potential in predicting the highly non-linear behavior of concrete. In this study, 119 datasets of HVFA control concrete compressive strength (CS) collected from literature are used to train SCT models such as artificial neural network (ANN), Support vector machine (SVM), particle swarm optimization-based ANN (PSO-ANN), and PSO-SVM models; and a dataset of 12 nos. from an individual experimental study is used for testing the models. Cement, fly ash, water-binder ratio, superplasticizer, fine aggregate, coarse aggregate, specimen type and fly ash type are the models’ input parameters for predicting the HVFA concrete CS. PSO is used to optimize the individual ANN and SVM parameters to improve their performance. Statistical parameters i.e., correlation coefficient, root mean square error and scatter index are used to measure the models’ efficacy. Both individual and hybrid model results show good predictions of the HVFA control concrete CS for an individual experimental study.
Keywords
Concrete, Fly ash, Compressive strength, ANN, SVM, PSO
Cite as
Rajeshwari Ramachandra and Sukomal Mandal, "Strength Assessment of HVFA Concrete using Soft Computing Techniques", In: Rahul Srivastava and Aditya Kr. Singh Pundir (eds), New Frontiers in Communication and Intelligent Systems, SCRS, India, 2022, pp. 713-725. 10.52458/978-81-95502-00-4-72