Open Access

ARTICLE

Multivariate Adaptive Regression Splines Model to Predict Fracture Characteristics of High Strength and Ultra High Strength Concrete Beams

P. Yuvaraj¹, A. Ramachandra Murthy², Nagesh R. Iyer³, Pijush Samui⁴, S.K. Sekar⁵

1 Bridges department, L&T Ramboll Consulting Engineers Lts, Guindy, Chennai, Tamilnadu, India.E-mail: yuvarajprakash@gmail.com
2 Senior Scientist, CSIR-SERC, Chennai-113. Email: murthyarc@serc.res.in
3 Director, CSIR-SERC, Chennai-600113, Tamilnadu, India. Email: nriyer@serc.res.in
4 Professor, CDMM, VIT University, Vellore, Tamilnadu, India. Email: pijush.phd@gmail.com
5 Principal, Annamalaiar College of Engineering, Modaiyur, Polur, Tamilnadu, India.

Computers, Materials & Continua 2013, 36(1), 73-97. https://doi.org/10.3970/cmc.2013.036.073

Abstract

This paper presents Multivariate Adaptive Regression Splines (MARS) model to predict the fracture characteristics of high strength and ultra high strength concrete beams. Fracture characteristics include fracture energy (GF), critical stress intensity factor (KIC) and critical crack tip opening displacement (CTODc). This paper also presents the details of development of MARS model to predict failure load (Pmax) of high strength concrete (HSC) and ultra high strength concrete (UHSC) beam specimens. Characterization of mix and testing of beams of high strength and ultra strength concrete have been described. Methodologies for evaluation of fracture energy, critical stress intensity factor and critical crack tip opening displacement have been outlined. MARS model has been developed by establishing a relationship between a set of predicators and dependent variables. MARS is based on a divide and conquers strategy partitioning the training data sets into separate regions; each gets its own regression line. Four MARS models have been developed by using MATLAB software for training and prediction of fracture parameters and failure load.MARS has been trained with about 70% of the total 87 data sets and tested with about 30% of the total data sets. It is observed from the studies that the predicted values of Pmax, GF, KIC and CTODC are in good agreement with those of the experimental values.

---

Keywords

---