ANFIS Application to Predict the Mechanical Properties of Self-Compacting Concrete from mix design proportions

Publication Type:
Conference Proceeding
Citation:
2014
Issue Date:
2014-07-24
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail201407-Hokkido_Proceedings - ICEAS, LSBE & ACCMES.pdf Published version59.29 MB
Adobe PDF
Thumbnailpeer review 3066.pdfAccepted Manuscript version201.02 kB
Adobe PDF
A lot of studies have been done to predict the compressive strength of conventional concrete based on hardened characteristics, but in the case of self-compacting concrete, these types of researches are very rare. Even there is no study presented to predict the compressive strength of self-compacting concrete base on mixing proportions and fresh concrete properties. This paper aimed to design ANFIS models to predict the relation between compressive strength, fresh properties and mixing proportions of self- compacting concrete. 18 combination of mixing design proportions and fresh concrete properties from 55 different case studies of experimental data, have been utilized by Adaptive neuro-fuzzy inference system (ANFIS) to evaluate the accurate relation, level of dependence and importance factor to predict the most important engineering property of self-compacting concrete, compressive strength. It’s obvious that modulus of elasticity, tensile strength and some other engineering parameters of self-compacting concrete could be accurately estimated from accurate predicted compressive strength. Obtained results indicate that, including all input data (fresh concrete property and mix design proportion) give the best adjustment with compressive strength. Ignoring fresh concrete properties from combination, affects the compressive strength but it’s not as much as effect of maximum aggregate size and volume in the mixing design. Based on errors in each combination, weighting factor and importance of parameters could be analysed in resultant compressive strength of self-compacting concrete that could have major influence the inclusion and coefficient of each parameter in formulation compressive strength.
Please use this identifier to cite or link to this item: