Buy article PDF
The purchased file will be sent to you
via email after the payment is completed.
US$ 35
|
Computers and Concrete Volume 21, Number 6, June 2018 , pages 649-659 DOI: https://doi.org/10.12989/cac.2018.21.6.649 |
|
|
|
Neuro-fuzzy model of concrete exposed to various regimes combined with De-icing salts |
||
Ahmed Ghazy and Mohamed. T. Bassuoni
|
||
| Abstract | ||
| Adaptive neuro-fuzzy inference systems (ANFIS) can be efficient in modelling non-linear, complex and ambiguous behavior of cement-based materials undergoing combined damage factors of different forms (physical and chemical). The current work investigates the use of ANFIS to model the behavior (time of failure (TF)) of a wide range of concrete mixtures made with different types of cement (ordinary and portland limestone cement (PLC)) without or with supplementary cementitious materials (SCMs: fly ash and nanosilica) under various exposure regimes with the most widely used chloride-based de-icing salts (individual and combined). The results show that predictions of the ANFIS model were rational and accurate, with marginal errors not exceeding 3%. In addition, sensitivity analyses of physical penetrability (magnitude of intruding chloride) of concrete, amount of aluminate and interground limestone in cement and content of portlandite in the binder showed that the predictive trends of the model had good agreement with experimental results. Thus, this model may be reliably used to project the deterioration of customized concrete mixtures exposed to such aggressive conditions. | ||
| Key Words | ||
| Neuro-fuzzy systems; concrete, De-icing salts; environmental conditions; durability | ||
| Address | ||
| Ahmed Ghazy: Public Works Department, City of Winnipeg, Canada and Department of Civil Engineering, Alexandria University, Egypt Mohamed. T. Bassuoni: Department of Civil Engineering, University of Manitoba, Winnipeg, MB, Canada | ||