A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Prediction of Particle Packing Density of Alternative Fine Aggregates by Artificial Neural Network Applications
Particle packing density mainly influences the behavior of concrete. Experimental determination of particle packing density is arduous and time-consuming. Thus, artificial neural network (ANN) is a tool that helps in determining particle packing density faster and accurately. ANN is used to predict the properties of the concrete with natural aggregates only. However, utilization of alternative fine aggregates is a need of hour. The aim of this paper is to apply ANN to predict particle packing density of combined aggregates, viz. coarse, intermediate and fine aggregates. In addition, alternative fine aggregates, namely crushed rock sand (CRS), recycled fine aggregate (RFA) and coal bottom ash (CBA) are used. In this study, the main constraints used to design ANN are size distribution and packing density of individual aggregates. Feedforward network is used to determine particle packing density of combined aggregates. A total of 231 experimental datasets are used to train ANN for each alternative fine aggregate, and the results were compared. It was observed that the error percentages are well below 5% of experimental and ANN optimization analyses. The optimum packing proportions for river sand, CRS, RFA and CBA were 40:20:40, 30:30:40, 30:30:40 and 30:20:50, respectively.
Prediction of Particle Packing Density of Alternative Fine Aggregates by Artificial Neural Network Applications
Particle packing density mainly influences the behavior of concrete. Experimental determination of particle packing density is arduous and time-consuming. Thus, artificial neural network (ANN) is a tool that helps in determining particle packing density faster and accurately. ANN is used to predict the properties of the concrete with natural aggregates only. However, utilization of alternative fine aggregates is a need of hour. The aim of this paper is to apply ANN to predict particle packing density of combined aggregates, viz. coarse, intermediate and fine aggregates. In addition, alternative fine aggregates, namely crushed rock sand (CRS), recycled fine aggregate (RFA) and coal bottom ash (CBA) are used. In this study, the main constraints used to design ANN are size distribution and packing density of individual aggregates. Feedforward network is used to determine particle packing density of combined aggregates. A total of 231 experimental datasets are used to train ANN for each alternative fine aggregate, and the results were compared. It was observed that the error percentages are well below 5% of experimental and ANN optimization analyses. The optimum packing proportions for river sand, CRS, RFA and CBA were 40:20:40, 30:30:40, 30:30:40 and 30:20:50, respectively.
Prediction of Particle Packing Density of Alternative Fine Aggregates by Artificial Neural Network Applications
J. Inst. Eng. India Ser. A
Singh, S. K. (author) / Pranathi, B. M. (author) / Kirthika, S. K. (author)
Journal of The Institution of Engineers (India): Series A ; 101 ; 127-140
2020-03-01
14 pages
Article (Journal)
Electronic Resource
English
Fine Recycled Concrete Aggregates Particle Morphological Parameters and Packing Properties
| Springer Verlag | 2020
Optimization of packing density of aggregates
| Tema Archive | 2008
Selecting aggregates for maximum packing density in low permeability concrete
Online Contents | 1998
Selecting aggregates for maximum packing density in low permeability concrete
British Library Online Contents | 1998