Factors Controlling Durability of Geopolymer Concretes in Chloride Determined via Growing Self-Organizing Maps
Geopolymer concrete offers a promising alternative to traditional Portland cement concrete,exhibiting comparable mechanical and durability performance while reducing environmental impacts. However, its mechanical and durability properties depend on many factors, such as the water/binder ratios, con...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
Universiti Putra Malaysia Press
2025
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/47720/3/Factors%20Controlling.pdf http://ir.unimas.my/id/eprint/47720/ http://www.pertanika.upm.edu.my/pjst/browse/regular-issue?article=JST-5102-2024 https://doi.org/10.47836/pjst.33.1.18 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Geopolymer concrete offers a promising alternative to traditional Portland cement concrete,exhibiting comparable mechanical and durability performance while reducing environmental impacts. However, its mechanical and durability properties depend on many factors, such as the
water/binder ratios, concentration of activator and curing temperatures. This study proposes using an unsupervised Artificial Neural Network (ANN) Self-Organizing Map (SOM) to predict the factors that control the durability of geopolymer concrete in a chloride environment based on experimental
datasets. This research aims to identify the impact of various water-to-binder ratios and molarity of
activators on the durability of geopolymer concretes by applying the Growing Self-Organizing Maps (GSOM) model to predict the durability of the design mix. A series of geopolymer concrete mixes with varying water-to-binder (w/b) ratios and activator molarity were prepared to achieve these
goals. These cylindrical samples of 100 mm height x 50 mm diameter size were cured for 24 hours at 80°C and subject to chloride migration test at 28-day curing age. The data collected was analyzed and modeled using statistical methods and machine learning techniques, i.e., SOM modeling. This
modeling approach effectively revealed patterns and relationships within the dataset, providing crucial insights into the chloride migration behavior. Based on the GSOM modeling, this study highlights efficient data analysis, pattern recognition, and optimization of outcomes, such
as geopolymer concrete durability prediction in a chloride environment based on the selected parameters. |
|---|