A review on material mix proportion and strength influence parameters of geopolymer concrete: Application of ANN model for GPC strength prediction. (21st November 2022)

• Record Type: Journal Article
• Authors: Paruthi, Sagar; Husain, Asif; Alam, Pervez; Husain Khan, Afzal; Abul Hasan, Mohd; Magbool, Hassan M.
• Is Part Of: Construction & building materials. Volume 356(2022)
• Publication Date: 2022-11-21
• Subjects: Alkaline activator -- FlyAsh -- Artificial Neural Network -- Ground Granulated Blast furnace slag -- Alccofine; Building materials -- Periodicals; 624.18
• DOI: 10.1016/j.conbuildmat.2022.129253
• Languages: English
• ISSNs: 0950-0618

Abstract

Graphical abstract: Highlights: Critical review on mix design, strength parameters, microstructures, and ANN application for GPC. Key parameters influencing strength of GPC are assessed. Artificial Neural Network – method used to predict the strength of GPC. ANN showed as a promising tool in predicting the GPC compressive strength. Abstract: Concrete is a combination of cement, sand, aggregate, and water. Cement manufacturing causes the generation of various gases, mainly greenhouse gases like CO2 in the atmosphere. This CO2 is the leading cause of global warming, so it becomes essential to find a replacement for cement in the construction industry and use more eco-friendly construction materials. Geopolymer concrete (GPC) has been growing in the last few decades due to several advantages, including improved strength, durability properties, and eco-friendly nature. The GPC consists of silica and alumina in large amounts with an alkaline solution. Due to the use of the alkaline solution to activate geopolymerisation reaction, it is called alkaline activated concrete (AAC). Herein, we reviewed the GPC material, mix proportion, strength influence parameters, and strength prediction method. In addition, an Artificial Neural Network (ANN) is proposed to predict the compressive strength of GPC incorporating various materials. The predicted results using varying machine learning tools such as ANN, GEP, DNN, ResNet, GEP etc., demonstrate the accuracy and performance evaluation of the model. Therefore, there is an urgent need to develop and employ machine learning tools to predict the strength parameters of GPC for various construction works. In summary, this literature review provides a direction to engineers involved in a wide range in the construction industry using GPC.