Elman Recurrent Neural Network for Durable Green Lightweight Aggregate Concrete Based on Plastic Polyethylene Terephthalate and Synthetic Rubber

Green Lightweight Aggregate Concrete (GLAC) is manufactured by recycled by-products or solid waste in building and construction materials. During Portland cement manufacturing, it has significant benefits in reducing negative impact on environment which minimizes natural resource consumption. Recently, natural resources and the environment have suffered from numerous problems which are associated with natural resource depletion besides the enhancement in environmental pollution that results from uncontrolled disposal of plastic waste. Hence, it is significant to determine feasible and effective solutions to employ this waste like utilizing them to generate friendly environment green concrete. This research analyzes GLAC using Plastic Polyethylene Terephthalate (PET) waste from plastic and synthetic rubber (tyres). Then, GLAC is generated at 100% replacement of lightweight coarse aggregate and the normal weight with plastic-based GLA (PGLA) and their durability-associated properties are compared. The Compressive Strength Test (CST) for 7, 14, and 28 days and the Chloride ion Penetration Test (CPT) for 28 days are carried out for GLACs which is compared to Normal weight Coarse Aggregate Concrete (NCAC), and Volcanic Lightweight Aggregate Concrete (VLAC) respectively. These data are analyzed using Elman Recurrent Neural Network (ERNN) which achieves a less Mean Absolute Percentage Error (MAPE) of 2.98 compared to existing methods like RNN.