Chaitanya Mishra
Department of Civil Engineering, Oriental University, Indore, Madhya Pradesh, INDIA

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Experimental Study and Physics-Informed Machine Learning on Sustainable Concrete Using Sugarcane Bagasse Ash and Distilled Sewage Water Mayuri Dnyaneshwar Ahirrao; Rakesh Patel; Chaitanya Mishra
Journal of the Civil Engineering Forum Vol. 13 No. 1 (January 2027)
Publisher : Department of Civil and Environmental Engineering, Faculty of Engineering, Universitas Gadjah Mada

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/jcef.28456

Abstract

The increasing environmental burden associated with cement production and freshwater consumption has created a need for sustainable alternatives in concrete production. This study investigates the combined utilization of Sugarcane Bagasse Ash (SCBA) as a partial cement replacement and Distilled SewageWater (DSW) as an alternative mixing-water source in M25–grade concrete. Seven concrete mixtures (M0–M6) were prepared by incorporating 0–30% SCBA and 0–40% DSW while maintaining a constant binder content of 400 kg/m3 and water-to-binder ratio of 0.50. Compressive strength was experimentally evaluated after 7, 14, and 28 days of curing. The results demonstrated that moderate replacement levels improved strength, with the mixture containing 10% SCBA and 20% DSW achieving the highest compressive strengths of 20.8, 26.1, and 34.2 MPa at 7, 14, and 28 days, respectively. The improvement was attributed to the pozzolanic activity and micro-filler effect of SCBA, together with adequate water availability. Higher replacement levels resulted in strength reduction because of cement dilution and reduced calcium hydroxide availability. Furthermore, Artificial Neural Network (ANN) and Physics-Informed Neural Network (PINN) models were developed to predict compressive strength using key mix–design parameters and curing age. Both models exhibited excellent predictive performance, with R2 values exceeding 0.999 during the evaluated stages. ANN provided slightly higher numerical accuracy, whereas PINN demonstrated more consistent validation and testing behaviour, indicating improved generalisation and physical consistency. The findings demonstrate that the integrated experimental, physics-informed modelling, and explainable–AI framework can support the development and optimization of sustainable SCBA–DSW concrete while reducing dependence on conventional cement and freshwater resources.