Investigation of coal bottom ash as pozzolan in binder and mortar along with strength prediction using machine learning model
摘要
This study focuses on the advancement and suggestion of employing pulverized bottom ash from thermal power plants as a pozzolanic material. The pozzolanic characteristics were investigated through binders and mortars of ordinary Portland cement (OPC) and pulverized bottom ash (PBA) at various replacement rates of 0%, 2.5%, 5%, 7.5%, 10%, 12.5%, 15%, 20%, 25%, and 30% by weight of cement. The consistency and setting time of binder paste and compressive strength of binders and mortars were determined along with a sulphate attack test on 1:3 mortar exposed to 90 and 180 days. The purpose of this research was to develop a new binder that can reduce the dependency on cement by replacing it with pulverized bottom ash. The maximum strength of binder and mortar was achieved with 7.5% PBA for 90 days of curing, and the strengths were 79.89 and 69.9 N/mm2 respectively. The morphology of the binder and mortar was studied with the help of SEM, and it was found that PBA has increased the density of paste. PBA mortar showed improved resistance to sulphate attack by 63% and 35% for 90 and 180 days of exposure. Furthermore, compared to the control binder, it was determined that a 6.5% cost reduction was achieved when 10% of the weight of OPC was substituted with PBA. Following the experimental work, machine learning models were created and trained to forecast the compressive strength of mortar based on the experimental data collected. The GBM model demonstrated superior performance with minimal errors for both the training and validation datasets, achieving correlation coefficients of 0.99 and 0.96, respectively.