Prediction of UCS and CBR behavior of fiber-reinforced municipal solid waste incinerator bottom ash composites using experimental and machine learning methods

Abstract

The present study deals with the evaluation of municipal solid waste incinerator bottom ash under the various design parameters considering Geotechnical and Geoenvironmental engineering applications of bottom ash. The effect of fiber-reinforced cement stabilized municipal solid waste incinerator bottom ash at varied cement and fiber content is studied. A set of 117 Unconfined compressive strength and 97 California bearing ratio tests were performed for 13 mm and 19 mm polypropylene fibers. The experimental results confirmed that the addition of 1 % fiber and 9 % cement leads to a maximum advancement of Unconfined compressive strength and California bearing ratio values. The results show the applicability of these materials for constructing roads, embankments, and compacted fills. Selecting a set of variables from the experimental results, the machine learning models- Artificial Neural Networks, Adaptive Neuro-Fuzzy Inference systems, and Multilinear regression models were used to predict these results. Varying the cement percentage, fiber content, fiber length, and curing period, the experimental unconfined compressive strength and California bearing ratio values were calculated, and the same is compared with machine learning models. The results were found to be suitable for predicting unconfined compressive strength and California bearing ratio values. During the study, the results produced from Multilinear regression were observed to be less significant. In addition, SEM analysis of the experimented specimens was conducted to comment on the strength development due to variations in cement content, fiber content, and fiber length, after 28 days of curing. The proposed models resulted in reliable results and thus can be used to avoid time-consuming mechanical lab tests.