Open this publication in new window or tab >>Show others...
2025 (English)In: Journal of Building Engineering, E-ISSN 2352-7102, Vol. 104, article id 112367Article in journal (Refereed) Published
Abstract [en]
Coal bottom ash (CBA) is recovered from thermal power plants; it is an essential byproduct of the coal industry, and dumping on open land is the most significant environmental risk. Sustainably using CBA can help in alleviating ecological problems. Therefore, this study investigates the possibilities of utilizing CBA as a sand replacement in concrete production. A series of tests, including slump test, compressive strength (CS), water absorption, and water sorptivity of various bottom ash-based concrete mixes, were evaluated at curing ages of 7, 28, and 90 days for the desired strength of M25 and M35 concrete. Additionally, to determine the CS of the CBA concrete support vector regression was used, and the hyperparameters were optimized using particle swarm optimization (PSO-SVR) and jellyfish search optimization (JSO-SVR). Besides the outcomes of experiments, the data from previously published studies was also compiled and utilized for the prediction models. Experimental results reveal that M25 and M35 concrete of Sahiwal ash exhibited 10 %, 11 % and 11.8 %, 10.3 % higher CS with 25 % and 50 % CBA at 90 days. Similarly, M25 and M35 concrete of Sheikhupura ash, the CS increased by 8.3 % and 10 % with 25 % CBA at 90 days. Higher CBA content raises water absorption and sorptivity, indicating decreased durability. Increasing CBA content reduces concrete workability due to the hygroscopic nature of CBA particles. The higher specific gravity of CBA enhances strength development, yielding better-quality concrete. In contrast, the outcomes of the JSO-SVR models exhibited R2 for the training, testing, and validation dataset, which were 0.974, 0.961, and 0.9601, respectively. Furthermore, the JSO-SVR predictions were interpreted using SHapley Additive exPlanations (SHAP). The SHAP analysis revealed that sand, curing age, and cement were the most influential features affecting CS.
Place, publisher, year, edition, pages
Elsevier Ltd, 2025
Keywords
Coal bottom ash, Compressive strength, Support vector regression, Particle swarm optimization, Jellyfish search optimization, SHapley Additive exPlanations
National Category
Infrastructure Engineering
Research subject
Structural Engineering
Identifiers
urn:nbn:se:ltu:diva-112269 (URN)10.1016/j.jobe.2025.112367 (DOI)001456234100001 ()2-s2.0-105000519076 (Scopus ID)
Note
Validerad;2025;Nivå 2;2025-04-07 (u5);
Funder: Natural Science Foundation of China (51378104); National Science Fund for Distinguished Young Scholars (52125802); Jiangsu Province (BZ2021011); Fundamental Research Funds for the Central Universities (2242022k30030, 2242022k30031);
2025-04-072025-04-072025-04-11Bibliographically approved