Analytical and innovative modeling investigations on the performance of nanoparticle-modified self-compacting mortars
The transition of a material from macro- to Nano-scale brings about significant changes in electron conductivity, optical absorption, mechanical properties, chemical reactivity, and surface morphology. These changes present opportunities for creating innovative composite mixtures. As there is a grow...
Uloženo v:
| Vydáno v: | European journal of environmental and civil engineering Ročník 29; číslo 5; s. 881 - 900 |
|---|---|
| Hlavní autoři: | , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Taylor & Francis
04.04.2025
|
| Témata: | |
| ISSN: | 1964-8189, 2116-7214 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Shrnutí: | The transition of a material from macro- to Nano-scale brings about significant changes in electron conductivity, optical absorption, mechanical properties, chemical reactivity, and surface morphology. These changes present opportunities for creating innovative composite mixtures. As there is a growing need for improved infrastructure, it becomes crucial to develop new, high-performance materials. To enhance the performance of concrete mixtures, various methods have been explored, including the utilization of nanoparticles (NPs). Incorporating NPs aims to improve the fresh and mechanical properties of self-compacting concrete (SCC) while also enhancing the permeability and absorption capacity of the composite by introducing extremely fine particles to fill micro-pores and voids. Numerous initiatives have been implemented to explore the mechanical characteristics of SCC. Typically, compressive strength (CS) serves as a crucial mechanical parameter for assessing concrete quality. Conventional methods for determining SCC's CS are costly, time-intensive, and restrictive due to the intricate interplay of various mixing proportions and curing processes. Thus, this investigation employs machine learning techniques, including artificial neural network (ANN), multi-expression programming (MEP), full quadratic (FQ), and linear regression (LR), to predict self-compacting mortar's CS. Approximately 292 CS values from the literature were extracted and analyzed to facilitate model development. Six influential variables were used as input parameters and one as an output during the modeling process. Four statistical metrics gauged model performance, and sensitivity analysis was conducted. Results indicate that the ANN model outperformed other models in predicting self-compacting mortar's CS. Meanwhile, the water-to-binder ratio, nanoparticle dosage, and concrete age significantly influence self-compacting mortar's CS. |
|---|---|
| ISSN: | 1964-8189 2116-7214 |
| DOI: | 10.1080/19648189.2024.2419433 |