In recent years, the study of high hydraulicity natural hydraulic lime (NHL5) mortars has been in the focus of many researchers, as it is considered a compatible, eco-friendly binding material, which can be used both for the restoration of culturally and historically significant structures, as well as for the construction of contemporary buildings. In the present study, artificial neural networks (ANNs) are used, aiming to simulate and map the development of NHL5 mortars' characteristics, such as compressive strength (CS), ratio of compressive to flexural strength (CS/FL) and consistency (CO), for selected mortar mix parameters, namely the binder to sand ratio (B/S), the water to binder ratio (W/B) and the maximum diameter of the aggregate (MDA) for different mortar specimen ages (AS). To this purpose, databases were developed, integrating experimental data from the international literature. Experimental verification of the developed ANN models revealed satisfactory fitting between theoretical and experimental results. This research highlights the potential of ANNs as a tool which can assist in mortar design and/or optimization, while mapping the development of mortar characteristics can assist in revealing the influence of the different mortar mix parameters on each characteristic. Furthermore, by combining the results of the three developed ANNs (CS, CO, CS/FL) targeted multi-parametric design of mortars can be assisted through a novel approach.
|Numero di pagine||15|
|Rivista||Cement and Concrete Research|
|Stato di pubblicazione||Published - 2020|
All Science Journal Classification (ASJC) codes