Title:
Towards a Rational Performance Based Mix-Design Approach for 3D Printable Concrete Mixes through AI Algorithms
Author(s):
Liberato Ferrara
Publication:
Web Session
Volume:
ws_F24_Ferrara.pdf
Issue:
Appears on pages(s):
Keywords:
DOI:
Date:
11/3/2024
Abstract:
3D Concrete Printing (3DCP) has been gaining momentum in the last few years, because of its great potentials from the social, environmental and economic point of view. Although nowadays the research carried out is extensive, also complemented by an increasing number of practical applications, ranging from decorative to structural, it is still difficult to steer 3DCP towards a standardization which would permit a wider use of the technology in large scale civil engineering applications. This deviation from going straightforward to standards can be due, among the other reasons, also to the variety of parameters involved in the design of “printable” mixes, which often opens to new questions on the study of the optimal mix composition and on the influence of the latter of the “printability” performance. In this sense, it results extremely powerful to perform a systematic survey of the literature on 3DCP in order to investigate the parameters of interest and establish correlations among them which may be helpful in the design of a mix from scratch. This is exactly the aim of the present work, which describes the steps followed during a systematic survey of the literature with the aim of developing a database on 3DCP, and using this last for practical application as predictive tool for the printability of selected mix designs through Artificial Intelligence (AI) approaches, as Artificial Neural Networks (ANN). Due to the complexity of the problem, a method to collect data in an orderly manner was considered necessary. A relational database based on Structured Query Language (SQL) was selected as an efficient way to visualize and select stored data than a single spreadsheet. Therefore, an entity-relationship model suitable for the case study was developed with the aim of providing a database structure that can be easily handled and modified in the next future by any researcher willing to integrate own results to broaden the size of the same database. The developed database