Title:
Using Artificial Neural Networks to Connect Concrete Composition and Rheology to Pumpability/Printability/Buildability Requirements in 3D Concrete Printing 3DCP Applications
Author(s):
Marcucci
Publication:
Web Session
Volume:
ws_S23_Marcucci.pdf
Issue:
Appears on pages(s):
Keywords:
DOI:
Date:
4/2/2023
Abstract:
Artificial Intelligence (AI) is tremendously expanding in the recent years, for its ability to solve complex problems even with lacking and chaotic data. Its applications are wide-ranging, covering several fields as economy, military and construction. AI can be implemented in Civil Engineering, for the study of different issues as the mix proportion, the workability, and the strength prediction. AI aims to mimic the human brain, that is made up of many nerve cells driven by neurons which are in control of the external stimuli. Several AI techniques already exist, namely Artificial Neural Network (ANN) and Fuzzy Logic (FL), which find terrific perspectives in the field of Civil Engineering topics. Digital Fabrication with Concrete (DFC) is gaining great momentum, driven by the need of technological advancement and innovation uptake in the construction industry. The rheology of concrete finds evident meaning, because of the necessity of working concrete in its early ages, when still fluid. To make the material suitable for the printing process, concrete must comply with the printability requirements that are governed by parameters such as yield, tensile and shear strength. The purpose of the paper is to analyze the printability through the implementation of AI techniques, designing a neural network between the parameters that control the properties and the rheology of various concrete mixes.