Interlayer bond strength prediction of 3D printable concrete using artificial neural network: Experimental and modeling study

Interlayer bond strength is a key factor in 3D concrete printing (3DCP); therefore, having a reliable prediction model for it is of great importance. Few studies offer models predicting interlayer bond strength. Given the sensitivity of 3DCP properties to material quality and mix design, conducting a comprehensive study with multiple mix designs using ingredients from the same supplier is crucial. This study developed a highly accurate artificial neural network model to predict interlayer bond strength. The model's input variables, in kg/m3 , included OPC, water, and sand content. Additionally, the water-cement ratio (W/C) and sand-cement ratio (S/C) were used as inputs. The wt% of viscosity-modifying admixture, fibers, and superplasticizer (SP) relative to the cement weight, along with three printing time gaps (min), were also considered as inputs. Sensitivity analysis highlighted wt% of SP, W/C, and printing time gap as key influencers on interlayer bond strength, contributing 18.5 %, 15 %, and 14.5 %, respectively.