The Effect of Wind on 3D Printed Concrete Inter-Layer Bond Strength Based on Machine Learning Algorithms (2022-09)¶

Over the past 5 years, research focused on 3D printed concrete (3DPC) has increased significantly. One specific issue that attracts significant attention under researchers, is the reduction in mechanical strength as a result of weak bond between filament layers. Deriving analytical expressions for this reduction in flexural capacity has proven to be extremely challenging, since there are multiple variables that affect the interlayer bond strength. For example, pass time (time between layer depositions), ambient temperature, wind, humidity, and concrete mixture impact the bond. Hence, a more sophisticated approach, such as Machine Learning (ML) models, could be beneficial to determine the effect of a large number of variables on, for example, the interlayer bond strength. It is with this background, that this paper seeks to investigate the effect of wind on the interlayer bond strength based on supervised ML algorithms, as a basis for future research. In this work, 29 3D printed concrete (3DPC) samples were printed under different wind conditions. The input data (wind speed) and output data (tensile capacity) were collected and cleaned using Python. The data was split into two categories, namely training data (data used to develop the model using a regression algorithm with the built-in Scikit package in Python) and test data (data used to test the accuracy of the model). It was found that an increase in wind speed up to approximately 25 km/h, correlates to a decrease in interlayer bond.