UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Experimental characterization, black-box modeling, and optimization of the fused deposition modelled Acrylonitrile Butadiene Styrene Vahed Mohammad Ghasemloo, Ronak


Additive manufacturing (AM) is becoming a mainstream manufacturing process in different engineering applications, owing to its capability to build 3-dimensional parts with complex geometries, while maintaining high production speed, low cost and a minimal material waste. Fused Deposition Modeling (FDM), as one of the AM techniques, is frequently used in industries and research laboratories to print parts from different thermoplastic filaments. The properties and quality of the FDM processed thermoplastic parts, however, is highly restricted by the proper selection of process parameters, and it remains as a challenging task for producing defect-free 3D printed parts. On the other hand, owing to the large number of FDM process parameters with highly nonlinear and interacting effects, the experimental optimization of FDM is costly and requires mathematical predictive models. This thesis presents an integrated experimental-black box modeling and optimization framework to study the viscoelastic and tensile properties of FDM processed Acrylonitrile Butadiene Styrene (ABS). The effect of selected process parameters (including nozzle temperature, layer height, raster orientation and deposition speed) as well as their interaction effects are studied. Specifically, in the first step, a Taguchi orthogonal array was employed to design the experiments with a minimal number of runs, while considering different working conditions (temperatures) for the final prints. The Dynamic mechanical analysis (DMA) as well as tensile testing were carried out to investigate the significance of the process parameters, measured by statistical hypothesis testing methods. Due to the observed complex nature of the interacting effects of the FDM parameters, a series of artificial neural networks were developed and employed to predict the propeorites of the 3D printed samples, and consequently the process parameters were optimized via a particle swarm optimization (PSO). The percent contribution and ranking of the process parameters were identified and linked to the underlying meso/micro level mechanisms, through visual inspections of the samples and a Raman spectroscopy analysis.

Item Citations and Data


Attribution-NonCommercial-NoDerivatives 4.0 International