Selection of Network Parameters in Direct ANN Modeling of Roughness Obtained in FFF Processes DOI Open Access
Irene Buj-Corral,

Maurici Sivatte-Adroer,

Lourdes Rodero de Lamo

et al.

Polymers, Journal Year: 2025, Volume and Issue: 17(1), P. 120 - 120

Published: Jan. 6, 2025

Artificial neural network (ANN) models have been used in the past to model surface roughness manufacturing processes. Specifically, different parameters influence fused filament fabrication (FFF) In addition, characteristics of networks a direct impact on performance models. this work, study about use ANN FFF processes is presented. The main objective paper discovering how key (specifically, number neurons, training algorithm, and percentage validation datasets) affect accuracy predictions. To address question, 125 3D printing experiments were conducted changing orientation angle, layer height temperature, measuring average Ra as response. A multilayer perceptron with backpropagation algorithm was used. evaluates effect three parameters: (1) neurons hidden (4, 5, 6 or 7), (2) (Levenberg–Marquardt, Resilient Backpropagation Scaled Conjugate Gradient), (3) data splitting ratios (70%–15%–15% vs. 55%–15%–30%). Mean Absolute Error (MAE) metric. 7 using 55% yielded best predictive performance, minimizing MAE. Additionally, dataset size prediction analysed. It observed that gets worse datasets reduced, emphasizing importance having sufficient data. This will help select appropriate values for processes, well define be roughness.

Language: Английский

Functionality Versus Sustainability for PLA in MEX 3D Printing: The Impact of Generic Process Control Factors on Flexural Response and Energy Efficiency DOI Open Access
Markos Petousis, Nectarios Vidakis, Nikolaos Mountakis

et al.

Polymers, Journal Year: 2023, Volume and Issue: 15(5), P. 1232 - 1232

Published: Feb. 28, 2023

Process sustainability vs. mechanical strength is a strong market-driven claim in Material Extrusion (MEX) Additive Manufacturing (AM). Especially for the most popular polymer, Polylactic Acid (PLA), concurrent achievement of these opposing goals may become puzzle, especially since MEX 3D-printing offers variety process parameters. Herein, multi-objective optimization material deployment, 3D printing flexural response, and energy consumption AM with PLA introduced. To evaluate impact important generic device-independent control parameters on responses, Robust Design theory was employed. Raster Deposition Angle (RDA), Layer Thickness (LT), Infill Density (ID), Nozzle Temperature (NT), Bed (BT), Printing Speed (PS) were selected to compile five-level orthogonal array. A total 25 experimental runs five specimen replicas each accumulated 135 experiments. Analysis variances reduced quadratic regression models (RQRM) used decompose parameter responses. The ID, RDA, LT ranked first time, weight, strength, consumption, respectively. RQRM predictive experimentally validated hold significant technological merit, proper adjustment per case.

Language: Английский

Citations

43
Research progress in polylactic acid processing for 3D printing DOI
Xiyue Wang, Lijie Huang,

Yishan Li

et al.

Journal of Manufacturing Processes, Journal Year: 2024, Volume and Issue: 112, P. 161 - 178

Published: Jan. 20, 2024

Language: Английский

Citations

33
Maximizing performance and efficiency in 3D printing of polylactic acid biomaterials: Unveiling of microstructural morphology, and implications of process parameters and modeling of the mechanical strength, surface roughness, print time, and print energy for fused filament fabricated (FFF) bioparts DOI
Ray Tahir Mushtaq,

Yanen Wang,

Chengwei Bao

et al.

International Journal of Biological Macromolecules, Journal Year: 2024, Volume and Issue: 259, P. 129201 - 129201

Published: Jan. 6, 2024

Language: Английский

Citations

21
Machine learning-driven 3D printing: A review DOI
Xijun Zhang,

Dianming Chu,

Xinyue Zhao

et al.

Applied Materials Today, Journal Year: 2024, Volume and Issue: 39, P. 102306 - 102306

Published: June 29, 2024

Language: Английский

Citations

16
Recent Advancements in Additively Manufactured Hip Implant Design Using Topology Optimization Technique DOI Creative Commons
Abdulla All Noman, M. S. Shaari, Hassan Mehboob

et al.

Results in Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 103932 - 103932

Published: Jan. 1, 2025

Language: Английский

Citations

3
Optimization of key quality indicators in material extrusion 3D printing of acrylonitrile butadiene styrene: The impact of critical process control parameters on the surface roughness, dimensional accuracy, and porosity DOI
Nectarios Vidakis, Constantine David, Markos Petousis

et al.

Materials Today Communications, Journal Year: 2022, Volume and Issue: 34, P. 105171 - 105171

Published: Dec. 17, 2022

Language: Английский

Citations

60
Mechanical strength predictability of full factorial, Taguchi, and Box Behnken designs: Optimization of thermal settings and Cellulose Nanofibers content in PA12 for MEX AM DOI
Nectarios Vidakis, Markos Petousis, Nikolaos Mountakis

et al.

Journal of the mechanical behavior of biomedical materials/Journal of mechanical behavior of biomedical materials, Journal Year: 2023, Volume and Issue: 142, P. 105846 - 105846

Published: April 9, 2023

Language: Английский

Citations

31
Biochar filler in MEX and VPP additive manufacturing: characterization and reinforcement effects in polylactic acid and standard grade resin matrices DOI Creative Commons
Nectarios Vidakis, Dimitrios Kalderis, Markos Petousis

et al.

Biochar, Journal Year: 2023, Volume and Issue: 5(1)

Published: July 4, 2023

Abstract The development of sustainable and functional biocomposites remains a robust research industrial claim. Herein, the efficiency using eco-friendly biochar as reinforcement in Additive Manufacturing (AM) was investigated. Two AM technologies were applied, i.e., vat photopolymerization (VPP) material extrusion (MEX). A standard-grade resin VPP also biodegradable Polylactic Acid (PLA) MEX process selected polymeric matrices. Biochar prepared study from olive trees. Composites developed for both 3D printing processes at different loadings. Samples 3D-printed mechanically tested after international test standards. Thermogravimetric Analysis Raman revealed thermal structural characteristics composites. Morphological fractographic features derived, among others, with Scanning Electron Microscopy (SEM) Atomic Force (AFM). proven to be sufficient agent, especially filament process, reaching more than 20% improvement 4 wt.% loading tensile strength compared pure PLA control samples. In results not satisfactory, still, 5% achieved flexural 0.5 loading. findings prove strong potential biochar-based composites applications, too. Graphical

Language: Английский

Citations

24
Energy Consumption vs. Tensile Strength of Poly[methyl methacrylate] in Material Extrusion 3D Printing: The Impact of Six Control Settings DOI Open Access
Nectarios Vidakis, Markos Petousis, Nikolaos Mountakis

et al.

Polymers, Journal Year: 2023, Volume and Issue: 15(4), P. 845 - 845

Published: Feb. 8, 2023

The energy efficiency of material extrusion additive manufacturing has a significant impact on the economics and environmental footprint process. Control parameters that ensure 3D-printed functional products premium quality mechanical strength are an established market-driven requirement. To accomplish multiple objectives is challenging, especially for multi-purpose industrial polymers, such as Poly[methyl methacrylate]. current paper explores contribution six generic control factors (infill density, raster deposition angle, nozzle temperature, print speed, layer thickness, bed temperature) to performance methacrylate] over its performance. A five-level L25 Taguchi orthogonal array was composed, with five replicas, involving 135 experiments. 3D printing time electrical consumption were documented stopwatch approach. tensile strength, modulus, toughness experimentally obtained. angle speed first second most influential strength. Layer thickness corresponding ones consumption. Quadratic regression model equations each response metric compiled validated. Thus, best compromise between achievable, tool creates value engineering applications.

Language: Английский

Citations

20
Box-Behnken modeling to quantify the impact of control parameters on the energy and tensile efficiency of PEEK in MEX 3D-printing DOI Creative Commons
Nectarios Vidakis, Markos Petousis, Nikolaos Mountakis

et al.

Heliyon, Journal Year: 2023, Volume and Issue: 9(7), P. e18363 - e18363

Published: July 1, 2023

Currently, energy efficiency and saving in production engineering, including Material Extrusion (MEX) Additive Manufacturing, are of key importance to ensure process sustainability cost-effectiveness. The functionality parts made with MEX 3D-printing remains solid, especially for expensive high-performance polymers, biomedical, automotive, aerospace industries. Herein, the tensile strength metrics investigated over three control parameters (Nozzle Temperature, Layer Thickness, Printing Speed), aid laboratory-scale PEEK filaments fabricated melt extrusion. A double optimization is attempted by consuming minimum energy, improved strength. three-level Box-Behnken design five replicas each experimental run was employed. Statistical analysis findings proved that LT most decisive setting mechanical An 0.1 mm maximized endurance (∼74 MPa), but at same time, it responsible worst (∼0.58 MJ) printing time (∼900 s) expenditure. statistical further discussed interpreted using fractographic SEM optical microscopy, revealing 3D quality fracture mechanisms samples. Thermogravimetric (TGA) performed. hold measurable engineering industrial merit, since they may be utilized achieve an optimum case-dependent compromise between usually contradictory goals productivity, performance, functionality.

Language: Английский

Citations

19