Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Major Losses in Pipes01:28

Major Losses in Pipes

When a fluid flows through a pipe, it experiences energy losses due to frictional resistance along the pipe walls, known as major losses. These energy losses result in a pressure drop, which varies based on the flow conditions — whether laminar or turbulent — and the specific physical properties of the fluid and pipe.
Fluid flow can be classified as laminar or turbulent, primarily based on the Reynolds number. This dimensionless number reflects the relative influence of inertial to viscous...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Characterization of 3D-Printed Glass Fiber-Filled and Calcium Carbonate-Filled Polypropylene Components for Surgical Planning.

Polymers·2025
Same author

Advanced Strategies for the Fabrication of Multi-Material Anatomical Models of Complex Pediatric Oncologic Cases.

Bioengineering (Basel, Switzerland)·2024
Same author

Patient-Specific 3D Printed Soft Models for Liver Surgical Planning and Hands-On Training.

Gels (Basel, Switzerland)·2023
Same author

Comparative Study about Dimensional Accuracy and Surface Finish of Constant-Breadth Cams Manufactured by FFF and CNC Milling.

Micromachines·2023
Same author

Perception of Recycled Plastics for Improved Consumer Acceptance through Self-Reported and Physiological Measures.

Sensors (Basel, Switzerland)·2022
Same author

3D Printing of Bioinert Oxide Ceramics for Medical Applications.

Journal of functional biomaterials·2022

Related Experiment Video

Updated: Jun 23, 2026

Experimental and Data Analysis Workflow for Soft Matter Nanoindentation
13:04

Experimental and Data Analysis Workflow for Soft Matter Nanoindentation

Published on: January 18, 2022

3.8K

Selection of Network Parameters in Direct ANN Modeling of Roughness Obtained in FFF Processes.

Irene Buj-Corral1, Maurici Sivatte-Adroer2, Lourdes Rodero-de-Lamo3

  • 1Department of Mechanical Engineering, Barcelona School of Industrial Engineering (ETSEIB), Universitat Politècnica de Catalunya, Av. Diagonal, 647, 08028 Barcelona, Spain.

Polymers
|January 11, 2025
PubMed
Summary

This study optimizes artificial neural network (ANN) models for predicting surface roughness in fused filament fabrication (FFF). The Resilient Backpropagation algorithm with 7 neurons and 55% training data achieved the best accuracy, highlighting the importance of sufficient data.

Keywords:
FDMFFFPolylactic acidartificial neural networksbackpropagation algorithmdatasets distributionmultilayer perceptronnumber of neuronssurface roughnesstraining algorithm

More Related Videos

Atomic Force Microscopy Cantilever-Based Nanoindentation: Mechanical Property Measurements at the Nanoscale in Air and Fluid
08:58

Atomic Force Microscopy Cantilever-Based Nanoindentation: Mechanical Property Measurements at the Nanoscale in Air and Fluid

Published on: December 2, 2022

2.9K
Surrogate Model Development for Digital Experiments in Welding
09:17

Surrogate Model Development for Digital Experiments in Welding

Published on: March 28, 2025

694

Related Experiment Videos

Last Updated: Jun 23, 2026

Experimental and Data Analysis Workflow for Soft Matter Nanoindentation
13:04

Experimental and Data Analysis Workflow for Soft Matter Nanoindentation

Published on: January 18, 2022

3.8K
Atomic Force Microscopy Cantilever-Based Nanoindentation: Mechanical Property Measurements at the Nanoscale in Air and Fluid
08:58

Atomic Force Microscopy Cantilever-Based Nanoindentation: Mechanical Property Measurements at the Nanoscale in Air and Fluid

Published on: December 2, 2022

2.9K
Surrogate Model Development for Digital Experiments in Welding
09:17

Surrogate Model Development for Digital Experiments in Welding

Published on: March 28, 2025

694

Area of Science:

  • Manufacturing Engineering
  • Materials Science
  • Computational Intelligence

Background:

  • Surface roughness is a critical quality indicator in manufacturing, particularly in additive manufacturing processes like fused filament fabrication (FFF).
  • Artificial neural networks (ANNs) offer a powerful approach for modeling complex relationships, such as those influencing surface roughness in FFF.
  • The performance of ANN models is sensitive to their internal parameters and the characteristics of the training data.

Purpose of the Study:

  • To investigate the impact of key artificial neural network (ANN) parameters on the accuracy of surface roughness predictions in fused filament fabrication (FFF).
  • To determine the optimal configuration of ANN parameters, including the number of neurons, training algorithm, and data splitting ratios, for surface roughness modeling.
  • To analyze the influence of dataset size on the predictive performance of ANNs in the context of FFF surface roughness.

Main Methods:

  • Conducted 125 fused filament fabrication (FFF) experiments, varying orientation angle, layer height, and printing temperature to measure average roughness (Ra).
  • Developed and evaluated multilayer perceptron neural network models using the backpropagation algorithm.
  • Systematically tested different ANN configurations: varying hidden layer neurons (4-7), training algorithms (Levenberg-Marquardt, Resilient Backpropagation, Scaled Conjugate Gradient), and data splits (70%-15%-15% vs. 55%-15%-30%).

Main Results:

  • The optimal predictive performance, indicated by the lowest Mean Absolute Error (MAE), was achieved using the Resilient Backpropagation algorithm with 7 neurons and a 55% training data split.
  • A significant decrease in prediction accuracy was observed as the size of the training dataset was reduced, underscoring the necessity of ample data.
  • The study identified specific ANN parameter settings that yield highly accurate surface roughness predictions for FFF processes.

Conclusions:

  • The selection of appropriate ANN parameters, specifically the training algorithm, number of neurons, and data partitioning strategy, is crucial for accurate surface roughness prediction in FFF.
  • Sufficient training data is essential for robust and reliable ANN model performance in predicting surface roughness.
  • This research provides valuable insights for optimizing both FFF printing parameters and ANN model design for enhanced surface quality.