Optimization of Extrusion-Based 3D Printing Process Using Neural Networks for Sustainable Development

Abstract

Technological and material issues in 3D printing technologies should take into account sustainable development, use of materials, energy, emitted particles, and waste. The aim of this paper is to investigate whether the sustainability of 3D printing processes can be supported by computational intelligence (CI) and artificial intelligence (AI) based solutions. We present a new AI-based software to evaluate the amount of pollution generated by 3D printing systems. We input the values: printing technology, material, print weight, etc., and the expected results (risk assessment) and determine if and what precautions should be taken. The study uses a self-learning program that will improve as more data are entered. This program does not replace but complements previously used 3D printing metrics and software.

A Review on Filament Materials for Fused Filament Fabrication

“…To a limited extent, machine learning algorithms are used in AM for process monitoring, quality control, part design, environmental impact analysis, and process parameter analysis. Machine learning algorithms can also be applied in material selection, finding future composites topology representation, material design, cybersecurity of AM, or cloud platform integration with AM [152][153][154][155]. The performance of machine learning algorithms depends on data quality, data size, feature selection, model architecture, input dimension, model parameters, and other factors related to the algorithm itself or the application of interest.…”

Reducing Waste in 3D Printing Using a Neural Network Based on an Own Elbow Exoskeleton

“…This could more significantly reduce air pollution, energy and material consumption, and provide a more optimal fit during 3D printing processes of medical devices such as the elbow exoskeleton presented here. An AI-based approach is required due to the need to adapt solutions to the Internet of Things and Industry 4.0 paradigms including the ability to technically monitor the entire production process, and respond to sensor signals and failures in real-time [ 31 ].…”

Artificial intelligence in the design and production of personalized 3D printed products – conclusions from the project “Things are for People”

“…W badaniu zastosowano program samouczący się, który będzie się doskonalił w miarę wprowadzania kolejnych danych. Program ten nie zastępuje, ale uzupełnia wcześniej stosowane metryki i oprogramowanie do druku 3D [10].…”