Manufacturing Energy Consumption Estimation Using Machine Learning Approach

2017 ◽  
Author(s):  
Ruoyu Song ◽  
Yanglong Lu ◽  
Cassandra Telenko ◽  
Yan Wang
Keyword(s):  
Machine Learning ◽  
Energy Consumption ◽  
Process Parameters ◽  
Cost Estimation ◽  
Fused Deposition Modeling ◽  
Design Stage ◽  
Regression Methods ◽  
Fused Deposition ◽  
Data Driven Approach ◽  
Efficient Data

Environmental impacts of manufacturing are often significant and influenced by part and process parameters. Energy consumption is one of the most critical factors for the overall environmental impact of manufacturing. To achieve energy reduction, one must estimate the manufacturing energy consumption throughout the design stage. This paper presents an efficient data-driven approach to utilize machine learning to estimate energy consumption of a manufacturing process from a CAD model. The approach enables quick cost estimation with limited knowledge about the exact process parameters. A case study of fused deposition modeling is used to illustrate the feasibility of this framework and test potential regression methods. Lasso and elastic net regressions were compared in this study. The potential application of this framework to other manufacturing processes is also discussed.

scholarly journals Effect of Process Parameters on Energy Consumption, Physical, and Mechanical Properties of Fused Deposition Modeling

Polymers ◽  
2021 ◽  
Vol 13 (15) ◽  
pp. 2406
Author(s):  
Emmanuel U. Enemuoh ◽  
Stefan Duginski ◽  
Connor Feyen ◽  
Venkata G. Menta
Keyword(s):  
Energy Consumption ◽  
Additive Manufacturing ◽  
Process Parameters ◽  
Fused Deposition Modeling ◽  
Function Optimization ◽  
Free Form ◽  
Production Time ◽  
Fused Deposition ◽  
Deposition Modeling ◽  
The Impact

The application of the fused deposition modeling (FDM) additive manufacturing process has increased in the production of functional parts across all industries. FDM is also being introduced for industrial tooling and fixture applications due to its capabilities in building free-form and complex shapes that are otherwise challenging to manufacture by conventional methods. However, there is not yet a comprehensive understanding of how the FDM process parameters impact the mechanical behavior of engineered products, energy consumption, and other physical properties for different material stocks. Acquiring this information is quite a complex task, given the large variety of possible combinations of materials–additive manufacturing machines–slicing software process parameters. In this study, the knowledge gap is filled by using the Taguchi L27 orthogonal array design of experiments to evaluate the impact of five notable FDM process parameters: infill density, infill pattern, layer thickness, print speed, and shell thickness on energy consumption, production time, part weight, dimensional accuracy, hardness, and tensile strength. Signal-to-noise (S/N) ratio analysis and analysis of variance (ANOVA) were performed on the experimental data to quantify the parameters’ main effects on the responses and establish an optimal combination for the FDM process. The novelty of this work is the simultaneous evaluation of the effects of the FDM process parameters on the quality performances because most studies have considered one or two of the performances alone. The study opens an opportunity for multiobjective function optimization of the FDM process that can be used to effectively minimize resource consumption and production time while maximizing the mechanical and physical characteristics to fit the design requirements of FDM-manufactured products.

Development of a health monitoring and diagnosis framework for fused deposition modeling process based on a machine learning algorithm

2019 ◽  
Vol 234 (1-2) ◽  
pp. 324-332 ◽  
Author(s):  
Jungsoo Nam ◽  
Nanhyeon Jo ◽  
Jung Sub Kim ◽  
Sang Won Lee
Keyword(s):  
Machine Learning ◽  
Health Monitoring ◽  
Fused Deposition Modeling ◽  
Learning Algorithm ◽  
Support Vector ◽  
Fused Deposition ◽  
Monitoring And Diagnosis ◽  
Modeling Process ◽  
Data Driven Approach ◽  
Deposition Modeling

In this article, a data-driven approach is applied to develop a health monitoring and diagnosis framework for a fused deposition modeling process based on a machine learning algorithm. For the data-driven approach, three accelerometers, an acoustic emission sensor, and three thermocouples are installed, and associated data are collected from those sensors. The collected data are processed to obtain root mean square values, and they are used for constructing health monitoring and diagnosis models for the fused deposition modeling process based on a support vector machine algorithm, which is one of machine learning algorithms. Among various root mean square values, those of acceleration data from the frame were most effective for diagnosing health states of the fused deposition modeling process with the non-linear support vector machine–based model.

scholarly journals Comparative Study of the Sensitivity of PLA, ABS, PEEK, and PETG’s Mechanical Properties to FDM Printing Process Parameters

Crystals ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 995
Author(s):  
Mohammed Algarni ◽  
Sami Ghazali
Keyword(s):  
Mechanical Properties ◽  
Layer Thickness ◽  
Process Parameters ◽  
Fused Deposition Modeling ◽  
Acrylonitrile Butadiene Styrene ◽  
Process Parameter ◽  
Design Stage ◽  
Future Research ◽  
Printing Process ◽  
Fused Deposition

Significant advances in fused deposition modeling (FDM), as well as its myriad applications, have led to its growing prominence among additive manufacturing (AM) technologies. When the technology was first developed, it was used for rapid prototyping to examine and analyze a product in the design stage. FDM facilitates rapid production, requires inexpensive tools, and can fabricate complex-shaped parts; it, therefore, became popular and its use widespread. However, various FDM processing parameters have proven to affect the printed part’s mechanical properties to different extents. The values for the printing process parameters are carefully selected based on the part’s application. This study investigates the effects of four process parameters (raster angle, layer thickness, infill percentage, and printing speed) on the mechanical behavior of printed parts that are based on available literature data. These process parameter’s influence on part’s mechanical properties varies depending on the FDM material. The study focuses on four FDM materials: polylactic acid (PLA), acrylonitrile butadiene styrene (ABS), polyether ether ketone (PEEK), and polyethylene terephthalate glycol (PETG). This paper summarizes the state-of-the-art literature to show how sensitive the material’s mechanical properties are to each process parameter. The effect of each parameter on each material was quantified and ranked using analysis of variance (ANOVA). The results show that infill percentage then layer thickness are the most influential process parameter on most of the material’s mechanical properties. In addition, this work identifies gaps in existing studies and highlights opportunities for future research.

scholarly journals Energy and Eco-Impact Evaluation of Fused Deposition Modeling and Injection Molding of Polylactic Acid

2021 ◽  
Vol 13 (4) ◽  
pp. 1875
Author(s):  
Emmanuel Ugo Enemuoh ◽  
Venkata Gireesh Menta ◽  
Abdulaziz Abutunis ◽  
Sean O’Brien ◽  
Labiba Imtiaz Kaya ◽  
...  
Keyword(s):  
Energy Consumption ◽  
Injection Molding ◽  
Polylactic Acid ◽  
Fused Deposition Modeling ◽  
Plastic Injection Molding ◽  
Measurement Models ◽  
Fused Deposition ◽  
Dog Bone ◽  
Deposition Modeling ◽  
Plastic Injection

There is limited knowledge about energy and carbon emission performance comparison between additive fused deposition modeling (FDM) and consolidation plastic injection molding (PIM) forming techniques, despite their recent high industrial applications such as tools and fixtures. In this study, developed empirical models focus on the production phase of the polylactic acid (PLA) thermoplastic polyester life cycle while using FDM and PIM processes to produce American Society for Testing and Materials (ASTM) D638 Type IV dog bone samples to compare their energy consumption and eco-impact. It was established that energy consumption by the FDM layer creation phase dominated the filament extrusion and PLA pellet production phases, with, overwhelmingly, 99% of the total energy consumption in the three production phases combined. During FDM PLA production, about 95.5% of energy consumption was seen during actual FDM part building. This means that the FDM process parameters such as infill percentage, layer thickness, and printing speed can be optimized to significantly improve the energy consumption of the FDM process. Furthermore, plastic injection molding consumed about 38.2% less energy and produced less carbon emissions per one kilogram of PLA formed parts compared to the FDM process. The developed functional unit measurement models can be employed in setting sustainable manufacturing goals for PLA production.

Effect of process parameters on flexural strength and surface roughness in fused deposition modeling of PC-ABS material

2021 ◽  
pp. 251659842110311
Author(s):  
Shrikrishna Pawar ◽  
Dhananjay Dolas1
Keyword(s):  
Surface Roughness ◽  
Flexural Strength ◽  
Response Surface ◽  
Layer Thickness ◽  
Process Parameters ◽  
Surface Finish ◽  
Fused Deposition Modeling ◽  
Build Orientation ◽  
Fused Deposition ◽  
Deposition Modeling

Fused deposition modeling (FDM) is one of the most commonly used additive manufacturing (AM) technologies, which has found application in industries to meet the challenges of design modifications without significant cost increase and time delays. Process parameters largely affect the quality characteristics of AM parts, such as mechanical strength and surface finish. This article aims to optimize the parameters for enhancing flexural strength and surface finish of FDM parts. A total of 18 test specimens of polycarbonate (PC)-ABS (acrylonitrile–butadiene–styrene) material are printed to analyze the effect of process parameters, viz. layer thickness, build orientation, and infill density on flexural strength and surface finish. Empirical models relating process parameters with responses have been developed by using response surface regression and further analyzed by analysis of variance. Main effect plots and interaction plots are drawn to study the individual and combined effect of process parameters on output variables. Response surface methodology was employed to predict the results of flexural strength 48.2910 MPa and surface roughness 3.5826 µm with an optimal setting of parameters of 0.14-mm layer thickness and 100% infill density along with horizontal build orientation. Experimental results confirm infill density and build orientation as highly significant parameters for impacting flexural strength and surface roughness, respectively.

Sign in / Sign up

Export Citation Format

Share Document