scholarly journals Modeling, analysis, and optimization of dimensional accuracy of FDM-fabricated parts using definitive screening design and deep learning feedforward artificial neural network

2021 ◽  
Vol 9 (1) ◽  
pp. 115-129
Author(s):  
Omar Ahmed Mohamed ◽  
Syed Hasan Masood ◽  
Jahar Lal Bhowmik
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Dimensional Accuracy ◽  
Second Order ◽  
Fused Deposition ◽  
Screening Design ◽  
Definitive Screening ◽  
Cylindrical Parts ◽  
Artificial Neural ◽  
Definitive Screening Design

AbstractAdditive manufacturing (AM) technologies such as fused deposition modeling (FDM) rely on the quality of manufactured products and the process capability. Currently, the dimensional accuracy and stability of any AM process is essential for ensuring that customer specifications are satisfied at the highest standard, and variations are controlled without significantly affecting the functioning of processes, machines, and product structures. This study aims to investigate the effects of FDM fabrication conditions on the dimensional accuracy of cylindrical parts. In this study, a new class of experimental design techniques for integrated second-order definitive screening design (DSD) and an artificial neural network (ANN) are proposed for designing experiments to evaluate and predict the effects of six important operating variables. By determining the optimum fabrication conditions to obtain better dimensional accuracies for cylindrical parts, the time consumption and number of complex experiments are reduced considerably in this study. The optimum fabrication conditions generated through a second-order DSD are verified with experimental measurements. The results indicate that the slice thickness, part print direction, and number of perimeters significantly affect the percentage of length difference, whereas the percentage of diameter difference is significantly affected by the raster-to-raster air gap, bead width, number of perimeters, and part print direction. Furthermore, the results demonstrate that a second-order DSD integrated with an ANN is a more attractive and promising methodology for AM applications.

scholarly journals Definitive Screening Design and Artificial Neural Network for Modeling a Rapid Biodegradation of Date Palm Fronds by a New Trichoderma sp. PWN6 into Citric Acid

Molecules ◽  
2021 ◽  
Vol 26 (16) ◽  
pp. 5048
Author(s):  
Maha S. Elsayed ◽  
Noha M. Eldadamony ◽  
Salma S. T. Alrdahe ◽  
WesamEldin I. A. Saber
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Citric Acid ◽  
Date Palm ◽  
Screening Design ◽  
Trichoderma Sp ◽  
Definitive Screening ◽  
Artificial Neural ◽  
Palm Fronds ◽  
Definitive Screening Design

Generally, the bioconversion of lignocellulolytics into a new biomolecule is carried out through two or more steps. The current study used one-step bioprocessing of date palm fronds (DPF) into citric acid as a natural product, using a pioneer strain of Trichodermaharzianum (PWN6) that has been selected from six tested isolates based on the highest organic acid (OA) productivity (195.41 µmol/g), with the lowest amount of the released glucose. Trichoderma sp. PWN6 was morphologically and molecularly identified, and the GenBank accession number was MW78912.1. Both definitive screening design (DSD) and artificial neural network (ANN) were applied, for the first time, for modeling the bioconversion process of DPF. Although both models are capable of making accurate predictions, the ANN model outperforms the DSD model in terms of OA production, as ANN is characterized by a higher value of R2 (0.963) and validation R2 (0.967), and lower values of the RMSE (13.44), MDA (11.06), and SSE (9749.5). Citric acid was the only identified OA as was confirmed by GC-MS and UPLC, with a total of 1.5%. In conclusion, DPF together with T. harzianum PWN6 is considered an excellent new combination for citric acid biosynthesis, after modeling with artificial intelligence procedure.

Knowledge-Based Design of Artificial Neural Network Topology for Additive Manufacturing Process Modeling: A New Approach and Case Study for Fused Deposition Modeling

2018 ◽  
Vol 141 (2) ◽  
Author(s):  
Hari P. N. Nagarajan ◽  
Hossein Mokhtarian ◽  
Hesam Jafarian ◽  
Saoussen Dimassi ◽  
Shahriar Bakrani-Balani ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Additive Manufacturing ◽  
Fused Deposition Modeling ◽  
Expert Knowledge ◽  
Fused Deposition ◽  
Knowledge Based ◽  
Deposition Modeling ◽  
Artificial Neural

Additive manufacturing (AM) continues to rise in popularity due to its various advantages over traditional manufacturing processes. AM interests industry, but achieving repeatable production quality remains problematic for many AM technologies. Thus, modeling different process variables in AM using machine learning can be highly beneficial in creating useful knowledge of the process. Such developed artificial neural network (ANN) models would aid designers and manufacturers to make informed decisions about their products and processes. However, it is challenging to define an appropriate ANN topology that captures the AM system behavior. Toward that goal, an approach combining dimensional analysis conceptual modeling (DACM) and classical ANNs is proposed to create a new type of knowledge-based ANN (KB-ANN). This approach integrates existing literature and expert knowledge of the AM process to define a topology for the KB-ANN model. The proposed KB-ANN is a hybrid learning network that encompasses topological zones derived from knowledge of the process and other zones where missing knowledge is modeled using classical ANNs. The usefulness of the method is demonstrated using a case study to model wall thickness, part height, and total part mass in a fused deposition modeling (FDM) process. The KB-ANN-based model for FDM has the same performance with better generalization capabilities using fewer weights trained, when compared to a classical ANN.

Dimensional Prediction for FDM Machines Using Artificial Neural Network and Support Vector Regression

2019 ◽  
Author(s):  
Jiaqi Lyu ◽  
Souran Manoochehri
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Linear Regression ◽  
Support Vector Regression ◽  
Process Parameters ◽  
Multivariate Linear Regression ◽  
Support Vector ◽  
Ann Model ◽  
Fused Deposition ◽  
Artificial Neural

Abstract With the development of Fused Deposition Modeling (FDM) technology, the quality of fabricated parts is getting more attention. The present study highlights the predictive model for dimensional accuracy in the FDM process. Three process parameters, namely extruder temperature, layer thickness, and infill density, are considered in the model. To achieve better prediction accuracy, three models are studied, namely multivariate linear regression, Artificial Neural Network (ANN), and Support Vector Regression (SVR). The models are used to characterize the complex relationship between the input variables and dimensions of fabricated parts. Based on the experimental data set, it is found that the ANN model performs better than the multivariate linear regression and SVR models. The ANN model is able to study more quality characteristics of fabricated parts with more process parameters of FDM.

scholarly journals Artificial Neural Network Approach for Modeling of Ni(Ii) Adsorption from Aqueous Solution by Peanut Shell

2018 ◽  
Vol 25 (4) ◽  
pp. 581-604 ◽  
Author(s):  
Sayiter Yildiz
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Kinetic Model ◽  
Second Order ◽  
Experimental Results ◽  
Order Kinetic ◽  
Peanut Shell ◽  
Order Kinetic Model ◽  
Pseudo Second Order ◽  
Artificial Neural

Abstract In this study, ANN (artificial neural network) model was applied to estimate the Ni(II) removal efficiency of peanut shell based on batch adsorption tests. The effects of initial pH, metal concentrations, temperature, contact time and sorbent dosage were determined. Also, COD (chemical oxygen demand) was measured to evaluate the possible adverse effects of the sorbent during the tests performed with varying temperature, pH and sorbent dosage. COD was found as 96.21 mg/dm3 at pH 2 and 54.72 mg/dm3 at pH 7. Also, a significant increase in COD value was observed with increasing dosage of the used sorbent. COD was found as 12.48 mg/dm3 after use of 0.05 g sorbent and as 282.78 mg/dm3 after use of 1 g sorbent. During isotherm studies, the highest regression coefficient (R2) value was obtained with Freundlich isotherm (R2 = 0.97) for initial concentration and with Temkin isotherm for sorbent dosage. High pseudo-second order kinetic model regression constants were observed (R2 = 0.95-0.99) during kinetic studies with varying pH values. In addition, Ni(II) ion adsorption on peanut shell was further defined with pseudo-second order kinetic model, since qe values in the second order kinetic equation were very close to the experimental values. The relation between the estimated results of the built ANN model and the experimental results were used to evaluate the success of ANN modeling. Consequently, experimental results of the study were found to be in good agreement with the estimated results of the model.

Sign in / Sign up

Export Citation Format

Share Document