Automated multi-objective optimization for thin-walled plastic products using Taguchi, ANOVA, and hybrid ANN-MOGA

Feng, Qingqing; Liu, Lingzhuo; Zhou, Xionghui

doi:10.1007/s00170-019-04488-2

Cited by 31 publications

(10 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As can be seen, the SRM made with recipe 4 has larger warpage [ 43 ] since it was attributed to the higher volumetric shrinkage [ 44 , 45 ] caused by higher gate solidification speed [ 46 , 47 ] during the solidification stage of the molded part. This disadvantage can be solved by controlling packing pressure [ 48 , 49 , 50 ] appropriately during the packing stage for an injection mold with high cooling efficiency before demolding of the molded parts.…”

Section: Resultsmentioning

confidence: 99%

Improving Cooling Performance of Injection Molding Tool with Conformal Cooling Channel by Adding Hybrid Fillers

Kuo

2021

Polymers

View full text Add to dashboard Cite

Silicone rubber mold (SRM) is capable of reducing the cost and time in a new product development phase and has many applications for the pilot runs. Unfortunately, the SRM after injection molding has a poor cooling efficiency due to its low thermal conductivity. To improve the cooling efficiency, the thermal conductivity of the SRM was improved by adding fillers into the SRM. An optimal recipe for fabricating a high cooling efficiency low-pressure injection mold with conformal cooling channel fabricated by fused deposition modeling technology was proposed and implemented. This study proposes a recipe combining 52.6 wt.% aluminum powder, 5.3 wt.% graphite powder, and 42.1 wt.% liquid silicon rubber can be used to make SRM with excellent cooling efficiency. The price–performance ratio of this SRM made by the proposed recipe is around 55. The thermal conductivity of the SRM made by the proposed recipe can be increased by up to 77.6% compared with convention SRM. In addition, the actual cooling time of the injection molded product can be shortened up to 69.1% compared with the conventional SRM. The actual cooling time obtained by the experiment is in good agreement with the simulation results with the relative error rate about 20%.

show abstract

Section: Resultsmentioning

confidence: 99%

Improving Cooling Performance of Injection Molding Tool with Conformal Cooling Channel by Adding Hybrid Fillers

Kuo

2021

Polymers

View full text Add to dashboard Cite

show abstract

“…The parameter curve obtained through parameter optimization can be used for continuous mass production and process monitoring. Ageyeva T et al [16]provided a comprehensive overview of the in-mold process monitoring tools and methods proposed for injection molding process control in the context of Industry 4.0 research. Established process monitoring techniques, direct temperature and pressure measurements with standard sensors and newly developed sensors, as well as techniques for measuring indirect process parameters such as viscosity, warpage deformation or volume shrinkage are reviewed.…”

Section: Introductionmentioning

confidence: 99%

A research method to improve the quality of box-type thin-walled parts by combining parameter optimization and i nverse-deformation design

Zhai

Chang²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

In this paper, a five-part study on the molding quality of box-type thin-walled injection molded parts with volume shrinkage and warpage deformation was conducted. In the first part, the molding parameters: filling time, plastic temperature, mold temperature (C), holding time (D), maximum holding pressure (E), and cooling time (F) were selected, and a six-factor, five-level L25 (5 6 ) orthogonal experimental design (OED) was designed. Then, the optimal combination of molding parameters was optimized by using gray correlation (GC) theory analysis. In the second part, the six molding parameters from the first part were further optimized based on the Box Behnken Design (BBD) response surface method (RSM) and BP neural network (BPNN) training, and the results showed that the fitted regression model after BPNN training had a smaller prediction error. In the third part, the regression models fitted with BBD and BPNN-BBD are further optimized globally. The optimization search method uses the NSGA-II genetic optimization algorithm. And comparing the above four optimization methods found that BPNN-BBD-NSGA-II prediction error < BPNN-BBD prediction error < BBD-NSGA-II prediction error < BBD prediction error. The fourth part is based on the anti-deformation design (ADD) theory. The NX/Moldex3D global optimization platform is created to finally almost completely eliminate warpage deformation of box-type thin-walled injection molded parts. In the last part, a trial mold verification analysis was performed, and the molding quality data was obtained by using optical scanning instruments, and the GOM software was used to compare and analyze the deviation values of each appearance surface and the flatness values.

show abstract

“…It was found that the plunge depth is the leading parameter for obtaining the best lap shear force. Feng et al [ 27 ] designed an automated two-staged multi-objective optimization tool for plastic injection molding and applied to industrial product using the DOEs method. It was found that the number of process parameters was compressed from 8 to 5 considering factors’ contribution percentage for reduction in time and computation cost.…”

Section: Introductionmentioning

confidence: 99%

Optimization of Process Parameters for Fabricating Polylactic Acid Filaments Using Design of Experiments Approach

2021

View full text Add to dashboard Cite

The amount of wasted polylactic acid (PLA) is increasing because 3D printing services are an increasingly popular offering in many fields. The PLA is widely employed in the fused deposition modeling (FDM) since it is an environmentally friendly polymer. However, failed prototypes or physical models can generate substantial waste. In this study, the feasibility of recycling PLA waste plastic and re-extruded it into new PLA filaments was investigated. An automatic PLA filament extruder was first developed for fabricating new PLA filaments. This paper also discusses the process, challenges, and benefits of recycling PLA waste plastic in an effort to fabricate new PLA filaments more sustainable. It was found that it was possible to fabricate PLA filament using recycled PLA waste plastic. The production cost is only 60% of the commercially available PLA filament. The tensile strength of the developed PLA filament is approximately 1.1 times that of the commercially available PLA filament. The design of experiments approach was employed to investigate the optimal process parameters for fabricating PLA filaments. The most important control factor affecting the diameter of PLA filament is the barrel temperature, followed by recycled material addition ratio, extrusion speed, and cooling distance. The optimal process parameters for fabricating PLA filament with a diameter of 1.7 mm include the barrel temperature of 184 °C, extrusion speed of 490 mm/min, cooling distance of 57.5 mm, and recycled material addition ratio of 40%.

show abstract

Automated multi-objective optimization for thin-walled plastic products using Taguchi, ANOVA, and hybrid ANN-MOGA

Cited by 31 publications

References 32 publications

Improving Cooling Performance of Injection Molding Tool with Conformal Cooling Channel by Adding Hybrid Fillers

Improving Cooling Performance of Injection Molding Tool with Conformal Cooling Channel by Adding Hybrid Fillers

A research method to improve the quality of box-type thin-walled parts by combining parameter optimization and i nverse-deformation design

Optimization of Process Parameters for Fabricating Polylactic Acid Filaments Using Design of Experiments Approach

Contact Info

Product

Resources

About