ANN-based prediction of surface and hole quality in drilling of AISI D2 cold work tool steel

Akincioğlu, Sıtkı; Mendi, Faruk; Çıçek, Adem; Akıncıoğlu, Gülşah

doi:10.1007/s00170-012-4719-6

Cited by 21 publications

(16 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(10) has been used to determine the performance of training and testing dataset. Figures 1 and 2 depicts the performance of training and testing datasets respectively.…”

Section: Resultsmentioning

confidence: 99%

“…Researchers use various methods for determination of surface and hole quality in drilling of AISI D2 cold work tool steel with uncoated TiN and TiAlN monolayer-and TiAlN/TiN multilayer-coated-cemented carbide drills. [1][2][3][4][5][6][7][8][9] Akıncıoglu et al 10 successfully adopted Artificial Neural Network (ANN) for prediction of surface and hole quality in drilling of AISI D2 cold work tool steel with uncoated TiN and TiAlN monolayer-and TiAlN/TiN multilayercoated-cemented carbide drills. However, ANN has some limitations such as black box approach, low generalization capability, arriving at local minima, overtraining problem, etc.…”

Section: Introductionmentioning

confidence: 99%

“…22 Researchers have successfully used LSSVM for solving different problems in engineering. [23][24][25] This study adopts the database collected from the work of Akıncıoglu et al 10 The dataset contains information about Cutting tool (t), Feed rate (fr) (mm/rev), Cutting speed, (v) (m/min), Surface roughness, (rs) (µm) and Roundness error, (re) (µm). The developed MPMR, MARS, and LSSVM have been compared with the ANN models developed by Akıncıoglu.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Determination of Surface and Hole Quality in Drilling of AISI D2 Cold Work Tool Steel Using MPMR, MARS and LSSVM

Samui

2014

J. Adv. Manuf. Syst.

View full text Add to dashboard Cite

This paper adopts Minimax Probability Machine Regression (MPMR), Multivariate Adaptive Regression Spline (MARS), and Least Square Support Vector Machine (LSSVM) for prediction of surface and hole quality in drilling of AISI D2 cold work tool steel with uncoated titanium nitride (TiN) and titanium aluminum nitride (TiAlN) monolayer-and TiAlN/TiN multilayer-coated-cemented carbide drills. MPMR is a probabilistic model. MARS is a nonparametric regression technique. LSSVM is developed based on statistical learning algorithm. Cutting tool (t), Feed rate (fr)(mm/rev), and Cutting speed (v)(m/min) have been adopted as inputs of MPMR, MARS, and LSSVM. The output of MPMR, MARS, and LSSVM is Surface roughness (rs) (µm) and Roundness error (re) (µm). A comparative study has been presented between the developed models. The results show that the developed model gives excellent performance.

show abstract

“…(10) has been used to determine the performance of training and testing dataset. Figures 1 and 2 depicts the performance of training and testing datasets respectively.…”

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Determination of Surface and Hole Quality in Drilling of AISI D2 Cold Work Tool Steel Using MPMR, MARS and LSSVM

Samui

2014

J. Adv. Manuf. Syst.

View full text Add to dashboard Cite

show abstract

“…In this context, 6 data for test and 26 data for training were randomly selected. The digits for the cutting tool to be entered into the ANN were denoted as TiN = 1 TiAlN, = 2, TiAlN/TiN = 3 and uncoated = 4 because they do not have numerical values [23]. All the values measured in the experiments are given in Table 5.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

Prediction of thrust forces and hole diameters using artificial neural networks in drilling of AISI D2 tool steel with cemented carbide tools

Akincioğlu¹,

Mendi²,

Çıçek³

et al. 2013

A P J Eng Sci

Self Cite

View full text Add to dashboard Cite

In this study, the effects of cutting speed, feed rate and different types of coating materials on thrust force and hole diameter were investigated in drilling of AISI D2 cold work tool steel. In addition, the thrust forces and hole diameters were predicted by artificial neural networks (ANN) using experimental data. Uncoated, TiN, TiAlN monolayer and TiAlN/TiN multi-layer coated cemented carbide drills with diameter of 5 mm were used in drilling experiments. The holes were drilled at different combinations of four cutting speeds (50, 55, 60, 65 m/min), two feed rates (0.063 and 0.08 mm/rev), and fixed depth of cut (7 mm). Experimental results showed that the lowest thrust forces and hole diameters were obtained with TiAlN/TiN multi-layer coated drills. After ANN training, it was found that the R 2 values are very close to 1 for both training and test sets. RMSE values are smaller than 0.03, and mean error values are smaller than 5% for the test set. This case shows that ANN is a powerful method for prediction of thrust forces and hole diameters.

show abstract

“…They found that the effects of spindle speed and feed rate on surface were larger than depth of cut for milling operation [26]. Göloğlu et al [27] investigated optimum cutting characteristics of DIN 1.2738 mould steel using highspeed steel end mills using Taguchi parameter design. The surface roughness on the strategy obtained using the optimal cutting conditions was less than that in others.…”

Section: Introductionmentioning

confidence: 99%

The effects of cryogenic-treated carbide tools on tool wear and surface roughness of turning of Hastelloy C22 based on Taguchi method

Akincioğlu

Gökkaya

Uygur

2015

Int J Adv Manuf Technol

Self Cite

108

View full text Add to dashboard Cite

In this study, Taguchi method has been applied to evaluate the effect of cryogenically treated tools in turning of Hastelloy C22 super alloy on surface roughness. The optimum parameters (cryogenic treatment, cutting speed, and feed rate) of turning were determined by using the Taguchi experimental design method. In Taguchi method, L9 orthogonal array has been used to determine the signal noise (S/N) ratio. Analysis of ANOVA was carried out to identify the significant factors affecting surface roughness. The statistical analysis indicated that feed rate, with a contribution percentage as high as 87.64 %, had the most dominant effect on machining performance, followed by the cryo-treated tools treatment and cutting speed, respectively. The confirmation tests indicated that it is possible to improve surface roughness significantly by using the Taguchi method. Surface roughness was improved by 28.3 and 72.3 % by shallow (CT1) cryogenic treatment and deep cryogenic treatment (CT2) applied on cementite carbide tools (UT). It found that wear resistance of tungsten carbide insert was increased by shallow and deep cryogenic treatments.

show abstract

ANN-based prediction of surface and hole quality in drilling of AISI D2 cold work tool steel

Cited by 21 publications

References 24 publications

Determination of Surface and Hole Quality in Drilling of AISI D2 Cold Work Tool Steel Using MPMR, MARS and LSSVM

Determination of Surface and Hole Quality in Drilling of AISI D2 Cold Work Tool Steel Using MPMR, MARS and LSSVM

Prediction of thrust forces and hole diameters using artificial neural networks in drilling of AISI D2 tool steel with cemented carbide tools

The effects of cryogenic-treated carbide tools on tool wear and surface roughness of turning of Hastelloy C22 based on Taguchi method

Contact Info

Product

Resources

About