J. Cent. South Univ. (2021) 28: 112-125

DOI: https://doi.org/10.1007/s11771-021-4590-4

Optimization of cutting parameters with Taguchi and grey relational analysis methods in MQL-assisted face milling of AISI O2 steel

Bilal KURSUNCU, Yasin Ensar BIYIK

Faculty of Engineering, Architecture and Design Mechanical Engineering Department, Bartin University,Bartin 74100, Turkey

Central South University Press and Springer-Verlag GmbH Germany, part of Springer Nature 2021

Abstract: This study aims to examine the usability of environmentally harmless vegetable oil in the minimum quantity of lubrication (MQL) system in face milling of AISI O2 steel and to optimize the cutting parameters by different statistical methods. Vegetable oil was preferred as cutting fluid, and Taguchi method was used in the preparation of the test pattern. After testing with the prepared test pattern, cutting performance in all parameters has been improved according to dry conditions thanks to the MQL system. The highest tool life was obtained by using cutting parameters of 7.5 m cutting length, 100 m/min cutting speed, 100 mL/h MQL flow rate and 0.1 mm/tooth feed rate. Optimum cutting parameters were determined according to the Taguchi analysis, and the obtained parameters were confirmed with the verification tests. In addition, the optimum test parameter was determined by applying the gray relational analysis method. After using ANOVA analysis according to the measured surface roughness and cutting force values, the most effective cutting parameter was observed to be the feed rate. In addition, the models for surface roughness and cutting force values were obtained with precisions of 99.63% and 99.68%, respectively. Effective wear mechanisms were found to be abrasion and adhesion.

Key words: hardmilling; minimum quantity of lubrication; tool wear; grey relational analysis; Taguchi method; AISI O2 steel

Cite this article as: Bilal KURSUNCU, Yasin Ensar BIYIK. Optimization of cutting parameters with Taguchi and grey relational analysis methods in MQL-assisted face milling of AISI O2 steel [J]. Journal of Central South University, 2021, 28(1): 112-125. DOI: https://doi.org/10.1007/s11771-021-4590-4.

1 Introduction

In the ever-developing technology of today, the need for materials with improved properties is increasing day by day, and this makes sustainable manufacturing even more important. Machining of workpieces with high hardness (HRC>45) and strength is called hard machining [1-4]. The major disadvantage of the hard machining process is the rapid wear of the cutting tool material due to extraordinary temperatures in the contact zone of cutting tools and workpiece material. Different methods are used to prevent this rapid tool wear in machining. Some of these methods are a minimum quantity of lubrication (MQL) [4, 5], using thin hard coating films on cutting tool materials [6, 7], cryogenic cooling [8-10], and cryogenic heat treatment [11, 12]. Different hard thin coatings are used so as to increase the tool life of cutting tools in machining dry conditions [6, 13-15]. Applications such as reducing the environmental impact of cutting fluids used in machining methods and increasing the tool life of cutting tools with several methods are frequently used in sustainable manufacturing approaches [16]. Environmentally friendly liquids are used to reduce the environmental damage of the cutting fluid [17-21].

In recent years, the MQL method has been frequently used in hard milling operations. This method is based on the creation of a protective zone between the workpiece and the cutting tool as a result of spraying small amounts of different liquids into the cutting zone [16]. In recent years, researchers have been doing researches related to the use of MQL system in the machining of different workpiece materials. In the majority of studies, selection of cutting fluids is carried out taking into consideration environmentalist approaches, and the effect of cutting fluids on the cutting performance of different materials is investigated. GUO et al [22] used nano-sized graphene and alumina-added cutting fluid in the MQL system and at the end of the machining operations, they achieved significant development in the surface roughness and cutting forces values. GUTNICHENKO et al [5] used the cutting fluid obtained by adding graphite nanoplatelets into a vegetable-based lubricant in MQL system in the machining tests of Alloy 718, and they improved workpiece surface roughness values and cutting tool life. SHARMA et al [23] obtained lower surface roughness and cutting zone temperature by using vegetable oil in MQL system for turning high carbon AISI D2 steel. They prepared hybrid cutting liquid containing multi-wallet carbon nanotubes and alumina-based nanoparticles and used this cutting fluid in MQL system for machining AISI 304 stainless steel, and they reduced cutting tool wear [24]. EKER et al [25] examined the turning of the magnesium alloy using the MQL system. According to the results obtained, the MQL system has provided higher performance in the machining of this alloy than that under dry conditions.

Experimental design and optimization methods are often used to optimize a complex process to save time and costs, and they provide information about optimum parameter levels by taking multiple parameters [26]. Design of experiment (DOE) is widely used in machining systems by researchers to determine optimum cutting conditions such as good surface quality, high cutting tool life, and optimum cutting parameters of low cutting force values. ASLANTAS et al [27] used Taguchi method to optimize cutting parameters in machining operation of titanium alloy. They found that lower feed rates and larger cutting depths lead to an increase in burr width. CALISKAN et al [28] used response surface methodology (RSM) for the optimization of cutting forces and surface roughness in milling operation of hard to cut material. PUSAVEC et al [29] used genetic algorithms to optimize and evaluate the machining process for each cooling and lubrication application in the machining of Inconel 718 superalloy. They have found that thanks to optimization, machining processes can be further improved with machined cutting tool life, surface quality, efficiency and power consumption, chip breakability.

VISWANATHAN et al [30] examined the optimization of cutting parameters in the turning process of a high strength magnesium alloy workpiece. They used Taguchi-based Gray relational analysis method for optimization of cutting parameters. After the tests applied with the optimum cutting parameters obtained, they obtained a minimum wear amount, surface roughness, and cutting zone temperature values. MIA et al [31] performed the optimization of the cutting parameters using the response surface methodology (RSM) and artificial neural network (ANN) methods in turning of Ti6Al4V alloy, which is frequently used in the aerospace industry. According to the models obtained, they found that the RSM method is more suitable than the ANN method in the optimization of cutting parameters. RAMESH et al [32] performed the optimization of cutting parameters in the processing of magnesium alloy using TOPSIS and grey relational analysis. They obtained similar results in both methods, demonstrating the usability of the methods. KIVAK [33] performed the optimization of the cutting parameters in the milling process of Hadfield steel using the Taguchi method. Along with the results, the Taguchi method has been shown to be applicable in the optimization of manufacturing cost and time. VISWANATHAN et al [34] determined by the ANOVA method that the most important factor affecting the temperature of the cutting zone during the turning of the magnesium alloy was the cutting speed. However, although there are studies related to the machining of different materials using vegetable oil in MQL system, there is no study on optimization of cutting parameters in milling AISI O2 steel using vegetable oil in MQL system.

In this work, the effects of feed rate, cutting speed, MQL flow rate on cutting force and surface roughness were investigated. All tests were carried out in dry conditions for comparison purposes. Then, the obtained values were optimized by using Taguchi, GRA and RSM methods. This study related to face milling of AISI O2 steel by consuming vegetable oil in MQL system is thought to contribute to on sustainable manufacturing.

2 Materials and methods

2.1 Workpiece material

AISI O2 steels with an annealed hardness of about HRC58 are provided in dimensions of 150 mm×100 mm×50 mm as workpiece materials, where surface milling tests are carried out. Then, in order to be able to measure the cutting forces in face milling operations, three stepped holes were drilled onto the workpiece and thus workpiece material is fixed on the dynometer.

2.2 Machine tool, cutting tools, MQL system and cutting fluid

The device used in the measurement of cutting forces consists of two parts. For measuring the cutting forces, the piezoelectric part (Kistler 9257B) to which the workpiece is connected, consists of a charge amplifier (Kistler 5070) and computer software (Dynoware).

Cutting tools (R390-11 T3 08M-PM 1010) used in milling operations and tool holder (R390-11 T3 08M-PM 1010) to which cutting tools are mounted, were supplied from Sandvick Company. There are two flutes for connecting the cutting tool on the tool holder. Only one flute is attached to the cutting tool in the face milling tests. This is because the cutting performance of a cutting tool can be accurately determined.

Surface roughness values were measured using Mitutoyo Surftest SJ 310. The average of the values obtained from five different points on the workpiece was used to evaluate the roughness results after each test parameter. In all cutting parameters, cutting performance tests were performed using the Falco VMC850 vertical machining center. To determine the amount of wear occurring on the flank surface of the cutting tool in the face milling tests, a computer-assisted stereo zoom microscope (VisionSX45) was used. The MQL system (Werte Micro STN 25), operating at pressures of 0.4-0.6 MPa bar with a lubrication range of 0.0021-0.028 mL, was used. Vegetable oil used as cutting liquid in MQL system is provided as sunflower oil.

2.3 Design of experiment

DOE provides meaningful results for processing with measured responses. The DOE methodology provides insight into the statistical significance of interaction effects and regression models of specific parameters alone and with each other. In this study, the purpose of surface regression modeling is to define the efficiency of surface roughness and cutting force during the milling of the workpiece and to develop a prediction ability by use of the parameters such as cutting speed, feed rate, and MQL flow rate. The predictive capacity of the regression models and the corresponding response graphs allow considering the effect of changing a parameter that is not in the experimental design [35]. The experimental design was prepared by using Taguchi’s L9 orthogonal array. While determining the cutting parameters, first, the manufacturer’s catalogs of the cutting tool were examined and then pretest was applied. Three different MQL flow rates, cutting speeds, feed rates and fixed axial and radial depths of cut, were used in the study. Tables 1 and 2 show the parameters and levels used in the experiments and the experimental design prepared with these parameters. The flowchart of the experimental setup is shown in Figure 1.

2.4 Grey relational analysis (GRA)

GRA is applied to specify the gray relational degree to express the effect of more than one criterion as a single effect in any optimization problem. The steps to be applied for the implementation of this method are as follows. The values measured in the first stage should be normalized to values between 0 and 1. In this work, surface roughness and cutting force values should be minimized; for this, the smaller-better approach will be used.

(1)

The N value in the formula is the normalized value of the responses obtained as a result of the experiments. In Eq. (1), max(y) is the highest value of the measured responses and y value is the value measured for each experiment number. The next step is to calculate the gray relational coefficient (GRC). This factor explains the relationship between the optimum and measured experimental results [36].

Table 1 Cutting parameters

Table 2 Experimental design

(2)

△=1-N (3)

In Eq. (2), △_min and △_max are the minimum and maximum values of the calculated △ values; ξ (weightage coefficient) can be determined to be between 0 and 1 [37], and ξ was determined 0.5 in this study. The next step in GRA is the calculation of the gray relational grade value (GRG). This value determines the relationship between the test parameters and the measured values.

(4)

In Eq. (4), n is determined as the number of responses measured in the experiments (in this study n=2). The higher degree of gray relational grade obtained means that it is close to the appropriate test parameter.

Figure 1 Flowchart of experimental setup

3 Results and discussion

In this section, tool life, surface roughness, cutting force values measured in milling operations are given. In addition, the results of the experiments were modeled and optimized by using Taguchi, RSM and GRA.

3.1 Tool life of cutting tool

In this section, the effect of MQL system on the tool life of cutting tools is examined. For comparison, the tests were performed without using any coolant. The cutting tool has been considered to have worn when the wear length on the flank face reaches 0.25 mm in accordance with the ISO 8688-1:1989 face milling standard [38]. So as to define the amount of tool wear, the cutting tool was taken from the CNC milling machine after every 0.15 m cutting length and the amount of wear was measured with optical microscope from flank face by using a software on computer.

Table 3 shows the cutting tool life as cutting length obtained by using both dry and MQL conditions according to the prepared experimental design according to Taguchi L9 orthogonal array after surface milling operations. In all test parameters, tool life after milling operations using MQL system increased compared to tests performed in dry conditions. Using the MQL system, the highest tool life was achieved in the test number 6. The highest tool life in dry conditions was reached in the test 9. Using the MQL system, the greatest increase in tool life was attained in the test 2 compared to dry condition. The optical microscopy images of the worn cutting tools in the test 2 are shown in Figure 2. At higher cutting speed and feed rate values, the positive effect of the MQL system on tool life was found to be lower. Thanks to the vegetable oil used in the MQL system, cutting tool life is increased by dropping the high temperature occurring in the cutting zone causing cutting tool wear [27].

Table 3 Cutting tool life as cutting length

Figure 2 Cutting tool life obtained in Test 2 and optical microscopy images of worn cutting tools

3.2 Surface roughness

The surface roughness was measured at five altered points of the machined surface of the workpiece material after each 0.15 m cutting length, and the averages of these five values were calculated as surface roughness value for the pass. This process was repeated for each 0.15 m cutting length up to 0.75 m and a graphic was created by taking the average of all values obtained. Thus, the probability of an error that may occur in the experiments was minimized. The surface roughness values were not measured from the machined surface of the first 30 mm and the last 30 mm cutting length. This is because the behaviors of the cutting tool in the entry and exit areas of the workpiece in the face milling operations are different. Figure 3 shows the average roughness values measured after 0.15 m cutting length of the milling operation according to the experimental design in both dry and MQL conditions. Consequently, in all cutting parameters a smooth surface was obtained with the effect of vegetable oil used in the MQL system [39]. In both situations, the highest amount of the surface roughness was measured in the test No. 1. The lowest value was measured in test No. 5 in MQL.

Figure 3 Comparison of surface roughness values in dry and MQL system

3.3 Signal to noise ratio (S/N) for surface roughness values

Table 4 shows the measured roughness and S/N values. The optimum levels of surface roughness are presented in Figure 4. An assessment of Figure 4 demonstrates that MQL flow rate of 150 mL/h, feed rate of 0.15 mm/tooth, cutting speed of 150 m/min were the optimal parameters that produced the lowest surface roughness values.

Table 4 S/N ratio values of surface roughness

3.4 Evaluation of experimental results for surface roughness with RSM

Analysis of variance table and a model for surface roughness values are shown in Table 5. A “Probe>F” value of less than 0.05 indicates that this term is important for the model [40]. In this case, all cutting parameters are significant model terms. R-squared 0.9998 means that 99.98% of the total variations are correctly estimated by the model [41].

The equation of response in terms of the significant factors for the surface roughness values of the ANOVA method is given as follows:

R_a=1.24419+4.98362×10^-3V_c-6.12044f_z-

6.935571×10^-3Q+3.43421×10^-5V_cQ+

0.021282f_zQ-2.09219×10^-5V_c²+10.0019f_z² (5)

Figure 4 S/N ratio graph for surface roughness values:

Figure 5 shows the comparison of the predicted and actual surface roughness values. Therefore, it is seen that the data obtained with the model rather coincide with the experimental data. By comparison of the calculated values and the obtained model with actual values, the lowest error percentage was obtained with the first test parameter with 0.049% and the highest error percentage was obtained with test No. 3 with 0.38%.

Table 5 ANOVA table for surface roughness

Figure 5 Comparison of actual and predicted roughness

Figure 6 (a) shows the effect of MQL flow rate and cutting speed on the surface roughness at feed rate of 0.1 mm/tooth. Accordingly, thanks to the increasing cutting zone temperature at high cutting speeds, a smoother surface was obtained with the workpiece material softening [42]. Similarly, with the increase of the MQL ratio, a smoother surface is obtained with the lubricating effect occurring in the cutting region [22, 43]. In Figure 6(b), the effect of the feed rate and MQL flow rate on the surface roughness at 125 m/min cutting speed can be appreciated. Accordingly, surface roughness values decrease with increasing feed rate value. Likewise, with increasing MQL flow rate, surface roughness value decreases, but this effect is not much compared to the rate of progress. The rise in the amount of wear in the cutting tool as a result of the increased feed rate and high chip volume values can be explained by the decrease in the nose radius. (R_a=f_z²/r_ε, where r_ε is the nose radius of cutting tool) [44]. Similarly, in Figure 7, the effect of two different parameters on the surface roughness value can be seen more clearly in the 3D surface graph.

Figure 6 Contour plots of surface roughness:

Figure 7 Surface plots of surface roughness:

3.5 Cutting forces

Figure 8 shows the measured cutting forces with different test parameters in face milling operations in dry and MQL conditions. Due to the lubricating effect of vegetable oil used in MQL system, the cutting force values were found to be lower in almost all cutting parameters [5, 45]. In all cutting parameters and in both conditions, due to the increased material removal rate, a rise in the cutting force was observed with the rise in the feed rate [46].

Figure 8 Resultant cutting forces obtained for both conditions

In the case of surface milling tests at low cutting speeds, more moderate cutting forces were obtained. This can be attributed to the fact that, the cutting fluid does not reduce the cutting zone temperature at the desired level, due to the very high temperature in the cutting zone. This shows that the MQL system has a lower effect at higher cutting speeds and feed rate in milling operation of AISI O2.

Figures 9 and 10 show the fluctuation of the F_x (Figure 9(a)), F_y (Figure 9(b)), F_z (Figure 10(c)) and F_r (Figure 10(b)) cutting forces measured in face milling operations with the first test parameter under dry condition and MQL condition at the same cutting length depending on cutting time. Because the milling process is intermittent, oscillations occur in measured cutting forces values, as milling is an intermittent machining process [2, 13]. The cutting force values are directly linked to the wear of the cutting tool. The cutting force values increase with increasing cutting tool wear amount [47]. The cutting force fluctuations of all force components are seen to be lower in tests using the MQL system. Thanks to the vegetable oil used in the MQL system, the reduction in the amount of wear of the cutting tools led to the reduction of the cutting forces [8, 48].

3.6 S/N ratio for cutting force values

Table 6 shows the measured roughness and S/N values. The optimum levels for cutting forces are presented in Figure 11. An evaluation of Figure 11 shows that MQL flow rate of 50 mL/h, feed rate of 0.05 mm/tooth, cutting speed of 150 m/min, were the optimum parameters.

3.7 Evaluation of experimental results for cutting force with RSM

The ANOVA of resultant cutting force values is shown in Table 7. A “Probe>F” value of less than 0.05 indicates that this term is important for the model. In this case, all cutting parameters, the interaction and MQL flow rate, and feed rate value and the square of feed rate, are significant model terms. R-squared 0.9963 means that 99.63% of the total differences are correctly estimated by the model. The equation of response in terms of the significant factors for the cutting force values of the ANOVA method is given as follows:

F_r=124.31211-0.60712V_c+470.51743f_z+0.48573Q-3.058xf_zQ+2890.65154f_z² (6)

Figure 12 shows the assessment of the resultant cutting force values calculated with the experiments and the results obtained with the model. Accordingly, it shows that the model obtained predicts the results accurately.

Figure 9 Cutting force fluctuation F_x (a) and F_y (b) measured for both conditions

Figure 10 Cutting force fluctuation F_z (a) and F_r (b) measured for both conditions

Table 6 Cutting force results and S/N ratios

Figure 11 S/N ratio graphs for cutting force values

Table 7 ANOVA table for resultant cutting force

Figure 12 Comparison of actual and predicted cutting forces

Figure 13 shows the variation of cutting forces depending on the MQL flow and feed rate values. The reason why the graphics are given only for these two parameters is that their interactions are important in the model obtained. It can be seen from the graphs that the increase in the feed rate increases the cutting force. With increasing the MQL flow rate at the lowest feed rate value, the cutting force value has increased. This is because the workpiece material does not soften due to the cooling of the cutting zone. In high feed rate values, there was a decrease in cutting force values through increasing MQL flow rate value. This reduction in cutting force values can be said caused by temperature, which makes the workpiece material softer during machining. Consequently, it can be said that the increase in the flow rate of MQL at high feed rates has a positive effect on cutting performance.

Figure 13 Cutting force contour plot (a) and surface plot (b) depending on MQL flow rate and feed rate

3.8 Grey relational analysis

In the machining of different engineering materials, the minimum surface roughness and cutting force values are indicative of a higher machining performance. In the procedure of processing gray relational analysis data, surface roughness and cutting force values were taken as the “lower is the better”. The gray relational coefficients and the degrees obtained using the gray relational analysis steps are shown in Table 8. The test parameter, which has the highest gray relational degree calculated according to the values attained as a result of the experiments performed, is the seventh experiment. Therefore, they are multiple performance machining parameters for the lowest response values.

3.9 Confirmation tests

The results attained from the experiments by the optimum cutting parameters obtained from Taguchi analysis are shown in Table 9. Along with the results obtained, it is seen that the optimum cutting parameters are determined appropriately.

3.10 Wear analysis

Figure 14 shows the SEM images and EDS analysis of worn tools as a result of MQL and dry condition milling tests in test No. 1 of experimental design. After the milling operation for both conditions, wear type was seen as notch wear. Wear mechanisms have been detected as abrasive and adhesive wear. Due to the extraordinary temperature in the zone of cutting, the workpiece material is welded to cutting tool and progressive cutting stages. The welded workpiece materials were broken and they damaged the cutting tool material on the flank face in notch wear type. According to the EDS analysis obtained from the notch wear zone of the cutting tools used in dry condition built up edge (BUE) formation is seen. In the milling process with MQL system, BUE formation is less than that under dry conditions in the worn cutting tools notch wear zone. In addition, fracture was detected at the edge of the cutting tool used in dry conditions in the milling tests. The seizure zones are seen in the cutting tools flank face used in dry conditions due to increased temperature [6, 14, 49-51].

Table 8 GRC and GRG values

Table 9 Confirmation tests results

Figure 14 SEM images of worn cutting tools for MQL and dry condition

4 Conclusions

The cutting performance of the carbide cutting tools for milling AISI O2 steel was examined experimentally. So as to improve the performance of carbide cutting tools, vegetable oil is used as the cutting liquid which is harmless to the environment in MQL system. In order to comprehend the influence of vegetable oil used in MQL system, all experiments are repeated in dry conditions. Then, Taguchi and RSM methods are used for optimization of parameters according to measured values.

In almost all tests performed to determine the cutting tool life of the carbide cutting tools, the tool life increases with the use of cutting fluid, surface roughness is improved with using vegetable oil in MQL system.

In the tests performed in all cutting parameters, cutting forces decrease with increasing cutting speed for both conditions. Compared to the cutting force fluctuations obtained in all directions, the cutting force fluctuations with the MQL system are obtained lower than those obtained in dry condition.

Dominant wear mechanisms are determined as abrasion and adhesion. The workpiece material adhered to the cutting tool is due to the high temperature occurring during milling. In addition, seizure zones are seen on the flank face of the carbide cutting tool, due to the high temperature occurring during milling in dry condition.

Vegetable oil used as cutting liquid in MQL system has a positive effect on cutting performance of carbide cutting tools. It is thought that vegetable oil used in MQL system would be a good alternative to cutting fluid in machining industry.

Contributors

Bilal KURSUNCU provided the concept and edited the draft of manuscript. Yasin Ensar BIYIK conducted the literature review and wrote the first draft of the manuscript. Bilal KURSUNCU edited the draft of manuscript.

Conflict of interest

Bilal KURSUNCU and Yasin Ensar BIYIK declare that they have no conflict of interest.

References

[1] CHINCHANIKAR S, CHOUDHURY S K. Machining of hardened steel-Experimental investigations, performance modeling and cooling techniques: A review [J]. Int J Mach Tools Manuf, 2015; 89: 95-109. DOI: 10.1016/ j.ijmachtools.2014.11.002.

[2] UMBRELLO D, PU Z, CARUSO S, OUTEIRO J C, JAYAL A D, DILLON O W. The effects of cryogenic cooling on surface integrity in hard machining [J]. Procedia Eng, 2011, 19: 371-376. DOI: 10.1016/j.proeng.2011.11.127.

[3] FOX-RABINOVICH G S, KOVALEV A I, AGUIRRE M H, BEAKE B D, YAMAMOTO K, VELDHUIS S C. Design and performance of AlTiN and TiAlCrN PVD coatings for machining of hard to cut materials [J]. Surf Coatings Technol, 2009, 204: 489-496. DOI: 10.1016/ j.surfcoat.2009.08.021.

[4] ASILTURK I, AKKUS H. Determining the effect of cutting parameters on surface roughness in hard turning using the Taguchi method [J]. Meas J Int Meas Confed, 2011, 44: 1697-1704. DOI: 10.1016/j.measurement.2011.07.003.

[5] GUTNICHENKO O, BUSHLYA V, BIHAGEN S, STAHL J E. Influence of GnP additive to vegetable oil on machining performance when MQL-assisted turning Alloy 718 [J]. Procedia Manuf, 2018, 25: 330-337. DOI: 10.1016/ J.PROMFG.2018.06.091.

[6] KURSUNCU B, CALISKAN H, GUVEN S Y, PANJAN P. Wear behavior of multilayer nanocomposite TiAlSiN/TiSiN/ TiAlN coated carbide cutting tool during face milling of inconel 718 superalloy [J]. J Nano Res, 2017, 47: 11-16. DOI: 10.4028/www.scientific.net/JNanoR.47.11.

[7] VEPREK S, HOLUBAR P, VEPREK-HEIJMAN M. Industrial applications of hard and superhard nanocomposite coatings on tools for machining, forming, stamping and injection molding [J]. Adv Mater Res, 2016, 1135: 218-233. DOI: 10.4028/www.scientific.net/AMR.1135. 218.

[8] KAYNAK Y. Evaluation of machining performance in cryogenic machining of Inconel 718 and comparison with dry and MQL machining [J]. Int J Adv Manuf Technol, 2014, 72: 919-933. DOI: 10.1007/s00170-014-5683-0.

[9] MUSFIRAH A H, GHANI J A, HARON C H C. Tool wear and surface integrity of Inconel 718 in dry and cryogenic coolant at high cutting speed [J]. Wear, 2017, 376-377: 125-133. DOI: 10.1016/j.wear.2017.01.031.

[10] PARK K H, SUHAIMI M A, YANG G D, LEE D Y, LEE S W, KWON P. Milling of titanium alloy with cryogenic cooling and minimum quantity lubrication (MQL) [J]. Int J Precis Eng Manuf, 2017, 18: 5-14. DOI: 10.1007/ s12541-017-0001-z.

[11] KURSUNCU B, CALISKAN H, GUVEN S Y, PANJAN P. Improvement of cutting performance of carbide cutting tools in milling of the Inconel 718 superalloy using multilayer nanocomposite hard coating and cryogenic heat treatment [J]. Int J Adv Manuf Technol, 2018, 97: 1-4. DOI: 10.1007/ s00170-018-1931-z.

[12] DESHPANDE R G, VENUGOPAL K A. Machining with cryogenically treated carbide cutting tool inserts [J]. Mater Today Proc, 2018, 5: 1872-1878. DOI: 10.1016/J.MATPR. 2017.11.288.

[13] KURSUNCU B, CALISKAN H. The effect of multilayer nanocomposite TiAlSiN/TiSiN/TiAlN coating on cutting forces in face milling of Inconel 718 superalloy [C]// 2nd International Conference on Advances in Mechanical Engineering. Istanbul, 2016: 11-14.

[14] CALISKAN H, KURSUNCU B, GUVEN S Y, KARAOGLANLI A C, SABRI G M, ALSARAN A. Effect of boron nitride coating on wear behavior of carbide cutting tools in milling of Inconel 718 [C]// Machining, Joining and Modifications of Advanced Materials. Singapore: Springer, 2016: 13-21. DOI: 10.1007/978-981-10-1082-8_2.

[15] KURSUNCU B, CALISKAN H, GUVEN S Y. Wear behavior of cryogenically treated TiN coated carbide cutting tool while face milling Inconel 718 Superalloy [C]// The International Conference on Engineering and Natural Sciences (ICENS). Sarajevo, 2016: 583.

[16] KURSUNCU B, YARAS A. Assessment of the effect of borax and boric acid additives in cutting fluids on milling of AISI O2 using MQL system [J]. Int J Adv Manuf Technol, 2018, 95: 2005-2013. DOI: 10.1007/s00170-017-1301-2.

[17] KUMAR G K, RAVI S M. Past and current status of eco-friendly vegetable oil based metal cutting fluids [J]. Mater Today Proc, 2017, 4: 3786-3795. DOI: https://doi.org/ 10.1016/J.MATPR.2017.02.275.

[18] NGUYEN T K, DO I, KWON P. A tribological study of vegetable oil enhanced by nanoplatelets and implication in MQL machining [J]. Int J Precis Eng Manuf, 2012, 13: 1077-1083. DOI: 10.1007/s12541-012-0141-0.

[19] MAHADI M A, CHOUDHURY I A, AZUDDIN M, NUKMAN Y. Use of boric acid powder aided vegetable oil lubricant in turning AISI 431 steel [J]. Procedia Eng, 2017, 184: 128-136. DOI: 10.1016/j.proeng.2017.04.077.

[20] AGRAWAL S M, PATIL N G. Experimental study of nonedible vegetable oil as a cutting fluid in machining of M2 steel using MQL [J]. Procedia Manuf, 2018, 20: 207-212.

[21] JEEVAN T P, JAYARAM S R. Tribological properties and machining performance of vegetable oil based metal working fluids—A review [J]. Mod Mech Eng, 2018, 8(1): 42-65.

[22] GUO S, LI C, ZHANG Y, WANG Y, LI B, YANG M. Experimental evaluation of the lubrication performance of mixtures of castor oil with other vegetable oils in MQL grinding of nickel-based alloy [J]. J Clean Prod, 2017, 140: 1060-1076. DOI:10.1016/j.jclepro.2016.10.073.

[23] SHARMA J, SIDHU B S. Investigation of effects of dry and near dry machining on AISI D2 steel using vegetable oil [J]. J Clean Prod, 2014, 66: 619-623. DOI: 10.1016/ j.jclepro.2013.11.042.

[24] SHARMA A K, KATIYAR J K, BHAUMIK S, ROY S. Influence of alumina/MWCNT hybrid nanoparticle additives on tribological properties of lubricants in turning operations [J]. Friction, 2019, 7(2): 153-168. DOI: 10.1007/s40544- 018-0199-5.

[25] EKER B, EKICI B, KURT M, BAKIR B. Sustainable machining of the magnesium alloy materials in the CNC lathe machine and optimization of the cutting conditions [J]. Mechanics, 2014, 20: 310-316.

[26] YU D, WANG C, CHENG X, ZHANG F. Optimization of hybrid PVD process of TiAlN coatings by Taguchi method [J]. Appl Surf Sci, 2008, 255: 1865-1869. DOI: 10.1016/ J.APSUSC.2008.06.204.

[27] ASLANTAS K, EKICI E, CICEK A. Optimization of process parameters for micro milling of Ti-6Al-4V alloy using Taguchi-based gray relational analysis [J]. Measurement, 2018, 128: 419-427. DOI: 10.1016/ J.MEASUREMENT.2018.06.066.

[28] CALISKAN H, KURBANOGLU C, PANJAN P, KRAMAR D. Investigation of the performance of carbide cutting tools with hard coatings in hard milling based on the response surface methodology [J]. Int J Adv Manuf Technol, 2013, 66: 883-893. DOI: 10.1007/s00170-012-4374-y.

[29] PUSAVEC F, DESHPANDE A, YANG S, M’SAOUBI R, KOPAC J, DILLON O W. Sustainable machining of high temperature nickel alloy-Inconel 718: Part 1-Predictive performance models [J]. J Clean Prod, 2014, 81: 255-269. DOI: 10.1016/j.jclepro.2014.06.040.

[30] VISWANATHAN R, RAMESH S, SUBBURAM V. Measurement and optimization of performance characteristics in turning of Mg alloy under dry and MQL conditions [J]. Measurement, 2018, 120: 107-113. https:// doi.org/10.1016/J.MEASUREMENT.2018.02.018.

[31] MIA M, KHAN M A, DHAR N R. Study of surface roughness and cutting forces using ANN, RSM, and ANOVA in turning of Ti-6Al-4V under cryogenic jets applied at flank and rake faces of coated WC tool [J]. Int J Adv Manuf Technol, 2017, 93: 975-991.

[32] RAMESH S, VISWANATHAN R, AMBIKA S. Measurement and optimization of surface roughness and tool wear via grey relational analysis, TOPSIS and RSA techniques [J]. Measurement, 2016, 78: 63-72. DOI: 10.1016/ J.MEASUREMENT.2015.09.036.

[33] KIVAK T. Optimization of surface roughness and flank wear using the Taguchi method in milling of Hadfield steel with PVD and CVD coated inserts [J]. Measurement, 2014, 50: 19-28. DOI: J.MEASUREMENT.2013.12.017.

[34] VISWANATHAN R, RAMESH S, ELANGO N, KAMESH K D. Temperature measurement and optimization in machining magnesium alloy using RSM and ANOVA [J]. Pertanika J Sci Technol, 2017, 25(1): 255-262.

[35] SARIKAYA M, GULLU A. Taguchi design and response surface methodology based analysis of machining parameters in CNC turning under MQL [J]. J Clean Prod, 2014, 65: 604-616. DOI: 10.1016/j.jclepro.2013.08.040.

[36] WOJCIECHOWSKI S, W MARUDA R, M KROLCZYK G, NIESLONY P. Application of signal to noise ratio and grey relational analysis to minimize forces and vibrations during precise ball end milling [J]. Precis Eng, 2018, 51: 582-596. DOI: 10.1016/j.precisioneng.2017.10.014.

[37] VARGHESE V, ANNAMALAI K, KUMAR K S. Optimization of machining parameters for high feed end milling on Inconel 718 super alloy [J]. Appl Mech Mater, 2014, 592-594: 584-590. DOI: 10.4028/www.scientific.net/ AMM.592-594.584.

[38] TS ISO 7005-2 1998. Tool life testing in milling – Part1: Face milling [S].

[39] CHINCHANIKAR S, SALVE A V, NETAKE P, MORE A, KENDRE S, KUMAR R. Comparative evaluations of surface roughness during hard turning under dry and with water-based and vegetable oil-based cutting fluids [J]. Procedia Mater Sci, 2014, 5: 1966-1975. DOI: 10.1016/ J.MSPRO.2014.07.529.

[40] LU Y M, WANG T, WANG C Q, WANG C W. Empirical modeling of high-speed WEDM finishing in gas [J]. Adv Mater Res, 2012, 486: 503-508. DOI: 10.4028/www. scientific.net/AMR.486.503.

[41] CHEN L, ZHAO Y, YAN F, HOU H. Statistical investigations of serpentine channel pouring process parameters on semi-solid ZL101 aluminum alloy slurry using response surface methodology [J]. J Alloys Compd, 2017, 725: 673-683. DOI: 10.1016/J.JALLCOM.2017.07.169.

[42] TIAN X, ZHAO J, ZHAO J, GONG Z, DONG Y. Effect of cutting speed on cutting forces and wear mechanisms in high-speed face milling of Inconel 718 with Sialon ceramic tools [J]. Int J Adv Manuf Technol, 2013, 69: 2669-2678. DOI: 10.1007/s00170-013-5206-4.

[43] LI M, YU T, YANG L, LI H, ZHANG R, WANG W. Parameter optimization during minimum quantity lubrication milling of TC4 alloy with graphene-dispersed vegetable-oil- based cutting fluid [J]. J Clean Prod, 2019, 209: 1508-1522. DOI: 10.1016/J.JCLEPRO.2018.11.147.

[44] TSCHATSCH H. Applied machining technology [M]. Springer Science & Business Media, 2010.

[45] ZHANG S, LI J F, WANG Y W. Tool life and cutting forces in end milling Inconel 718 under dry and minimum quantity cooling lubrication cutting conditions [J]. J Clean Prod, 2012, 32: 81-87. DOI: 10.1016/j.jclepro.2012.03.014.

[46] SURESH R, BASAVARAJAPPA S, GAITONDE V N, SAMUEL G L. Machinability investigations on hardened AISI 4340 steel using coated carbide insert [J]. Int J Refract Met Hard Mater, 2012, 33: 75-86. DOI: 10.1016/j.ijrmhm. 2012.02.019.

[47] TEBALDO V, DI CONFIENGO G G, FAGA M G. Sustainability in machining: “Eco-friendly” turning of Inconel 718. Surface characterization and economic analysis [J]. J Clean Prod, 2017, 140: 1567-1577. DOI: 10.1016/ j.jclepro.2016.09.216.

[48] LIAO Y S, LIAO C H, LIN H M. Study of oil-water ratio and flow rate of MQL fluid in highspeed milling of Inconel 718 [J]. Int J Precis Eng Manuf, 2017, 18: 257-62. DOI: 10.1007/s12541-017-0033-4.

[49] CALISKAN H, KURBANOGLU C, PANJAN P, CEKADA M, KRAMAR D. Wear behavior and cutting performance of nanostructured hard coatings on cemented carbide cutting tools in hard milling [J]. Tribol Int, 2013, 62: 215-222. DOI: 10.1016/j.triboint.2013.02.035.

[50] KURSUNCU B, YARAS A. AISI O2 Tak1m Celiginin Frezelenmesinde Minimum Miktarda Yaglama (MQL) Sisteminin Kesme Performans1na Etkisi [J]. Mühendislik ve Teknol Bilimleri Dergisi, 2018, 5: 125-128. (in Turkish)

[51] CALISKAN H, CELIL C C, PANJAN P. Effect of multilayer nanocomposite TiAlSiN/TiSiN/TiAlN coating on wear behavior of carbide tools in the milling of hardened AISI D2 Steel [J]. J Nano Res, 2016, 38: 9-17. DOI: 10.4028/ www.scientific.net/JNanoR.38.9.

(Edited by FANG Jing-hua)

中文导读

采用Taguchi法和灰色关联分析法优化AISI O2钢MQL辅助面铣削的切削参数

摘要：本文研究了对环境无害的植物油在AISI O2钢面铣加工最小润滑量(MQL)系统中的可行性，并利用不同的统计方法对切削参数进行优化。用植物油作切削液，试验模式采用Taguchi法。测试结果表明，由于MQL系统的存在，在干燥条件下所有参数下的切削性能都得到了改善。在切削长度为7.5 m，切削速度为100 m/min，MQL流量为100 m/h，进给率0.1 mm/齿的条件下，刀具寿命最长。根据Taguchi分析确定最佳切削参数，并得到试验验证。此外，还采用灰色关联分析法确定了最佳试验参数。对测量的表面粗糙度和切削力进行方差分析，发现最有效的切削参数是进给率。本研究还建立了表面粗糙度和切削力的模型，其精度分别为99.63%和99.68%。有效的磨损机制是磨损和黏附。

关键词：硬铣；最小润滑量；刀具磨损；灰色关联分析；Taguchi法；AISI O2钢

Received date: 2019-10-17; Accepted date: 2020-07-08

Corresponding author: Bilal KURSUNCU, PhD, Associate Professor; Tel: +90-378-5011000(1665); E-mail: bkursuncu@bartin.edu.tr; ORCID: https://orcid.org/ 0000-0002-2304-2962