Journal of Harbin Institute of Technology (New Series)  2021, Vol. 28 Issue (2): 38-46  DOI: 10.11916/j.issn.1005-9113.2019063
0

Citation 

Suha K Shihab, Ethar Mohamed Mubarak, Rawaa Hamid Al-Kalali. Influence and Optimization of Surface Roughness on Surface Integrity during Turning Using Grey Relational Analysis[J]. Journal of Harbin Institute of Technology (New Series), 2021, 28(2): 38-46.   DOI: 10.11916/j.issn.1005-9113.2019063

Corresponding author

Suha K Shihab, E-mail: suhakshihab@gmail.com

Article history

Received: 2019-10-09
Influence and Optimization of Surface Roughness on Surface Integrity during Turning Using Grey Relational Analysis
Suha K Shihab1, Ethar Mohamed Mubarak2, Rawaa Hamid Al-Kalali2     
1. Department of Materials Engineering, College of Engineering, University of Diyala, Diyala 32001, Iraq;
2. Institute of Technology, Middle Technical University, Baghdad 10001, Iraq
Abstract: Current machining studies have reported effects of prevalent and common factors, while ultra-high finish requires holistic approach to identify all factors and investigate their effects on machining of hard to machine materials. In this work, a less investigated yet important factor, roughness of the uncut surface, was studied, and its effects on the individual response, i.e., surface finish of the machined part, were found to be significant. AISI 316, which is mainly applied in strategic areas, was selected and three effective turning factors, cutting speed (A), feed rate (B), and roughness of the uncut surface (C) on three output responses including surface roughness of the machined surface (Ra), microhardness(HV), and material removal rate (MRR), were reported. Further, single response optimization of the individual output response and multi-response optimization of all the three responses were carried out. Taguchi L9 orthogonal array based signal-to-noise (S/N) ratio method was used for individual response optimization, and grey relational analysis (GRA) was employed for multi-response optimization. Effects of the process factors on the output responses were evaluated through inclusive statistical analyses. The individual response optimization revealed that there was a considerable effect of roughness of the uncut surface on the machining performance. Results of the GRA illustrated that the speed during the cutting process and the feed rate had substantial trace on the surface integrity (indicated by Ra and HV) and production rate (indicated by MRR), while roughness of the uncut surface did not have a significant effect.
Keywords: turning    surface integrity    Taguchi method    optimization    GRA    
0 Introduction

AISI 316 austenitic stainless steel is extensively utilized in industries such as naval, aeronautical, and nuclear power plants due to its excellent mechanical properties and corrosion resistance[1]. The requirements of improved surface integrity (SI) including surface hardness and surface finish are becoming demanding. However, problems are associated with the machining of this steel owing to its high strain hardening and tensile strength with poor thermal conductivity, which may increase the cutting forces and the temperature during machining and lead to poor surface finish[2-3]. Most machining studies report effects of prevalent and common factors, while ultra-high finish requires holistic approach to identify all factors and investigate their effects on the machining of hard to machine materials. Thus, a less investigated yet important factor, roughness of the uncut surface, was discussed in this study, and its effects on the individual response, i.e., surface finish of the machined part, were found to be significant. SI enhancement for critical structural components is essential by combining multi-responses such as surface roughness of the machined surface (Ra) and surface flaws, microstructural characteristics, and surface mechanical properties such as microhardness and residual stresses[4]. Wang and Liu[5] provided a comprehensive review on the effects of structure, material, and tool wear on SI. Ezilarasan et al.[6] studied different responses of SI of C-263 Nimonic Alloy during turning process including its microhardness (HV), Ra, and residual stress, where Ra is the foremost among the SI characteristics and a predominant parameter in machining processes and product quality control[7]. Reducing surface roughness affects the finishing accuracy and increases the hardness of the workpiece surface, which in turn induces compressive residual stresses leading to improved fatigue life, increased wear, and corrosion resistances[8]. Researchers have investigated the suitability of various experimental models to calculate and analyze the surface roughness of the machined surfaces, including finite element models and design of experiments (DoE) such as response surface methodology (RSM), full factorial design, and Taguchi design[9]. Sharma and Gupta[10] applied Taguchi L9 in dry turning of SS 304 and used uncoated and multilayer coated carbide tools to study the effect of input machining parameters on tool wear and Ra. Palanisamy et al.[11] made an attempt to optimize the turning parameters for Incolloy 800H under cryogenic condition using multilayer CVD-coated tool. In their study, the degree of work hardening, Ra, HV, and material removal rate (MRR) were regarded as the response variables. Manav and Chinchanikar[12] implemented multi-response optimization using genetic algorithm to obtain minimum cutting temperature (T), cutting forces (F), and Ra with maximum tool life during turning of hardened AISI 4340 steels with different hardness. Moreover, it was found from recent studies that exploring the sustainability and machinability of stainless steel using multi-response optimization is imperative for manufacturers and customers to achieve the desired SI and productivity. Willert et al.[13] investigated the SI of 42CrMoS4 steel during turning process. Vishwas et al.[14] performed a comprehensive analysis of SI, namely Ra and tool-workpiece interface temperature, when machining AISI 410 martensitic stainless steel using Taguchi orthogonal array. Sachin et al.[15] explored the impact of main control factors (i.e., surface hardness, surface topography and morphology, Ra, and residual stresses of 17-4 PH SS) on the characteristics of SI under minimum quantity lubrication (MQL) condition utilizing modified tool. Zerti et al.[16] carried out multi-objective optimization for maximizing productivity and minimizing F, Ra, and power at different machining parameters including cutting speed (A), feed rate (B), and depth of the cut (D) during hard turning of AISI 420.

In the turning process, it is crucial but challenging for manufacturers to select the optimum machining parameters that can lead to high machining performance for a specific environment and machine. The machining parameters A, B, and D were observed to be the main factors that affect SI[4, 17-18]. Further, apart from using the common input turning parameters, the focus now changes to the investigation of the effects of uncommon parameters such as roughness of the uncut surface of the workpiece and tool wear on the machining performance. The demand for higher surface finish is increasing, which is one of the challenging areas of metal machining. During turning process, B affects the advance of tool of each revolution, which directly influences the surface finish of the machined component. A close observation on the machining theory reveals that variations in the roughness of the uncut surface(C) will change the shear place area, which consequently affects the cutting forces. Greater roughness of the uncut surface will cause larger variations in the cutting forces and contribute to vibrations, which in turn leads to poor surface finish of the machined surface. It might also have a significant effect on the efficiency of machining processes through its impact on the tool wear, and its robust connection with the selected values of the feed rate may affect the quality of the machined workpiece in terms of Ra. In addition, the roughness of the uncut surface may increase the generated temperature during machining, which can induce residual stresses and affect HV.

Grey relational analysis (GRA) is a simple but valuable multi-response optimization technique, which estimates the effectiveness of different input parameters[19]. In addition, it is an efficient approach for the determination of the grade of approximation between sequences employing grey relational grade (GRG)[20]. Therefore, the use of GRA for the experimental results obtained using Taguchi method can significantly reveal the single optimization of response to the optimization of multiple responses[21].

Based on the literature and the consideration of the uncommon parameters during machining, a research on the effect of roughness of the uncut surface as uncommon machining parameter and other common machining parameters including A and B on SI is presented in this paper. The novelty of this paper is that in addition to common machining parameters, roughness of the uncut surface is discussed in the experimental investigation of SI, which is a less studied but an important factor. Analysis of variance (ANOVA) was employed to analyze significance of the turning factors on the output responses Ra, HV, and MRR. Also, the optimization of individual response on the basis of signal-to-noise (S/N) ratio and the optimization of multiple responses using GRA were applied to achieve better quality in terms of Ra and HV, and enhance the productivity defined by higher MRR while turning AISI 316 austenitic stainless steel.

1 Material and Method

Austenitic stainless steel 316 with bulk hardness of 192 HV was used as the workpiece material in this study. Table 1 presents the chemical composition of the workpiece material. The workpiece material was in the form of a cylindrical bar of 28 mm diameter, and the cutting length was kept at 50 mm for each experiment. Fig. 1 shows the overall work plan and the experimental setup of the study, comprising of a CNC lathe machine (type KNUTH) and a multilayer coated carbide cutting tool insert (type TNMG 160408 TN2000, manufactured by WIDIA, INDIA).

Table 1 Chemical composition of the workpiece material 

Fig.1 Experimental setup and the Taguchi based GRA steps

1.1 Measurements of Output Responses

The surface roughness Ra of the machined samples was measured using an automatic digital device (MAHR FEDERAL POCKET SURF 44100 model). The measurement was repeated for four times for the same machined surface of each experiment and the average value was recorded for further analysis.

The digital micro-vickers hardness machine (Model: TH714) was employed to measure the micro-hardness values of the machined samples. The measurement was repeated for three times at several positions on each machined sample and the average value was recorded for further analysis. MRR is an important measure of productivity that the higher MRR is, the higher the productivity is, and vice-versa, which is expressed as follows[22]:

$ \operatorname{MRR}=(A)(B)(D) $ (1)

where A denotes the cutting speed (m/min), B denotes the feed rate (mm/rev), and D denotes the depth of cut (mm).

1.2 Taguchi Design and GRA

Three turning factors A, B, and C at three levels were considered in this study. Throughout the experimental investigations, D was kept constant at 0.3 mm. Table 2 shows the turning factors and their levels. Taguchi design is an efficient approach to assess main effects of the factors using minimum number of experimental runs. Taguchi's L9 is a suitable orthogonal array (OA) to investigate main effects of the three turning factors with three levels. Therefore, L9 OA was selected to conduct the experimental study, and S/N ratio and ANOVA were used to analyze the experimental results. The used Taguchi's L9 OA in terms of coded and actual values are represented in Table 3.

Table 2 Turning factors and their levels

Table 3 Taguchi's L9 OA in terms of coded and actual values

In general, there are three quality characteristics, i.e., the lower-the-better (LTB), the nominal-the-better (NTB), and the higher-the-better (HTB) to compute S/N ratio. For Ra, LTB is the suitable quality characteristic since it is required to be minimized. S/N ratio for Ra was calculated using Eq.(2). On the other hand, HTB is the desired quality attribute for HV and MRR, and for these output responses the S/N ratio was computed using Eq.(3)[22-23].

$ \mathrm{S} / \mathrm{N}=-10 \log \frac{1}{m} \sum\limits_{i=1}^{m} y_{i}^{2} $ (2)
$ \mathrm{S} / \mathrm{N}=-10 \log \frac{1}{m} \sum\limits_{i=1}^{m} \frac{1}{y_{i}^{2}} $ (3)

where m represents the number of the experiment, and yi signifies the value of the output response of the ith experiment.

1.3 Grey Relational Coefficient (GRC) and Grey Relational Grade (GRG)

GRA was employed for multi-response optimization to determine the optimal setting of the turning factors that yielded optimal values of the output responses simultaneously. This method is used to convert a multi-objective assignment with conflicting objectives into a single-objective problem and illustrates the correlation between the input and the output variables of the manufacturing process. Commonly, the execution attributes of this method depend on the GRG which is obtained by calculating the average value of GRC of the output responses Ra, HV, and MRR. Measured values of the output responses were normalized to decrease unpredictability, known as data preprocessing, which was required because the diversity and unit of each response were different from others. Normalization of the Ra values was performed using Eq.(4) and HV, and MRR values using Eq.(5)[24].

$ y_{i}^{n}(k)=\frac{\max \left(y_{i}(k)\right)-y_{i}(k)}{\max \left(y_{i}(k)\right)-\min \left(y_{i}(k)\right)} $ (4)
$ y_{i}^{n}(k)=\frac{y_{i}(k)-\min \left(y_{i}(k)\right)}{\max \left(y_{i}(k)\right)-\min \left(y_{i}(k)\right)} $ (5)

where yi(k) is the target value, yin(k) is the later grey relational formation value, and min(yi(k)) and max(yi(k)) are the smallest and largest values of yi(k) for the kth response correspondingly.

The GRC values which describe the relationship between the demanded and the actual experimental data were computed based on the normalized values as follows:

$ \gamma\left(y_{0}(k), y_{i}(k)\right)=\frac{\delta_{\min }+\zeta \delta_{\max }}{\delta_{i}(k)+\zeta \delta_{\max }} $ (6)

where 0 < γ(y0(k), yi(k))≤1, ξ is the recognition coefficient with acceptable values between 0 and 1, ξ∈[0, 1]; δi(k) is the deviation sequence of the absolute values between y0(k) and yi(k), namely,

$ \delta_{i}(k)=\left| y_{0}(k)-y_{i}(k)\right| \\ \begin{array}{l} \delta_{\max }=\max\limits_{\forall j \in i} \max\limits_{\forall k}\left|y_{0}(k)-y_{i}(k)\right| \\ \delta_{\min }=\min\limits_{\forall j \in i} \min\limits_{\forall k}\left|y_{0}(k)-y_{i}(k)\right| \end{array} $

The mean value of GRC was taken, which represents the GRG(γi) using Eq.(7)[24] as

$ \gamma_{i}=\frac{1}{n} \sum\limits_{k=1}^{n} \zeta_{i}(k) $ (7)

where n represents the number of quality characteristics.

The correlation between the reference sequence and the comparison sequences was characterized by GRG. Higher GRG value revealed that experimental values were in strong relational grade with the normalized values. Thus, higher relational grade indicated that the collection of the process parameters moved toward the optimum value.

2 Results and Discussion 2.1 Effects of Turning Factors on Ra, HV, and MRR

The experimental data for different output responses Ra, HV, and MRR are shown in Table 4.

Table 4 Experimental results for different output responses

The probability plots in Fig. 2 confirmed the normal distribution of the experimental data of Table 4, which were obtained for a confidence level of 95%. The significant effect of each factor on the output responses Ra, HV, and MRR was verified by the ANOVA results, as shown in Tables 5-7. In the ANOVA tables, model term is significant if its p-value is less than 0.05. Based on the F-value and p-value, it is evident from Table 5 that B was the most important factor to affect Ra, followed by C and then A. It is attributed to the theory of metal cutting[22], which suggests that increase in B and C results in the increase of vibration acceleration leading to high Ra. The results of this study were in line with those in Refs.[25-27]. From Table 6, it can be found that the impact of A on the microhardness was greater than B and C. T increased with the increase of A, resulting in the hardening of the machined surface, i.e., an increase in the microhardness. In addition, severe plastic deformation of the material occurred at high value of B due to the decrease of the grain size of the machined surface, which resulted in the increase of the microhardness. Similar results were reported in previous studies[28]. Table 7 shows the ANOVA results for MRR. It is obvious that within the investigated range of the turning factors, B had the most significant influence on the MRR while C had no effect on MRR, which was justified by Eq.(1), where the material removal rate mainly depends on primary machining parameters.

Fig.2 Probability plots for Ra, HV, and MRR

Table 5 ANOVA results for Ra

Table 6 ANOVA results for HV

Table 7 ANOVA results for MRR

The optimization of individual output response was achieved using S/N ratio plots, as exhibited in Fig. 3(a)-(c). The optimum level of the turning factors was selected based on the highest S/N ratio. It can be found from Fig. 3(a) that high value of A and low value of B and C resulted in the minimum Ra. Thus, the highest A (level 3, 120 m/min), the lowest B (level 1, 0.05 mm/rev), and the lowest C (level 1, 1.07 μ m) were the optimum turning factors for minimum Ra. Also, according to Fig. 3(b), highest A (level 3, 120 m/min), highest B (level 3, 0.15 mm/rev), and lowest C (level 1, 1.07 μ m) were the optimum turning factors for the maximum value of HV. Similarly, as evident in Fig. 3(c), the highest level of A and B (A=120 m/min and B= 0.15 mm/rev) were suggested for the optimum production rate (maximum MRR) and there is no predominant variation in the MRR with the changing C.

Fig.3 S/N ratio plots for Ra, HV, and MRR

2.2 Optimization of Multiple Responses Based GRA

The study employed the Taguchi based GRA to optimize multiple responses of turning factors. Table 8 shows the calculated S/N ratio and normalized values of each investigated response. Table 9 presents all the normalized values obtained using Eqs. (4)-(5). The reference sequence is represented as y0 (k) and the comparison sequences as yi (k). Moreover, $\partial_{\min }, \partial_{\max }$, and $\partial_{i}$ for the nine runs and for each response were computed, and the value of ζ was taken as 0.5. Eq.(7) was employed to calculate GRG values. Table 10 records the GRC and the GRG for all the experiments.

Table 8 S/N ratio and normalized values of the output responses

Table 9 The sequence after executing data preprocessing

Table 10 GRC and GRG for the sequences

Table 11 shows the response table of the Taguchi method, where the average value of the GRG at each level of the turning factors is listed. The maximum value of GRG showed that the product quality would be closer to the ideal value. Therefore, a larger GRG was preferred for optimum performance. As shown in Table 11, the optimum level of the turning factors to obtain the best SI with higher production rate was (A3, B3, and C2). It can be observed in Table 10 that the value of the GRG was the maximum (0.8546) for experiment number 9. Thus, the factor setting of experiment number 9 was the optimum. The ANOVA was employed to examine the effect of turning factors on multi-responses. The ANOVA results for GRG are presented in Table 12, in which the two turning factors (A and B) had considerable effects on SI and MRR at 95% significance level because p-value was less than 0.05. Moreover, it was observed that B was the most significant turning factor with the highest contribution of 56.70%, followed by A with the contribution of 31.52%. It can also be seen in Table 12 that effects of C was not significant. The effect of base metal roughness, namely C on the SI, reduced at the time of the effect of other conflict responses (indicated by HV and MRR) using multi-response optimization based on GRA. The effect of feed and initial roughness was significantly large as discussed in Section 2.1. It is noteworthy that for smaller values of B, the roughness reduced sharply. Also, at smaller levels of feed, greater roughness occurred, which caused large variation in the shear plane thickness. Variation in the shear plane thickness greatly changed the cutting forces. Machining with varying forces induced vibrations and caused the roughness of the machined surface to increase. However, in the case of multi-response, the three responses (Ra, HV, and MRR) were equally weighted and the effect of B and A on HV and MRR was very high. Consequently, the effect of C on the multi-response turned out to be insignificant. While when Ra was the sole response criteria, B and C still remained significant factors. In this study, confirmation test was not required since the optimal turning factor combination (A3, B3, and C2) was the same as the parameter setting of experiment number 9.

Table 11 Response table for GRG

Table 12 ANOVA results for GRG

3 Conclusions

This study investigated the influence of two common and one uncommon turning factors (A, B, and C) on three output responses Ra, HV, and MRR while machining AISI 316 austenitic stainless steel. Taguchi's L9 OA was adopted to perform experiments, and results for response variables were obtained. The significant effect of the turning factors on the output responses was determined using S/N ratio and ANOVA. Multi-response optimization was carried out using GRA to achieve minimum Ra and maximum MRR and HV. The following conclusions are drawn:

1) Taguchi design is an efficient and valuable approach for single response optimization with least number of experiments.

2) Optimization of individual response based on S/N ratio revealed that highest value of A and lowest values of B and C resulted in the minimum Ra. There was a significant effect of roughness of the uncut surface on the individual output response. The highest values of A and B were responsible for the maximum HV and MRR.

3) GRA is a relatively simple multi-response optimization method. The ANOVA results for GRG showed that B dominantly affected SI, which was followed by A, while the influence of C was negligible.

4) Results of the GRA based multi-response optimization showed that A at 120 m/min, B at 0.15 mm/rev, and C at 2.17 mm led to the optimum SI with higher production rate.

The study only considered three levels of each of the three turning factors, cutting speed, feed rate, and roughness of the uncut surface, and investigated their effects on SI and on the production rate, whereas various other parameters such as rake angle, tool geometry, and tool wear may also affect SI, and thus should be included in future studies. Further, other DoE such as response surface methodology may be applied for optimization.

References
[1]
Vasumathy D, Meena A, Duraiselvam M. Experimental study on evaluating the effect of micro textured tools in turning AISI 316 austenitic stainless steel. Procedia Engineering, 2017, 184: 50-57. DOI:10.1016/j.proeng.2017.04.070 (0)
[2]
Bagaber S A, Yusoff A R. Multi-objective optimization of cutting parameters to minimize power consumption in dry turning of stainless steel 316. Journal of Cleaner Produdction, 2017, 157: 30-46. DOI:10.1016/j.jclepro.2017.03.231 (0)
[3]
Naves V T G, Da Silva M B, Da Silva F J. Evaluation of the effect of application of cutting fluid at high pressure on tool wear during turning operation of AISI 316 austenitic stainless steel. Wear, 2013, 302(1-2): 1201-1208. DOI:10.1016/j.wear.2013.03.016 (0)
[4]
Zhang W, Wang X, Hu Y, et al. Quantitative studies of machining-induced microstructure alteration and plastic deformation in AISI 316 stainless steel using EBSD. Journal of Materials Engineering and Performance, 2018, 27(2): 434-446. DOI:10.1007/s11665-018-3129-9 (0)
[5]
Wang B, Liu Z. Influences of tool structure, tool material and tool wear on machined surface integrity during turning and milling of titanium and nickel alloys: a review. International Journal of Advanced Manufacturing Technology, 2018, 98(5-8): 1925-1975. DOI:10.1007/s00170-018-2314-1 (0)
[6]
Ezilarasan C, Senthil Kumar V S, Velayudham A. Effect of machining parameters on surface integrity in machining nimonic C-263 super alloy using whisker-reinforced ceramic insert. Journal of Materials Engineering and Performance, 2013, 22(6): 1619-1628. DOI:10.1007/s11665-012-0439-1 (0)
[7]
Petropoulos G P, Pandazaras C N, Davim J P. Surface texture characterization and evaluation related to machining. Davim J P. Surface Integrity in Machining. London: Springer, 2010: 37-66. DOI:10.1007/978-1-84882-874-2_2 (0)
[8]
Saldaña-Robles A, Plascencia-Mora H, Aguilera-Gómez E, et al. Influence of ball-burnishing on roughness, hardness and corrosion resistance of AISI 1045 steel. Surface and Coatings Technology, 2018, 339: 191-198. DOI:10.1016/j.surfcoat.2018.02.013 (0)
[9]
Shunmugesh B J K, Kumar M P, Arun K. Optimization in turning of 11sMn30 through process capability index. Materials Today: Proceedings, 2019, 11(3): 961-970. DOI:10.1016/j.matpr.2018.12.025 (0)
[10]
Sharma N, Gupta K. Influence of coated and uncoated carbide tools on tool wear and surface quality during dry machining of stainless steel 304. Materials Research Express, 2019, 6(8): 086585. (0)
[11]
Palanisamy A, Selvaraj T, Sivasankaran S. Optimization of turning parameters of machining Incoloy 800H superalloy using cryogenically treated multilayer CVD-coated tool. Arabian Journal for Science and Engineering, 2018, 43(9): 4977-4990. DOI:10.1007/s13369-018-3287-y (0)
[12]
Manav O, Chinchanikar S. Multi-objective optimization of hard turning: a genetic algorithm approach. Materials Today: Proceedings, 2018, 5(5): 12240-12248. DOI:10.1016/j.matpr.2018.02.201 (0)
[13]
Willert M, Riemer O, Brinksmeier E. Surface integrity in precision turning of steel. International Journal of Advanced Manufacturing Technology, 2018, 94(1-4): 763-771. DOI:10.1007/s00170-017-0696-0 (0)
[14]
Vishwas C J, Girish L V, Naik G M, et al. Effect of machining parameters on surface integrity during dry turning of AISI 410 martensitic stainless steel. IOP Conference Series: Materials Science and Engineering, 2018, 376(1): 012127. DOI:10.1088/1757-899X/376/1/012127 (0)
[15]
Sachin B, Narendranath S, Chakradhar D. Sustainable diamond burnishing of 17-4 PH stainless steel for enhanced surface integrity and product performance by using a novel modified tool. Materials Research Express, 2019, 6(4): 046501. DOI:10.1088/2053-1591/aaf900 (0)
[16]
Zerti A, Yallese M A, Meddour I, et al. Modeling and multi-objective optimization for minimizing surface roughness, cutting force, and power, and maximizing productivity for tempered stainless steel AISI 420 in turning operations. International Journal of Advanced Manufacturing Technology, 2019, 102(1-4): 135-157. DOI:10.1007/s00170-018-2984-8 (0)
[17]
Puh F, Jurkovic Z, Perinić M, et al. Optimization of machining parameters for turning operation with multiple quality characteristics using Grey relational analysis. Tehnički Vjesnik-Technical Gazette, 2016, 23(2): 377-382. DOI:10.17559/TV-20150526131717 (0)
[18]
Patole P B, Kulkarni V V. Optimization of process parameters based on surface roughness and cutting force in MQL turning of AISI 4340 using nano fluid. Materials Today: Proceedings, 2018, 5(1): 104-112. DOI:10.1016/j.matpr.2017.11.060 (0)
[19]
Zuo W, E J, Liu X, et al. Orthogonal experimental design and fuzzy grey relational analysis for emitter efficiency of the micro-cylindrical combustor with a step. Applied Thermal Engineering, 2016, 103: 945-951. DOI:10.1016/j.applthermaleng.2016.04.148 (0)
[20]
E J, Li Y Q, Gong J K. Function chain neural network prediction on heat transfer performance of oscillating heat pipe based on grey relational analysis. Journal of Central South University of Technology, 2011, 18(5): 1733. DOI:10.1007/s11771-011-0895-z (0)
[21]
Ghetiya N D, Patel K M, Kavar A J. Multi-objective optimization of FSW process parameters of aluminium alloy using Taguchi-based grey relational analysis. Transactions of the Indian Institute of Metals, 2016, 69(4): 917-923. DOI:10.1007/s12666-015-0581-1 (0)
[22]
Mia M, Dey P R, Hossain M S, et al. Taguchi S/N based optimization of machining parameters for surface roughness, tool wear and material removal rate in hard turning under MQL cutting condition. Measurement, 2018, 122: 380-391. DOI:10.1016/j.measurement.2018.02.016 (0)
[23]
Shi K, Zhang D, Ren J. Optimization of process parameters for surface roughness and microhardness in dry milling of magnesium alloy using Taguchi with grey relational analysis. International Journal of Advanced Manufacturing Technology, 2015, 81(1-4): 645-651. DOI:10.1007/s00170-015-7218-8 (0)
[24]
Kalyon A, Günay M, Özyürek D. Application of grey relational analysis based on Taguchi method for optimizing machining parameters in hard turning of high chrome cast iron. Advances in Manufacturing, 2018, 6(4): 419-429. DOI:10.1007/s40436-018-0231-z (0)
[25]
Trent E M, Wright P K. Coolants and lubricants. Trent E M, Wright P K. Metal Cutting (Fourth Edition). Oxford: Butterworth-Heinemann, 2000. DOI: 10.1016/B978-075067069-2/50012-7. (0)
[26]
Shihab S K, Khan Z A, Mohammad A, et al. Optimization of surface integrity in dry hard turning using RSM. SĀDHANĀ - Academy Proceedings in Engineering Sciences, 2014, 39(5): 1-19. DOI:10.1007/s12046-014-0263-4 (0)
[27]
Savas V, Ozay C, Ballikaya H. Experimental investigation of cutting parameters in machining of 100Cr6 with tangential turn-milling method. Advances in Manufacturing, 2016, 4(1): 97-104. DOI:10.1007/s40436-016-0134-9 (0)
[28]
Uddin M S, Rosman H, Hall C, et al. Enhancing the corrosion resistance of biodegradable Mg-based alloy by machining-induced surface integrity: influence of machining parameters on surface roughness and hardness. International Journal of Advanced Manufacturing Technology, 2017, 90(5-8): 2095-2108. DOI:10.1007/s00170-016-9536-x (0)