Microsoft Word - cet-01.docx CHEMICAL ENGINEERING TRANSACTIONS VOL. 46, 2015 A publication of The Italian Association of Chemical Engineering Online at www.aidic.it/cet Guest Editors: Peiyu Ren, Yancang Li, Huiping Song Copyright © 2015, AIDIC Servizi S.r.l., ISBN 978-88-95608-37-2; ISSN 2283-9216 Orthogonal Test Analysis and Modeling of GCr15 Bearing Steel Machined Surface Residual Stress in Hard Precision Turning Chen Guangjuna,b, Wang Liangb, Chen Weidongb,Zhang Jinduob, Zhou Xiaoqina a School of Mechanical Science and Engineering , Jilin University , Changchun 130025, China b School of Mechanical Engineering, Jiamusi University, Jiamusi 154007, China chenguangjun@126.com Orthogonal experimental analysis of GCr15 bearing steel residual stress in machined surface is carried out; Based on 1stOpt software, Using the Marquardt method and universal global optimization algorithm established Multivariate nonlinear prediction model of residual stress in machined surface. There sults of the experiment show that With the increase of cutting speed, the residual tensile stress showed a trend of increasing first and then decreasing; With the increase of feed rate, the residual tensile stress decreases first and then increases, and the residual compressive stress first increases and then decreases; With the increase of cutting depth, residual tensile stress increased slowly (compressive stress decreases); With the increase of the radius, the residual tensile stress increases (compressive stress decreases). The increase of cutting vibration makes the residual tensile stress increase or decrease of the residual compressive stress. Through the test and analysis, the establishment of the prediction model fitting degree is excellent and has statistical significance. 1. Introduction Hard cutting has the advantages of good machining flexibility, machining efficiency and surface quality, and it has become a new way to hardened aluminium processing. However, control of surface integrity in hard cutting, especially the prediction and modeling of surface residual stress also needs to be studied systematically. Related scholars have been carried out a lot of researches on the surface residual stress of the machining. Through the hard cutting experiment, Dahlman contrasts the distribution of hard cutting and grinding residual stress. It is indicated that compared with grinding, hard cutting can make the residual stress distributed to a deeper metal layer, the residual stress in the surface layer is greater. With cutting parameters as the influencing factors, Liu M and Barbacki et al. carry out the hard cutting experiments. Through the analysis of the experimental results, it is pointed out that the influence of feed rate on the surface roughness is significant, Increase the feed rate will cause the surface residual compression stress decreases gradually, and finally transformed into tensile stress, and can distribute the residual stress to the distance from the surface. Hua et al. study the influence of cutting tool radius on the residual stress distribution through the hard cutting experiments, through the analysis of the test results, for making the residual compressive stress to distribution in the surface, should choose smaller tip blunt edge of the circular mouth tool guide circle. With the cutting parameters and cutting parameters as the test factors, Thiele et al studies the effect of different factors on the distribution of residual stress. By studying the results of the experiments, they know that, the formation of the surface residual tensile stress is related to the cutting speed and the flank wear of the cutting tool. The cutter parameters have the function of the residual stress in the depth. Professor Chen Ming of Shanghai Jiao Tong University studies the influence of cutting parameters on the surface residual stress; the results show that the influence of the rake angle on the surface residual compressive stress is significant. With taking cutting and cutting tool wear as test factors, Yang Bo carry out the experimental study of high speed milling, The research result is that the cutting speed and cutting depth are significant factors to obtain the surface residual DOI: 10.3303/CET1546195 Please cite this article as: Chen G.J., Wang L., Chen W.D., Zhang J.D., Zhou X.Q., 2015, Orthogonal test analysis and modeling of gcr15 bearing steel machined surface residual stress in hard precision turning, Chemical Engineering Transactions, 46, 1165-1170 DOI:10.3303/CET1546195 1165 compressive stress, When the tool wear is small, we can obtain the surface of the uniform distribution of the residual stress. Through the orthogonal hard turning test, the surface residual stress is analyzed. based on 1stOpt software, Using the Marquardt method and universal global optimization algorithm to establish a multivariate nonlinear prediction model of Residual stress on machined surface. 2. Hard cutting test 2.1 Test scheme The orthogonal test was designed with L16 (44) Orthogonal array. The cutting speed v, feed rate f, cutting depth ap, and tool radius rε , as the four test factors, Each of these factors were taken four identical values. A total of 16 sets of experiments were conducted. 2.2 Cutting test condition The cutting tests are carried out on the CAK4085 numerical control lathe. Figure 1 is a device for cutting tests; in these tests, the GCR15 bearing steel with a geometrical size of Φ110×200mm is chosen. After quenching treatment, the hardness of the material reaches 62~64HRC. The cutting tool is the most basic blade in PCBN7025 negative rake angle Turning Inserts of Sandvik Coromant Company. The cutter bar is CoroTurn RC Rigid clamp DCLNR/L2020K12 conventional tool bar. Table 1 is blade model and its parameters. In the experiment, the vibration acceleration of the cutting tool Ag is measured by the SD1403 piezoelectric acceleration sensor. The vibration signal acquisition and processing is carried out under the Vib'SYS software program, which is in the development of Beijing's pop company. Figure 1 Experimental Cutting Set-up Table 1: Model and Parameters of the Blade 2.3 Test of Residual Stress In order to establish the prediction model of residual stress, the residual stress in the surface of the machined surface was measured by X—350A X ray stress measurement system. X ray method is used to testing the residual stress. Basic equation about sin2ψ is as follows: ( ) ( ) ψ θ θσ ϕ 20 sin 2 cot 12 ∂ ∂ ⋅⋅ + −= v E (1) ( ) 01 cot12 θ⋅+−= v E K , ( ) ψ θ 2sin 2 ∂ ∂ =M Then MK ⋅= 1ϕσ (2) In the formula; K1—stress constant; M—The slope of 2θ-in2ψ diagram; E—Elastic modulus; v—Poisson ratio; 2θ—Angle between the diffracted rays and the rays; θ—Bragg angle; ψ—Angle between Crystal surface normal that produces diffraction and Material surface normal. Table 4 is orthogonal experiment. 3. The Experiment Results and Analysis 3.1 The Experiment Results Table 2 are the test data and test results Types of Blades rε(mm) βr Thickness of Blades S(mm) The form of Cutting edge Angle of the Chamfer rn brn (mm) CNGA120404S01030A 0.4 80 4.76 Negative groove and Fillet 30° 0.10 CNGA120408S01030A 0.8 CNGA120412S01030A 1.2 CNGA120416S01030A 1.6 1166 Table 2: Design and Test Results of Orthogonal Test Test parameters The root mean square value of acceleration residual stress v（m/min） f（mm/r） ap（mm） rε(mm) Ag（g） ϕσ （MPa） 1 101 0.03 0.05 0.4 0.02 -1005.98 2 101 0.05 0.1 0.8 0.06 -649.65 3 101 0.08 0.15 1.2 0.03 -229.87 4 101 0.12 0.2 1.6 0.03 49.75 5 202 0.03 0.1 1.2 0.23 109.46 6 202 0.05 0.05 1.6 0.34 -199.82 7 202 0. 08 0.2 0.4 0.02 -106.23 8 202 0.12 0.15 0.8 0.05 90.32 9 301 0.03 0.15 1.6 0.27 300.11 10 301 0.05 0.2 1.2 0.11 350.35 11 301 0.08 0.05 0.8 0.03 -100.28 12 301 0.12 0.1 0.4 0.06 50.43 13 370 0.03 0.2 0.8 0.23 70.64 14 370 0.05 0.15 0.4 0.04 -60.21 15 370 0.08 0.1 1.6 0.31 89.56 16 370 0.12 0.05 1.2 0.03 -200.03 3.2 Range analysis Table 3 is a table for the analysis of residual stress range. Table 3: Intuitive Table for the Analysis of Residual Stress Range v f ap rε K1 -1835.75 -525.77 -1506.11 -1121.99 K2 -106.27 -559.33 -400.2 -588.97 K3 600.61 -346.82 100.35 29.91 K4 -100.04 -9.53 364.51 239.6 k1 -458.9375 -131.4425 -376.5275 -280.4975 k2 -26.5675 -139.8325 -100.05 -147.2425 k3 150.1525 -86.705 25.0875 7.4775 k4 -25.01 -2.3825 91.1275 59.9 R 609.09 137.45 467.65 340.3975 Table 4 is the Range analysis table about Root mean square of Vibration acceleration Table 4: The Range Analysis Table about Root Mean Square of Vibration Acceleration v f ap rε K1 0.14 0.75 0.42 0.14 K2 0.46 0.55 0.66 0.37 K3 0.47 0.39 0.39 0.4 K4 0.61 0.17 0.39 0.95 k1 0.035 0.1875 0.14 0.035 k2 0.16 0.1375 0.165 0.0925 k3 0.1175 0.0975 0.0975 0.1 k4 0.1525 0.0425 0.0975 0.2371 R 0.125 0.145 0.0675 0.2021 (1)The influence of cutting speed on the Residual stress and the Vibration of cutting process As shown in Figure 2, with the increase of cutting speed, the residual tensile stress appears to be increased and then decreased. When the cutting speed is less than 301m/min, the residual tensile stress increases with the increase of cutting speed; When the cutting speed exceeds 301m/min, the residual tensile stress decreases with the increase of cutting speed. When the cutting speed is between 101 and 202m/min, the 1167 vibration acceleration increases with the increase of cutting speed. When the cutting speed is between 202 and 301m/min, the vibration acceleration will decrease with the increase of cutting speed; When the cutting speed is between 301 -370 m/min, the vibration acceleration increases with the increase of cutting speed. Figure 2: The Influence of cutting speed on the Residual Stress and the Vibration Acceleration of Cutting Process (2) The influence of feed rate on the residual stress and the vibration of cutting As shown in Figure 3, with the increase of feed rate, the residual tensile stress appears to be decreased and then increased. When the feed rate is less than 0.05 mm/r, the residual tensile stress decreases with the increase of the feed rate; When the feed rate is greater than 0.05 mm/r, the residual tensile stress increases with the increase of the feed rate. The vibration acceleration of the cutting process gradually decreases with the increase of the feed rate. Figure 3 ： The Influence of Feed Rate on the Residual Stress and the Vibration Acceleration of Cutting Process (3)The influence of the cutting depth on the Residual stress and Vibration of cutting process As shown in Figure 4, with the increase of the cutting depth, Residual tensile stress appears to be slowly increasing; (Residual compressive stress appears to decrease); However, the vibration acceleration of the cutting process firstly increases and then decreases. When the cutting depth is less than 0.1mm, the vibration acceleration will increase with the increase of cutting depth. When the engagement was between 0.1 and 0.15mm, the cutting depth will increase the vibration acceleration decrease, when the cutting depth is more than 0.15mm, the vibration acceleration will no longer change. Figure 4: The influence of the cutting depth on the Residual stress and Vibration Acceleration of cutting process (4) The influence of corner radius on the Residual stress and Vibration of cutting process 1168 As shown in figure 5, with the increase of the corner radius, the Residual tensile stress and Vibration of cutting process gradually increases; however when the corner radius is between 0.8 and 1.2mm, the Vibration acceleration increases slowly. Figure 5: The Influence of the Corner Radius on the Residual Stress and Vibration Acceleration of Cutting Process 4. Modeling and Analysis of Residual Stress According to the test results of Table 2, A prediction model of residual stress in machined surface is established by using the method of Marquardt, widely used in 1stOpt software and General global optimization algorithm. 4.1 The establishment of nonlinear model of residual stress. The established prediction model is shown in the following formula： ϕσ =p1+p2*x1^2+p3*x2^2+p4*x3^2+p5*x5^2+p6*x1+p7*x2+p8*x4+p9*x5+p10*x1* x2+p11*x3*x4+p12*x4*x5+p13*x2*x3+p14*x2*x4 (3) Using 1stOpt software to fit the above model： ϕσ =-2134.746+7.919v-7774.413f+1609.562rε+1038.642Ag+26.911vf+ 1207.950aprε- 139.502rεAg+126570.896fap-10801.238frε-0.013v 2-35895.264f2-6105.159ap 2+17111.474 Ag 2 (4) 4.2 Prediction of model simulation Figure 6 is a model simulation for predicting the residual stress, through the analysis of the pictures we learn that, the increase of cutting vibration makes the residual tensile stress increase or decrease of the residual compressive stress. Through the forecast error table 5, we can see Maximum error is4.57%, Minimum error is -0.03%, Mean error is 0.01%, Forecast model has higher precision. Error analysis of the residual stress prediction model, as shown in Table 6, coefficient of determination is0.999979134, which shows that the fitting is good; Chi square coefficient is 0.145762, which shows that the deviation degree of theoretical inference and practical observation is very small; F Statistical value is 670928.2, which shows that the above models have statistical significance. 1169 Table 5: Error of Residual Stress Prediction Serial number experimental values predictive value residual errors 1 -1005.98 -1004.76 -1.21878 0.12% 2 -649.65 -651.523 1.873171 -0.29% 3 -229.87 -227.461 -2.4088 1.05% 4 49.75 47.4754 2.274595 4.57% 5 109.46 109.0422 0.41782 0.38% 6 -199.82 -200.649 0.829464 -0.42% 7 -106.23 -106.878 0.648187 -0.61% 8 90.32 93.89815 -3.57815 -3.96% 9 300.11 300.2061 -0.09613 -0.03% 10 350.35 350.4396 -0.08961 -0.03% 11 -100.28 -100.554 0.273729 -0.27% 12 50.43 48.44903 1.980968 3.93% 13 70.64 71.29634 -0.65634 -0.93% 14 -60.21 -60.9258 0.715753 -1.19% 15 89.56 90.40327 -0.84327 -0.94% 16 -200.03 -199.907 -0.12262 0.06% Table 6: Error Analysis of Residual Stress Prediction Model Mean Square Deviation Residual Sum of Squares Squared Correlation Coefficient (R^2) Coefficient of Determination Chi Square Coefficient F Statistics 1.4946111 35.741797 0.9999791 0.999979134 0.145762 670928.2 5. Conclusion Orthogonal experimental analysis of GCr15 bearing steel residual stress in machined surface is carried out. With the increase of cutting speed, the residual tensile stress showed a trend of increasing first and then decreasing. With the increase of feed rate, the residual tensile stress decreases first and then increases, while the residual compressive stress first increases and then decreases; With the increase of cutting depth, residual tensile stress increased slowly (compressive stress decreases); With the increase of the radius, the residual tensile stress increases (compressive stress decreases). The increase of cutting vibration makes the residual tensile stress increase or the residual compressive stress decrease. Establish a multiple linear regression model about the residual stress In the process of precision cutting of hardened steel; After examination and analysis, It is proved that the forecasting model has good fitting degree, and it has statistical significance. Acknowledgments This work was supported by National Natural Science Foundation of China (51175227) and China Postdoctoral Science Foundation (2015M571358). References Dahlman P, Gunnberg F, Jacobson M. ,2004,The influence of rake angle, cutting feed and cutting depth on residual stresses in hard turning[J]. Journal of Materials Processing Technology, 147(2):181–184. Meng L, Takagi J I, Tsukuda A., 2004, Effect of tool nose radius and tool wear on residual stress distribution in hard turning of bearing steel[J]. Journal of Materials Processing Technology, 150(3):234-241. Hua J, Shivpuri R, Cheng X, et al., 2005,Effect of feed rate, workpiece hardness and cutting edge on subsurface residual stress in the hard turning of bearing steel using chamfer + hone cutting edge geometry[J]. Materials Science & Engineering A, 394:238–248. Jeffrey D. Thiele, Shreyes N. Melkote, Roberta A. Peascoe, et al., 2000,Effect of cutting-edge geometry and workpiece hardness on surface residual stresses in finish hard turning of AISI 52100 steel[J]. Journal of Manufacturing Sciences& sengineering, 122(4):642-649. Fan Ning, ChenMing, WangHui, 2009, Simulation of Cutting Tool Geonetry Parametery Impactn Residual Stress [J]. MACHINE TOOL&HYDRAULICS, Vol.37 No.11:31-48. Yang Bo, .2010, 1, Surface Integrity and Cutting Parameters Optimization in Machining of New Titanium Alloys [D]. Nanjing University of Aeronautics and Astronautics. 1170