Vous êtes sur la page 1sur 5

Improvement by Genetic Algorithm in Finding the

Optimum Surface Finish When End Milling Ti64 using


SNTR coated tools
A.S Mohrunia , S. Sharif, M.Y. Noordinb A .Novliantaa

Faculty of Engineering,
Sriwijaya University,
Indralaya, 30662 OI-Indonesia
E-mail : Anggaimoetskali4@gmail.com

ABSTRACT
In this works, surface roughness for end milling of Ti-6Al-4V under wet conditions were optimized.
Genetic algorithm (AG) was used for finding the optimum cutting conditions such as cutting speed
(V), feed per tooth (fz), and radial rake angle (γo). The optimized results were compared to that
had been generated using response surface methodology (RSM). It has been proven that AG-
results showed more accurate than RSM-results which have been validated using data taken
according to the design of experiments (DOE).

Keywords: Surface Roughness, End Milling, Titanium Alloys, Genetic Algorithm, Response
Surface Methodology

materials. Base on these facts, it is necessary to take part in


contribution of providing such lack in information
1. Introduction

Titanium alloys are used widely known as difficult to 2. Procedure of Experiments


cut materials, especially at higher cutting speeds, due to
their several inherent properties. Among the titanium MAHO 700S CNC machining center for side milling
alloys, Ti-6Al-4V is the most widely used in the aerospace, operation was used. The grade K-30 solid carbide end mill
chemical and ship building industry because of their cutters, with PVD Supernitride coated which were prepared
superior mechanical properties, heat and corrosions with different radial rake angle according to DOE, were
resistance, so it has been chosen as the workpiece in this used for experimentation.
study.[1].
Surface roughness of the machined surface was
Due their low machinability of the alloy under study, measured using portable Taylor Hobson Surftronic +3.
selecting the machining conditions and parameters is Before conducting the measurement, the instrument was
crucial. According to the past reports, the range of feeds calibrated using a standard specimen roughness delivered to
and cutting speeds which provide a satisfactory tool ensure the consistency and accuracy of surface roughness
performance is very limited. On the other hand, adequate values. .
tool, coating, geometry and cutting flow materials should
be used [2]. There are three cutting parameters used in this
study, such as cutting speed, feed rate, and radial rake
The study of [3] has pioneered in finding of the angle. Machining conditions used in this optimization study
optimum cutting conditions for machining processes using for each cutting parameters are :
response surface methodology, which are followed by [4]- • Cutting speed V : 130 - 160 m/min.
[5]. After that, [6]-[8] have begun with the researches using • Feed rate fz : 0,03 - 0,07 mm/teeth.
titanium alloy as workpiece. Recently, it has begun to • Radial rake angle : 7 - 13o
explore the study using non-conventional algorithm in [9]-
[11]. Furthermore, according to the previous studies, there
is no researcher employed genetic algorithm in searching Cutting parameters such as cutting speed, feed rate, and
the optimum cutting conditions for machining of aerospace radial rake angle are coded using transformed equation (1)
according to circumstance of limitation of the milling usually in pairs through the application of genetic operator.
machine. These operators are applied to copulate of individuals with
ln xn − ln xn0 a given probability, and result in new offspring. The
x= (1) offspring from reproduction are then evaluated by mutation
ln xn1 − ln xn0
and elitism probability, and then these new individuals are
prime for the next generation. Selection, reproduction and
Where x the coded variable of any factor
evaluation processes are repeated until some termination
corresponding to its natural is xn, xn1 is the natural value at
criteria are achieved. The flow chart of AG method is
the +1 level and xn0 is the natural value of the factor
showed by figure 1.
corresponding to the base or zero level. The level of
independent variables and coding identification are shows
To solve the problem in this optimization study, it’s
in Table 1.
important to pick out the following parameters that take
crucial part of AG, such as population size, maximum
Table 1. Levels of independent variables for end milling Ti6Al4V
number of generation, total string length, crossover
Level in coded form probability, mutation probability, and elitism probability.
It’s important to acquire the best solutions. To solve the
Independent Variable
-α -1 0 1 α problem in this optimization study, it’s important to pick
out the following parameters that take crucial part of GA,
V (mm. min −1 ) x1 124.53 130 144.22 160 167.03 such as population size, maximum number of generation,
total string length, crossover probability, mutation
−1
fz (mm.tooth ) x 2 0.025 0.03 0.046 0.07 0.083 probability, and elitism probability. It’s important to
acquire the best solutions.
γο (°) x3 6.2 7.0 9.5 13.0 14.8 Parameters used in this study using GA that must determine
are [13]:
• Population size = 80
• Maximum generation = 15
• Crossover probability (Pc) = 0.45
• Mutation probability (Pm) = 0.25
3. Research Methodology • Elitism probability (Pe) = 0.5

The mathematical model used in this study is 2nd CCD


surface roughness model. Genetic algorithm result
compared to the response surface methodology. The
mathematical model shown by equation (2) Initial Population
2
Ў2 = -1.0014 – 0.085531x1 + 0.43082x2 - 0.070215x1 +
2
0.057597x2 -0.016614x1x3 - 0.020616x2x3 - Gen =0
0.0752385x12 + 0.0787339x1x2 (2)

Thus model is valid for end milling of titanium alloy, Evaluate Solution
Ti-6Al-4V using Supernitride coated carbide tools under
wet conditions with the following range of respective
cutting speed (V, fz, and γ) : 130 ≤ V ≤ 160 m/min; 0.03 ≤ Gen=gen+1 Reproduction
fz ≤ 0.07 mm/teeth; 7 ≤ γ ≤ 13 (o) for std order 1 to std
order 24.
2nd order CCD surface roughness model also used for std Crossover
order 13 to std order 24 with the following range of
respective cutting speed (V, fz, and γ) : 124.53 ≤ V ≤ No
167.03 m/min; 0.025 ≤ fz ≤ 0.083 mm/teeth; 6.2 ≤ γ ≤
Mutation
14.8 (o).

Genetic Algorithm (AG) inspired from biological Termination


evolution where the evolution is the method of searching yes
among en enormous number possibilities for solutions. AG
is the algorithm of searching base on selection and genetic Stop
mechanism. Figure 1: Genetic algorithm flow chart

The solution found by Genetic Algorithm is coded to


binary numbers called chromosomes. The fitness value of
each chromosome evaluated by an objective function.
Selected individuals are then reproduced, the selecting
4. Results and Discussion 6 0,190 0,193 0,190

Surface roughness experimental result for Supernitride 7 0,612 0,639 0,612


coated solid carbide tools showed by table 2. This result 8 0,576 0,600 0,576
used for validate the comparison between Response Surface
Methodology and Genetic Algorithm. 9 0,329 0,380 0,346

10 0,416 0,380 0,415

Table 2: Surface roughness result for Supernitride coated solid 11 0,352 0,380 0,352
carbide tools
12 0,400 0,380 0,403
Std V fz (mm/
Type γ (⁰) Ra (µm)
Oder (m/min) tooth) 13 0,344 0,360 0,362

1 Factorial -1 -1 -1 0,284 14 0,320 0,360 0,361


2 Factorial 1 -1 -1 0,196 15 0,272 0,282 0,283
3 Factorial -1 1 -1 0,668 16 0,288 0,282 0,285
4 Factorial 1 1 -1 0,624
17 0,230 0,224 0,224
5 Factorial -1 -1 1 0,280
18 0,234 0,224 0,225
6 Factorial 1 -1 1 0,190
19 0,640 0,758 0,641
7 Factorial -1 1 1 0,612
20 0,696 0,758 0,696
8 Factorial 1 1 1 0,576
21 0,361 0,367 0,368
9 Center 0 0 0 0,329
10 Center 0 0 0 0,416 22 0,360 0,367 0,368

11 Center 0 0 0 0,352 23 0,368 0,367 0,368

12 Center 0 0 0 0,400 24 0,360 0,367 0,368

13 Axial -1.1412 0 0 0,344


14 Axial -1.1412 0 0 0,320 Table 4: The Optimization Result for RSM
15 Axial 1.1412 0 0 0,272
fz
16 Axial 1.1412 0 0 0,288 NO V (m/min) γ (0) Ra (µm)
(mm/tooth)
17 Axial 0 -1.1412 0 0,230
1 160,00 0,030000 8,3130 0,19298
18 Axial 0 -1.1412 0 0,234
2 160,00 0,030000 9,3537 0,19298
19 Axial 0 1.1412 0 0,640
3 160,00 0,030000 12,8220 0,19298
20 Axial 0 1.1412 0 0,696
21 Axial 0 0 -1.1412 0,361 4 160,00 0,030000 11,0570 0,19298

22 Axial 0 0 -1.1412 0,360 5 160,00 0,030000 10,9900 0,19298

23 Axial 0 0 1.1412 0,368 6 160,00 0,030000 10,4470 0,19298


24 Axial 0 0 1.1412 0,360 7 160,00 0,030000 10,4160 0,19298

Table 3 shows the optimization result of RSM and AG, 8 160,00 0,030000 7,9021 0,19298
and then compared to find out root mean square error 9 160,00 0,030000 12,9200 0,19298
(RMSE) of RSM and AG method.
10 160,00 0,030000 9,0691 0,19298
Table 3: The optimization result for RSM and GA
Std Experimental
RSM Ra AG Ra Table 5 : The Optimization Result for AG
Order Ra
fz
1 0,284 0,282 0,284 NO V (m/min) γ (0) Ra (µm)
(mm/tooth)
2 0,196 0,193 0,197
1 132,906315 0,031630894 7,246147479 0,284
3 0,668 0,639 0,635
2 158,5590309 0,030337631 7,027378405 0,197
4 0,624 0,600 0,617 3 130,9776529 0,069577432 8,164525475 0,635
5 0,280 0,282 0,280 4 146,0167593 0,069891315 9,53713297 0,617
5 131,3774996 0,030624535 9,623519611 0,280 21 0,361 0,367 0,368 4,05103E-05 4,48784E-05
6 159,8029413 0,030762691 9,655391314 0,190 22 0,360 0,367 0,368 5,42398E-05 5,97878E-05
7 134,1258457 0,067612164 12,65851627 0,612
23 0,368 0,367 0,368 4,03517E-07 9,05191E-08
8 145,5432982 0,065918166 12,42482688 0,576
24 0,360 0,367 0,368 5,42398E-05 5,92767E-05
9 158,2682586 0,046188643 9,568714806 0,346
Mean Square Error 0,0011641 0,000161348
10 153,709782 0,051902779 10,42444599 0,415
RMSE 0,034118912 0,0127023
11 155,1514483 0,046037666 9,545735942 0,352
12 156,813422 0,051416531 10,35263801 0,403 The result from Table 4. shows that Genetic Algorithm is
13 124,526509 0,04616723 9,54 0,362
better than response Surface Methodology. It can be seen
from RMSE from RSM and AG method.
14 124,526509 0,046003217 9,553855719 0,361
15 167,0279571 0,046193011 9,542933855 0,283 Comparison of RSM and AG method about experimental
data showed by Figure 2.
16 166,7334247 0,046 9,542933855 0,285
17 144,2349736 0,025454132 9,542933855 0,224
18 144,2349736 0,025492529 9,542933855 0,225
19 144,2349736 0,073569095 9,542933855 0,641
20 144,2349736 0,078154435 9,542933855 0,696

RMSE comparison of RSM and GA about experimental


data from the previous research showed by Table 4.

Table 6: Comparison between Response Surface Methodology


validate with experimental result
Experime
Std RSM AG Estimated Estimated
n
Order Ra Ra Error RSM Error AG
tal Ra
Figure 2: Comparison of RSM and AG method validate out with
1 0,284 0,282 0,284 4,03254E-06 1,86539E-07 experimental result.
Based on the results above, optimum combination of
2 0,196 0,193 0,197 9,13705E-06 1,05855E-06
cutting conditions such as cutting speed, feed rate, and
3 0,668 0,639 0,635 0,000818617 0,001059942 radial rake angle for achieving minimum surface roughness
value for machining Ti-6Al-4V, can be found by AG
4 0,624 0,600 0,617 0,000599008 4,79605E-05
method.
5 0,280 0,282 0,280 3,96759E-06 3,17621E-08

6 0,190 0,193 0,190 8,86399E-06 6,29325E-05


5. Conclusions
7 0,612 0,639 0,612 0,000750131 5,12766E-08

8 0,576 0,600 0,576 0,000553443 1,69157E-07 1) Overall performance of optimizing the cutting
conditions using genetic algorithm has shown slightly
9 0,329 0,380 0,346 0,002605034 0,000300692 better results than those using response surface
10 0,416 0,380 0,415 0,001293155 1,37661E-06 methodology. This can be recognized from the root
mean squared error (RMSE) of AG which is 0,0127023,
11 0,352 0,380 0,352 0,000786215 5,11355E-08 when compared to the RMSE of RSM 0,034118912.
12 0,400 0,380 0,403 0,00039842 7,9734E-06 2) Additionally AG showed also more precise than RSM in
finding of the minimum surface roughness value.
13 0,344 0,360 0,362 0,00026509 0,00031128 3) Genetic algorithm can accomplish the optimization of
14 0,320 0,360 0,361 0,001622606 0,001657032 surface roughness in machining of aerospace materials
with adequate accuracy, which is required in industry..
15 0,272 0,282 0,283 0,000101215 0,000124301
4) The optimum cutting condition found using genetic
16 0,288 0,282 0,285 3,52771E-05 1,21589E-05 algorithm is as follows : cutting speeds V = 159,8
m/min, feed per tooth fz = 0,0307mm/tooth and radial
17 0,230 0,224 0,224 3,43118E-05 3,16323E-05
rake angle γ0 = 9,6550.
18 0,234 0,224 0,225 9,71729E-05 8,8526E-05

19 0,640 0,758 0,641 0,013947146 9,31132E-07

20 0,696 0,758 0,696 0,003856166 4,19497E-08


4V Using Uncoated PCD Inserts”, Proceeding of The
4th International Conference of Leading Edge
References Manufacturing in 21st Century, CP 7A(103), Fukuoka,
7-9 November, Japan, 2007.
[1] Z.A. Zoya and R. Krishnamurty, “The Performance of [8] A.S. Mohruni, S. Sharif, and M.y. Noordin,
CBN Tools in the Machining of Titanuim Alloys”, “Mathematical Modeling of Cutting Force un End
Journal of Materials Processing Technology, Vol. 100, Milling Titanuim Ti-6Al-4V using Supernitride Coated
pp. 80-86, 2000. Carbide Tools”, Proceedings of International
[2] L.N. Lopez de lacalle, J. Perez, J.I. Llorente, J.A. Conference on Mechanical & Manufacturing
Sanchez, “Advanced Cutting Condtions for the Milling Engineering (ICME2008), ISBN: 97-98-2963-59-2,
of Aeronautical Alloys”, Journal of Materials Johor Bahru, 21-23 May, MFG ID 0394, 2008.
Processing Technology, Vol. 100, pp. 1-11, 2000 [9] N. Baskar, P. Asokan, r. Saravanan, G. Prabhaharan,
[3] K.C. Lo and N.N.S. Chen, “Prediction of Tool Life in “Optimization of Machining Parameters for Milling
Hot Machining of Alloy Steel”, International Journal Operations Using Non-Conventional Methods:,
fro production research, Vol. 15, pp. 47-63, 1977. International Journal of Advanced Manufacturing
[4] M. Alauddin, M.A. El-Baradie, M.S.J. Hasjhmi, Technology, Vol. 25, pp. 1078-1088, 2005
“Modeling of Cutting Force in End Milling Inconel [10] N.S.K. Reddy and P.V. Rao, “A Genetic Algorithm
718”, Journal of Materials Processing Technology, Approach for Optimization of Surface Roughness
Vol. 58, pp. 100-108, 1996. prediction Model in Dry Milling”, Machining Science
[5] M. Alauddin,-M.A. Mazid, M.A. El-Baradie, M.S.J. and Technology, Vol. 9, pp. 63-84, 2005.
Hasjhmi, “Cutting Forces in End Milling Inconel 718”, [11] N.K. Jain, V.K. Jain, K. Deb, “Optimization of process
Journal of Materials Processing Technology, Vol. 77, Parameters of Mechanical Type Advanced Machining
pp. 153-159, 1998. Processes using Genetic Algorithms”, International
[6] Z.G. Wang, M. Rahman, Y.S. Wong, “Tool Wear Journal of Machine Tools & Manufacture, Vol. 47, pp.
Characteristics of Binderless CBN Tools Used ib High 900-919, 2007.
Speed Milling of Titanium Alloys”, Wear, Vol.258, pp.
752-758, 2005.
[7] T.L. Ginta, A.K.M.N. Amin, A.N.M. Karim, M.h.
Istihtiyaq, “Modeling for Surface ted WC-Co and
Roughness in End Milling of Titanium Alloy Ti-6Al-

Vous aimerez peut-être aussi