Scholarly article on topic 'Parametric Optimization in Drilling EN-8 Tool Steel and Drill Wear Monitoring Using Machine Vision Applied with Taguchi Method'

Parametric Optimization in Drilling EN-8 Tool Steel and Drill Wear Monitoring Using Machine Vision Applied with Taguchi Method Academic research paper on "Materials engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Materials Science
OECD Field of science
Keywords
{"Drill Wear" / "Taguchi Method" / Anova / "Machine Vision"}

Abstract of research paper on Materials engineering, author of scientific article — Y.D. Chethan, H.V. Ravindra, Y.T. Krishne Gowda, G.D. Mohan Kumar

Abstract Tool wear is highly correlated to production cost and efficiency. In this research, a drill wear monitoring system based on parametric optimization and machine vision technique is developed. A drilling model of cutting parameters (drill diameter, spindle speed and feed rate) and tool condition (focusing on drill wear measurement and analysis) is developed. To increase the tool life and for the required surface roughness in machining the parameters are to be optimized. The experimental design methods developed in this study can be used to optimize cutting parameters efficiently and reliably. The drilling model based on cutting parameters was constructed using Taguchi method. The derived relation is useful for in-process wear monitoring. Tool wear dynamics are extremely complex and not yet fully understood. Therefore, machine vision-based tool wear monitoring techniques can help elucidate wear progression. In this study, a drilling model based on the machine vision technique is used to establish a direct relation between cutting parameters and tool wear. Experimental results indicate that tool condition monitoring can be successfully accomplished by analyzing texture feature information extracted from the drill image.

Academic research paper on topic "Parametric Optimization in Drilling EN-8 Tool Steel and Drill Wear Monitoring Using Machine Vision Applied with Taguchi Method"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedia Materials Science 5 (2014) 1442 - 1449

International Conference on Advances in Manufacturing and Materials Engineering,

AMME 2014

Parametric optimization in drilling EN-8 tool steel and drill wear monitoring using machine vision applied with taguchi method

Y.D.Chethan3*, H.V.Ravindrab, Y.T.Krishne Gowda3, G.D.Mohan Kumarb

a AssistantProfessor, Department of Mechanical Engineering, Maharaja Institute of Technology, Mysore-571 401, India b Department of Mechanical Engineering, P.E.S. College ofEngineering, Mandya-571 401, India

Abstract

Tool wear is highly correlated to production cost and efficiency. In this research, a drill wear monitoring system based on parametric optimization and machine vision technique is developed. A drilling model of cutting parameters (drill diameter, spindle speed and feed rate) and tool condition (focusing on drill wear measurement and analysis) is developed. To increase the tool life and for the required surface roughness in machining the parameters are to be optimized. The experimental design methods developed in this study can be used to optimize cutting parameters efficiently and reliably. The drilling model based on cutting parameters was constructed using Taguchi method. The derived relation is useful for in-process wear monitoring. Tool wear dynamics are extremely complex and not yet fully understood. Therefore, machine vision-based tool wear monitoring techniques can help elucidate wear progression. In this study, a drilling model based on the machine vision technique is used to establish a direct relation between cutting parameters and tool wear. Experimental results indicate that tool condition monitoring can be successfully accomplished by analyzing texture feature information extracted from the drill image. © 2014ElsevierLtd.Thisisanopenaccessarticleunder the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of Organizing Committee of AMME 2014 Keywords:Drill Wear;Taguchi Method;Anova;Machine Vision

1. Introduction

Tool wear has a direct effect on the quality of surface finish, dimensional precision and ultimately the costs of the parts produced. Information about tool wear, if obtained on-line, can be used to establish tool change policy, adaptive control, economic optimization of machining operations and full automation of Machining operations.

* Corresponding author. Tel.: +91 9620228113; fax: +0-000-000-0000 . E-mail ö^^re55:ydcgowda@gmail.com

2211-8128 © 2014 Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of Organizing Committee of AMME 2014 doi: 10.1016/j.mspro.2014.07.463

The important goal in the modern industries is to manufacture the products with lower cost and with high quality in short span of time. There are two main practical problems that engineers face in a manufacturing process. The first is to determine the values of process parameters that will yield the desired product quality (meet technical specifications) and the second is to maximize cutting tool performance. Drilling is the principal operation to make bolted or riveted assemblies in industry. Drilling is the final operation during assembly of the aircraft and automotive structures. For example, for a small plane with one engine over 100.000 holes are drilled while for a large transport aircraft millions of holes are necessary, Ersan Asian et al. (2007) in this publication, performance of A1203-based ceramic tools when turning hardened AISI 4140 steel (63 HRC)have been discussed. However, their high degree of brittleness usually leads to inconsistent results and sudden catastrophic failures. This necessitates a process optimization, by employing Taguchi techniques. Combined effects of three cutting parameters, namely cutting speed, feed rate and depth of cut on two performance measures, flank wear (VB) and surface roughness (Ra), were investigated employing an orthogonal array and the analysis of variance (ANOVA). J.A. Ghani et al.(2004) have outlines theTaguchi optimization methodology. The study shows that the Taguchi method is suitable to solve the stated problem with minimum number of trials as compared with a full factorial design. Mondal et al. (1992)evaluated the performance of their developed ZTA inserts during machining of low and medium carbon steel. Condition Monitoring is an act of extracting information about the condition of a machine or component with the use of special instrumentation". Condition Monitoring requires selection of some suitable and measurable parameter to carry information about the condition of the machine or component like vibration amplitude, temperature, voltage, force etc. There are two main methods of estimating tool wear: (i) Indirect method and (ii) Direct method. Indirect methods include those based on sensing of the cutting forces S.N. Huang et al.(2007), vibrations Nicolas Guibert et al.(2009), acoustic emission Martin P. Gomez et al.(2009), and motor current H.Y. Kim et al.(2002). Successful tool wear monitoring requires that a number of technical tasks are understood and handled. Predicting the interrelationships among tool wear, tool life and cutting conditions is problematic. The use of machine vision in the determination of tool wear is fairly wide spread in the manufacturing literature. Several researchers have examined the usage of machine vision for the measurement of tool wear. For example, S. Kurada et al.(1997) presented image processing techniques and machine vision systems can enable direct tool wear measurement to be accomplished in-cycle. The Comprehensive literature reviews have been published by,C.K.Huang (2006),have discussed about the development of an automated flank wear measurement scheme using vision system for a micro drill. It is evident from literature survey that there exists a need to study on the effects of various parameters on drilling of EN8 Tool Steel since lot of works are reported pertaining to drilling of composite materials, aluminum alloys and very rarely on drilling of tool steels. It is reported that the combination of process parameters optimization and tool wear monitoring can give satisfactory response for achieving desired results.

2. Experimental procedure

The experimental work consists of drilling EN8tool steel using High-Speed Steel drill bit. The machining was carried out in an automatic drilling machine tool; drill wear is measured using Tool Makers Microscope. The experiments were conducted for different cutting speeds and feeds combinations. The cutting speeds considered are 360 rpm, 490 rpm and 680 rpm. Feeds considered are 0.095 mm/rev, 0.190 mm/rev and 0.285 mm/rev. In all the cutting conditions for each hole tool wear is measured. Experimental set-up is as shown in the Fig. 1.

2.1. The Taguchi method and Design of experiments

In this study, the settings of drilling parameters were determined by using taguchi experimental design method. Orthogonal arrays of taguchi, the Signal- to -Noise (S/N) ratio, the analysis of variance (ANOVA), and regression analysis are employed to analyze the effect of the drilling parameters on tool wear .In order to reduce time and cost, experiments are carried out using L27 orthogonal array. For the purpose of observing the degree of influence of the cutting conditions in drilling process three factors (cutting speed, feed and drill diameter), each at three levels are taken into account as shown in Table 1

Fig. 1 Experimental Set-up

Table 1. Process parameters and their levels

Parameter Level 1 Level 2 Level 3

A:Cutting Speed (Rpm) 360 490 680

B :Feed(mm/rev) 0.095 0.190 0.285

C:Drill Diameter (mm) 6 8 10

In the taguchi method, the term signal represents the desirable value (mean) for the output characteristic and the term noise represents the undesirable value (deviation, SD) for the output characteristic. Therefore the S/N ratio is the ratio of the mean to the SD. Taguchi uses the S/N ratio to measure the quality characteristic deviating from the desired value. Usually there were three categories of the performance characteristics in the analysis of the S/N ratio, i.e. the lower-the-better (LB), the higher-the-better (HB) and the nominal-the-better (NB). The S/N ratio for each level of process parameters was computed on the S/N analysis. In drilling, lower flank wear is the indication of better performance and higher tool life. Furthermore, a statistical Analysis of Variance (ANOVA) was performed to see which parameters were significant. By both S/N and ANOVA analysis, the optimal combination of process parameters were predicted. Finally, a confirmation experiment was conducted to verify the optimal process parameters obtained from the parameter design.

The equation for calculating S/N ratio for low band (LB) characteristics (indB)

n=-Wlog\l(Z?=1TWft\ (1)

Table 2. L27 Orthogonal array and the Desired Parameter Values

Trials Speed Feed Dia of drill TW S/N Ratio

(rpm) (mm/rev) (mm) (mm) (dB)

1 360 0.095 6 0.35 9.1186

2 360 0.095 8 0.31 10.1728

3 360 0.095 10 0.30 10.4576

4 360 0.19 8 0.18 14.8945

5 360 0.19 10 0.15 16.4782

6 360 0.19 6 0.17 15.3910

7 360 0.285 10 0.10 20.0000

8 360 0.285 6 0.19 14.4249

9 360 0.285 8 0.14 17.0774

10 490 0.095 8 0.28 11.0568

11 490 0.095 10 0.35 9.1186

12 490 0.095 6 0.38 8.4043

13 490 0.19 10 0.13 17.7211

14 490 0.19 6 0.22 13.1515

15 490 0.19 8 0.21 13.5556

16 490 0.285 6 0.20 13.9794

17 490 0.285 8 0.23 12.7654

18 490 0.285 10 0.28 11.0568

19 680 0.095 10 0.43 7.3306

20 680 0.095 6 0.49 6.1961

21 680 0.095 8 0.47 6.5580

22 680 0.19 6 0.25 12.0412

23 680 0.19 8 0.24 12.3958

24 680 0.19 10 0.26 11.7005

25 680 0.285 8 0.25 12.0412

26 680 0.285 10 0.21 13.5556

27 680 0.285 6 0.23 12.7654

Main Effects Plot for SN ratios

Data Means

6 8 10 Signal-to-noise: Smaller is better

Fig.2 Effect of Cutting speed, Feed and Dia of drill bit on tool wear

Spindle speed, followed by feed rate and drill diameters are the most important factor for tool life. Fig.2. shows the main effect plot for tool wear. Tool wear decreases with increased speed, Tool wear increases with increased feed, further, the tool wear increases with increase in diameter therefore; to obtain a low value of tool wear should be set at high cutting speed, low feed and drill diameter of 6mm.The land is the area remaining after fluting. In order to reduce the amount of land that creates friction with the hole wall (thus generating heat), drill bits are margin relieved. The amount of land remaining in contact with the hole wall during drilling is referred to as the margin. The wider the margin, the greater the friction area and the higher the drilling temperature, resulting in higher extents of heat-related hole quality defects. Besides, among the three types of drill diameters,6mm drills generates least wear and lower friction against En8 steel. It can also be seen from the experiment results as show in Table

3 that A3B1C1 is the best combination of parameters for achieving desired cutting performance. In other words, to obtain the best cutting performance, drill dia, feed rate, and spindle speed should be 10mm, 0.0.0095mm/rev, and 680 rpm, respectively.

2.2. Development of regression model

In metal cutting research, attempts have been made to fit appropriate equations to the experimental data to understand the effect of cutting parameters on the measured tool wear, in the drilling of EN8 tool steel also, similar attempts are made to predict the tool wear. It is seen in the literature that linear or polynomial functions are fitted sometimes for the tool wear modelling. The cutting speed, feed and drill tool diameter are considered in the development of mathematical models for, tool wear. The correlation between the considered drilling parameters for drilling conditions on EN8 tool steel are obtained by linear regression. The linear polynomial models are developed using commercially available Minitab 13 software for drilling parameters are given below.

Tool wear (mm) = 0.220 + 0.000349 speed - 0.976feed + 0.00696 drill diameter (2)

The predicted values are also compared with experimental values and shown in Fig. 3.

a- Experimental

H-1-I-1-1-1-I-1-I-[-1-1-I-

0 5 10 16 20 25 30

Number of Trails

Fig.3.Experimental v/s predicted

3. Monitoring of optimized drill using blob analysis technique by machine vision

From S/N analysis the optimum levels of control factors were calculated as A3Bland C3.hence, the monitoring of drill wear was done using blob analysis technique by machine vision.

3.1. Cuttingconditions

The values of the cutting conditions based on optimization as shown in Table.3

Table 3. Optimized Process parameters

Parameter Optimum level unit

A:Cutting Speed (Rpm) 680 rpm

B:Feed(mm/rev) 0.0095 mm/rev

C:Drill Diameter (mm) 10 mm

The uniqueness of the TCM strategy proposed here is that features are extracted and monitored only from certain sections of the drilling cycle, and not the entire cycle. This idea is based on the fact that the measured drill wear exhibited a very obvious change in relation to machining time at some parts of the drilling cycle, whereas they hardly exhibited any changes at others. By focusing on those sections with the most distinct change, the information contained within the feature does not get weak. This method can distinguish the presence of tool wear from magnified digital images of the HSS drill bit. The blob analysis and texture-based segmentation methods are superior as they generate an output indicative of the progressive tool wear.

Figuxe.4:a) Fresh drill Image (Before processing)

b) Drill with region of interest(After processing)

c)Drill Wear as seen under machine vision system

d)Texture-based segmentation

The sequence of processed image is as shown in fig.4 (a) and 4(b) in each process signal as well as to extract suitable features. Finally, small area blobs that are present in the image must be eliminated, leaving only the tool wear blob. Morphological erosion, with a 5 X 5 structuring element, was performed on the binary image.worn out drill,textured based segmentation is as shown in fig.4(c) and fig.4(d).The method is effective in identifying the feature attributable to the tool wear. This is the principal difference compared to the previously reported techniques.

The flank wear area can be written as follows:

flank -wear

=D0-Dw

Where, Do = Area of fresh tool image, Dw = Area of worn tool

The images of drills are captured and the boundary of the cutting plane is extracted using edge detection. The computed flank wear area in Eq.(3) was included in the experiments of drilling operations. As the machining time increases, the drill wear area decreases.

The signals and tool wear data were recorded from the first cutting until the cutting tool wear reached 0.3 mm The overall results of the machining processes were recorded and the related figures were shown in Fig. 5. From these figures, it can be clearly seen that the flank wear of the cutting tools can be divided into three stages, break-in period, steady state and failure region. The flank wear measurement reaches 0.3 mm faster with the increase of machining time.

Fig. 4 shows the Variation of Tool wear with Time (speed-680 rpm, feed-0.095 mm/rev) for EN-8 material of thickness 10 mm, for the tool wear as the response variable. And Figure.5 shows the Variation of Area with Time (speed-680 rpm, feed-0.095 mm/rev) for EN-8 material of thickness 10 mm, for the tool wear as the response variable. According to Fig.6 progressing tool wear seems to have three effects on the machining time: Firstly, it shows a moderate increases in magnitude, which can be attributed to the rapid tool wear from 102 pixels to 352 pixels. Secondly, the magnitude become steadier from 352 pixels to512 pixels, this is the indication of steady wear. Thirdly, the slope of the tool wear in a drilling cycle changes from 512pixels to 700 pixels, so much so that the point at which the drill wear (AREA) begins to raise, Another indicator for severe tool wear. The maximum slope yield

useful results that determined the maximum value of tool area in terms of pixels and could be well correlated to tool wear in general.

Fig.5The plot of VB mm v/s machining time in sec (a) feed=0.095mm/rev Speed=680rpm

Fig.6 The plot of VB pixels v/s machining time in sec (a) feed=0.095mm/rev Speed=680rpm

4. Conclusion

Based on the various experimental results, the following conclusions were drawn.

• In this study, the Taguchi method was used to analyze the effect of spindle speed, feed rate and drill diameter on flank wear.

• In Taguchi method, a three level and three factors, L27 orthogonal array has been used to conduct experiments and to determine S/N ratio.

• The results obtained will be used to determine a set of parameters that presents the minimum amount of flank wear.

• ANOVA was used to determine the most significant process parameters affecting the tool wear.

• In this study, the analysis of the confirmation experiment for tool wear has shown that Taguchi parameter design can successfully verify the optimum cutting parameters (A3B1C3), which are speed=680rpm) feed rate= 0.0095 mm/min) and drill diameter=10mm generate least tool wear and best hole quality.

• Tool wear monitoring is done by first captured images of drill bit, extracting image feature before and after cutting using machine vision technology.

• The flank wear area increased with the number of holes drilled.

• Better correlation of actual tool wear and estimated tool wear using machine vision was observed

• The combination of process parameters optimization and tool wear monitoring can give satisfactory response for achieving desired results.

References

Ersan Asian, Necip Camuscu, Burak Birgo'ren, 2007, "Design optimization of cutting parameters when turning hardenedAISI 4140 steel (63

HRC) with A1203 + TiCN mixed ceramic tool" Materials and Design, 28, 1618-1622. J.A. Ghani, I.A. Choudhury, H.H. Hassan, 2004, "Application of Taguchi method in the optimization of end milling parameters" Journal of Materials Processing Technology, 145, 84-92.

B. Mondai, A.B. Chattopadhya, A. Virkar, A. Paul, 1992, "Development and performance of Zirconia toughened alumina ceramic tools", Wear 156,365-383.

Martin P. Gomez, Alfredo M. Hey, José E. Ruzzante, Carlos E. D'Attellis, 2010, "Tool wear evaluation in drilling by acoustic emission" 819825

Nicolas Guibert, HenriParis, Joe Rech, Christophe Claudin, 2009, "Identification of thrust force models for vibratory drilling", 730-738. H.Y kim et.al 2002, "Real Time Drill Wear estimation Based On Spindle Motor Power" (2002) 267-273

S.N. Huang_, K.K. Tan, Y.S. Wong, C.W. de Silva, H.L. Goh, W.W. Tan, 2007, "Tool wear detection and fault diagnosis based on cutting force Monitoring", 444-451.

Kurada and Bradley, 1997, "a review of machine vision sensors for tool condition monitoring computer in industry", 34 (1). J.C. Su, C.K. Huangl, Y.S. Tarng, 2006, "Automated Flank Wear measurement Of Micro drills Using Machine Vision", 328-33