Design and Optimization of CNC Milling Process Based

on Vibration System Detector Using Particle Swarm Op-

timization Algorithm Method
Chairul Anam, Abdul Rohman, Eka Mistiko Rini, Mahros Darsin

Department of Mechanical Engineering Politeknik Negeri Banyuwangi Banyuwangi,

Indonesia, Department of Me-chanical Engineering Universitas Jember Jember,
Indonesia
e-mail address :[email protected], e-mail address :
[email protected], e-mail address : Ekamrini
@poliwangi.ac.id, e-mail address : [email protected]

Abstract. Machine maintenance in industry requires speed and

convenience, one method is vibration analysis. Engine vibrations
cause a pattern of sound emitted by the engine, where the sound
of one engine mixes with the sound of another engine. Excessive
vibration levels in the engine indicate damage to engine
components. If this excessive vibration is not acted upon, the
machine will experience more serious damage. In order for it to
work optimally, the machine requires maintenance or
maintenance. Machine maintenance systems are very important
in industry to extend machine life. One maintenance method that
is often used is predictive maintenance based on vibration
signals. Predictive maintenance is a type of maintenance that can
be carried out by monitoring the vibration conditions caused by
the machine. One way that can be done to overcome damage to
the machine is to analyze the vibration level in the machine in
the form of the amplitude value of the vibration speed. This
method can predict machine damage based on the vibration
signals that arise, so that serious damage can be avoided. The
research designed

© The Author(s) 2024

M. U. H. Al Rasyid and M. R. Mufid (eds.), Proceedings of the International Conference on Applied Science and
Technology on Engineering Science 2023 (iCAST-ES 2023), Advances in Engineering Research 230,
https://doi.org/10.2991/978-94-6463-364-1_95
 Design and Optimization of CNC Milling Process 1045

and made a CNC Milling machine prototype which is a

combination of two outputs of vibration detection and process
optimization. The aim of this research is to find out and
determine parameter settings that are able to produce an
optimum response. The experimental design used is full
factorial, orthogonal array, and response surface methodology,
with the optimization methods being back propagation neural
network (BP ANN) and particle swarm optimization (PSO)
algorithm.

Keywords. CNC Milling; vibration detector; PSO ; BPANN-

GA.

1. INTRODUCTION

Machines in industry that have a rotational function often

experience vibrations. Engine vibrations cause a pattern of sound
emitted by the engine, where the sound of one engine mixes with
the sound of another engine (1). Changes in engine vibration will
result in changes in the sound patternemitted by the engine
1046 C. Anam et al.

Excessive vibration levels in the engine indicate damage to engine

components. If this excessive vibration is not acted upon, the
machine will experience more serious damage (2). In order to work
optimally, the machine requires maintenance. The machine
maintenance system is very important in industry to extend the life of
the machine. One maintenance method that is often used is
predictive maintenance based on vibration signals. Predictive
maintenance is a type of maintenance that can be carried out by
monitoring the vibration conditions caused by the machine (5). One
maintenance method that is often used is predictive maintenance
based on vibration signals. Predictive maintenance is a type of
maintenance that can be carried out by monitoring the vibration
conditions caused by the machine. One way that can be done to
overcome damage to the machine is to analyze the vibration level in
the machine in the form of the amplitude value of the vibration
speed.
By measuring the vibration level values, it can be seen whether the
bearings on the machine are suitable for use or not (6). Problems or
types of machine damage that can be detected based on vibration
signals are unbalance, looseness, misalignment and bearing damage.
This method can predict machine damage based on the vibration
signals that arise, so that serious damage can be avoided.
The CNC operating system uses a program that is controlled directly
by a computer. In general, the construction of CNC machine tools
and their working system is synchronized between the computer and
the mechanic. This CNC milling machine is a machine tool that
works on 3 axes X, Y and Z. This CNC milling machine will work
according to the pattern of the workpiece drawing that is made and is
equipped with a system Control. The control system on this CNC
milling machine is a combination of several components that are
connected using cables to one another. Several important
components contained in the CNC milling machine control system
include the computer, breakout board, motor driver, stepper motor
and power supply. faster wear rate with applied load.
 Design and Optimization of CNC Milling Process 1047

2. THEORETICAL ANALYSIS

Surface roughness is defined as the irregularity of the surface

configuration on an object or plane. This occurs because various
deviations occur during the machining process, so that a surface that
has a perfect shape cannot be created. The position of Ra, profile
shape, sample length and measurement length read by the surface
roughness measuring instrument can be seen in Figure

Figur 1. Parameter in Profil roughness (Rochim,2001)

Surface roughness is defined as the irregularity of the surface

configuration on an object or plane. This occurs because various
deviations occur during the machining process, so that a surface that
has a perfect shape cannot be created. The position of Ra, profile
shape, sample length and measurement length read by the surface
roughness measuring instrument can be seen in Figure
All mechanical systems have a certain mass and elasticity that is
capable of moving relatively, so that almost all machine tool
structures will experience vibration. Vibration is a back and forth
movement within a certain time interval. Back and forth movement
within a certain time interval is called periodic movement. This
periodic movement can always be expressed in a sine or cosine
function, therefore periodic movement is alsocalled harmonic
An illustration of harmonic periodic motion in the time domain is
shown in Figure 1. Displacement is depicted in the vertical direction
while time is depicted in the horizontal direction
1048 C. Anam et al.

3. EXPERIMENTAL METHOD AND PROCEDURE

The experimental equipment used is a CNC Milling machine,

accelerometer, power supply, Picotech ADC (analog to digital
converter), and computer as well as a surface roughness test. The
workpiece used is the SKD 11 tool (300 x 60 x 20) (mm), carbide
tool. The process variable is the spindlerotation which is determined:
(2000, 3000,4000,5000) (rpm). Depth of cut 0.3 mm, 0.4 mm 0.5
mm. The response variables are vibration level (g.rms) and average
surface roughness (Ra) (μm). Vibration measurements are carried
out during cutting tests, namely installing two accelerometers
(sensors at points A and B) on the workpiece in the x-axis direction
(channel A) and in the z-axis direction (channel B) and connecting
them to a power supply and ADC (Analog to Digital) Converter),
then connected to the computer. Set process variables such as
depth ofcut and cross feed speed. Then a cutting test iscarried out on
each combination of process variables, until the cutting data is
obtained in the form of amplitude which is stored on the computer.
The data is processed in mathCAD software to obtain the time
domain, then using the formula (fast Fourier transform) the FFT is
converted into a frequency domain. The
cutting test scheme to obtain vibration amplitude values can be seen.

Figur 2. experimental scheme

 Design and Optimization of CNC Milling Process 1049

4. RESULTS AND DISCUSSION

The results of the cutting test data arein the form of vibration
level amplitude values and surface roughness measurement
data. Research experiments were carried out on aluminum
material using the CNC process with a Carbide Tool with a
diameter of 6 variations in spindle rotation and cutting depth.
Results were obtained as shown in Figure 1 for vibration level
amplitude data for surface roughness data, where for rms (root
mean square) vibration calculation data in the time domain the
values will be the same as calculations in the frequency
domain. For peak and peak to peak vibration calculation data,
the time domain is used, where peak amplitude is the
maximum value of vibration and peak to peak amplitude is the
total amplitude value from the positive point to the negative
point of vibration. Examples of vibrations in the time domain
and frequency domain in the flat grinding process with cutting
depth parameters of 0.3.0.4.0.5 mm and spindle rotation of
2000, 3000, 4000, 5000, 6000 rpm can be seen in the
following figure.

Figure 3. Vibrations in the Time Domai

1050 C. Anam et al.

Figure 4. Vibrations in the Frequency Domain

 Design and Optimization of CNC Milling Process 1051

Research experiments were carried out with 4 variations of sp

spindle rotation (2000, 3000, 4000, 5000 and 6000) and 3
variations of
cutting depth (0.3, 0.4 and 0.5 mm). From these machining
process variables, vibration is then tested and the surface
roughness is measured. Data from measurements by the
accelerometer are calculated using mathCAD with the formulas
for rms, peak and peak to peak on each x-axis and y-axis as
follows:
Tabel 1.Data from testing and calculation

Dept Amplitudo
Spindel rotation

Ra max. Axis y
off cut

Ra max. axis x
axis x (A) axis y (B)
(a)
l(rpm)

mm m/s2 m/s2

a = 0.3 183,7417 188,1417 0,33 0,36

2000

a = 0.4 286,0418 266,6726 0,43 0,40

a= 0.5 346,4123 337,6609 0,52 0,48
a = 0.3 242,8280 246,7232 0,55 0,53
3000

a = 0.4 339,6239 316,7831 0,63 0,64

a= 0.5 385,3925 403,2435 0,71 0,74
a = 0.3 297,4235 291,2057 0,67 0,69
4000

a = 0.4 427,7960 360,5500 0,78 0,81

a= 0.5 459,7460 454,2341 0,89 0,98
a = 0.3 337,6735 325,4557 0,76 0,73
5000

a = 0.4 492,2763 432,9695 1,05 1,05

a= 0.5 577,0734 531,1888 1,19 1,16
a = 0.3 422,4194 414,0373 0,96 0,99
6000

a = 0.4 586,2794 561,3089 1,35 1,22

a= 0.5 650,0344 592,1781 1,41 1,42
result
1052 C. Anam et al.

Figure 5. Relationship between amplitude and spindle rotation on peak amplitude in the x
axis for cutting depths of 0.3 mm, 0.4mm, 0.5 mm

Figure 6. Relationship between amplitude and spindle rotation on peak amplitude in the x
axisfor cutting depths of 0.3 mm, 0.4 mm, 0.5 mm
 Design and Optimization of CNC Milling Process 1053

Figure 7. Relationship between amplitude, spindle rotation, and surface roughness on

peak amplitude on the x axis for a cutting depth of 0.3mm

Figure 8. Relationship between amplitude, spindle rotation, and surface roughness on peak
1054 C. Anam et al.

amplitude on the y-axis for a cutting depth of 0.3 mm

Figure 9. The relationship between amplitude, spindle rotation, and surface roughness on
peak amplitude on the y-axis for a cutting depth of 0.4 mm

Figure 10. Relationship between amplitude, spindlerotation, and surface roughness on

peak amplitude on the y-axis for a cutting depth of 0.5 mm

ACKNOWLEDGMENT

Praise the presence of Allah SWT for all His mercy and grace, so
that the author was given the opportunity to complete this research.
 Design and Optimization of CNC Milling Process 1055

The author would like to express his thanks to all parties for all the
help, encouragement and dedication that has helped. The author
would like to express his thanksto:
Mr.Prof. Dr–Ing. Ir. Suhardjono, M.Sc.who always guides in research
Special to my mother, my wife and mychildren who always pray for me
1056 C. Anam et al.

Friends at the ITS Mechanical Engineering Manufacturing Process Laboratory

who study the field of vibration and fellow lecturers at Poliwangi
The author realizes that there are still many shortcomings in writing this
research journal.Therefore, the author really hopes for constructive criticism
and suggestions. We hope that this journalcan provide benefits and contribute
to the advancement of science, especially in the field of engineering and
manufacturing systems. Amen

REFERENCES

1. CASTILLA, J., FORTES, J.C., DÁVILA, J.M., MELGAR,

S., SARMIENTO, A., “Predictive maintenance of mining machinery based on vibrational
analysis” In: Proceeding of 18th International Multidisciplinary Scientific
GeoConference Surveying Geology and Mining Ecology Management (SGEM), pp. 663-668,
Bulgaria, July. 2018.
2. LAISSAOUI, A., BOUZOUANE, B., MILOUDI, A.,
HAMZAOUI, N., “Perceptive analysis of bearing defects (Contribution to vibration
monitoring)”, Applied Acoustics.140,248-255, 2018.
3. LEE, C.Y., SU, HUANG, T. S., LIU, M. K., LAN, C. Y.,
“Data Science for Vibration Heteroscedasticity and Predictive Maintenance of Rotary
Bearings”, Energies. 12(5), 801-818, 2019.
4. EDE, K. N., OGBONNAYA, E. A., LILLY, M. T.,
“Vibration Monitoring of Rotating Systems”, Engineering.
2(1), 46-54, 2010.
5. EROL, S. S., “A Research Study on Vibrating Elements and Consuming Electricity in
Predictive Maintenance”, European Journal of Interdisciplinary Studies. 2(3), 45-50, 2016.
6. DIVEKAR, P. MANUGADE, P. KAMBLE, A.
HARAGAPURE, A.A., “Vibration Measurement Us-ing Accelerometer and Arduino”
International Journal of Recent
 Design and Optimization of CNC Milling Process 1057

Innovation in Engineering and Re-search. 2(3) : 177-179, 2017.

7. Anda, A R. Tugas Akhir. 2006. Penggunaan Frekuensi Sesaat untuk Deteksi Pola Suara
Kerusakan Motor Listrik. [ITS, Surabaya.
8. Putu, I.P.W., 2015, “Sistem Pakar untuk Mendeteksi Kerusakan Sepeda Motor Berbasis
Android , ” Konferensi Nasional Sistemdan Informatika, STMIK STIKOM Bali.
9. Pawar P, Kumar A, Ballav R. Development and manufacturing of arduino based
electrochemical discharge machine. J Mach Eng 2018; 18: 45–60.
10. Antil P, Singh S, Manna A. Genetic algorithm based optimization of ECDM process for
polymer matrix composite. Mater Sci Forum 2018; 928 MSF: 144–149.
11. Malik A, Sanghvi N. Optimization of laser-assisted jet electrochemical machining
parameters by grey relationalanalysis and fuzzy logic. World J Eng.
12. Singh R, Hussain SAI, Dash A, et al. Modelling and optimizing performance parameters in
the wire-electro discharge machining of Al5083/B4C composite by multi-objective response
surface methodology. Springer Berlin Heidelberg. Epub ahead of print 2020. DOI:
10.1007/s40430-020- 02418-y.
13. Antil P. Modelling and Multi-Objective Optimization during ECDM of Silicon Carbide
Reinforced Epoxy Composites. Silicon 2020; 12: 275–288.
14. Abuzied HH. Prediction of Electrochemical Machining ProcessParameters using Artificial
Neural Networks. 2012; 4: 125–
132.
15. Teixidor D, Ferrer I, Ciurana J, et al. Optimization of process parameters for pulsed laser
milling of micro-channels on AISI H13 tool steel. Robot Comput Integr Manuf 2013; 29: 209–
218.
16. Sibalija T V. Particle swarm optimisation in designing parameters of manufacturing
processes: A review (2008– 2018). Appl Soft Comput J 2019; 84: 105743.[
17. Rao RV, Kalyankar VD. Optimization of modern machining processes using advanced
optimization techniques: A review. Int J Adv Manuf Technol 2014; 73: 1159–1188.
18. Li JG, Yao YX, Gao D, et al. Cutting parameters optimization by using particle swarm
optimization (PSO). In: Applied Mechanics and Materials. Trans Tech Publ, 2008, pp. 879–
883.
19. Kennedy J, Eberhart R. Particle swarm optimization. In: Proceedings of ICNN’95-
international conference on neural networks. IEEE, 1995, pp. 1942–1948.

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-
NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/),
which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any
medium or format, as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's
Creative Commons license, unless indicated otherwise in a credit line to the material. If material
is not included in the chapter's Creative Commons license and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain
permission directly from the copyright holder.