You are on page 1of 4

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882

Volu me 4, Issue 12, December 2015

Wear Studies on Composite Bearing Balls An ANN Approach


Mr. Lalit N. Patil1 , Prof. A. V. Patil2 , Prof. R. B. Barjibhe 3
1

PG Scholar, S.S.G.B.C.O.E.T. Bhusawal, Dist. Jalgaon, Maharashtra, India.


Associate Professor, S.S.G.B.C.O.E.T. Bhusawal, Dist. Jalgaon, Maharashtra, India.

2,3

ABSTRACT
Planned maintenance of machinery is very crucial for
reduction in downtime. Bearing element plays vital role
in rotating machines for friction reduction. In the present
paper attention is focused on prediction of wear in ball
bearing under pure sliding contact conditions in order to
improve the performance of ball bearing. According to
ASTM G99 standard Pin on Disc Apparatus is used to
determine relative wear in microns. Composite Ball
materials such as Silicon Nitride and Alumina Oxide are
used and are compared with baseline case of bearing
steel Chrome Steel balls sliding against 52100 Steel disc.
A load of 10N, 60N and 120N is applied on balls for
observing nature of wear. Application of artificial neural
network technology is used to validate wear and friction
behavior of ball bearing materials under consideration.
Matlab program is successfully developed.
Keywords Artificial Neural Network s, Alumina, Ball
Bearing, Composite ball, Silicon Nitride.

I.

INTRODUCTION

An Artificial Neural Network (ANN) is an information


processing paradigm that is inspired by the way
biological nervous systems, such as the brain, process
information. The key element of this paradigm is the
novel structure of the information processing system. It
is composed of a large number of highly interconnected
processing elements (neurons) working in unison to
solve specific problems. ANNs, like people, learn by
example. An ANN is configured for a specific
application, such as pattern recognition or data
classification, through a learning process. Learning in
biological systems involves adjustments to the synaptic
connections that exist between the neurons. This is true
of ANNs as well.
Neural networks, with their remarkable ability to derive
meaning from complicated or imprecise data, can be
used to extract patterns and detect trends that are too
complex to be noticed by either humans or other
computer techniques. A trained neural network can be
thought of as an "expert" in the category of information
it has been given to analyze. This expert can then be
used to provide projections given new situations of

interest and answer "what if" questions. Other


advantages include:
1. Adaptive learning: An ability to learn how to do
tasks based on the data given for training or initial
experience.
2. Self-Organization: An ANN can create its own
organization or representation of the information it
receives during learning time.
3. Real Time Operation: ANN computations may be
carried out in parallel, and special hardware devices
are being designed and manufactured which take
advantage of this capability.
4. Fault Tolerance via Redundant Information Coding:
Partial destruction of a network leads to the
corresponding degradation of performance.
However, some network capabilities may be
retained even with major network damage.

II.

ARTIFICIAL NEURAL NETWORK

Artificial Neural Networks (ANN) can be likened to


collections of identical mathematical models that
emulate some of the observed properties of biological
nervous systems and draw on the analogies of adaptive
biological learning. The key element of an Artificial
Neural Network is its structure. It is composed of a
number of interconnected processing elements tied
together with weighted connections, which take
inspiration from biological neurons. Learning like in a
biological system takes place through training, or
exposure to a set of input and output data where the
training algorithm adjusts the weights iteratively.
Artificial Neural Networks are good pattern recognition
engines and robust classifiers, with the ability to make
decisions about imprecise input data.
A. Artificial Neuron and Activation Function
The artificial neuron like the biological neuron described
is a processing element. An output for this artificial
neuron is calculated by multiplying its inputs by a
weight vector. The results are then added together and an
activation function is applied to the sum. The activation
function is a function used to transform the activation
level of a unit or rather a neuron into an output signal.
Typically, activation functions have a squashing
effect; they contain the output within a range.

www.ijsret.org

1208

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volu me 4, Issue 12, December 2015

(20W40) oil was used at flow rate of 50 ml/min on the


rubbing surfaces using oil lubrication system during the
test. It was ensure that lubrication will be continuously
between Pin and counterface during the wet test.
The experiments are conducted as per the standard
orthogonal array. The selection of the orthogonal array
IS based on the condition that the degrees of freedom for
the orthogonal array should be greater than or equal to
sum of those wear parameters. In the present
investigation an L9 orthogonal array is chosen, which
has 9 rows and 3columns as shown in Table I the wear
parameters chosen for the experiments were (1) sliding
speed, (2) load, (3) sliding Time. Table I indicates the
factors and their level.

Fig. 1 Neural Network Training


B. Neural Network Mat lab Toolbox
These activation functions were built using Mat lab.
MATLAB, which stands for matrix laboratory is an
interactive system, which was originally written as
software for matrix computation. It has evolved into a
testing and analysis research tool used in engineering,
Mathematics and science. MATLAB 2009a toolboxes
are collections of functions used to solve particular
classes of problems. In this project the Mat lab Neural
Network Toolbox is used to build, train and test system
identification neural network models.

III.

Fig. 2 Experimental Setup of Tribometer


The experiment consists of 9 tests (each row in the L9
orthogonal array) and the columns were assigned with
parameters. The first column in Table I is assigned to
sliding speed (V); second column was assigned to load
(L); third column was assigned to sliding Time (T). The
experiments were conducted as per the orthogonal array
with level of parameters given in each array row.

EXPERIMENTAL SETUP

This test is conducted as per G99 Standard of ASTM.


Chrome Steel discs were polished with metallographic
abrasive papers (C-400) and (C-600) respectively.
Chrome Steel disc rotating at a selected speed slide
against a ball according to velocity track diameter of ball
on disc was varied accordingly. This pre-rubbing process
ensured a full contact of the ball and disc surfaces. The
surface roughness Ra of disc specimens was 0.09
0.11m. All the specimens are manually cleaned in
petrol and then thoroughly dried. The friction and wear
tests were performed at room temperature (280 C) in
atmosphere. Applied loads ranged from 10 N to 120N
and rotation speeds of discs ranged from 7m/s to 14m/s,
time ranged from 30 to 90 minute, and the sliding
distance was varied accordingly. The Servo engine oil
www.ijsret.org

TABLE I. TEST PARAM ETER


Level

Velocity (m/s)

Load (N)

Time (min)

V1=7

L1=10

T1=30

V2=10

L2=60

T2=60

V3=14

L3=120

T3=90

TABLE II. L 9 ORTOGONAL ARRAY


L9 Test

P1

P2

P3

6
7

2
3

3
1

1
3

1209

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volu me 4, Issue 12, December 2015

IV.

ANN RESULTS

Fig. 6 Comparison of ANN Results with Experiment for chrome steel


Fig. 3 Regression for Chrome Steel

Fig. 7 Comparison of ANN Results with Experiment for Alumina

Fig. 4 Regression for Alumina

Fig. 8 Comparison of ANN Results with Experiment for Si3N4

Fig. 5 Regression for Silicon Nitride

Above results shows the comparative analysis of wear


for various materials.
www.ijsret.org

1210

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volu me 4, Issue 12, December 2015

V.

CONCLUSION

1. ANN program is successfully developed and


applied to predict wear of various materials.
2. This means a considerable saving in cost and time,
which could benefit the industry to build more
general and particular databases of material
properties for ball materials.
3. The ANN prediction and the experimental
observation show a good agreement.
4. It has been observed that Silicon Nitride Balls have
negligible wear and best material for ball bearing.
5. By Increasing number of Layers and Neurons ANN
result can be optimize.

[8] Mr. Lalit N. Patil, Prof. A. V. Patil . " Wear


Prediction Model for Composite Bearing Balls under
Pure Sliding Contact Condition ", Volume. 4 - Issue. 12
, December - 2015, International Journal of Engineering
Research & Technology (IJERT) , ISSN: 2278-0181 ,
www.ijert.org

REFERENCES
[1] Lada A. Gyurova, Klaus Friedrich, Artificial
neural networks for predicting sliding friction and wear
properties of
polyphenylene sulfide composites,
Tribology, International University of Kaiserslautern,
Germany, 23 Dec 2010, pp.603-609.
[2] Xu Liujie, Xing Jiandong, Wei Shizhong, Zhang
Yongzhen, Long Rui, Optimization of chemical
composition of high speed steel with high vanadium
content for abrasive wear using an artificial neural
network, Materials and Design, 20 December 2005,
pp.1031-1037.
[3] Jiahua Zhu, Yijun Shi, Xin Feng, Huaiyuan Wanga,
Xiaohua Lu, Prediction on tribological properties of
carbon fiber and TiO2 synergistic reinforced
polytetrafluoroethylene composites with artificial neural
networks, Materials and Design, Institute of Science.
[4] Lada A. Gyurovaa, Modeling the sliding wear and
friction properties of polyphenylene sulfide composites
using artificial neural networks, Germany, Wear, Spain
Gottlieb Daimler Str. 44, 67663 Kaiserslautern,
Germany, 16 Dec 2009, pp.708-714.
[5] L. Wang, R.J.K. Wood, T.J. Harvey, S. Morris,
H.E.G. Powrie & I. Care, 2003, Wear performance of
oil lubricated silicon nitride sliding against various
bearing steels, Elsevier Wear 255(2003) 657-668.
[6] Varade B. V., Dr. Kharde Y. R., Prediction of
specific wear rate of glass filled PTFE composites by
Artificial Neural Networks and Taguchi Approach,
International Journal of Engineering Research and
Applications, ISSN: 2248-9622 Vol. 2 Issue 6 PP 679683.
[7] David J. Mitchell, John J. Mecholsky Jr. & James H.
Adair,1999, All-steel and Si3N4 -steel hybrid rolling
contact fatigue under contaminated conditions,
Elsevier, Wear 239 _2000 176188.

www.ijsret.org

1211

You might also like