Fault Diagnosis of Automobile Gearbox Using Artificial Neural Network

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

International Journal of Scientific Research in Engineering and Management (IJSREM)

Volume: 07 Issue: 07 | July - 2023 SJIF Rating: 8.176 ISSN: 2582-3930

FAULT DIAGNOSIS OF AUTOMOBILE GEARBOX USING


ARTIFICIAL NEURAL NETWORK

1ANJALI RAJ, 2ANILA JOSE


1Msc Scholar, 2Assistant Professor
1,2
Department of Computer Science
1,2
St.Joseph’s College (Autonomous), Irinjalakuda, Thrissur, India

ABSTRACT: Gears are very critical and essential device in power transmission systems, industries etc.. to
vary speed and load conditions according to the requirements. It has a wide area of application in the field
of rotating machinery. In recent years, the operational accuracy of gears are increasing as demand
increases. Some minor errors like gear misalignment, backlash and tooth cracks after running for some
period of time, will create some form of vibration which will affect the supporting components like
bearings, shafts etc.. to fail. So it is very important to monitor the condition of gears to avoid the run time
breakdown of machines. Vibration measurements are used to monitor the condition of the machine for
predictive maintenance and to predict the gearbox faults successfully.

In this work propose the use vibration signal for automated fault diagnosis of gearbox and the
statistical features that extracted from the vibrational signal are given as an input to the Deep learning
algorithm ANN (Artificial neural networks) for fault identification. We can collect the vibrational datas
from vibrational sensors that placed on gears. By using vibrational data and the load in the vehicle we can
predict the fault in the gear box accurately.

KEYWORDS: Deep Learning, Artificial Neural Network, Automobile Gearbox,

INTRODUCTION: Gears are very critical and a controlling component in power transmission systems,
it is having wide area of application in the field of rotating machinery. In recent years the operational
accuracy of gears are increasing as demand increases. Some of the minor errors like gear misalignment,
backlash, and tooth cracks after running for some period of time this defect will create some form of

© 2023, IJSREM | www.ijsrem.com | Page 1


International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 07 Issue: 07 | July - 2023 SJIF Rating: 8.176 ISSN: 2582-3930

vibration which will affect the supporting components like bearings, shafts etc.. to fail. So it is very
important to monitor the condition of the gears to avoid run time breakdown of the machine. The time-
domain diagnostic algorithm is developed for monitoring the gear faults which has improvement in the
extraction capability.

Vibration is the behavior of machine’s mechanical component as they react to internal or external
forces. Vibration signal analysis is mainly employed as it is an effective approach in detecting and
diagnosis of faults in gears in a power transmission system. Continuous monitoring of gearbox vibration
signals detects the deterioration due to propagation of faults which tells the defect in gear, in order to
schedule the proper shutdown to avoid unexpected shutdown cost and time. Excess load, manufacturing
error and installation error affects the dynamic behaviour of gears in transmissions. The fault in gear
creates periodic impulses which in turn transmitted to adjacent parts. So the vibration signal should be
detected early to avoid failure of whole system. The future of condition based maintenance research
would be a design of intelligent device, which continuously monitor its own health using online data
acquisition, online signal processing and online diagnostic tools. Compared to other monitoring
techniques vibration signals are better and it is a very popular technique to reduce cost and enable
diagnosis.

LITERATURE REVIEW:
[1] The paper "A time domain approach to diagnose gearbox fault based on measured vibration signals" by
Hong and Dhupia (2014) proposes a new time-domain method for detecting and identifying gear faults in
gearboxes. The method combines the fast dynamic time warping (Fast DTW) and correlated kurtosis (CK)
techniques to extract the periodic impulse excitations caused by the faulty gear tooth and identify the
corresponding faulty gear and its position. Fast DTW is a technique for aligning two time-series signals that
may have different lengths and speeds. It is used in this paper to align the measured vibration signal with a
reference signal that has the same frequency as the nominal gear mesh harmonic. This allows the periodic
impulse excitations caused by the faulty gear tooth to be extracted from the measured vibration signal. The
extracted signal is then resampled and analyzed using the CK technique. CK is a measure of the non-
Gaussianity of a signal. In the context of gear fault detection, it can be used to identify the presence of
periodic impulse excitations in the signal. The position of the faulty gear can then be estimated by finding
the peak of the CK curve. The authors evaluated the performance of their method on simulated and

© 2023, IJSREM | www.ijsrem.com | Page 2


International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 07 Issue: 07 | July - 2023 SJIF Rating: 8.176 ISSN: 2582-3930

experimental vibration signals. The results showed that the method was able to detect and identify gear faults
with high accuracy. The method was also shown to be robust to noise and variations in operating speed. The
main contribution of this paper is the development of a new time-domain method for detecting and
identifying gear faults. The method is based on the combination of Fast DTW and CK techniques, which
allows it to extract the periodic impulse excitations caused by the faulty gear tooth and identify the
corresponding faulty gear and its position. The method was shown to be effective in detecting and
identifying gear faults in simulated and experimental vibration signals.

[2] The paper proposes a statistical model of gear vibration signals that can be used to detect and diagnose
gear faults. The model is based on the assumption that the gear vibration signal can be decomposed into a set
of sinusoidal components with different frequencies and amplitudes. The parameters of the sinusoidal
components are estimated using the least-squares method. The authors evaluated the performance of their
model on simulated and experimental vibration signals. The results showed that the model was able to detect
and diagnose gear faults with high accuracy. The model was also shown to be robust to noise and variations
in operating speed. The main contribution of this paper is the development of a statistical model of gear
vibration signals that can be used for fault detection and diagnosis. The model is based on a simple and
intuitive assumption, and it is relatively easy to implement. The model was shown to be effective in
detecting and diagnosing gear faults in simulated and experimental vibration signals.

[3] The paper investigates the performance of two statistical fault detection indicators, the root mean square
(RMS) and kurtosis, for three different series of crack propagation scenarios in a spur gear. The crack
propagation scenarios are:

• Crack propagating through the whole tooth width with a uniform crack depth distribution.

• Crack propagating through the whole tooth width with a parabolic crack depth distribution.

• Crack propagating in both the depth and the length directions simultaneously.

The authors used a dynamic simulation model to generate vibration signals for each crack propagation
scenario. The vibration signals were then analyzed using the RMS and kurtosis indicators. The results
showed that the RMS indicator was more sensitive to the first crack propagation scenario, while the kurtosis

© 2023, IJSREM | www.ijsrem.com | Page 3


International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 07 Issue: 07 | July - 2023 SJIF Rating: 8.176 ISSN: 2582-3930

indicator was more sensitive to the second and third crack propagation scenarios. The authors concluded that
the choice of fault detection indicator should be based on the specific crack propagation scenario. The main
contribution of this paper is the investigation of the performance of different statistical fault detection
indicators for different crack propagation scenarios. The results of this study can be used to improve the
accuracy of gear fault diagnosis.

[4] The paper proposes an analytical model to quantify the reduction in gear mesh stiffness caused by a tooth
crack. The model is based on the potential energy method, which considers the elastic deformation of the
gear teeth and the contact forces between the teeth. The crack is modeled as a line defect in the tooth, and the
model is used to calculate the change in the gear mesh stiffness as a function of the crack length and
location. The authors validated the model against finite element simulations. The results showed that the
model was able to accurately predict the reduction in gear mesh stiffness caused by a tooth crack. The model
was also shown to be insensitive to the choice of material model. The main contribution of this paper is the
development of an analytical model to quantify the reduction in gear mesh stiffness caused by a tooth crack.
The model is simple and easy to use, and it can be used to quickly assess the impact of tooth cracks on gear
performance.

[5] The paper proposes a novel method for detecting faults in moving machines using sound. The method
uses a microphone to record the sound emitted by the machine, and then analyzes the sound signal to
identify the presence of any abnormal frequencies or patterns. The authors evaluated the performance of
their method on a simulated gear fault. The results showed that the method was able to detect the fault with
high accuracy, even when the machine was moving at high speed. The method was also shown to be robust
to noise and variations in operating conditions. The main contribution of this paper is the development of a
novel method for detecting faults in moving machines using sound. The method is non-contact and can be
used to detect faults in a wide range of machines. The method is also relatively inexpensive and easy to
implement.

The main disadvantages of these papers are in the accuracy. They all have the accuracy around 75-85%. So,
in order to improve the accuracy we propose another model that aims to detect the faults in the gearboxes
using vibrational sensors. In this system we place 4 vibrational sensors on the gearbox to collect the

© 2023, IJSREM | www.ijsrem.com | Page 4


International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 07 Issue: 07 | July - 2023 SJIF Rating: 8.176 ISSN: 2582-3930

vibrational data. The vibrational sensor consists of piezoelectric crystal that has a seismic mass attached to it.
When the machine/equipment under study experiences vibrations, the crystal is subjected to stress and an
electric signal is generated, which is then converted to valuable data. The collected vibrational data is
preprocessed and then features are extracted from the data. After that training the data, then evaluate the
final result ie, the gearbox is faulty or healthy. Vibration signal analysis is an effective approach in detecting
and diagnosis of fault in gears. Here we use Artificial Neural Network for the identification of fault in
gearbox. This system has some advantages: -

• The prediction accuracy is achieved around 90.58%.


• The data training time is less.
• The time-domain diagnostic algorithm is developed for monitoring the gear faults which has
improvement in the extraction capability.

BLOCK DIAGRAM OF PROPOSED SYSTEM:

Sensor Data
& Data Pre-Processing
Load

Sent the status to Generate result


user

SYSTEM OVERVIEW:

Data gathering is the 1st step and this process is followed by Data pre-processing. The statistical
feature extraction from vibration data, required features are extracted for fault detection from 4
vibration sensors and load. Statistical features were extracted from the acquired data and it is passed
to the ANN for feature classification. Gearbox diagnosis using trained ANN and evaluates the status

© 2023, IJSREM | www.ijsrem.com | Page 5


International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 07 Issue: 07 | July - 2023 SJIF Rating: 8.176 ISSN: 2582-3930

of gearbox.
• Data Gathering
• Data pre-processing
• Fault diagnosis detection
• Feature extraction
• Training
• Evaluation

a) Data Gathering

The dataset used is Open Energy Data Initiative(OEDI). It consists data of both healthy and fault
diagnosis. This dataset contains csv files. These files are aggregated into single one and most
relevant features are extracted from it.

b) Data Pre-Processing
Each csv file in the dataset contains raw data. This data should be preprocessed before feeding it to the
algorithm. For this, first we aggregate all those csv files and parsing is also done. The file created in this way is
treated as the master file. From this master file, features are extracted. This data contains integers and strings.
So, it is to be encoded before giving it to the algorithm. The missing values should be filled with mean value of
the feature.
c) Fault Diagnosis Detection
We find the fault from the attributes done by that data.
d) Feature Extraction
Csv files are parsed and relevant data fields are extracted. We predict the fault diagnosis of gearbox with
which are dealing. All relevant features are encoded. The encoded values are given to the algorithm to make
prediction presence of a feature is represented by ‘1’ while absence by ‘0’.
e) Training and Evaluation
The model is trained. The difference between input and output is calculated. This input is reconstructed to the
output in the training phase. The model is trained during testing phase, when we give a new data of a fault
diagnosis, the reconstruction error will be high. A threshold value is set to segregate healthy and faulty
diagnosis. If value is higher than the threshold, it is predicted as fault data, otherwise as healthy data.

© 2023, IJSREM | www.ijsrem.com | Page 6


International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 07 Issue: 07 | July - 2023 SJIF Rating: 8.176 ISSN: 2582-3930

CONCLUSION:
Fault diagnosis of automobile gearbox based on deep learning techniques is one of the important studies
in the field of condition monitoring of rotating machinery. Classification efficiency of healthy and faulty
gear with two different speeds and load conditions was carried out and the classification efficiency of four
gears was quoted. ANN shows better classification ability in identification of a various faults in the
gearbox and it can be used for automated fault diagnosis.

REFERENCES:

[1]Liu Hong, Jaspreet Singh Dhupia, A time domain approach to diagnose gearbox fault based on
measured vibration signals, J. SoundVib. 333(2014)2164–2180.
[2] Juliang Yin, Wenyi Wang, ZhihongManc, SuiyangKhoo, Statistical modeling of gear vibration
signals and its application to detecting and diagnosing gear faults, Inf. Sci. 259 (2014) 295–303.
[3] Omar D. Mohammed, MattiRantatalo, Jan-OlovAidanpaa, UdayKumar, Vibration signal analysis
for gear fault diagnosis with various crack progression scenarios, Mech. Syst. Sig. Process. 41 (2013)
176–195.
[4] F.Chaari, T.Fakhfakh, M.Haddar,Analyticalmodeling of spur gear tooth crack and influence on
gear mesh stiffness, Eur. J. Mech. A. Solids. 28(2009) 461–468.
[5] Dae-HoonSeon, Jong-HoonJeon,Yang-Hann Kim, A novel sensing method of fault in moving
machine, Mech. Syst. Sig. Process. 45 (2014) 154–169.
[6] Andrew K.S. Jardine, Daming Lin, DraganBanjevic, A review on machinery diagnostics and
prognosticsimplementing condition-based maintenance, Mech. Syst. Sig. Process. 20 (2006) 1483–
1510.
[7] G.Diwakar, Dr. M. R.S.Satyanarayana, P. Ravi Kumar, Detection of Gear fault using vibration
analysis, Int. J. Emerg. Technol. Adv. Eng. 2 (9) (2012)ISSN 2250-2459.
[8] B. Samanta, K.R. Al-Balushi, S. A. Al-Araimi, Artificial neural networks and support vector
machines with genetic algorithm for bearing fault detection, Eng. Appl. Artif. Intell. 16 (2006) 657–
665.

© 2023, IJSREM | www.ijsrem.com | Page 7


International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 07 Issue: 07 | July - 2023 SJIF Rating: 8.176 ISSN: 2582-3930

[9]M.A. Mohandes, T. O. Halawani, S. Rehman, Ahmed A. Hussain, Support vector machines for
wind speed prediction, Renewable Energy.29 (2004) 939–947.
[10]AchmadWidodo, Bo-Suk Yang, Tian Han, Combination of independent component analysis and
support vector machines for intelligent faults diagnosis of induction motors, Expert Syst. Appl. 32
(2007) 299–312.
[11] Victor L. Brailovsky, OfirBarzilay, Rabin Shahave, On global, local, mixed and neighborhood
kernels for supportvector machines, Pattern Recognit. Lett.20 (1999) 1183-1190. [12] L. B. Jack, A.
K. Nandi, Fault detection using support vector machines and artificial neural networks augmented by
genetic algorithms, Mech. Syst. Sig. Process. 16(2–3) (2002) 373–390. [13] M. Saimurugan, K.I.
Ramachandran, V. Sugumaran, N.R. Sakthivel, Multi component fault diagnosis of rotational
mechanical system basedon decision tree and support vector machine, Expert Syst. Appl.38 (2011)
3819–3826.

© 2023, IJSREM | www.ijsrem.com | Page 8

You might also like