The vast majority of past research has focused on signal processing and

spectral analysis for fault diagnostics

Emerging technologies such as the Internet of things (IoT) devices have formed a
gateway to connect to
machines and its subcomponents to not only collect the process data and its
parameters but also to collect the
physical health aspects of the machine such as vibration, pressure, temperature,
acoustics, viscosity,
flow rate and many as such. This information is widely used for early fault
detection, fault identification of the
machine and to predict the future state of the machine. Some of this is made
possible due to
machine learning algorithms available across different learning domains
"A Research Study on Unsupervised Machine Learning Algorithms for Early Fault
Detection in Predictive Maintenance"

Hence, providing advance failure warning and precise fault detection in such
components are pivotal and cost-effective.

Some previous studies have overcome the

problem by applying data-mining algorithms and machine learning classification
technologies, which use a
historical database of the system to predict failures.

Among various methods that have been used machine learning, artificial neural
networks (ANN) have experienced
the fastest development over the past few years[6]

The aim of this research is to propose a fault diagnosis method that is able to
overcome all the above mentioned
drawbacks, provide system with higher sensitivity in fault detection and the most
important point is it does not need
huge historical data with fault samples. This method is based on anomaly detection
approaches to create models of
normal data and then attempts to detect abnormalities from the normal model in the
observed data
"A Data Mining Approach for Fault Diagnosis: An Application of Anomaly Detection
Algorithm"

Machine learning
technique develops algorithms that are able to find different patterns in data and
adjust program actions according to
the training dataset

Frequency-domain techniques have previously been

employed to show that a localized defect can generate a periodic signal with a
singular characteristic frequency [15]
"Y. Li, S. Billington, C. Zhang, T. Kurfess, S. Danyluk, and S. Liang, "Adaptive
prognostics for rolling
element bearing condition," Mechanical systems and signal processing, vol. 13, pp.
103-113,
1999."

Support vector machines (SVM) are one of the supervised learning approaches used
for classification and regression
which has recently emerged as a popular machine learning method. The SVM training
algorithm creates a model
based on the training examples with the ability to predict a new example would
belong to which category. SVMs
have been applied in various diagnostic applications, such as bearing faults [20],
induction motors[21], machine
tools [22, 23], rotating machines [24], etc. Generally, in fault diagnostics SVMs
are combined with other feature
selection techniques and kernel functions [25]. In this paper both Anomaly
Detection (AD) and SVM are conducted
on the experimental data set to obtain a comparison of these techniques.

Vibration data was

collected every 10 min for 164hwith a sampling rate of 20 kHz

The Fast Fourier Transform (FFT) spectrum of the signal for 8 days
before the test rig stop working of data collection has been plotted in the
following figures which the horizontal axes
indicate frequency to give a clear perspective of the data patterns in this test to
failure experiment

The results clearly prove that the anomaly detection techniques have greater
sensitivity and a 95% detection
accuracy

For many engineering and science problems, learning techniques have been widely
used when there is no direct
mathematical solution for different issues.