Scribd
Upload
137 views21 pages

Artificial Neural Network and Its Applications

Explore the applications of ANN in the various fields of Mechanical Engineering and its future scope.

Uploaded by

HImanshu Dabral
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
137 views21 pages

Artificial Neural Network and Its Applications

Explore the applications of ANN in the various fields of Mechanical Engineering and its future scope.

Uploaded by

HImanshu Dabral
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 21

Artificial Neural Network and

its Applications
By:-

Himanshu Dabral
Himanshu Dhiran
Himanshu Solanki
Hitesh Bikonia
Swati Agarwal

Under the guidance of:-

[14134003]
[14135031]
[14135032]
[14135033]
[14135088]

Dr. Nilanjan Mallik

Objectives :
To study the fundamentals of Artificial
Neural Network.
To study the Single Layer Perceptron
algorithm.
To study Back Propagation algorithm.
To explore the applications of ANN in the
various fields of Mechanical Engineering
and its future scope.

Introduction :
Inspired by the sophisticated functionality of human brains
where hundreds of billions of interconnected neurons
process information in parallel, researchers have
successfully tried demonstrating certain levels of
intelligence on Silicon.
Artificial Neural Networks are the fundamental
components required in such machine learning.
A Neural Network model is a group of
interconnected artificial units (neurons) with linear or nonlinear tranfer functions. Neurons are arranged in different
layers including input layer, hidden layer and output layer.

A simple structure of a typical Artificial Neural Network[ANN]


with input, hidden, and output layers

How does ANN works ?


Most applications require three normal types of layers - input,
hidden, and output.
The layer of input neurons receive the data either from input
files or directly from electronic sensors.
The output layer sends information directly to devices such as
a mechanical control system.
Between these two layers can be many hidden layers.
ANNs learn the relation between inputs and outputs of the
system through an iterative process called training.
Each input into the neuron has its own associated weight.
Weights are adjustable numbers which are determined
during training the network.

+
)

= (
=

h : Output
: Transfer function
N : Number of input neurons
Vij : The weights
xj : Inputs to the input neuron
Tihid : the threshold terms of hidden neurons

Machine Traning Algorithms :


Single Layer Perceptron

Back Propagation
Algorithm

Single Layer Perceptron :


In machine learning, the perceptron is an
algorithm for supervised learning of binary
classifiers: functions that can decide whether an
input belongs to one class or another.
It is a type of linear classifier, i.e. a classification
algorithm that makes its predictions based on a
linear predictor function combining a set of
weights with the feature vector.

A simple neuron

Algorithm :
Initialize the weights and the threshold. Weights
may be initialized to 0 or to a small random value.
In the example below, we use 0.
For each example j in our training set D, perform
the following steps over the input and desired
output dj
Calculate the actual output:
yj(t)= = f ( wo(t) + w1(t)j,1+ wn(t)j,n)

Update the weights:


for all features ,

Back Propagation Algorithm :


A Back Propagation network learns by example.
You give the algorithm examples of what you
want the network to do.
It changes the networks weights so that, when
training is finished, it will give you the required
output for a particular input.
Back Propagation networks are ideal for Simple
Pattern Recognition and Mapping Tasks.
The backpropagation learning algorithm can be
divided into two phases: propagation and
weight update.

Phase 1 : Propagation
Each propagation involves the following steps:
1. Forward propagation of a training pattern's
input through the neural network in order to
generate the propagation's output activations.
2. Backward propagation of the propagation's
output activations through the neural network
using the training pattern target in order to
generate the deltas (the difference between
the input and output values) of all output and
hidden neurons.

Phase 2 : Weight update


For each weight-synapse follow the following steps:
1. Multiply its output delta and input activation to get the
gradient of the weight.
2. Subtract a ratio (percentage) of the gradient from the
weight.

This ratio (percentage) influences the speed and quality of


learning; it is called the learning rate.
The greater the ratio, the faster the neuron trains; the lower
the ratio, the more accurate the training is.

Typical Back Propagation Algorithm

Error formula :
E is the error
t is the target output for a training
sample
y is the actual output of the output
neuron.

The input
to a neuron is the weighted sum of outputs
previous neurons

of

The variable

denotes the weight between neurons

and

Observations :
In our project we have use a code for back propagation algorithm that
usually takes 12-13 iterations and reduces the error value to minimum
by varying the weights in the respective iterations. We can see Back
propagation algorithm can train fast and effectively.
Graph of error vs no. of iteration for
input 6 and output 12

Graph of error vs no. of iteration


for input 7 and output 18

Applications in Mechanical Engineering :


Condition Monitoring
Condition monitoring is
considered as a major part of
predictive maintenance. It
assesses the operational
health of rotating equipment,
in order to provide early
warning of potential failure
such that preventative
maintenance action may be
taken.

Fault Diagnosis
Fault diagnosis plays an
important role in the efforts
for gas turbine owners to
shift from preventive
maintenance to predictive
maintenance, and
consequently to reduce the
maintenance cost . It
concerns with monitoring a
system in order to identify
when a fault has occurred as
well as to determine the type
and location of the fault.

Robotics
A robot that walks on two legs
will have some sort of gyro or
accelerometer system that is
equivalent to the human innerear.
A robot that responds to a variety
of voice commands must analyze
the time, amplitude, and
frequency components of what it
hears; and compare it to a known
vocabulary.
A game-playing robot must
respond to the unpredictable
behaviour of its opponent. Also,
it may want to "learn from
experience" how to play a better
game.

Conclusion :
Artificial Neural Network prove out to amazingly helpful in
replacing a human operating a machine by just some bunch of
codes, they train fast and work sometimes better than human
mind. So ANN can be deployed in every field by careful
observation to increase efficiency and reduce man power.

Future Scope :
ANN can be used to make safe automatically driven vehicles
for faster and safer journeys.
A human companion robot for everyday work.
Exoskeleton Suits for extreme jobs that require human care as
well as unhuman power and efficiency.

You might also like