As environmental monitoring
constraints imposed federal agencies become stricter, demand for gas sensor with high sensitivities for sensing of various
gases are increasing now a days. To meet the demand of low volume gas
detection, gas sensors should be enhanced in sensitivity, selectivity, recovery
time and response time 1. At the same time they should be cost effective and
reliable over long term2. In Metal
oxide semiconductor sensor explore measurement  of electric conductivity for sensing the gases
of interest. SnO2 is most widely used material among
semi-conductor oxide for making sensors due to its low development cost, long
life and good reproducibility 3,4 ,thick film SnO2 device are most
studies and most candidate due to their high level of sensitivity ,simple
design, low weight and cost effectiveness. SnO2 is an n –type ,wide
–band gap (3.6 Ev) semiconductor 5.Its electrical conductivity id due to the
non-stoichiometric compositions as a result of oxygen deficiency 6.The
sensing properties of the thick film gas sensor are based on the adsorption of
the gas molecules on it surface which produce changes in their conductivity
7. When gas sensor exposed to atmospheric
air, freshly prepared tin-oxide particles adsorbed oxygen atoms on the sensor
upper layer surface 8. These oxygen atoms pick up the e-s  from the conduction band of tin oxide and are
adsorbed on the particle surface as O- ions. Each tin oxide particle
is covered with negativity charged O- 
ions on the surface. On the other hand ,due to depletion e-s ,there
exits a layer of positively charged donor atoms just below the particle
surface. The  O-  adsorbents react with the gases and
release the  e-s   to the conduction band at higher temperature,
when reducing gases came in contact with sensor. Consequently, the depth of the space –charge layer decreases,
which result’s in a decrease in the height of the potential barrier
for  the electronic conduction at  the grain boundaries.

The concept of ANN
analysis have been discovered nearly 50 years ago, but in handing the practical
problem it is used only from last 2 –decades9.ANN are collections of small
individually interconnected processing 
units. Information is passed between these units along interconnections.
An incoming connection has two value associated with it, an input value and a
weight. The output of the unit is a function of the summed value. Once an ANN
is trained for a prescribed data it may be ready to be used then for the
prediction or classification ANNs can automatically learn to recognize pattern
in the data real system or from physical models, or other sources. An ANN can handle many input and produce
answer that are suitable for designers in required proper format 10.

Artificial Neural Network
(ANN) model may be used as alternative method for technological analysis and
matlab based calculation. Artificial Neural Networks have two main components-
the processing element called neurons and the connection between them, each
connection have their own weights.The neurons are the information processors
and the connection functions are the information storage. Each processing
element first calculates a weighted sum of the input signals and then applies
the transfer functions .The term ‘Feed Propagation’ comes due to the training
method used during the training process-back propagation of error. A Gradient
Descent Backpropagation with adaptive learning rate algorithm is used to adjust
the weights in the hidden and output layer nodes. The result is a network that
produces the mapping between the input values and output values with help of
the neurons. In this model perception, Feed-Forward Propagation is one of
suitable method of artificial neural network, designed for the testing and
training of data. Training methodologies
used in forward propagation are purelin, logsin and tansin network transfer
function for all the neurons, which reflects the relationship between
concentration as input and sensitivity for different concentration as output of
SnO2 based 1% Pd-doped thick film gas sensor. Sensitivity is tested
by artificial neural network. In neural network architecture one layer acts as
input layer, ten neurons acts as the hidden layer and other layer output layer.
In this model input is concentration of methanol and output is the sensitivity
of sensor. Though in present work single sensor is exposed to single gas or
vapor at a time and ANN is utilized to confirm it with experiments so that the
data collected can be used to train the network when sensor is replaced by
sensor array and single gas is replaced by group of gases or vapors to achieve
high selectivity. This model was trained to generate a mapping between the
input concentration of the methanol and output as the sensitivity of the
methanol.Sensitivity is dimension less quantity which is obvious from its
expression In the present work feed
forward network used to cross verification of gas sensor sensitivity at
different temperature & sample concentration after training & testing
from practical lab data for the sensitivity of the 1 % Pd-doped SnO2
sensor. Feed Forward network uses the Gaussian activation function. The
importance of such function is that it is non negative for all value of x.

     

x

Hi!
I'm Katy!

Would you like to get a custom essay? How about receiving a customized one?

Check it out