E-NOSE

PRESENTED BYK SOMANATH PRADHAN

REG:- 0501227358

INTRODUCTIONWHAT IS E-NOSEE-Nose is a “sensing system” or instrument that consist of-Sample conditioning inlet systemGas sensor arrayPattern classifier software operating in a computerNEED The availability of a miniature, portable instrument capable of identifying contaminants in the breathing environment at part-per-million levels would greatly enhance the ability to monitor the quality of recycled air as well as providing notification of the presence of potentially dangerous substances from spills and leaks.

GENERATION

ARCHITECTURE

Continued…Electronic Nose uses an array of 32 Sensors(two each of 16 different polymers) to simulate the human nose's sensor cells. The sensors are insulators, but are impregnated with carbon particles to enable them to conduct electricityThey are connected to a small computer that takes the information from the sensors and figures out just what the smell is, similar to how our brain, through experience, learns to identify what your nose is smelling.

HOW IT WORKSBaseline resistanceFilm swelling and shrinkingPattern recognition algorithm1. statistical approaches

pca, partial least squares, cluster analysis 2.ANN approaches supervised, unsupervised 3.neuromorphic approaches

All of the polymer films on a set of electrodes (sensors) start out at a measured resistance, their baseline resistance. If there has been

no change in the composition of the air, the films stay at the baseline resistance and the percent change is zero

e- e- e- e- e- e-

BASELINE RESISTANCE

If a different compound had caused the air to change, the pattern of the polymer films' change would have been different:

Each polymer changes its size, and therefore its resistance, by a different amount, making a pattern of the change

THE ELECTRONIC NOSE SMELLS SOMETHING

e- e- e-

e- e-e-

e-

e-

e-

e-e-

e-

PATTERN RECOGNITION

After measurement procedure the signals are transformed by a processing blockThe results obtained are input for PCA,CADATA MATRIXEach sample is characterized by a unique and typical set of data forming a finger print of an analyte in a m-dimensional pattern spacePREPROCESSINGTypical techniques: manipulation of sensor baseline, normalization and scaling of response for all sensors

PRINCIPAL COMPONENT ANALYSIS AND CLUSTER ANALYSISPCA AND CA are multivariate pattern analysis

technique reducing dimensionality of problem and reducing high degree of redundancy

PCALinear feature extraction technique finding most

influencial,new direction in pattern space explaining as much of variance in data set

The new directions called principal components are the base for a new data matrix

Usually 2 or 3 of them are sufficient to transfer more than 90% of the variation of the sample

CLUSTER ANALYSISThe base principle of cluster analysis is the

assumption of close position of similar samples in multidimensional pattern space

Similarity between each two samples is calculated as a function of the distance between them usually in Euclidean sense an displayed in dendrogram

ARTIFICIAL NEURAL NETWORKInformation processing structure imitating

behaviour of human brainMain advantage: adaptive structure,complex

interaction between input and output data, parallel data processing and handling high noise level data implies useful pattern recognition tool

Many possible architecture and algorithm available in literature

The most common in measurement application is feed-forward network and back-propagation algorithm

CONTINUED…The base units of ANN are neurons and

synapses. Neurons are organized in layers and connected by synapses.

Their task is to sum up their inputs and non linear transfer of the result, which is then transmitted via synapses with modification by means of synapses weights.

This signal in turn is the input for the next layer of the network.

ADVANTAGEIt can detect chemicals in concentration so

small that people could not smell or so large that they would overwhelm our noses.

E-nose sensors do not fatigue.It can detect toxic and hazardous situations

that human may wish to avoid.It can detect carbon monoxide(co) which is

odorless to humans.

APPLICATIONMedical diagnosticsSafety purposesFood and beverages industryEnvironmental monitoring identification of toxic and hazardous

wastes, analysis of fuel mixtures, detection of oil leaks, monitoring factory emission.

FUTURE DEVELOPMENTFight against crimeRecognition of terroristsTelesurgeryVirtual reality and virtual environment fire fighter, dangerous discharge.

CONCLUSIONDespite some very interesting analytical

capabilities, the road to success for the E-nose has been rocky at best.

Depending upon the application, many of the issue remain a concern today and may limit E-nose success unless addressed and solved.

In spite of that ,it can perform some very complex analytical tasks not get addressed by even the largest and most expensive systems and is currently the subject of research.

REFERENCES1. P.ciosek,Z.brzozka,W.wroblewski,classification of

beverages using a reduced sensor array,Sens.Actuators B,103(2004),76-83.

2. P.ciosek,W.wroblewski,The analysis of sensor array data with various pattern recognition techniques,Sens.Actuators B,in press.

3. Sensors and Sensory Systems for an Electronic Nose, J. W. Gardner and P. N. Bartlett (editors), Kluwer Academic Publishers, Dordrecht, Netherlands 1992. (The famous "Iceland Symposium" that marks the beginning of the worldwide research effort on the electronic nose.)

QUESTION AND ANSWER

THANK YOU