Neural Network based Multi-Sensor Data Fusion Architecture for Temperature Measurement

Santhosh, K V (2015) Neural Network based Multi-Sensor Data Fusion Architecture for Temperature Measurement. In: International Conference on Emerging Trends in Technology and Applied Science, Kottayam, India.

[img] PDF
paper 7.pdf - Published Version
Restricted to Registered users only

Download (559kB) | Request a copy

Abstract

Multi sensor data fusion based architecture is proposed in this paper for measurement of temperature. Two dissimilar temperature sensors like thermistor and Resistance Temperature Detector (RTD) are fused together to have the following objectives (i) characteristics like linearity and sensitivity should be improved as compared to measurement carried out using individual sensor. (ii) Identification of sensor fault if any and carryout measurement accurately without interruption. Data conversion circuits are used to convert signals from RTD and thermistor to voltage. Output of data conversion circuit is fused by adding a neural network block in cascade. Once ANN is trained to produce output which is linear and sensitive it is subjected to tests. Results produced by proposed technique show fulfillment of set objectives.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Temperature measurement; Multi-Sensor Data Fusion; Artificial Neural Network; Fault detection; Fault diagnosis
Subjects: Engineering > MIT Manipal > Instrumentation and Control
Depositing User: MIT Library
Date Deposited: 06 Nov 2015 11:51
Last Modified: 06 Nov 2015 11:51
URI: http://eprints.manipal.edu/id/eprint/144501

Actions (login required)

View Item View Item