Self calibrating thermocouple using neural network algorithm for improved sensitivity and fault identification

Santhosh, K V (2018) Self calibrating thermocouple using neural network algorithm for improved sensitivity and fault identification. In: ICANI, 04/06/2018, Silchar.

[img] PDF
1296.pdf - Published Version
Restricted to Registered users only

Download (361kB) | Request a copy

Abstract

Automated calibration of temperature measurement technique is proposed in this work. The objective of the work is to design a technique which will be able to automatically calibrate the temperature sensor (thermocouple) output to obtain higher sensitivity and can also detect fault in sensor if any. Signal from the thermocouple is amplified using an instrumentation amplifier. Output of instrumentation amplifier is acquired on to the system using a general purpose voltage data acquisition card (USB 6008). Based on user defined range the sensor output is calibrated to produce an output which has the highest sensitivity using neural network algorithms. Designed support vector algorithm is also trained to identify fault in sensor. The trained system is tested for evaluating its performance with both simulated and practical data. Results produced show successful achievement of set objective

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Calibration, Intelligence, Support Vector Machine, Temperature, Thermocouple
Subjects: Engineering > MIT Manipal > Instrumentation and Control
Depositing User: MIT Library
Date Deposited: 03 Jul 2019 04:05
Last Modified: 03 Jul 2019 04:05
URI: http://eprints.manipal.edu/id/eprint/154084

Actions (login required)

View Item View Item