Design of intelligent temperature measuring system

Dissertant

Ubayd, Ahmad Jabbar

University

University of Technology

Faculty

-

Department

Department of Electrical Engineering

University Country

Iraq

Degree

Master

Degree Date

2004

English Abstract

Neural Network, Genetic Algorithm and other intelligent tools have been used by many authors to design an intelligent measuring different types of systems. In this thesis, a method / has been proposed to design a temperature reader and controller for different types of thermocouples.

This suggested system is based on Baekpropagation type Neural Network. The BPNN is used to recognize the induced voltage on the.

Thermocouples for many types of transducers.

As an extension to this system, a general PWM controller has been, simulated to work adaptiveiy with the reading system.. The performance of this suggested system has been demonstrated by using a test example which contains two heaters and one- cooling fan.

One of these two heaters is used for fin temperature control and the other is used for fast heating up. Thermocouples of types (E.

B, T.

J, K.

R and S) have been simulated in this system by using ANN's.

The input to this ANN is the induced EMF signal at the thermocouple end, after a preprocessing stage, a special code is used to represent the type of each T / C-Two types of ANN'S have been tried, the first one simulate the function of (7) T / C types and the second type is simulate the controller Which is responsible on generate aft controlling signals in the system These controlling signals are controlling on the driving circuit which is drive the electrical power into the two heaters and the cooling fan.

The output of the suggested measuring system shows a good match with the actual temperate values.

Main Subjects

Electronic engineering

Topics

• Algorithms
• Neural networks(Computer science)

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-305996