Liquid level control modern control techniques

Dissertant

Amid, Nihad Muhammad

University

University of Technology

Faculty

-

Department

Department of Computer Engineering

University Country

Iraq

Degree

Master

Degree Date

2004

English Abstract

Two methods of control, P-I and Fuzzy control presented theoretically for a laboratory water tank level control system.

P-I controller lafctsthe system performance to obtain the good behavior of the control system.

Fuzzy logic controller design is presented by using Matlab and 100 mm gunk.

A neural controller is presented practically, where a process control system built with the artificial neural network modeling approach, this model formed the major component of software controller using the approach illustrated by Levenberg-Marguardt algorithm.

Interfacing with the process constructed through the parallel printer port with necessary data acquisition system.

The range of correct operation obtained by using Runge-Kutta algorithm Fuzzy controller can be used for process control after adding an integral at the input path to integrate the deviation of error from the set point, this model give a good system response for a given process in less settling time (approximately 30 min) with zero steady state error.

Forward Neural controller gives better process response at higher settling time (approximately 100 min), the actual level will never overshoot the set point, that make this type of controller becomes very useful in real processing.

Neural Networks learning based on empirical data.

This makes neural networks much more reliable in processing models.

Main Subjects

Information Technology and Computer Science

Topics

• Control systems
• Neural networks(Computer science)

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-305513