Dynamic modelling of a water distribution laboratory set-up system with SCADA capabilities

Level control systems are widely implemented in various processes in the process industries, water distribution system (WDS) in particular. An efficient WDS is represented by a system which is able to transfer a desired amount of water automatically, within a specified time. In current practice, the...

Full description

Saved in:
Bibliographic Details
Main Authors: Sahlan, Shafishuhaza, Eek, Rickey Ting Pek, Anuar, Aiman Najmi
Format: Article
Published: Penerbit UTHM 2022
Subjects:
Online Access:http://eprints.utm.my/id/eprint/101285/
http://dx.doi.org/10.30880/ijie.2022.14.01.018
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Level control systems are widely implemented in various processes in the process industries, water distribution system (WDS) in particular. An efficient WDS is represented by a system which is able to transfer a desired amount of water automatically, within a specified time. In current practice, the distribution of water is done manually. Although the system has proven sufficient so far, the current increase in general population has resulted in increasing demand of clean water, hence the need for a more efficient water distribution system has become the priority in the water industry. One method to resolve this is through automation of the distribution system. Hence, in this research, an automated WDS is set-up in the laboratory, which dynamically represents a real system. The laboratory set-up is utilized to design a control algorithm by obtaining the mathematical model of the system, i.e. the highlight of this paper. Since the laboratory set-up is part of a supervisory, control and data acquisition (SCADA) system of the WDS, the mathematical model of the system is obtained from the logged data, utilizing system identification technique, prediction error method (PEM) in MATLAB. Band pass filter is used to remove electrical noises. From the results obtained, a mathematical model of 6th order is obtained with best fits of 99.9998% and a mean square error of 0.00000000388. The mathematical model obtained represents the relationship between the reservoir tank level and the incoming motor speed.