Title :
Non-linear modelling and identification of a coagulant dosage system for water treatment plants
Author :
Bello, Oladipupo ; Hamam, Yskandar ; Karim, Dinar
Author_Institution :
Dept. of Electr. Eng., Tshwane Univ. of Technol., Pretoria, South Africa
fDate :
Feb. 26 2014-March 1 2014
Abstract :
Non-linear modelling and estimation of a coagulant dosage system for water treatment plants based on Hammerstein-Wiener models is presented in this paper. Different configurations of the Hammerstein-Wiener model for the system are developed and compared with each other. The most suitable model configuration is selected and linearised at an estimated operating point. Simulation results further show that the Hammerstein-Wiener model configuration with both saturation and dead-zone as input channels and piecewise linear functions as output channels has remarkable improvement over Hammerstein, Weiner, other configurations of Hammerstein-Wiener and linear models. The proposed model is in a form that the widely used controllers in the industry could be applied to improve automatic control of coagulation process.
Keywords :
estimation theory; stochastic processes; water treatment; Hammerstein-Wiener model; coagulant dosage system; nonlinear modelling; piecewise linear function; water treatment plant; Chemicals; Coagulation; Equations; Estimation; Mathematical model; Predictive models; Vectors;
Conference_Titel :
Industrial Technology (ICIT), 2014 IEEE International Conference on
Conference_Location :
Busan
DOI :
10.1109/ICIT.2014.6894964