|
Title:
|
USE OF THE NEURAL NETWORK FOR ESTIMATING THE MUD S QUANTITY GENERATED BY EFFLUENT TREATMENT STATIONS: A CASE STUDY |
|
Author(s):
|
Paulo Bousfield , Cladir Zanottelli , Paulo Olivieira , Cátia Ganske , Sidney Schossland , Edson Torrens |
|
ISBN:
|
978-972-8924-87-4 |
|
Editors:
|
António Palma dos Reis |
|
Year:
|
2009 |
|
Edition:
|
Single |
|
Keywords:
|
Artificial Neural Networks, Effluent Treatment, Textile Industry. |
|
Type:
|
Short Paper |
|
First Page:
|
195 |
|
Last Page:
|
199 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
The main objective of this study is the development of an intelligent system for estimating the quantity of mud generated
by an Effluent Treatment Station (ETS) at a Textile Industry; using the artificial neural networks (ANNs). The efficiency
of removing the potential ETS pollutant was analyzed through a study of the main components, verifying if the organic
parameter color and material require more studies, and alternatives to reduce their concentrations, in accordance with
current legislation. The creation of an ANN involved the drawing up of a database, through a series of data from 2002 up
to 2007 of the parameters analyzed by the company and the mud volume (in m3) generated after effluent treatment.
Aimed at obtaining an ANN of good generalization capacity, a series of feedforward architecture were elaborated; the
supervised learning and the backpropagation algorithm, for the purpose of finding the number of neurons in the concealed
layer. The outcome of the tests was based on the mean quadratic error and the linear regression of the results obtained
from the network prediction. It was evident that the use of ANNs for modeling effluent treatment stations, are adequate
and are highly relevant in improving the process modeling techniques. |
|
|
|
|
|
|
