Neuro-Fuzzy Modelling of the Effluent Neutralisation Process in a Paper Mill



In paper industry, the effluent treatment is a major concern nowadays since it is extremely important to avoid pollution and simultaneously minimize the impact of process loss in the environment. Effluent neutralisation is vital to achieve this goal. However, the effluent neutralisation process presents high complexity, time-varying dynamics and non-linear characteristics, which makes its modelling not an easy task. Data mining techniques, particularly neuro-fuzzy, can be used to obtain transparent models with reasonable accuracy. In this paper, Takagi-Sugeno fuzzy models and RBF network models are used to predict the pH values in the effluent neutralisation process. Some preliminary results for the second stage of the process are presented.


Effluent neutralisation, neuro-fuzzy modelling, fuzzy systems, neural networks, hierarchical structures


Neuro-Fuzzy Modelling


Controlo'02, September 2002

Cited by

No citations found