Adesola I. Oke and Adesola G. Aderounmu
Fuzzy model, carbon resin electrode, water treatment
This paper demonstrated the utilization of fuzzy logic in the selection of factors that influence stability of carbon resin electrode during electrochemical treatment of water and wastewaters. Carbon resin electrodes were developed from used dry cell and resin using non- heat treated technique. A fuzzy model was the developed to relate the input (particle size of the powdered carbon, percentage of resin used, compressive press and current through the electrode) and output (stability of the electrode) variables. Results of the factorial experiments were used to train developed fuzzy model. Performance of the proposed model was compared with respect to the actual stabilities and statistical model using model of selection criterion (MSC), absolute error (absErr), total error (err), mean error (MnErr), error (Err), correlation coefficient (CD) and relative error. The study revealed that fuzzy model predicts stability of carbon resin electrode and the model compared well with statistical model. Statistical evaluation shown that average MSC, err2 , absErr, MnErr, Err, CD, relative error and mean squared error for fuzzy model and statistical models were 0.98 and 0.73; 203.03 and 264.66, 50.48 and 51.91, 3.16 and 3.52, 14.50 and 16.54, 78.69 % and 60.45%, 3.29 and 3.65, 14.37 and 16.54 respectively. It was concluded that the proposed model can be used with the highest degree of accuracy to identify the factors that influence stability of carbon resin electrodes in comparison with the previous models.
Important Links:
Go Back
