Application of RSM and ANN models for optimization of photoelectrocatalytic oxidation process for fulvic acid removal
Artificial neural network(ANN)and response surface methodology(RSM)were used to build a predictive model of the combined effects of independent parameters(pH,K2S2O8,applied potential)for TOC removal efficiency of fuIvic acid(FA)using a Ti/TiO2 electrode.The optimum operating conditions obtained from the quadratic form of the RSM and A NN models were pH 3.8,K2S208 88.40 mg/L,and applied potential 0.88V of predicted TOC removal efficiency within 2h of photoelectrocatalytic degradation.The square mean-square error obtained from RSM was 5.544.whereas square mean-square error obtained from ANN in training set and test set were 1.7439 and 4.6367 respectively.The results demonstrated an higher prediction accuracy of ANN compared to RSM.This superiority of ANN over other multi factorial approaches could make this estimation technique a very helpful tool for photoelectrocatalytic oxidation process.
Fulvic acid photoelectrocatalytic oxidation response surface methodology artificial neural network parameters
J.F.FU Q.L.WU X.D.XUE
School of Energy and Environment,Southeast University,Nanjing,210096,P.R.China Southeast University Medical School,Nanjing 210009 P.R.China
国际会议
西安
英文
801-808
2008-05-15（万方平台首次上网日期，不代表论文的发表时间）