Study On The Simulation And Optimization Of Biological Filter Using Biowin Software

A pilot-scale Denitrification Biological Filter(DNBF) was built and operatedunder two different conditions, filtration velocity was8m/h, C/N of influent was4and5respectively by adding methanol as external carbon source to study itsdenitrification performance. The results of experiments show that theconcentration of effluent TN decreased to about7.14mg/L under the condition ofinfluent C/N=5, which could satisfy to the new discharge standard of TN<10mg/L.BioWin software was used to simulate the pilot-scale DNBF, based on theoperation data of filtration velocity of8m/h and influent C/N of5, normalizedsensitivity coefficient measure () and mean square sensitivity measure () were usedto conduct sensitivity analysis for the412parameters of BioWin software. Thesensitivity analysis results indicated that the most sensitive parameters are thestoichiometric parameters related to methylotrophs and the parameters related tobiofilm. According to the sensitivity analysis results, five most sensitiveparameters were selected for model calibration by adjusting their values to obtainproper simulation of the operating condition of filtration velocity of8m/h andinfluent C/N of5. The operating condition of filtration velocity of8m/h andinfluent C/N of4was simulated using the calibrated model for model validationafter model calibration. The results of experiments and simulations show that thepresented model of pilot-scale DNBF and the model calibration protocol of thisstudy can be used for the simulation of DNBFs.To further study whether the BioWin model and model calibration methodscan be used to simulate actual biological filters, BioWin software and twosensitivity analysis methods are used to simulate the Denitrification BiologicalFilter(DNBF)+Biological Aerated Filter(BAF) process in Yuandang WastewaterTreatment Plant. Based on the BioWin model of DNBF+BAF process, operationdata of09/2013is used for sensitivity analysis and model calibration, operationdata of10/2013is used for model validation. The results indicate that thecalibrated model can accurately simulate practical DNBF+BAF processes, and themost sensitive parameters are the parameters related to biofilm, OHOs andaeration. After model validation, calibrated model is used for process optimization bysimulating the operation results of pilot DNBF and practical DNBF+BAF processunder different conditions. The results show that, the best operation condition ofpilot DNBF is filtration velocity=8m/h, influent C/N=5, and the influent DOshould be decreased by properly controlling the secondary treatment of WWTP.The best operation condition of DNBF+BAF process for discharge standard B isreflux ratio=50%and ceasing methanol adding, while the best operation conditionfor discharge standard A is reflux ratio=50%and influent COD=155mg/L afteradding methanol. For example, in September2013, the original influent COD is134.32mg/L, so the adding COD is almost20mg/L.