Study On The Optimal Operation Of The Step Feeding AO Nitrogen Removal Process

With the rapid development of society and the constant improvement of the standard ofliving in China, nitrogen contamination of water bodies becomes more and more serious.Conventional biological nitrogen removal processes have disadvantages such as a largeamount of excess sludge, high operating cost, large footprint and big investment. Researchersengage in the development of new biological nitrogen removal processes and theimprovement of conventional biological nitrogen removal processes actively. With theadvantages of high total nitrogen (TN) removal efficiency, less excess sludge, less investmentand lower operating cost, step feeding Anoxic/Oxic (A/O) biological nitrogen removalprocess has attracted a great deal of interest.Based on the experiment research of step-feeding A/O biological nitrogen removalprocess, a mathematical model of the process was build, and steady-state operations weresimulated. By experiments and simulations, following conclusions were drawn:1. When the experiment process was three-step feeding A/O activated sludge biologicalnitrogen removal process, volumes of anoxic zones and oxic zones of the process wereidentified by equal sludge loadings of nitrate nitrogen and total Kjeldahl nitrogen respectively.The ratio of the volume of anoxic zone to that of oxic zone of activated sludge three-stepfeeding A/O biological nitrogen removal process was2:3, return sludge ratio was0.75andinfluent flow rate distribution ratio was0.46:0.32:0.22, removal efficiencies of ammonianitrogen(NH4N), total nitrogen(TN) and chemical oxygen demand (COD) were97.60%,82.34%,94.52%and99.82%,87.18%96%when the hydraulic retention time was24h and12h respectively.2. When the experiment process was three-step feeding A/O biofilm biological nitrogenremoval process, volumes of anoxic zones and oxic zones of the process were identified byequal sludge loadings of nitrate nitrogen and total Kjeldahl nitrogen respectively. Flow ratedistribution coefficient of the process was identified by α (the value of the amount of CODconsumed for denitrifying1mg nitrate nitrogen in the anoxic zone, mgCOD/mgNO3N).When the value of α was7mgCOD/mgNO3N, the influent flow rate distribution ratio was0.46:0.32:0.22, removal efficiencies of NH4N, TN and COD were96.71%,78.42%and98.84%. When the value of α was4mgCOD/mgNO3N, the influent flow rate distributionratio was0.64:0.26:0.10, removal efficiencies of NH4N, TN and COD were94.85%,75.81%and99.62%. 3. Based on the Activated Sludge Model No.1, and programmed by MATLAB language,the mathematical model for step-feeding A/O biological nitrogen removal process wasobtained by model modification, sensitivity analysis and model calibration. When HRT was24h, errors of NH4NandNO3Nbetween the simulations and the experiments were14.29%and3.93%respectively, and when HRT was12h, errors of those between thesimulations and the experiments were0and1.75%respectively.4. Factors affected the flow rate distribution results were analyzed by simulation. Whenthe flow rated was distributed by the principle that COD and nitrate nitrogen in anoxic zoneswere reacted exactly, the ratio of COD to nitrate nitrogen should be equal to7mgCOD/mgNO3N.5. Based on the simulations, when HRT were24h，16h，12h，8h，6h and4h, removalefficiencies of each treatment zones to pollutants was different, and as a result, the processcan be operated HRT with6h.6. Based on the simulations, when return sludge ratio were1/3,1/2,3/4,1and3/2,removal efficiencies of each treatment zones to pollutants change heavily, consequently,return sludge ratio should within the range of3/4to1.7. By simulating operations of the process under two flow rate distribution modes, it wasfound that, compared with the equal flow rate distribution mode, better effluent quality can beobtained when the flow rate was distributed by the principle that the COD-to-nitrate nitrogenratio (C/N ratio) of the influent of anoxic zones equaled to the C/N ratio completedenitrification required.8. Operations of the process under different volume distribution modes were simulated.Compared with ordinary equal volume distribution modes, better effluent quality can beobtained when volumes of anoxic zones were distributed by the principle of equal volumeloading or equal sludge loading of nitrate nitrogen and volumes of oxic zones were distributedby the principle of equal sludge loading or equal sludge loading of nitrate nitrogen.