Modeling And Predicting Sludge Reduction Process Of Static Sequencing Batch Worm Reactor

Biological wastewater treatment process has generated a large quantity of excess sludge,which caused a severe environmental pollution. The costs of conventional sludge treatmentand disposal technology are generally too high to solve the problem fundamentally.Biological predation technology uses the energy transfer efficiency principle of food chain toartificially increase the number of predators in activated sludge system to reduce the excesssludge, it has advantages in energy-saving and no secondary pollution, thus has beenattracting more and more attention recently.Aquatic worms’ predation is an effective process for ecological sludge reduction, but itssludge reduction efficiency can be affected by the fluctuation of operational conditions. In thisstudy, dissolved oxygen （DO） concentration, temperature, worm density and sludge loadwere identified to be the main parameters in sludge consumption process. With theseparameters, the adaptive neurofuzzy inference system （ANFIS） and artificial neural network（ANN） were employed to predict the sludge consumption rate in the worm reactor.Compared with the ANN model, the ANFIS model was proved to be more effective.According to the prediction of ANFIS model, the optimal operating condition range of wormreactor had been defined.Extend these appropriate operating conditions to The Static Sequencing Batch WormReactor（SSBWR） and run it in cycle of2d using the excess sludge of a MBR system whichwas diluted to certain concentration（3000⁴000mg/L）, sample in12h、24h、36h、48hfrom SSBWR thereafter and test them. It had been repeated34times. The results showed thatMLSS decreased by22.89%averagely, VSS/SS ratio also decreased, indicating that theworms prey on bacteria and their secretions in the sludge mixed liquor; In sludge supernatantthe concentration of NH₄⁺-N increased almost28times, NO3--N was removed effectivelywhereas NO2--N didn’t changed much in whole cycle, COD increased more than160%, therelease of NH₄⁺-N and COD are mainly due to worm excretion; Microbial related SMPincreased a little because its polysaccharides increased, the amount of proteins andpolysaccharides in EPS both decreased more than50%, while Pr/Po（proteins/polysaccharides）ratio increased, this could be explained by integrated analysis that worm feces contains moreamino acids. The trend of change on storage of organic substrates（STO） was accumulatedfirst and then consumed, it reflects the phenomenon that heterotrophic bacteria stores organic substrates into cells when they are sufficient and reuse these intracellular polymers whennutrients in the system are depleted. In the process of worm predation, the conversion rates ofvarious substances all gradually slowed down with the passage of time whether they wereincreasing or decreasing, this suggested that the activity of worms and other microorganismwere in decline caused by the accumulation of harmful metabolites such as NH₄⁺-N.After worm’s predation every molecular weight section of organic compounds decreased,indicating that worms prey on substances of different molecular weight widely, proved onceagain that aquatic worms are suitable for sludge reduction. Number-average MolecularWeight（M_n）increased, Weight-average Molecular Weight（M_w） and Coefficient of molecularweight distribution（M_w/M_n） both decreased, therefore the molecular weight distribution ofsubstances in sludge mixed liquor became narrow. When using three-dimensionalfluorescence spectroscopy （3DEEM） to analysis EPS of predation process, the characteristicpeak intensity of tyrosine and tryptophan both have been found to decrease by more than50%, meanwhile, it can be deduced that worms may prefer to utilize more tryptophanproteins.Furthermore, an ’ASM3-growth-storage-predation’ model was established to describethe mechanism of SSBWR and to predict its sludge consumption effect, this model adoptedthe concept of simultaneous growth and storage of organic substrates by heterotrophicbacteria, included process of worms’ predation on bacteria and their EPS, it contains11components,16reactions and41model parameters all together. Apply Runge-Kuttanumerical algorithm to solve the model for stationary solution and employ single factorsensitivity analysis method to confirm the sensitive parameters of MLSS、COD、NH₄⁺-N、NO₃^--N and EPS. All initial parameters are selected based on the literature or reasonableassumptions. Sensitive parameters in the model will be calibrated according to part of theexperimental data, then use the rest to test the revised model and evaluate its predictionperformance by calculating the statistical criteria RMSE and MAPE, the results showed thatmodel’s simulated value coincide well with observed value, so ’ASM3-growth-storage-predation’ model has some direct significance for engineering practice.