| Removal of carbonaceous organic matter in wastewater continues to be a pressing issue but the control of nutrients discharge into the environment has gained importance over the last decade. New technologies are being sought to achieve strict effluent discharge requirements for nutrients and micropolluents, which have adverse impacts such as entrophication and deteriorated water quality in receiving water bodies. In certain watersheds, government agencies have imposed strict polluents discharge limits such that conventional activated sludge (CAS) systems are unable to meet. Membrane bioreactor (MBR) is a novel technology for wastewater treatment, obtained by combing traditional suspended biomass processes with membrane filtration. Compared with CAS processes, MBR process has great advantages including a smaller footprint, less sludge production and better effluent quality. Based on the above advantages, the MBR is widely being applied.The objectives of this study were to evaluate the feasibility of treating the municipal sewage to meet the applicable discharge limits consistently using pilot-scale submerged MBR technology and further to optimize the opetating conditions of the system. Bench scale studies were conducted under diffent conditions with an airlift submerged MBR and the primary goal was to analyse the mechanism of nitrogen removal in the reactor. The main results and conclusions of this study are as follows.(1) Performance of the pilot-scale submerged MBR for treatment of municipal sewage was investigated. The removal efficiency for COD, NH3-N, TN and TP was 86.3%,95.0%,25.0%and 26.4%, respectively. The effluent SS was almost zero and the effluent turbidity was always below 2NTU due to effective rejection of the membrane module. The removal of COD mainly depended on the biodegradation, but the membrane separation also played an important role in providing the stable effluent quality. During the early stage of the run, the contribution of membrane module to COD removal was caused dominantly by membrane inself filtration. As the amount of activated sludge increased, the contribution was caused by the membrane filtration and the biofilm on the surface. The removal of NH3-N was greatly influenced by water temperature. When the water temperature was lower than 15℃, it became the restriction factor of NH3-N removal. The fouling resistance was found to decrease significantly with intermittent suction. The back flushing was particularly effective in decreasing membrane fouling(2) BP artificial neural network (ANN) was employed to predict COD and NH3-N in the effluent using the monitoring parameters (influent COD, influent NH3-N, HRT, SRT, MLSS, air-water ratio and temperature) with the pilot-scale submerged MBR. The results indicated that the R of COD and NH3-N was 0.8902 and 0.9915, respectively, and RMSE of COD and NH3-N was 0.2117 and 0.0512, respectively, which indicated the simulation accuracy and prediction stability. The BP model was found to provide an efficient and a robust tool in preciting the effluent quality.(3) In order to decrease membrane fouling of pilot-scalt submerged MBR system, genetic algorithms (GA) has been first combined with BP artificial neural network to select the optimal ruuning parameters of water treatment process. The BP artificial neural network was used to describe the relations of membrane fouling and running parameters. The GA was employed to select the optimal solution of BP artificial neural network model established. The results showed that the optimal running parameter (Jv, MLSS, Q, T and t) was 12.5L/m2.h,4112.5mg/L,30:1,8.3min and 1.0min, respectively, and the optimal K was 0.44942kPa/d. Compared with non-optimal parameters, membrane flux played the most important role in influencing membrane fouling.(4) A single-stage continuously aerated MBR with inter-loop airlift reactor configuration has been established and applied for nitrogen removal. The environment conditions from aspects of the reactor (macroenvironment) and flocs (microenvironment) were analyzed for better understanding the mechanism of nitrogen removal. It was demonstrated that simultaneous nitrification and denitrification (SND) is the mechanism leading to nitrogen removal. The results of floc size distribution showed that the effect of microenvironment on SND was not significant. Macroenvironment analyses showed that gradient distribution of DO level in MBR imposed a significant effect on SND by developing considerable anoxic zone spatially. (5) The effects of C/N ratio, aeration intensity and HRT on SND were investivated using the bench scale MBR. The results showed that either low or high C/N ratio, aeration intensity or HRT could all restrain SND. When the operating conditions (C/N ratio, aeration intensity and HRT) was 10,50m3/m2.h and 8～10h, the removal efficiency of TN was 66.8%,52.6% and 61.8%～67.5%, respectively. A MBR filled with carriers instead of activated sludge was also investivated for the performance. The results showed that adding suspended carriers could effectively improve the performance of pollutant removal and decrease membrane fouling.(6) A model that predicts DO distribution based on S-P model and ASM1 model was established. The downcomer DO profiles predicted by the model were in close agreement with the experimental results. The model should prove useful in the design and optimization of the airlift submerged MBR. A simulation model for calculating the cross-flow velocity was also developed. Effects of aeration intensity and bioreactor structure on the cross-flow velocity were investigated. Under the same aeration intensity, a high cross-flow velocity could be achieved, which has a higher height, narrow riser, wider downcomer and wider bottom passage.
|
PDF Full Text Request