Research On Membrane-based Water Treatment System Simulation And Membrane Fouling Prognosis

The increasing shortage of water resources has become the biggest obstacle to the economic development of China and the worldwide countries,and it is urgent to solve the problem of water resources scarcity.The development of wastewater treatment and desalination technologies such as reclaimed water and desalinated water is gradually becoming an important method to solve the water resource problems.Membrane bioreactor(MBR),which combines the membrane separation technology and bioreactor technology together,has great advantages and significant development potential in wastewater treatment and seawater desalination processes.While the continuous development of membrane bioreactors,membrane fouling has become the most important factor that seriously hinders the development of MBR applications.The direct influences caused by membrane fouling are the reduction of membrane flux,increased feed pressure,reduced productivity,increased maintain-shutdown time period and the operating costs.Therefore,study on the mechanism of membrane fouling and predictive control methods are the keys to promote the popularization and application of MBR to ensure that the membrane bioreactor to obtain a stable large membrane flux under low energy consumption conditions.A membrane fouling state prognosis method is proposed in this thesis aiming at the above issues.Sequentially,in order to overcome the two major problems of the increase in the volume of pretreated sludge and the low overall water recovery rate of the seawater reverse osmosis(SWRO)desalination technology,a new hybrid reverse osmosis-forward osmosis(RO/FO)system is developed and the software of RO/FO hybrid system is coded and packaged in MATLAB in this thesis.(1)The classification,mechanism,influencing factors and control of membrane fouling in MBR system are studied.The classification of membrane fouling is studied based on the location of membrane fouling,the degree of cleanability,and the composition of pollutants,and the mechanism of membrane fouling is also briefly described.According to the four aspects of membrane properties,operating conditions,mixed liquid properties,and influent water quality,an in-depth research on the generation,operating principles,and influencing processes of various membrane fouling factors is carried out,which laid the foundation for the followed membrane fouling state prognosis research.By introducing and analyzing in detail of the various methods of membrane material body modification,membrane material surface modification,membrane module optimization,material liquid property change,operation condition controlling and membrane fouling cleaning,etc,the energy consumption reduction of the seawater desalination process is studied.(2)Due to the complexity of the membrane fouling influencing factors,it is hard to accurately predict membrane fouling state.To do this effectively,a PCA-based GA-RBF membrane fouling prognosis model is proposed in this thesis.First of all,the principalcomponent analysis(PCA)is employed to preprocess the membrane fouling data to obtain the three main influencing factors which have the greatest impacts on membrane flux.Secondly,the genetic algorithm(GA)is introduced to optimize the RBF neural network parameters to fine the prediction accuracy after the structure of the RBF neural network is determined.Finally,the obtained experimental data is used to train the network to establish the PCA-based GA-RBF model,Simulation results proves that the proposed optimized model is superior to the traditional RBF neural network both in prediction accuracy and convergence speed.It achieves the research purpose of improving the membrane fouling state prediction accuracy.(3)To solve the two main problems of increased volume of pretreated sludge and low overall water recovery rate in the seawater reverse osmosis(SWRO)desalination technology,a new hybrid reverse osmosis-forward osmosis(RO/FO)system is proposed.The sludge liquid discharged from the desalination pretreatment of seawater is further concentrated by forward osmosis,the centrifuge is replaced by FO,and the reverse osmosis concentrated water is used as the FO draw liquid,then the desalination is recycled to increase the water recovery rate,reduce the volume of discharged sludge material liquid,and increase a certain freshwater yield.(4)The software of the above RO/FO hybrid system was developed and packaged in MATLAB,and it is verified under the different conditions of water flux values.By easily setting the different operating parameters in the RO/FO system software interface,the percentage reduction of pretreated sludge volume,the RO overall water recovery rate,the required FO membrane area and the dilution rate of the RO waste residue in a variety of different options can be quickly estimated.