Font Size: a A A

Application Of CGA-BPNN And CFD In MBR Membrane Fouling

Posted on:2021-04-26Degree:MasterType:Thesis
Country:ChinaCandidate:S GaoFull Text:PDF
GTID:2518306494496604Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Membrane bioreactor(MBR)is a new wastewater treatment process which is produced by the organic combination of membrane technology and wastewater biological treatment technology.When MBR system is used for sewage treatment,a large amount of sludge is often trapped on the membrane surface,which will cause membrane pollution and affect the sewage treatment effect of MBR system.It was found that sludge yield(sludge apparent yield coefficient)and system yield were important parameters in MBR membrane fouling.The purpose of this paper is to establish a network model to predict sludge yield and use CFD simulation system to calculate the water yield,so as to measure the degree of membrane contamination.In this paper,the back propagation neural network(BPNN)was used to establish the prediction model when calculating the apparent yield coefficient of sludge.BPNN algorithm is one of the most widely used neural network algorithms.BPNN algorithm has the advantages of simplicity and plasticity.However,when using the gradient descent method to train BPNN,the initial weights and thresholds in the BPNN prediction model are randomly generated,which leads to the slow convergence rate of BPNN and easy to fall into the local minimum value.Therefore,in this paper,genetic algorithm(GA)is first used to optimize the initial weight threshold of BPNN,and the calculated results of the optimized model are more accurate.But in the process of searching,genetic algorithm is easy to fall into local optimal,leading to premature.Therefore,chaos optimization GA(chaos genetic algorithm(CGA)is introduced in this paper.CGA perfectly solved the problems encountered in GA optimization by using its chaotic characteristics,and the BNNN prediction model obtained the optimal initial weight and threshold values.The sludge apparent yield coefficient calculated by the model was more accurate and had less error than the results obtained after GA optimization.In this paper,the water production of the system is calculated by using the serial MBR membrane module as the model and the CFD simulation software ANSYS to build the serial MBR model.Results show that the tandem MBR system model is established by using CFD software,water rate for the water rate is the actual system of relative error is small,high calculation accuracy,achieved the purpose of accurate budget MBR system water rate,to solve the water rate by computing system to measure degree of membrane fouling,the study of MBR membrane fouling has significant value.
Keywords/Search Tags:Back propagation neural network, Genetic algorithm, Chaos genetic algorithm, MBR membrane fouling, Computational fluid dynamics
PDF Full Text Request
Related items