In the field of MBR sewage treatment, membrane fouling affects the performance of MBR process. The direct consequence of it is the decline in membrane flux. The prediction of MBR flux is of great significance to the application of MBR. Because the neural network can arbitrarily approximate any continuous function with arbitrary precision, and it can establish a complex nonlinear relationship between input and output model, it is widely used in the field of prediction. We use a feedback type Elman neural network to model the MBR membrane flux, and use this model to predict membrane flux and evaluate the degree of membrane fouling. The results show that the prediction results of membrane flux obtained by Elman neural network prediction model are higher than BP neural network model. The prediction model has a guiding effect on the design and application of MBR system.
[1]
V. Chen,et al.
Fouling in Membrane Bioreactors
,
2013
.
[2]
Xue Wang,et al.
Ad Hoc network traffic prediction based on the Elman neural network
,
2015
.
[3]
曹扬,et al.
Predication of plasma concentration of remifentanil based on Elman neural network
,
2013
.
[4]
Keyu Jiang,et al.
Analysis of LPG Engine PID Parameter Control of Transient Air-fuel Ration Based on Improved Elman Neural Network
,
2010,
J. Softw..
[5]
Pierre Le-Clech,et al.
Fouling in membrane bioreactors used in wastewater treatment
,
2006
.