Prediction Model of MBR Membrane Flux for Elman Neural Network Based on Particle Swarm Optimization

Abstract

Membrane fouling restricts the efficiency of sewage treatment to some extent. Membrane flux acts as an important parameter to measure membrane fouling, which has become an important research field in MBR simulation prediction. According to the predicted results of membrane flux, determine the degree of membrane pollution, make a decision to clean or replace the membrane which can save invest and reduce energy consumption. Particle swarm Optimization(PSO) is an optimization algorithm, which can quickly get the global optimal value. In this paper, using PSO optimize the weights and thresholds of Elman neural network to establish PSO-Elman membrane pollution prediction model. Compared with Elman neural network prediction model, this model has few steps and fast convergence. Experimental result shows that the neural network prediction model proposed in this paper has higher prediction accuracy than Elman neural network model and BP neural network model. The application of this prediction model has significance for MBR simulation.