Application of PSO-RBF Neural Network in MBR Membrane Pollution Prediction

In order to predict MBR fouling more accurately, based on the research of various forecasting models in the field of MBR, principal component analysis is used to determine the main factors influencing membrane fouling firstly, such as MLSS, operating pressure and temperature, and then the prediction model based on RBF neural network was established. Aiming at the existing problems of traditional radial basis function(RBF) neural network in MBR membrane pollution prediction,a new prediction model is proposed:Particle Swarm Optimization Based on RBF Neural Network. The global search ability of Particle Swarm Optimization(PSO) is used to optimize the three input parameters of RBF neural network and the optimized RBF network is applied to MBR membrane pollution prediction. The results show that PSO-RBF based fouling simulation simulator is better than pure RBF neural network prediction model in terms of convergence rate and prediction accuracy,which can predict MBR fouling better.