IQPSO neuralnetwork algorithm and application for fermentation process modeling

Due to BP neuralnetworks is defective in rapidity of convergence and apt to trap into local minimum,this article by integrating BP algorithm with IQPSO for estimation model of penicillin concentration.Training and testing the model with penicillin fermentation data-set,the experiment results indicated that this model had lots of advantage features,such as high learning speed,small training error,strong generalization ability,high estimation precision.The model is able to realize multi-step pre-estimate.