FICA-PNN fault diagnosis for penicillin fermentation process

A novel ensemble approach based on fast independent component analysis and probabilistic neural network (FICA-PNN) is presented to diagnose faults in the fed-batch penicillin fermentation process. FICA is used to extract fastly the information of a non-Gaussian process. PNN is used as a classifier for diagnosing faults. The experimental results clearly demonstrate that the proposed approach is faster and more efficient and has higher accuracy rate compared to conventional fault diagnosis approaches.