Improvements on Bidirectional Associative Memories and Application Simulation in Intelligent Fault Diagnosis
Shanguang Chen · 1997
In this paper,improvements on basic Bidirectional Associative Memory(BAM)were discussed and a BAM model with robust EBP(Error Back Propagation)algorithm(BAM-EBP)was proposed.In BAM-EBP model,hidden layers were introduced in the net structure and an improved EBP algorithm was ap- plied.The simulation of the basic BAM and BAM-EBP models in application to intelligent fault diagnosis for a typical control system and the comparison between the two models were carried out.The analysis and results showed that both models were effective,while the BAM-EBP had better performance especially in fault toler- ance,storage capacity,application adaptability.