Research of Fault Diagnosis System for Pickling and Cold-rolling Electric Drives Based on BP Neural Network

摘要

The Paper has exploited a system of on-line monitoring and fault diagnosis for pickling and cold-rolling electric drives, making use of configuration technologies, Matlab simulation tools and BP neural network. Based on the theory of BP neural network, a fault diagnosis model is designed, which is for electro-hydraulic servo valve, the key execute machine of pickling and cold-rolling electric drives. The model is directed at analyzing the fault possibility and fault type of electro-hydraulic servo valve. After tests on the well-trained BP-based fault diagnosis system, the system proves to have the ability of diagnosing any type of fault of electro-hydraulic servo valve accurately. The result of experiment gives the evidence that this method is very efficient for fault diagnosis system for pickling and cold-rolling electric drives.