Application of BP Neural Network in Faults Diagnosis of Process Valve Actuator

Abstract

As modern process plants become more complex, the ability to detect and identify the faulty operation of pneumatic control valves is becoming increasingly important. This chapter investigates Back Propagation (BP) neural network with improved algorithm to diagnose the pneumatic control valve faults. The particular values of six measurable quantities are shown to depend on the severity of commonly occurring faults. The relationships between these parameters from fault signatures for each operating condition that are subsequently learned by a multilayer BP neural network. Through the activity of the Development and Application of Methods for Actuator Diagnosis in Industrial Control Systems (DAMADICS) research to train the network. The simulation experiments’ results prove the BP neural network has the capability to detect and identify various magnitudes of the faults of interest and can isolate multiple faults. In addition, it is observed that the network has the ability to estimate fault levels not seen by the network during training.