A study of condition monitoring of check valve using acoustic emission and neural network technique

The check valve is one of typical component being extensively used in safety systems of nuclear power plants. The failures of check valves have resulted in significant maintenance efforts, on occasion, have resulted in water hammer, over-pressurization of low-pressure systems, and damage to flow system components. The objective of this research is to demonstrate condition-monitoring system based on acoustic emission, AE detection that can provide timely detection of check valve degradation and service aging so that maintenance/ replacement could be preformed prior to loss of safety function. This work is also focused on the capability of neural network technique to provide diagnostic information useful in determining check valve aging and degradation, check valve failures and undesirable operating modes