Health monitoring of a pneumatic valve using a PIT based technique

Abstract

This paper is concerned with the development of a health monitoring system for a pneumatic valve employed in pressure regulation systems. The proposed method is based on the statistical analysis of deviations of the controlled pressure signal from a baseline behavior. For this purpose, the Probability Integral Transform is employed to calculate an index of dissimilarity between the distributions of monitored and baseline data. The proposed method was applied to field records of 15 units, which were monitored during eight months. In the case of failed units, the degradation index showed an increasing trend prior to the failure occurrence. It is worth noting that the failure level was similar in all cases, which is an important characteristic for the future development of prognostic solutions. In addition, no false alarms were observed for the healthy units. The results found in the case study are realistic and fit within practical requirements to support maintenance decision.