Risk Based Maintenance for Compressor Systems

Abstract

This thesis reviewed part of the maintenance activities in a compressor system using the risk based maintenance philosophy. The goal was to verify if a contribution could be made to the maintenance plan of the system.

The risk based maintenance philosophy is applied according to the recommendations in NORSOK Z-008. A reliability model is proposed to verify the impact of the new maintenance approach and to make better informed decisions in the maintenance management process. The model used data from both industry database and from the maintenance history of the FPSO from the case study. The model simulated the maintenance behavior of the compressor system using the MAROS software from DNVGL.

The reciprocating compressor was found to be the equipment with the lowest mean time to failure from the main equipment in the high pressure compressor trains, according to data from the reliability database. For this reason it was selected to be modeled for preventive maintenance. Simulation results indicate that the frequency of preventive maintenance interventions being used in the FPSO of the case study could be reviewed and possibly reduced, with advantages to the availability of the system. The results also indicate that it is beneficial to include conditional monitoring in the reciprocating compressors.