Condition based reliability simulator for power stations

Abstract

The SA Power Utility operates 27 power stations with a total nominal capacity of 41,995 MW, comprising 35,726 MW of coal-fired stations, 1,860 MW of nuclear, 2,409 MW of gas–fired, 2,000 MW hydro and pumped–storage stations as well as a 3 MW wind–farm. There are significant concerns as well as efforts by the Power utility to improve the current generation plant performance decline indicated below. There is also consensus on the significant opportunity for the power utility to leverage Condition Monitoring (CM) and advanced Analytics (AA) technologies to assist in the turn–around of technical performance. Most power stations in the Power Utility have a formal Condition Monitoring program in place, but still experience unacceptably high levels of unplanned (UCLF) and planned (PCLF) downtime for maintenance. The current approach is use of a blanket implementation of the ISO standards as the primary input to set up the condition monitoring program – the view of the researcher is that this not the way to go, but that the CM Program and decisions regarding plant condition should be made taking into consideration all aspects, including Plant Design Basis, plant age, process design and influences. The research study developed a proof of concept Condition Based Reliability Simulator (CBRS) that uses a standard analytical framework and the associated fault and failure models, based on the Design Basis of plant systems and equipment. By coupling this to operating philosophy and control system data; as well as maintenance strategy, it was demonstrated how it is possible to significantly improve the identification and prediction of impending failures, or identify unacceptable asset management practices. By developing and implementing this capability in a prototype integrated simulation environment, it could be demonstrated how scenarios can be introduced to determine their impact on plant condition and reliability, without affecting actual plant operations. It was also demonstrated how this CBRS can significantly enhance predictive capability and assist with asset management activities.