Is Your Power Plant Headed for a HILP? How to Avoid, Detect or Mitigate High Impact–Low Probability (HILP) Events

Abstract

In recent years some power companies have instituted programs aimed at reducing or eliminating their power plants’ unreliability caused by abnormal events that occur infrequently but result in extended unplanned outages when they do occur, i.e. High Impact–Low Probability events (HILPs). HILPs include catastrophic events such as turbine water induction, boiler explosions, generator winding failures, etc. Many of these successful programs have relied on the detailed reliability data contained in the North American Electric Reliability Corporation’s (NERC) Generating Availability Data System (GADS) that contains data collected over the past 25 years from 5000+ generating units in North America. Using this data, these companies have been able to 1) benchmark their fleet’s unreliability due to HILPs against their North American peers, 2) prioritize their peer group’s susceptibility to various HILP modes and 3) use root cause data contained within the NERC-GADS data base to help identify and evaluate ways to proactively prevent, detect and/or mitigate the consequences of HILP events. This paper will describe the methods used in these successful programs in sufficient detail to enable others to adopt the techniques for application at their own generating plants.