Gas Turbine Common Issues, Failure Investigations, Root Cause Analyses, and Preventative Actions

Abstract

Gas turbines are a type of internal combustion engine and are used in a wide range of services powering aircraft of all types, as well as driving mechanical equipment such as pumps, compressors in the petrochemical industry, and generators in the electric utility industry. Similar to the reciprocating internal combustion engine in an automobile, energy (mechanical or electrical) is generated by the burning of a hydrocarbon fuel (i.e., jet fuel, diesel or natural gas). The core of a gas turbine engine is comprised of three main sections: the compressor section, the combustor section, and the turbine section. To ensure that a gas turbine operates safely, reliably, and with optimum performance, all gas turbines are provided with a control system designed either by the OEM or according to the OEM’s specification. The OEM-provided control systems will typically include complex and integrated subsystems such as (but not limited to): a graphic user interface, an engine management system (EMS or ECS), a safety related system (SRS), and a package control system (PCS) that may interface with a facilities’ existing computerized control systems. Any failure of the mechanical systems, electro-mechanical systems, or logic based control systems of a gas turbine can result in forced outage. A forced outage of a gas turbine, whether in a mechanical service, such as pipelines, or in either a simple cycle or combined cycle power generation installation results in a reduction of system availability and therefore a loss in revenue. The significant capital investment in a gas turbine system necessitates a high degree of reliability and system availability while reducing forced outages. A power plant can minimize occurrences of forced outages and optimize recovery of capacity by effectively combining proactive and reactive solutions. This paper will discuss both proactive and reactive programs as well as their implementation in order to answer the key questions that often surround an outage: How is outage time minimized while increasing reliability and system availability? What went wrong and who or what is responsible? How soon can the unit or the plant get back online? And what operational or maintenance considerations are needed to prevent a similar recurrence. Proactive approaches to be discussed include process hazard analyses (PHA) such as hazard and operability studies (HAZOP), hazard identification (HAZID), layer-of-protection analyses (LOPA), what-if analyses, and quantitative risk assessments (QRA) in addition to failure mode and effects analysis (FMEA); and failure mode, effects and criticality analysis (FMECA). Reactive approaches to be discussed include various root cause analysis (RCA) and failure analysis (FA) techniques and methodologies such as fault-tree analysis. Case studies and some lessons learned will also be presented to illustrate the methods.