Generally, gas turbine failures are a result of several different causes: design or manufacturing errors, problems during assembly or commissioning, operating errors or even natural catastrophes. Most of the failures occur suddenly and are unforeseen. However, some issues leading to future failures might be found in advance by the identification of critical trends. Taking a reasonable intervention in good time, the failure can thus be avoided.

So called “condition monitoring” tries to identify these trends within the operational data or by using the vibration monitoring systems. An advancement is “predictive maintenance” where the usual behavior of engines or plants is assessed by machine learning with regard to life time and damage potential.

In this paper possibilities and limitations of early fault detection systems in gas turbines are discussed, based on the following chosen damage experience:

·      -  Flood event affecting a gas turbine during standstill

·      -  Sand ingress into a gas turbine during a sand storm as a result of an assembly error

·      -  Explosion of a waste heat steam generator during the commissioning of a gas turbine due to an assembly error

·      -  Vane fractures in a compressor due to the temporary occurrence of resonant vibrations

·      -  Rubbing event in the turbine as a result of an operating error within the lubricant system

·      -  Blade and vane fractures in a compressor due to increased wear and tear within the variable guide vane bushings

·      -  Rubbing event in the bearing seals due to increased leakage of compressor exit air into the bearing housing

·      -  Rubbing event in the rotor seals during a hot restart due to a critical cooling down cycle

For each example the failure mechanism and the root cause are shown. Afterwards, it is discussed

·     -   if it was generally possible to avoid the failure and

·     -   which monitoring systems would have been necessary to identify the damage occurrence by the condition monitoring or the predictive maintenance system.


Possibilities and Limitations of Early Fault Detection in Gas Turbines

Division / Sponsor:
International Gas Turbine Institute (IGTI)

Paper Proceedings: https://asmedigitalcollection.asme.org/GT/GT2020/volume/84058

Turbo Expo - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.