In our site, we have turbogenerators for the production of electricity (max 7MW). One in service which is synchronized with the national network and others in stand-by. The objective of our study is to evaluate the risks during operation of a machine exceeding the limit of the hours of march (requires revision MI and HGPI), the tolerances of the risks will be shared with the service exploitation
Thank you for advising us for:
-the paramount parameters to watch out for the actions to be performed on the instrumentation parts (jet pump, exhaust transmitters, valves)
and for the mechanical part which is essential because our rotor exceeds their hours of service (vibration, paper temperatures, axial displacement ...)
Thank you for helping us find the potential risks and actions we should be doing to ensure safe operation until the major inspection.
-Evaluation of the risks for the operation of a turbogenerator exceeds the limit of hours of operation.
-The actions and strategy to adopt for the control and monitoring of these PGT.
A concise and thorough evaluation is impossible without MUCH more information.
Suffice it to say that this happens more often than one might think, but it has its risks. Many types of failures, including catastrophic failures can occur.
It would be necessary to know, among other things, the type(s) of fuel(s) being burned; the number of fired starts; the usual load of the unit (Base Load; Part Load; Peak Load; etc.); the number of trips, what was replaced during the last maintenance outage, what the condition of the machine was during the last maintenance outage--just so many things (the above and so much more).
Again, a thorough evaluation can't be given here on a free World Wide Web forum; it requires too many details.
And, this is a controls forum--not a mechanical forum. We deal with mechanical issues as they relate to control systems, but not mechanical issues in any depth.
Best of luck!
I am asked to evaluate all the mechanical contingencies related to the control systems during operation this turbine generator that exceeds the limit of hours of operation. Example what are the major risks and the actions to take with the system of control - vibration tree / overheating bearing ......
More infos :
Type of fuel burned:Naturel gas
the number of fired starts : 554;
the usual load of the unit: Part Load (3.5 MW--MAX: 06MW)
the number of trips: More than 04 trips in the last 4 years by overload due to the disruption of the electricity network that is synchronized with.
1 to 2 trips of turbogenerator by IVP (intervalvulaire presssure)
what was replaced during the last maintenance outage
what the condition of the machine : Fonctionnement en partiel load avec quelque anomalies exemple basse pression air comprimé de nettoyage filtre, défaut sur la vanne principale SDV de fuel gaz lors démarrage de la machine.
model : PGT-10 / POWER : 7MW at 45°C
manufacturer : NUOVO-PIGNONE
Thanks for support.
From this thread:
you have posted the unit is experiencing HIGH exhaust temperature spreads. That's NOT good for turbines, and the longer this persists the more damage can be done (is being done).
I don't really think the typical control system (of which a Mark VIe is) can be used to anticipate or predict a failure, especially a catastrophic failure. Operating a machine past the typical, scheduled outage regime is risky--and turbine control systems aren't made for that. They will trip the machine on high-high vibration, but the damage may already be done, and be made worse during the coast-down. Again, a steady increase in vibration over some period of time, along with a condition such as high exhaust temperature spreads, can be an indicator of some kind of failure and/or impending failure. And, again, often these kinds of failures happen in a couple of seconds, and the damage which occurs is worsened during the coast-down.
Turbine parts are designed with a life (hours of operation) in mind. Some parts are better than others. We don't know where the parts in your machine originated and what condition they were in after the last outage, or even the previous outage.
Turbine control systems are designed to control the turbine to produce as much power as possible while trying to limit wear on and achieve the desired life of turbine parts. Yes; the do try to protect the turbine, also. But, if a large enough piece of the first stage turbine is liberated because the metal just couldn't take any more heat and pressure because it was past its prime life, and strikes a first stage turbine bucket, well then there's gonna be a lot of damage which will result in high vibration which will cause the turbine control system to trip the turbine. And the turbine is running at several thousand RPM. The damage is already done. And the turbine control system couldn't have predicted that it would occur; that's not it's purpose nor does it have the sensors to do that.
Also, we have no idea how long you want to run the unit past the manufacturer's recommendations.... A month? Six weeks? Six months? And you already have high exhaust temperature spreads, and haven't really found the problem (at least not that you have shared with us; you've told us that you have found black (carbonaceous?) deposits on the fuel nozzles--but you didn't tell us if you cleaned the deposits off or is you inspected the fuel nozzles and/or fuel piping to see if the deposits are pervasive and be causing plugging (choking) of the fuel nozzle orifices).
What you are asking of the typical turbine control system is really not what the typical turbine control system was designed to do. It will do it's job is the turbine is operated normally and the parts are replaced in a reasonable period in conjunction with the parts manufacturer's recommendations. That's all the typical turbine control system can do.
High exhaust temperature spreads can lead to worse problems with the combustion hardware (fuel nozzles; combustion liners; cross-fire tubes; transition pieces; first stage turbine nozzles; to name a few).
The major risks are cracked and/or broken combustion liners and transition pieces, including hula skirt seals on the combustion liners. Turbine buckets (and nozzles) can also be negatively impacted by excessive hours of operation combined with high exhaust temperature spreads. Problems with turbine nozzle degradation can cause further damage to buckets (FOD - Foreign Object Damage) which can be catastrophic. None of this can be detected fast enough by the turbine control system to be of any good to limiting the damage. The exception to this statement is that often as nozzles and/or buckets lose metal pieces the vibration can increase--but that doesn't usually happen gradually, and using the typical "seismic" (velocity) sensors mounted on bearing caps aren't all that good for detecting this type of vibration increase, until it's too late and the damage has been done.
Monitoring exhaust temperature spreads is another way to try to predict damage which might be occurring or has occurred. Simply pulling fuel nozzles and inspecting for blockage (choking) is not enough. It's customary to use a borescope to inspect the insides of the combustion liners (including the hula skirt seal), cross-fire tubes, and transition pieces, and in some cases the borescope can be used to inspect the first stage turbine nozzles--all while the fuel nozzles are removed.
A borescope can be used to inspect all portions of the axial compressor and turbine to check for signs of damage or impending damage. Some sites do this when they want to extend the period for the next scheduled outage; it has mixed results. Borescopes can't really see or predict problems with degrading metal grain structures and internal damage to air-cooled nozzles and/or buckets.
A "sudden" decrease in CPD can be indicative of some kind of problem or impending problem. Especially in machines which have been operating long past a scheduled maintenance outage.
Basically, what would probably be best for your site would be to do a condition-based assessment of the unit, but that would involve a shutdown and some "invasive" inspection techniques (like a borescope) would be used, along with an analysis of the past inspection records (and photographs, if available). And even then, there are NO guarantees. (I worked on a GE-design Frame 5 a few years back that the owner wanted to put off a Major Inspection (MI) for six months. A shutdown was taken, and a condition-based assessment was performed and it was decided the unit, based on number of fired hours, starts and trips, as well as the borescope and visual inspections, could run for another six months. Two months later, a first stage turbine bucket tip was liberated (broke off) and the unit came crashing down, causing a big scramble to get parts and people together to perform a forced Major Inspection outage. There was no gradual increase in vibration--just a very sudden increase in vibration which caused the Mark* to trip the turbine, which coasted down to zero speed in less than 60 seconds.... Risk was assumed to be low, but that assumption proved to be incorrect.)
Supervisors and managers, and consultants, get PAID to perform risk assessment and management. Contributors to this World Wide Web forum get paid exactly the same as you paid to post your query here. That is: Nothing. We can offer advice, but that's about it.
Life is full of risks. We can mitigate some, and we live with others, and we try to avoid as many as we can. This dilemma is no different. But, with the available instrumentation and a typical turbine control system what you are asking is next to impossible to achieve.
Best of luck!