Load rejection test survey

[byrdj]

After hearing about the accident at Eskom’s Duvha power station, there were other related articles (from googling) that seem to imply a load rejection test was something performed routinely (annually?).

Distinguishing between an actual over speed trip test, where the turbine speed is brought up under control of the governor, past the high speed stop override, till the emergency governor trips; A load rejection test being opening the breaker while the generator is at load and allowing the mechanical torque of the turbine to rapidly accelerate the turbine, usually at 10% per second. Thus testing the governor's response to see if it can control the over speed during this 1 second before the emergency governor setting.
The load rejection test I assisted with were planned and performed under the strictest conditions and fully instrumented. Prior to testing all the components were tested just prior and only key personnel allowed on site.
These were part of commissioning testing and required by contract. Recently, I have assisted with minimum load rejection testing they were performing to determine moment of inertia for system modeling, but they were also a onetime event and not a routine test.

So my question, does anyone perform load rejection test routinely and if so, how seriously are they monitored?

 

[GTstartup]

In my experience (GTs>100MW), just during initial commissioning and only using plant instrumentation and even then only if required by contract. We normally start with partial load rejection and work up to full load.

 

[ScottyUK]

Not aware of it as a routine test.

I do (or rather did) take an interest in how well the governors respond to a load dump event caused by external factors: a couple of times I've noted a bad response where one or more governors did not close as rapidly as commanded and the fault was investigated. I don't recall a true overspeed occuring as a result of a load dump - which is a good thing!


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[abeltio]

There are differences if you consider Gas Turbine and Steam Turbine.

In GT world, in the simplest of cases: the Load Rejection test requires that upon breaker opening the unit should "survive" and come to 100% Speed and no load (Full Speed No Load - FSNL), i.e. ready to resynchronize with the grid.
BUT:
Typically there are two breakers the Generator breaker and the Line Breaker, if the load rejection is performed with the Line breaker, the unit should provide "domestic loads" and EVEN might be required to go into "island operation" (sometimes even going into "isochronous" mode)

For ST units the concern is to avoid overspeed and the valves trip closed tripping the unit as well.

In both cases, the grid dispatch usually gets very nervous if the unit to be tested quite large - this is why those tests are usually carried out on weekends during the small hours of the morning (what a joy)
Typically the station instrumentation is used to record all the parameters.
This test could be a grid dispatch requirement depending on the contract between the utility and the grid (specially in the case of GT's to prove that the unit will not trip in case of a grid upset - adding a couple of hours to the recovery) on the other hand the overspeed test is a typical requirement from the insurance company.

saludos.
a.

 

[rmw]

For GT units, overspeed during full load rejection can result in problems with station loads if the line breaker trips. The GT and generator may be designed to withstand the overspeed immediately following, but other plant equipment connected to the generator might now.

rmw

 

[byrdj]

the GE LSTG "design" of the steam turbine's control was such that the governor would control the overspeed below trip speed so that the house load would be carried durring blow down and once settled speed is reached, the unit could be closed back in.

So far on my survey , it seems a load rejection test is not a routine (annual) test. I think the media must be confussing with actual overspeed test.

as a side to this quest, I have been researching to why some steam turbines designs have experanced destructive runaway when performing the emergancy governor testing and/or online excersizing. this was addressed in NUREG-1275 vol 11 4/95. this report has also provided me with an insight into why the religous zeal of overspeed protection is not shared by some of the other controls engineers

 

[ScottyUK]

byrdj,

The Westinghouse design was similar, relying on the governors to close in to control the overspeed, then using the MSV / throttles to cut off steam and trip the machine if the governors didn't catch the overspeed.


----------------------------------

If we learn from our mistakes I'm getting a great education!