Hello dear fellows, I need your

[Kevin8686]

Hello dear fellows,

I need your help please,

A power generation plant consists of 3 steam turbines each ( 4.8 MW - 6600 V - 50 Hz ) these turbines feed a network of an oil facility which demand currently 7 MW and when all equipment it demands 11 MW but mainly demand 7 MW.

This power plant is also connected to a public grid which is 220 KV - almost 5000 MW - 50 Hz by two main breakers to the busbar.

So, now we have 4 power sources ( 3 Breakers from 3 Steam Turbines and 1 breaker from national grid as the other breaker is usually disconnected; this busbar is 6600 V - 50 Hz - 7 MW )

The 3 steam turbines are controlled by Woodward 505E, each unit is totally separated the other, No common HMI, No DCS, No Woodward 505E HMI workstation, and No power management system.

In the Configuration documents the following is noted :

Speed Control Droop Settings:

Droop % 5

Use KW Droop Yes

Gen Load Units No



The Problem :

During and after the commissioning of Woodward505E control System ( before units were controlled by pneumatic classic control system ) we suffer form the following:



Whenever we have frequency and load fluctuation at the national grid trend Breaker A1 ( Electric Shedding sys. is used and has graphs ) then immediately we suffer bad fluctuation in the output power ( Active power trend output ) of the 3 steam turbines by almost 1 MW and each Generator start producing power and decreasing power and for 1 or 2 seconds the reading of the National Grid MW become Minus (- ....)

until the operator start Rising the speed of one or two steam turbine at the same time so the output load of the 3 Generators and the feeder come to stability.



This happens multiple time a day...



And when we shut down 1 unit the network and loads are relatively stable and when we put the Two National Girds (220 KV) Feeders on the main busbar (6600) with the other units then the frequency and loads are relatively stable.



Please find attached Trends,



Please help me with your technical opinions and experience in analyzing this problem.



Kind Regards.

 

[TugboatEng]

5% droop, that's ugly. Do the governors have the ability to run isochronous? It's normal practice to have one set in isochronous mode with the rest in droop. This gives your system a constant frequency, let's you set a constant load on the droop units, and only the isochronous sees a varying load.

 

[waross]

This is not a Tugboat.

5% droop, that's ugly.

A lot of utilities demand that co-generation sites run at 5% speed droop.

It's normal practice to have one set in isochronous mode with the rest in droop.

That's called the swing set, Tug.
The utility will have ONE swing set.
When a co-generation site is on-line, it does not use a swing set. It becomes part of the grid which already has a swing set.
Two swing sets on one system do not work and play well together.
Using an isochronous or swing set on a small system may cause more problems than it causes.
When in doubt, run in droop.
That small system will probably work well in droop when islanded.
If the droop causes issues when islanded, the last solution that I would use to address those issues would be an isochronous set.
By the way, when I was with the small utility we discussed the possibility of going from droop control to isochronous control.
On the one hand we considered the issues that may arise if an operator did not respond to a load change quickly enough.
And we considered that we had been running in droop for years and no customer had ever noticed that we were in droop rather than isochronous.
We remained in droop control.
And, by the way, the classic isochronous control responds in droop and then corrects to the original set point so you can still get the same over and under shoot in response to a step change in loading or frequency.

Back to the problem at hand.

Going back to basic control theory, a step change in load or in setpoint will cause a typical repeating over-shoot and undershoot until the fluctuations are damped out.
You will probably have good results tweaking the stability adjustment.
If this was a temperature control or a flow control I would increase the proportional band. (Proportional band is very closely related to droop. It is a different way of expressing the same effect with a very small difference in the numerical value.)
Unfortunately the utility will not want to see more than 5% droop so you need some damping.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[dik]

I had to look up 'droop'...

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[davidbeach]

Or take them out of droop and run them in PV (nothing to do with solar) mode with no governor action. Let the national grid deal with frequency response and simply have your plant supply its desired power. As long as you are paralleled to the national grid you have no frequency control, so trying to do so is just asking for trouble.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[crshears]

PV - valve positioning? If so, I don't consider this a very good idea, absent some manner of topology analysis to determine whether or not a grid connection exists; this is challenging.

Better in my view to always have a governor in operation in droop mode with the droop setting pegged at that required by the AHJ. When synched to a grid with a very stable frequency, adjustment of the speeder gear performs the same function as a valve positioner but without having to introduce some other form of turbine speed / frequency control in the event of loss of grid connection.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[davidbeach]

P=Power, V=Voltage. A common operational mode for generation running in parallel with the larger grid. Most units run to produce a set amount of power, regardless of frequency. Some units, generally hydro units in my neck of the woods are under governor control and respond to ongoing frequency deviations to keep things where they need to be.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[catserveng]

I have a number of cogeneration plants in parallel to the grid I work at regularly, in fact was at one this morning with a 3MW steam turbine using a Woodward 505D with a load fluctuation problem while operating against the grid. All but one of these systems use a Woodward DSLC and in most cases also a Woodward Real Power Sensor. Also most of these system use a form of steam pressure monitoring to manage load with available steam pressure. The one system that does not use a DSLC in conjunction with the 505 uses a pretty well done DCS/SCADA system that uses the input of the 505 that the DSLC would normally connect to.

All of these systems typically run in baseload or remote load control depending on how their systems are setup while in parallel to the utility. The systems I am familiar with are in the range of 2 to 6MW. Some of these sites can also operate in isochchronous load share in an island mode in parallel with the combustion turbines and in some cases available emergency diesel standby sets.

Take a look at the attached manual, page 14 has a table that discusses several example applications.

Properly getting a 505 setup on a steam turbine usually requires quite a bit of legwork, and serious experience to get it done right, even though I have worked on several I am by no means an expert, steam turbine generators are not my wheelhouse.

I would suggest you either contact Woodward or a company like TRi-GEN in Houston (and other offices in the world)that have the experience needed to properly deal with your questions and get your installation issues resolved.

My 2 cents worth, MikeL

 

[waross]

Old school.
Take Mike's advice on the new equipment.
Historically, droop was an easy way to set the power output.
It had the advantages of:
Avoiding frequency runaways if a large part of the load was suddenly lost.
Helping the swing set to limit frequency excursions during large load changes.
Allowing the grid to continue in operation with a slight and usually unnoticeable frequency error in the event that the swing set went offline.
dik:
nominal frequency 60 Hz.
5% Droop; no load frequency 63Hz, +5%
5% Proportional band full load frequency 57Hz, -5%
or
5% proportional band plus 5% offset = 5% droop.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[crshears]

P=Power, V=Voltage. A common operational mode for generation running in parallel with the larger grid. Most units run to produce a set amount of power, regardless of frequency.

Ah.

"Regardless of frequency" - what prevents unit over speed and damage if these units become isolated with less load than the unit's [s'] output? I'm serious; I've never encountered PV control before.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[waross]

The swing set handles load changes within its capability.
I'm sorry.
I forgot this old darling.
Getting old. grin
I did once visit a power plant that used a manual setting of the water flow, following telephone orders from the central load dispatch center.
This hydro plant was completely manually controlled.Sorry. From a picture I found today, I see that even in 1911 droop control was used.
It typically ran at either 10% output or 90% output.
When I arrived, one set was running at 10%.
While I was there a call came through to go to 90%.
The operator turned a control and a hydraulic ram slowly extended to increase the water flow.
He was watching the PF meter.
As the output increased, the PF dropped.
He would stop extending the ram and manually increase the excitation to bring the PF up.
This was repeated three or four times before he got the output up to 90%.
At the time I thought that the control was completely manual but I was wrong.
I have since found a photo of part of the control panel and realize that the operator was not controlling the ram directly, but was raising the set-point of the droop control and the droop covernor was controlling the water flow in response to the changing set point.

Powerhouse at Stave Falls
Powerhouse at Stave Falls closed due to COVID-19

To ensure the safety of the public and our staff, the Powerhouse at Stave Falls Visitor Centre remains closed. Updates on the 2021 season will be posted as they become available. We look forward to seeing you when we open.
Powerhouse at Stave Falls Visitor Centre

Take a walk through a 100 year old powerhouse and enjoy the historical displays, electricity demonstrations and many interactive exhibits. The Powerhouse at Stave Falls has something for everyone, from individuals to families and school groups. This former generating facility is also a National Historic Site of Canada filled with archival photos, artifacts and the original mechanical and electrical components which helped power the province.

In 2000, the dam's powerhouse was replaced after a four-year upgrade. The powerhouse was once British Columbia's largest hydroelectric power source and is a National Historic Site of Canada.[1][2]

Wiki

Link

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[redlinej]

When a droop generator is connect to unstable grid the droop when react with every frequency excursion the grid sees. This is inherent to droop mode of a governor. IF the grid frequency goes below 50(nominal) the Turbine will take on load and vise versa. the best operating mode for unstable grid is base-load (KW control) because the governor will be fix and will not change due to excursion.

 

[waross]

There are more things in heaven and earth, Horatio, Than are dreamt of in your philosophy.

Turbine will take on load and vise versa.

That is as it should be and contributes greatly to the stability of the grid.
With one swing set and the rest of the connected generators in droop, the whole grid responds to and limits frequency excursions due to sudden load changes.
With the kW control, only the swing set will react to sudden load changes and the frequency excursions may be an order of magnitude or greater in the event of a sudden load change.
But I was only responsible for the operation of a very small system.
What do I know.
Neither was Shakespeare an expert in power generation.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[redlinej]

Waross,
When connected to a unstable national grid the genset should be in a fix KW mode. The turbine should not be reacting to grid disturbance (you can operate in droop if the grid is very stable) . If the grid do become unstable and out side the limits of the genset protection, the plant can decouple from the grid via the main breaker and operate in droop or droop/isochronous or isochronous load share.

The reason for this if you are exporting or importing a fix KW you normally set that value and regardless of the grid frequency the genset will continue to carry the same load.

 

[waross]

Many systems use one swing set to control the frequency and all other sets are fixed kW output.
One way to set the output is to adjust the setpoint in droop mode.
As long as system load swings are within the capability of the swing set, the swing set adjusts it's output for varying loads.
In the event that a load variation exceeds the limits of control of the swing set, all the generators work together in droop to limit system swings.
As far as the swing set control;
Swing set control may be the Proportional plus Integral control.
P and I of a PID controller.
The first action of a PID controller to a system upset is Proportional or droop.
The integral feature detects the setpoint error and then corrects to offset.
When all the fixed kW generators in a system have the output set by a droop controller, all the generators will work in support of the swing set to minimize system disturbances.
Many system operators demand that c-ogeneration operators run in 5% droop for this reason.
If the National Grid is chronically unstable, it may well have all the sets other than the swing set in fixed kW control rather than droop control.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[waross]

If you are stuck with a widely varying grid, you are correct, you may have to use kW control.
If the system operator does not mandate droop control, one generator only in droop mode may be pointless.


Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!