Power grid affect on steam turbines

[54fytng]

I work for a large power company at a pretty old gas fired steam turbine plant.
I started working there about 5 years ago but I have 30 years of power production experience in the military and civilian world before getting a job at this plant.
Most of my experience is with Diesel power engine driven generators and I have a very good idea of how droop in governors allow generators paralleled to share a given load.

My problem is this: We can have our 250MW turbine generator and another 120MW turbine running along on the grid and not on load control (generation dispatch has no control of our generators)

Anyway, we are sitting at around 200MW total and everything running smooth and all of a sudden the gas and air start swinging and our load goes up or down for no reason at all.

My theory is that the ebb and flow of the grid has an influence on our turbines at all times. If there is a gain or a loss of load on the net, it affects our turbines just enough to cause this excursion of gas and air and upset our boilers.

The old timers laugh at me because they think just because there is no loss of frequency or they cant see the governor oil pressure go up or down drastically, this cant be true.

I think that even minute changes can upset things enough that you can see a change of a couple megawatts up or down.

What do you all think?

 

[ScottyUK]

Movements in grid frequency can certainly have an influence - a small movement of a MW or two isn't unusual on a steam set of that size. Major disturbances to the boiler would require a fairly major upset in grid conditions to swing the voltage and / or frequncy about, and you would almost certainly notice that. It's not something I'm used to in the UK on our dense and heavily interconnected grid, but I can imagine it could occur on other grid topologies. Where are you located?


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

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

 

[waross]

Quite a bit left out. Can you describe the relationship between the governor oil pressure and the steam flow?
Is your gas supply stable? If you are on droop control, it will take a noticeable frequency shift to cause a noticeable shift in load. If you are controlling the generators, your control scheme will be causing the swings, not the grid. Your power output is closely related to your power input. If your controls vary the power input, the power output will change accordingly.
Do you get pressure swings in the gas supply? Are any of your mechanical control devices experiencing "sticktion"? Look at the P&IDs of your control scheme to see what may cause a swing in the gas and air.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[54fytng]

The controls are very old. These two units were put in, in towards the late 1950s. They are old pneumatic controls with modern P-I and I-P controls behind them.
We are a large company that has around 6000MW of capability for our generation area. We have 2 large coal burners, 500mw each about an hours drive from here and 4 large coal burners 4 hours from here. We have two combined cycle plants on opposite ends of the city area I live in, and a couple other large gas burners down south of the state, plus several wind farms.
We are also tied to the SW Power Pool that pretty much controls our demand, so we are tied to several states around us.

So if everyone else is on load control from our power generation dispatch and our dispatch just tells us to stay at a certain load like 200MW; our other plants are following the up and down demands of the grid as we just sit there we are not influenced by those other turbines on the grid as they raise and lower their load according to demand?

 

[waross]

Something is commanding your controls to vary the gas input, which I would expect the air to follow.
Are you using reset or integral settings in your controllers? If so, and the controller tries to make a small adjustment but is frustrated by "sticktion" in an actuator, the integral may continue to change the output until the actuator jumps to a new position.
I suspect that this is a much more likely cause than any grid influence on your operation.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[byrdj]

Determining cause for MW swings blamed on older steam turbine controls is my hobby.
You mentioned governor oil pressure so I’ll assume your turbines are WH. My background is only GE, but the principles are the same
First, is the unit being operated on the governor or on load limit?
If on load limit, the CVs are positioned by the operator and only a very large frequency INCREASE will close the CVs. If on the governor, then the CVs are constantly modulating with the minute grid frequency swings. I like checking the response of a governor in the morning. A significant frequency swing would be+/- 2 rpms. If the regulation is set for the nominal 5%, the CVs will only be responding +/- 1%. Thus the expected MW swing for a 200MW unit would be +/- 2MWs, IF the governor was in excellent condition and had no dead band. Thus normally it will be less.
Another factor for the actual MW swing will be how linear the CV camming. The deviation in linearity of CV opening is referred to as incremental regulation. On a recent investigation, the governor was moving the CVs just under the 1% expected, BUT the MW swing was over 5% due to the miss adjustment of a CV opening at the load they wanted to run at,
Next, is the boiler in some “element control” or totally manual. I would doubt totally manual, so I would expect the boiler pressure control occasionally pulses the governor to control the minute changes in the boiler.
The interface with the governor is usually the primary problem I encounter. Mostly the governor’s motor speed changer linkage will have dead band and binding. Thus the boiler control will give a pulse to change the CVs, but they don’t respond. Then the boiler controls gives 2 bumps and again nothing happens. Now it gives 4 bumps and the bind breaks free and the dead band allows the CVs to respond with a magnitude for all 7 pulses.
Another problem seen is the setting of the motor speed that drives the motor speed changer. As mentioned, the boiler controls pulses the MSC for a fraction of a second. The expected valve opening per pulse depends on the boiler control tuner. They normally like to get about ½% per pulse. However they seem to over look the speed setting circuit for the MSC I have seen a .2 second pulse give over 5% CV change. Once the MSC speed was set back to design, the pulse was the desired ½%

So when you have your MW swings
Is there a grid frequency swing, if so calculate the expected response and determine if the governor is doing what it should?
Is the MSC being pulsed during the swing, If so, what is pulsing it. Measure the MSC change and determine if it corresponded to the MW swing
If the CVs are under load limit and not moving, then the MW change is due to a change in steam conditions. A 5% increase in MS pressure will give a 5% increase in MWs, but pressure swings are normally slow.
Is there something going on in feed water, is a heater level going high and its drains dumping. Besides being an immediate effect on MWs due to steam flow through the turbine, the final feed water temp, will affect the boiler level

 

[rmw]

There isn't an electric arc steel mill anywhere nearby on your system is there? I was in a scrap fed steel mill connected to the utility just to the east of SWPP and the plant's electrical engineer mentioned that their instantaneous load swing was 155 MW. He was describing the system he had just designed to compensate for the unpredictability of the phase rotation each time the 3 probes lowered into the bucket of scrap - which was several times a day.

As he said it I pitied the poor unit nearby whose governor tried to chase that load.

Later I was in a oil/gas fired ~800 MW plant a few hours away and casually asked why they were allowed to run when it appeared that there were more optimum units on the grid that could run (coal, nuclear plants on their system nearby) and the answer was that they were the only unit on that end of their system able to change fast enough to take the swings caused by that steel mill. Seems the coal burners and nuclear plants wouldn't touch that.

Anything like that out there on your system?

rmwd

 

[waross]

As I understood the description, the unit was on fixed loading. With that control the unit should be immune to grid loading unless the grid frequency changed, and with some control schemes, immune even then.
byrdj's suggestion is along the lines of mine, but he has better first hand experience. When the control element doesn't respond to a small command the PID reset will wind up and eventually result in a large change.
54fytng;
How long does the condition last? Does it eventually correct itself? What measures do the operators take when an issue occurs?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[54fytng]

Sorry for the late answers to this blog, I have been in Aruba for a week
The units are on the governor as soon as they hit around 3500 rpm and yes these are WH turbines.
We do push MVARS at our end of town when necessary.

Yes our smaller gas units could be taking the load swings faster than the larger units.

Our closest large coal burner is almost 2 hours drive from here but we do have combined cycle units closer to us and they can handle things very fast.

This actually brings back memories of when I used to run old diesel units in parallel with mechanical governors set at 60 droop and 0 droop. The 0 droop unit would take the load swings and distribute those swings to the other diesel units next to them. Sometimes you had to coax those 60 droop units along and manually distribute the load.

Could our old steam turbines could be doing the same type of thing on our net.