Paralleling with Utility Theory

[buddy91082]

Anyone know of a good book or article that discusses how the paralleling sources know, such as utility and distributed generation (pv, turbines, generators, and etc.)how much load generation to share? For example, with gas or steam turbine systems operating in parallel with the utility, how does the utility know how much power to provide to supplement load that the turbines cannot provide.? How do the sine waves from both sources interact with each other, etc?

Looking for control and stability of paralleling sources.
Thanks
b

 

[davidbeach]

The throttle of the turbine determines how much power the turbine puts on the shaft to spin the generator; that power, less losses, becomes the electric power out of the generator. That power output is dumped into the vast pool of power that is the "grid". If there's local load, the power to supply that load is drawn from the same pool that the generator is adding to. In theory the load power could come from anywhere, in more practical terms, local load is supplied by local generation so long as the generation exceeds the load.

 

[waross]

Well said David.
A few more comments:
Excitation; With a stand alone or islanded set the excitation or voltage adjustment controls the voltage.

When the same set is operated in parallel with the grid, the voltage adjustment controls the voltage in the ratio of​

set capacity/grid capacity This ratio is so small that in practical terms there is no change in grid voltage. The excitation level or voltage adjustment does control the amount and direction of reactive power.
Real power; The throttle, gas valve or steam valve (power in to the prime mover) controls the speed and frequency of an islanded set. The throttle setting or power in to a grid connected set controls the real power delivered to the grid.
Expanding on David's comments, when the generator output is less than the local load, the grid supplies the power deficit. When the power output is more than the local load, the flow of power reverses and the excess power is delivered to the grid.
How do the sine waves interact?
The sine waves synchronize and lock together, almost. If the sine wave of the generator is slightly ahead of the sine wave of the grid, power will be transferred to the grid. If the generator sine wave drops slightly behind the grid the generator will become a motor and the grid will drive the machine.
Control; Co-generation sets are generally operated in droop mode. The utility will specify the droop setting. This gives good control. Droop allows the set to track with minor frequency variations and also acts to correct frequency variations. Droop control also gives an acceptable response to major system disruptions. Droop control prevents runaway should the set become disconnected from the grid.
Example; If the droop setting is 2% then a frequency setting of 102% or 61.2 Hz will result in 100% loading. A frequency setting of 101% or 60.6 Hz will result in 50% loading.

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

 

[xnuke]

A few additions:

Once synchronized, adjustment of the real power into the grid is adjusted by changing the no-load frequency of the generator to cause the voltage produced by the generator to lead the voltage of the grid by a small angle, known as the power angle. Power transferred is proportional to the sine of the power angle. This is accomplished by changing the power into the prime mover, as he said, for an electromechanical generator. For a PV system, it is accomplished by changing the firing angle of the grid-tied inverter switches so that its sine wave advances ahead of the grid's sine wave. Since PV output power can vary significantly and therefore the inverter input power can vary significantly, this angle is adjusted automatically by the inverter's controller as it tries to keep the input DC voltage to the inverter constant. As power out of the PV array rises, the DC voltage on the inverter's input capacitors will try to rise. In response to this, the firing angle of the inverter advances, and more power is transferred to the grid, lowering the DC voltage on the inverter input to maintain a constant value.

Droop allows load sharing between generators or between generators and the grid. If the droop percentage is the same for all connected machines, they will share load proportional to their ratings. Assume a 4 MW rated generator paralleled with a 1 MW rated generator with equal droop settings in an isolated grid serving 4 MW of load. The larger generator will carry 80% of the load due the the rating ratio of 4 MW/(4 MW + 1 MW), or 3.2 MW. The smaller generator will carry the remaining 20% of the load due to the rating ratio of 1 MW/(4 MW + 1 MW), or 0.8 MW. Note that both machines are carrying 80% of their rated capacity. Now, if the load increases to 5 MW, the larger machine will automatically increase its output to 4 MW, and the smaller machine will automatically increase its output to 1 MW. Assuming no other changes are made, the frequency of the islanded grid will drop as well, and the no-load frequency settings of each machine will have to be increased to restore frequency to its initial value. See Link for more details on load-sharing between generators or between a generator and the grid.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Compositepro]

"Once synchronized, adjustment of the real power into the grid is adjusted by changing the no-load frequency of the generator"

I would clarify this statement by saying "by changing the set-point of the no-load frequency". A grid-tied generator will always be turning at grid frequency, whatever its load. By increasing this set-point the generator will work harder to try to bring the frequency up but the frequency will not actually increase. The lead in the phase angle between the generator and the grid sine waves will change microscopically.

 

[ScottyUK]

"The lead in the phase angle between the generator and the grid sine waves will change microscopically."

Across the GSU transformer it is quite normal to have a significant phase difference between generator output voltage and grid voltage due to the GSU transformer reactance. This frequently causes difficulties when paralleling up a unit auxiliary supply tapped off the generator main terminals with a station auxiliary supply derived from the grid. Typically a check-sync relay will block the closure even though voltages match; de-loading the set reduces the machine's load angle relative to the grid and paralleling conditions are met.

 

[buddy91082]

Thanks for all the informative posts. Where can one read up on the topic?

 

[catserveng]

You may find the attached Basler paper helpful, it covers the topic pretty well in general terms. Basler has some other excellent papers on the subject but since they "improved" their website they have been difficult to find.

Another good piece of information is the Woodward Governor is their Manual 26260, here,

http://www.otherpower.com/images/scimages/3409/Governing_Fundamentals_and_Power_Management.pdf

And here is a short introductory article from NETA,

http://www.netaworld.org/sites/default/files/public/neta-journals/Charleswtr04.pdf

The EGSA Onsite Power Reference Manual also covers the topic in a couple of different chapters,

http://www.egsa.org/Publications/OnSitePowerGenerationBook.aspx

Hope that helps, MikeL.

 

[redlinej]

Woodward have a power generation learning module part number 8447-1012.All the theory explained.

 

[MatthewDB]

In response to Compositepro's post:

"I would clarify this statement by saying "by changing the set-point of the no-load frequency". A grid-tied generator will always be turning at grid frequency, whatever its load. By increasing this set-point the generator will work harder to try to bring the frequency up but the frequency will not actually increase. The lead in the phase angle between the generator and the grid sine waves will change microscopically."

That is only valid for a simple proportional response governor (intentional droop). Governors on island systems often have intentional droop in speed as load increases so generators automatically load share. More complex governors will have a power output setting that can dial in an exact power setting. This keeps the unit at a constant load even if the grid frequency drifts a little.

 

[buddy91082]

Thanks for the reading links Mike and redlinej.