[SOLVED] Theoretical Question about frequency behaviour

[danilorj]

Hello guys

I have a basic doubt about the behaviour of the frequency during a mismatch of power and demand in a distributed system. For example if the generation is higher than the demand the frequency will rise, and if the generation is lower than the demand the frequency will drop. So far so good. The problem is when this distributed source is connected to the grid, what happens to the frequency when there is a mismatch between generation an demand? Does the frequency still go up or down? I was thinking that the surplus of generation, in the case of a load reduction, would go the grid and there would be no increase of frequency. But I don't know if it's right or not.

 

[KlausST]

Hi,

You speak of mains grid frequency?
Then it is true is there is too much power, then there is an increase of frequency and vice versa.

In europe there are special "power plants" compensating this.

For more information:
https://www.mainsfrequency.com

Klaus

 

[D.A.(Tony)Stewart]

it is much more regulated than this.

No only does frequency matter but phase lag/lead when grid connected.

There is a central authority who manages phase frequency specs and outputs of each supplier to grid.

Adjustments are made to efficiently transfer ( buy/sell) power to also compensate for power factor ( lag/lead) and adjust frequency so overall all daily error is minimized.

At one time in Canada I lived in a major supplier area which sold to USA and I could calibrate my 10MHz clock to the 60Hz within 1e-10 and my voltage is 1~5% stable out of 10% limits.

Today it is far more complex with lag/lead compensation and buy/sell to grid.

Your scenario might apply to a 4 cycle gas generator with poor mechanical governor for RPM with load variations.

Since the majority of loads are slip frequency industrial induction motors, if the transient grid load approaches the grid supply capability, and fault is anticipated, it can recover by reducing frequency so motors reduce demand (coast) and then drop voltage if that fails.

This infrastructure supply/demand failure is what causes India and similar countries to have wide power quality instability, which costs billions to repair in a wide area grid and is being upgraded continually.

 

[FvM]

A small power plant connected to the grid or grid connected solar inverter will simply follow the mains frequency and adjust it's phase for zero reactive power.

 

[danilorj]

This is what I mean:



For example, if the pv panel is supplying loads A and B and for some reason load B is disconnected from the system, there will be a surplus of generation. Let's say power from PV panel = 6 kW, Load A= 3 kW and Load B=3 kW. Will this surplus of generation make the frequency of the utility grid, depicted in the picture, rise?
Let's look to an other scenario: and if the utility grid is disconnected? Let's say the pv panel is operating in a standalone mode. Where would this surplus of generation go, would there be a rise in frequency as well? I guess if it is in a grid-connected mode this surplus of generation would go to the utility grid. ok?

 

[KlausST]

Hi,

Let's look to an other scenario: and if the utility grid is disconnected? Let's say the pv panel is operating in a standalone mode. Where would this surplus of generation go, would there be a rise in frequency as well?

No, in standalone mode the PV will supply only as much power as the load draws. Expect only little and short rise in voltage. No chang in freqency in standalone mode. Frequency usually is fix.

When connected to the grid the PV delivers as much power as possible. If load is disconnected, then the power is pushed immediately to the grid. This power will not significantely change grid frequency.

Klaus

 

[BradtheRad]

Your equipment will never be capable of altering the grid frequency. The electric company is at very low impedance, which overrides your equipment which is at higher internal impedance.

To send power onto the grid, your output voltage must be 100 or 200 V greater than grid amplitude. It will require a different inverter than the one which powers your conventional loads. Or, you must install circuitry which steps up or down, as required.

Your grid-tied inverter needs to match grid frequency precisely. If it does not, the grid will force current in the wrong direction through your equipment at some part of the cycle. This puts your equipment at risk of destruction.

 

[danilorj]

ok. But why in some grid codes is written that the power delivered by the distributed generation should decrease if there is a rise in frequency? Let's say for some reason the frequency is within the interval of 50.5<f<52 Hz, then it is said that the distributed generation should decrease its power so the frequency drops below 50.5 Hz. And it is also said that the generation should increase its power only if the frequency is whitin 50 +/- 0.05 Hz for at least 300 s.

 

[FvM]

For example, if the pv panel is supplying loads A and B and for some reason load B is disconnected from the system, there will be a surplus of generation. Let's say power from PV panel = 6 kW, Load A= 3 kW and Load B=3 kW. Will this surplus of generation make the frequency of the utility grid, depicted in the picture, rise?
Let's look to an other scenario: and if the utility grid is disconnected? Let's say the pv panel is operating in a standalone mode. Where would this surplus of generation go, would there be a rise in frequency as well? I guess if it is in a grid-connected mode this surplus of generation would go to the utility grid. ok?

To take up the question title, yes changing the load will theoretically affect the instantaneous grid frequency because it accelerates or decelerates the generators. According to the relation of 3 kW to xx GW grid power, it won't by a detectable amount.

A standalone (island) electromechanical generator will show a frequency variation as primary response before it's possibly compensated by the controller. The behaviour of an electronical generator (inverter) depends on the implemented controller. It can be e.g. constant frequency.

But why in some grid codes is written that the power delivered by the distributed generation should decrease if there is a rise in frequency?

That's intentional frequency control. It's superimposed to the "natural" behaviour of the generator.

 

[BradtheRad]

ok. But why in some grid codes is written that the power delivered by the distributed generation should decrease if there is a rise in frequency?

The story I once heard, is that when the electric company let the mains AC drop in frequency, the town's electric clocks slowed down. They showed the wrong time. Therefore the electric company would speed up their generators for a while, until the clocks showed the correct time again.

Generator rpm is changed by feeding more or less coal to the burners, or by opening or closing water flow to turbines, etc.

When there is more than one generator, the waveforms must be kept in sync. I'm not sure how they accomplish that, to alter the speed of enormous generators spinning at several rotations per second.

 

[FvM]

Generator rpm is changed by feeding more or less coal to the burners, or by opening or closing water flow to turbines, etc.

When there is more than one generator, the waveforms must be kept in sync. I'm not sure how they accomplish that, to alter the speed of enormous generators spinning at several rotations per second.

One a generator is synchronized to the grid, it's rpm and generated frequency can't be individually varied. Only the power delivered to the grid can be changed in a first order, but at the same time, the grid frequency is pushed up or pulled down by a very small amount by accelerating or decelerating the distributed generators. The primary and secondary control mechanism described in the link in post #2 is intentionally varying the power of a number of power plants to achieve a grid frequency stabilization.

 

[danilorj]

What I still did not understand is why if I decrease the power injected by the pv panel, the frequency of the grid will also decrease. Let's say I have a pv panel at the roof of my house injecting power to the grid so that I can amortize the invoice amount a the end of the month. And if for some reason the grid frequency increases to a value within the interval 50.5<f<52 Hz, so if I decrease the power injected by my pv panel (at the roof of my house) I would decrease the frequency of the whole grid. That sounds very weird to me!

 

[BradtheRad]

Your equipment must detect the grid waveform. Your output must match its frequency.

This thread displays a schematic of a grid-tied inverter. It has a transformer which is driven by grid frequency. In turn it drives the internal operation. That is how the inverter stays in sync with the grid.

https://www.edaboard.com/threads/307293/

 

[KlausST]

Hi,

ok. But why in some grid codes is written that the power delivered by the distributed generation should decrease if there is a rise in frequency?

You can find the explanation when you follow my link in post #2

Klaus

 

[The Electrician]

What I still did not understand is why if I decrease the power injected by the pv panel, the frequency of the grid will also decrease. Let's say I have a pv panel at the roof of my house injecting power to the grid so that I can amortize the invoice amount a the end of the month. And if for some reason the grid frequency increases to a value within the interval 50.5<f<52 Hz, so if I decrease the power injected by my pv panel (at the roof of my house) I would decrease the frequency of the whole grid. That sounds very weird to me!

If you decrease the power injected by your equipment, yes, the frequency of the grid will decrease, in theory, by about .00000001 Hz, an undetectable amount. Similarly, when you increase the power you inject, the grid frequency will increase by about .00000001 Hz in theory.

The power you can inject is so small compared to the power the grid involves, that your contribution will have a totally undetectable effect on the overall grid frequency or voltage.

The way a small generator (alternator actually) controls the power it delivers to the grid, is that it tries to increase the voltage or frequency, but the grid is too powerful for that to happen. The alternator's attempt to increase the voltage or frequency results in power from the alternator being injected into the grid. The alternator's attempt to increase the grid voltage results, in theory, in an increase of the grid voltage by .00000001%, and a similar theoretical increase in frequency; the changes that a several kilowatt alternator or inverter can make in grid voltage and frequency are so small that they can't be measured. The effects might be measurable at your house, but the larger overall grid will only see effects so small that they are not measurable.

I hope you understand that the number I used for the increase or decrease of grid voltage and frequency as a result of your equipment injecting power into the grid (.00000001% or .00000001 Hz) is just a made up number intended to show how small is the effect of your equipment.

 

[KlausST]

Hi,

In germany the installed solar peak power was at about 4.9% in 2013 (increasing). So it definitely influence in grid frequency.
This results in several problems, because it is not that predictable than other power sources. Also the decentralized and uncontrolled power generation creates new problems. Frequency shift, domino effect and a power black out...


Klaus

 

[FvM]

I hope you understand that the number I used for the increase or decrease of grid voltage and frequency as a result of your equipment injecting power into the grid (.00000001% or .00000001 Hz) is just a made up number intended to show how small is the effect of your equipment.

Acording to a simulation for the UCTE area (covering Western and Central Europe), the initial frequency variation caused by a 3 kW load step would be factor 25 larger, about .00000025 Hz (0.25 µHz). Which supports your statement.

 

[The Electrician]

Acording to a simulation for the UCTE area (covering Western and Central Europe), the initial frequency variation caused by a 3 kW load step would be factor 25 larger, about .00000025 Hz (0.25 µHz). Which supports your statement.

What can I say other than :thumbsup:

 

[D.A.(Tony)Stewart]

• A small step drop in frequency causes a voltage phase lag that increases at a steady rate until it is leading.
• This would cause 100% reactive power cycling of out of frequency sync, lag, wrong phase, lead,in phase etc. ( N.G. i.e. fault)

There is a central grid clock reference , which is very precise and adjustments are sent by SCADA to each supplier.

PV autonomous systems require will require "Intelligent" control if they ever become pervasive to avoid disturbances.

 

[Vbase]

In Europe they invest in a very reliable power generation only because they know that black outs increase the birth rate.