3
\$\begingroup\$

Is the frequency behavior equal in very distant interconnected points of the grid? For example, if there is a sudden generation drop in Portugal, is the frequency behavior in Portugal exactly the same as the one in Poland? (Poland and Portugal are interconnected inside the CE power grid). If not, which are the differences in terms of amplitude and time?

\$\endgroup\$
7
  • 2
    \$\begingroup\$ As you've done a little research, perhaps you'd tell us how they are interconnected, AC or DC connection? The former requires synchronisation, the latter doesn't. It's not only physics that dictates how connected systems behave, but politics and commerce control the power flows that would synchronise them. \$\endgroup\$
    – Neil_UK
    Feb 25 at 10:00
  • 1
    \$\begingroup\$ @Neil_UK it is an AC connection. Yes, for sure there would be a small difference in oscillations between the two parts but, what I wanted to underline is if the frequency behaviors are more or less the same in amplitude and time trend. Probably there would be an oscillation mode with opposite directions between the two countries, but are them the same in amplitude and time trend? \$\endgroup\$ Feb 25 at 10:21
  • \$\begingroup\$ Power is randomly injected or consumed from the grid. it is therefore impossible for there to be a "direct relationship" between two distant points... \$\endgroup\$
    – Antonio51
    Feb 25 at 10:55
  • 2
    \$\begingroup\$ The "grid" is really a big "matrix" where the "neighbor's nodes" are more important, for regulation, than the other far nodes. So, frequency and amplitude at two different far nodes can't be easily related. Note that frequency is specified in a "close" interval and that frequency is not the only variable but also the close number of "periods" in a specified lapse of time (for the "old" "mechanical" clocks). \$\endgroup\$
    – Antonio51
    Feb 25 at 11:19
  • 5
    \$\begingroup\$ It may help to consider those two points as being connected by a very long springy shaft. The two generators could go out of synch but the more they do the more torsion will be on the shaft pulling them back into alignment. In practice you'd have multiple units along the shaft and multiple shafts joined in by 1:1 gearboxes as various angles. \$\endgroup\$
    – Transistor
    Feb 25 at 12:16

2 Answers 2

9
\$\begingroup\$

The frequency trend is similar but not equal in two different distant points of the network after a perturbation. The trends are different in amplitude and time, and this is due to numerous variables (inertia, control actions, different damping, etc...). To summarize, it is different because the electric grid is a very complex system with several variables and regulatory actions involved. But at steady-state, the frequency will have the same value due to the fact that the electric grid is synchronized. In the picture, there is an example of 3 different frequency behavior after an imbalance of 1 GW in Spain: enter image description here

\$\endgroup\$
6
  • \$\begingroup\$ Was the 1 GW a change in a load or source? Also, what's the source? I'd love to explore this dataset. \$\endgroup\$
    – LShaver
    Feb 25 at 15:12
  • \$\begingroup\$ I would add also that perhaps the most important variable of the grid is the "phase" of the generators nodes, because only the "phase" of these nodes is the key that can control the flow of reactive power through the grid. \$\endgroup\$
    – Antonio51
    Feb 25 at 15:31
  • 1
    \$\begingroup\$ @LShaver unfortunatelly I don't know what is the source, I took it from notes of my past course of Electric Power System. It was given by my professor in the slides. I want to find a source to check the frequency too, I cannot find it :( If anyone knows a source to check the frequency please tell us \$\endgroup\$ Feb 25 at 17:16
  • 2
    \$\begingroup\$ @LShaver If I had to guess, it's a power source going offline: drop in frequency means it's either more load or less generation, and an abrupt drop in generation (earthquake or wildfire or whatever knocking out the connection from source to grid) seems more plausible than an abrupt jump on the gigawatt scale, but I;m no expert \$\endgroup\$
    – llama
    Feb 25 at 22:20
  • 1
    \$\begingroup\$ Open source paper: nature.com/articles/s41467-020-19732-7 Open database analysis of scaling and spatio-temporal properties of power grid frequencies. This paper points to this database: Power grid frequency data base osf.io/by5hu \$\endgroup\$
    – D Duck
    Feb 25 at 23:03
4
\$\begingroup\$

No, it's not. Speed of light wouldn't allow for things to be "exactly" the same.

Much more importantly, you need to think about how "frequency adjustment" looks like, even if it happens with two generators standing close to each other: either one is the first to become slower, which means there's a phase difference in currents, linearly growing with time, which means there builds a blind power component between, which grows (trigonometrically related to phase difference), and then that leads to a slow down of the other, which has inertia, so, which leads to the same process in the other direction, so, there's going to be some oscillations in the frequency.

I'd expect you see frequency "dips" to "swoosh" back and forth through the European grid until they're completely dampened away – it's a distributed control system like any other :)

\$\endgroup\$
2
  • \$\begingroup\$ Muller, yes for sure some generators will rotate with a certain speed at the beginning and others not, ok, ok also the fact that there would be some oscillations, I know, but what I want to point out is the frequency of the grid electrical components (voltage and current). Is the frequency the same (ok, not exactly the same, but approximately equal in amplitude and time trend) between two very distant points of the grid? \$\endgroup\$ Feb 25 at 10:26
  • 1
    \$\begingroup\$ I explicitly answered that question in my answer. Please re-read my answer, especially the first sentence. \$\endgroup\$ Feb 25 at 10:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.