Can anyone give me some pointers on the ways used by electricity suppliers to control voltage levels and frequency in power networks?


Rotor windings control output voltage and armature rotational speed sets the output frequency. The devil is in the detail of course.

enter image description here

The field winding in the above picture is on the rotor - a DC voltage sets the excitation current and effectively produces a rotating magnet when the rotor is spun. This induces voltages in the three stator windings and these voltages, become the three phase supply outputted from the generator.

This is very simplified of course.

  • \$\begingroup\$ This is correct in terms of generator voltage, but I'm not sure it answers OP's question (the electricity suppliers). Also, "the armature rotational speed sets the frequency", in a power system it's pretty much the other way around (unless it's very small) \$\endgroup\$ – Stewie Griffin Jan 11 '17 at 12:42
  • \$\begingroup\$ @StewieGriffin maybe you can paraphrase the question so I can see what you mean? I'm not following what you meant in your second sentence i.e. what does "the other way around" mean? \$\endgroup\$ – Andy aka Jan 11 '17 at 12:51
  • \$\begingroup\$ I think OP is asking for what actions the system operators (or as OP calls it, the electricity suppliers) take in order to control the voltage magnitude and frequency, not how you would control the voltage at the generator terminals or the rotor frequency. I.e. how can the operators ensure the frequency is 50/60 Hz, and close to nominal voltage. \$\endgroup\$ – Stewie Griffin Jan 11 '17 at 13:05
  • \$\begingroup\$ If I'm right and OP is interested in the "system", not one or a few components, then it's the grid frequency that determines the rotor speed. There are no ways you can control the rotational speed of a single generator. Increasing the input torque might give you 0.001 Hz or something, but not enough to call it "frequency control". \$\endgroup\$ – Stewie Griffin Jan 11 '17 at 13:05
  • \$\begingroup\$ Note: I'm not saying I'm right and you're wrong. It's just a question about what OP is asking for, and I think he/she is asking about the grid/system. By the way, it's of course correct that you can boost the voltage locally by increasing the field current, even in a large stiff system. :) \$\endgroup\$ – Stewie Griffin Jan 11 '17 at 13:09

Here's a short and simple answer:

The grid operators control frequency by ordering more or less generation (watts) to be delivered to the grid. If the grid frequency is dropping below the target (usually 50Hz or 60Hz), they'll request more generation from those generators that are online and scheduled for frequency regulation duty.

The grid operators control voltage by ordering more or less reactive power to be delivered to the grid. If the grid voltage is dropping below the target voltage, they'll request more reactive power (kVARs) to be injected into the grid. This could be from generators, or from capacitive reactance banks.

In reality, grid operators don't really control voltage per se. Generally, at any point in the system, the grid characteristics are known, and either the generators are contracted to supply a certain number of VARs with their power (watts), or there are capacitors installed at strategic locations on the grid that supply the required VARs. The only thing that grid operators normally worry about is frequency regulation, and they do that by scheduling generators to start and stop. If the grid voltage changes significantly, then the system might be collapsing, but one would likely see the frequency collapsing first.

All generators are contracted to maintain voltage at their point of interconnection at some agreed value to offer the best grid voltage. In order for the generator to control voltage, that single generator will vary the VARs as necessary to maintain voltage at their point of connection. Likewise, all other generators are doing the same.

  • \$\begingroup\$ Yes! Well written! \$\endgroup\$ – Stewie Griffin May 5 '17 at 19:49

I'll answer this from a transmission system operator's point of view (the electricity suppliers), as I believe that's what you're asking about.


Frequency control involves a number of factors, from the control loops at individual generating stations (governors and the like) to the overall grid management. My answer will focus on the overall grid management.

The frequency in the grid directly correlates with the power balance. You've probably heard that electricity must be produced and consumed at the same time. You can't store electricity in the power system.

This is partially true.

Suppose you have a 50 Hz power system in balance, meaning your power injection (production and import) equals your power consumption (consumption, losses and export).

If you suddenly lose a large load, for instance a large industry plant, you'll have more production than consumption. The immediate result of this is that the frequency in the grid starts increasing.

(Analogy: You drive a car towing a loaded trailer at constant speed along a straight road. Now, if that trailer suddenly falls off and you don't compensate by pressing lighter at the gas pedal you'll start accelerating.)

Luckily, most grids are stiff. They have hundreds, thousands or millions of rotating motors and generators connected to it. You have lost the industry plant, giving you overproduction. That excess energy will cause the frequency to rise. Since it's impossible to increase the frequency in only one part of the grid (except transient behavior), you need to increase the frequency of all motors and generators in the entire grid. Increasing the grid frequency by only a fraction of a Hertz takes immense amount of energy giving the system operators some time to take corrective actions (Remember, there's a difference between power and energy).

The system operators will now look at how they can restore the balance. They will, after assessing the system reliability, the N-1 principle etc. ask one or several producers to turn down their production. They'll simply ask (or rather tell) producers to produce less power.

Similarly, if you want to connect the industry plant to the grid again, the producers will be asked to turn up the production again.

You might be interested in the speedometer shown on the front page of Statnett, the Norwegian transmission system operator. It shows the frequency in the Nordic power grid in real time.


This is a very different story. The voltage varies a lot and you can have over voltage one place, and just some kilometers away have under voltage.

I'll disregard the short term fluctuations, and rather look at the longer time perspective (the dynamic and transient response is quite complex).

The voltage in a power system depends (almost solely) on reactive power. Reactive power is a the part of the power that doesn't result in actual work being done (this is called the Active power, and is the part we discussed in the frequency "chapter").

The system operators have installed capacitor banks, static var compensation and more, that can be used to inject the desired amount of reactive power into the system. Some of these are static, as the names imply, while some are automatically (or manually) adjusted based on the voltage level in the station it's located.

In addition to this, the system operators can adjust the taps on transformers. Some transformers have manual tap changers, while some have automatic changing. By changing the tap on transformers you can typically turn the voltage up or down up to 10% (+/-5%).

Then we have voltage regulators that acts directly on the generators, that will increase or decrease the magnetization of the stator or rotor to adjust the output voltage.

All these combined (and more) are at the system operator's disposal (although some are static).

  • \$\begingroup\$ Nice wall of text unfortunately a few major errors. Just one example, generating plant rpm is controlled by a governor which maintains the plant rpm and hence frequency very tightly. If Power needed drops the amount of steam being fed to the turbines drops to maintain RPM. Given the inertia of the turbine/generator this is not a difficult control loop. \$\endgroup\$ – RoyC Jan 11 '17 at 14:09
  • \$\begingroup\$ The grid frequency is not governed by governors. What you describe is true if the system consists of one single generating plant, and it's true on a very short time frame. It's a bit more complex when we connect the entire country together, or when we're talking about minutes, hours and days, not seconds. Do you feed all turbines (in the country) less steam, or only one? You can easily tune this in a single plant, but what about all the other plants? I doubt their turbines are part of your control loop? How do you pick? \$\endgroup\$ – Stewie Griffin Jan 11 '17 at 19:50
  • \$\begingroup\$ If you have such a simple control loop and can control the frequency very tightly, then how come it fluctuates? Live coverage of the frequency in the Nordic power system. \$\endgroup\$ – Stewie Griffin Jan 11 '17 at 19:53
  • \$\begingroup\$ No it is not. The automated response systems take care of the second by second minute by minute response. The human mediated response takes over on the longer term. This involves shutting down the more expensive generating plants like gas turbines and falling back on the base load stations like nuclear and oil fired steam. \$\endgroup\$ – RoyC Jan 11 '17 at 19:56
  • \$\begingroup\$ Yet, the automated systems don't manage to keep the frequency constant, it goes up and down all the time. This is what I describe in the paragraph starting with: "Luckily, most grids are stiff." How would you explain the fluctuations? \$\endgroup\$ – Stewie Griffin Jan 11 '17 at 19:59

The frequency is still controlled very precisely of more interest is the fact that every generator on the grid has to be synchronized with all the others. This involves very precise and minor tweaks to the rotation of a generator until all the phases match and then switching it into the grid.Once brought into synch it tends to stay there. The result of bringing in a 500MVA generator and turbine out of synch are very serious. I was told stories of generator and turbine flattening a path miles long through forest in Canada when its bearings failed due to the shock of this.

  • \$\begingroup\$ This is how you would connect a generator to the grid, but I'm not sure if it answers OP's question about how the suppliers control voltage levels and frequency. \$\endgroup\$ – Stewie Griffin Jan 11 '17 at 12:44

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.