1
\$\begingroup\$

The drawing below shows an induction motor (IM), powered from the grid through a variable frequency drive (VFD). This IM, through a gearbox, rotates a permanent magnet synchronous machine (PMSM) working as an electric generator.

The PMSM is an experimental model on a test bench. The IM, also on the same test bench, just simulates the (wind) turbine that is supposed to turn the electric generator.

The big issue is that the IM, like the PMSM, is of high power and draws a lot of energy from the grid. As the tests will last about one year, the cost of electricity will run to unacceptable high levels.

The power generated by the PMSM can not be pumped back into the grid because this operation will attract penalties from the electricity company and the bill will be higher than in the case when the PMSM dissipates its output on a 3-phase resistor.

The only solution that I see is to reuse the energy generated by the PMSM (which is about 75-80% of that consumed by the IM from the grid) and feed it back to the induction motor.

Question: How can I reuse the energy outputted by the PMSM? Is the schematic of such a feedback circuit complicated?

(The PMSM and the IM have powers of the order of 750 kW. However, even an answer that applies to an IM and PMSM of about 1kW will be welcomed.)

enter image description here

An induction motor rotating an electric generator.

UPDATE

Possible schematic made starting from the suggestions of the user mkeith:

enter image description here

GRID -> Rectifier -> IGBT (that only allows energy from the grid if the voltage across the smoothing capacitors drops below a certain level) -> IGBT block -> Induction Motor -> Gearbox -> Permanent Magnet Synchronous Generator -> Rectifier -> Energy back to the capacitors.

The question is will the schematic work? Are there fundamental mistakes I made?

\$\endgroup\$
4
  • \$\begingroup\$ @DKNguyen why not make that an answer? It's the simplest most obvious way to do it. \$\endgroup\$
    – Neil_UK
    May 21, 2022 at 5:15
  • \$\begingroup\$ @Neil_UK is that a complete answer? I don't feel like it is personally. \$\endgroup\$
    – DKNguyen
    May 21, 2022 at 5:15
  • \$\begingroup\$ @DKNguyen It's as complete as you want to make it. A better diagram than the OP's would help, showing where the variable boost converter or whatever matches the genmerator to the bus \$\endgroup\$
    – Neil_UK
    May 21, 2022 at 5:16
  • \$\begingroup\$ Just curious, is information (details) about this project published anywhere? Or will it be done? \$\endgroup\$
    – Vladimir
    May 21, 2022 at 23:26

3 Answers 3

1
\$\begingroup\$

Whatever you do, you need something that meets all safety regulations in your locale. At this power level, you'll get very large sparks if something goes wrong, and a very upset insurance company if you haven't followed all the rules.

To unify and extend the existing answers from DKN and TT ...

Once grid power has been through the VFD, IM, generator and its converter, you will have grid-level AC power again.

The electrical power output from the generator can be routed back into the mechanical input to the generator. You have several different ways to do this, depending on the details of what your electricity supply authority will allow, what you want to test, and where you want to spend your money.

The very simplest connection is, as TT says, route the power output of your final converter back to the VFD input terminals, it's likely to be directly grid-tie-able. This has the advantage that it involves little extra engineering. It may not be permitted by your electricity supplier. It might be allowed by them if you include some sort of fast acting breaker between grid and your equipment to shut off a backfeed should it occur.

The next connection as suggested by NKD has isolation between the grid and your equipment. Your VFD will be one of two types, a matrix converter (not suitable for this suggestion as it has no DC bus), and one that uses a DC bus fed from the input 3 phase with with an input rectifier. If you feed this DC bus with a rectifier from the output of your generator's converter, then the VFD input rectifier positively prevents any backfeed into the grid. This needs an accessible DC bus in your VFD, or a DC-input VFD with a separate input rectifier.

You could conceivably obtain even more isolation with no electrical connection by running two IMs into your gearbox. Use a 250 kW machine fed from a small VFD from the grid. Use a 750 kW machine fed from a VFD driven by the output of your generator converter. This would be more expensive, needing a second (but small) IM, and a gearbox or common shaft mounting that could combine both, but worth a thought if the electrical routes are not permitted, and you're happy with the mechanical issues. The feedback VFD would need to be happy with driving a spinning motor, regardless of the amount of power it was trying to push into it.

\$\endgroup\$
1
\$\begingroup\$

It's easiest if your VFD has the ability operate off a shared DC bus. Then you can just feed into the DC inputs after making sure the voltage is at the correct level. It's way more complicated if your VFD can only accept AC because then you basically have to synch up and track the mains (though it seems you already have a converter capable of that).

\$\endgroup\$
9
  • \$\begingroup\$ I do not have the VFD yet. What I can say is that the block labeled "Convertor", in the image I posted, works like this: PMSM -> AC (of variable frequency) -> DC -> AC (50 Hz) -> Grid. From what you say, I understand that the VFD must have two inputs, one AC to get energy from the grid and the other DC to collect the power of the PMSM. This is an unusual VFD. Where can I find it? \$\endgroup\$ May 21, 2022 at 5:26
  • \$\begingroup\$ At the 750 kW level, you specify to the supplier what supplies your VFD is to take, so you specify DC, it will be cheaper than an AC-input model. Then you buy an AC->DC converter, called a bridge rectifier, to drive it from the AC grid. If you want to test your entire converter as well, you buy another bridge rectifier to take the AC output of your converter to the DC required by the VFD. You can (or should be able to) adjust the output voltage level of your converter to control the current going to the VFD. For safety and testing, you also need a 1 MW load you can use. \$\endgroup\$
    – Neil_UK
    May 21, 2022 at 5:46
  • \$\begingroup\$ I have no experience with such high power levels. But small VFDs generally have exposed DC terminals and you can make DC connections with other VFDs. So if the DC link voltage (or rectified voltage) of the PMSM is compatible with the VFD DC link, this should work. See this document. support.automationdirect.com/docs/hitachi_dcpower.pdf \$\endgroup\$
    – user57037
    May 21, 2022 at 8:40
  • 1
    \$\begingroup\$ @RobertWerner I, too, don't have experience with VFDs that big. The ones I work with are AC input but sometimes have the DC inputs available so they can be used with a shared DC bus so you can hook both up at the same time. But if you must choose between only AC or DC input for the larger VFDs, pick the DC input. It is easier to rectify the AC input from the grid than it is to invert the DC output from your generator. In the smaller VFDs being able to run off a shared DC link it is not always an advertised feature. You have to look in the manual and look at the terminals to know. \$\endgroup\$
    – DKNguyen
    May 21, 2022 at 15:31
  • 1
    \$\begingroup\$ @mkeith Yeah you need to make sure the DC feed is equal (or maybe a tad bit higher to account for rectifier voltage drop) than the grid after DC rectification. \$\endgroup\$
    – DKNguyen
    May 21, 2022 at 21:10
0
\$\begingroup\$

Just wire both ends to the mains in parallel.

Assuming that your "Converter" is capable of synchronizing with the grid,you should already have all you need to reuse the generated power. Just connect the VFD and the output in parallel. The generated power will always be strictly less that the consumption of the VFD,so there's no way you could end up with a net supply to the grid that would make the power company unhappy.

There are two potential concerns to keep in mind:

  • Power factor. Likely a non-issue with a decent VFD.
  • Transient backfeeding on sudden shutdowns. If the grid fails, the rotational inertia will keep the system generating for a while longer. Either your Converter is already prepared to handle this (stop generating if the grid is lost),or you will have to add a load to dump the residual energy into.
\$\endgroup\$
3
  • \$\begingroup\$ This is where physics meets regulations. 'Just' (that engineering 4-letter word) wire your 1 MW output to your input, and the electricty supply authority will come down on you like a ton of bricks. Even if it works, at that power level, and running for a year, you are going to get caught. \$\endgroup\$
    – Neil_UK
    May 21, 2022 at 11:41
  • \$\begingroup\$ @Neil_UK Would you mind explaining what I got wrong? I can delete my answer if it completely missed the mark. I'm not a power engineer,but at a former employer we had a 400 kW diesel generator wired in parallel with the grid connection. The power company didn't mind us running it together with the grid,we just had to make sure we won't ever backfeed the grid (the main "breakers" would trip in that case). \$\endgroup\$
    – TooTea
    May 21, 2022 at 14:38
  • \$\begingroup\$ It depends on location and local regulations. Your power company didn't mind you parallel feeding. It appears you had breakers that would prevent that. With no location in the OP, and no warning in your answer, it would be safer to assume that the OP's local power company would object to a direct connection. He did say that they would charge him prohibitively to dump the power back into the grid. Maybe if be backfeeds 'inside the machine' (however that is defined), and uses fuses and breakers grid-side, they would permit it? How strong does 'make sure' have to be? \$\endgroup\$
    – Neil_UK
    May 21, 2022 at 15:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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