Key Concepts developing a test bench bed for an inverter connected to the public grid

[factory37]

Hey there!

I need to write a short document about how I should approach the develop of a test bench for the control of various inverters conected to the public grid.
I attached a picture of something looking like I mean. It should be like HEMS in the picture.


Firstly I thought in these points:

Hardware:

• Check the correct sensing of the voltage and current of the different inverters and plugs.
• Check the isolation of the sensing lines.
• Check its immunity to EMC, both conducted and radiated.

The control

• Add test routines to check the correct operation of the FPGA/DSP.

Interface

• Try the proper working of the differents modes of configuration.
• Debug the interface.

I am a bit lost about this kind of things and I need to know if there are something more interesting to check or if I've wrote it is a nonsense.
Also any technical key concept would be welcome.

Regards

 

[chuckey]

What happens when the grid gets struck by lightning? In general I thought that wind turbines and solar cells charged up a battery, which is then inverted and fed into the grid. This means that there is one LV-mains inverter and not three as in your diagram, also it has an easier life as its input volts will change by 15% at the most and its just one synchronising circuit.
Frank

 

[factory37]

So I have to add "Check the HEMS front overcurrent and overvoltage" in Hardware testing.
What do you mean with just one synchronising circuit?

Thanks
Regards

 

[betwixt]

You missed out the most important concept - "anti islanding", the ability to disconnect all power sources if the incoming grid connection fails. It is a legal requirement that all sources of power are immediately isolated from the grid to preven it being 'back driven' from the inverters. In view of it being a wireless controller, you would also have to ensure the latency in communications didn't slow it's reaction time too much.

Brian.

 

[mtwieg]

In what world does every outlet have a wifi connection? Seriously now, even new tech like solar inverters won't have that (or even a wired network).

In general I thought that wind turbines and solar cells charged up a battery, which is then inverted and fed into the grid.

This is sometimes the case, though it's something we're trying to abandon in favor of microinverters.

 

[factory37]

So finally it should be something like this:

Hardware

• Check the correct sensing of the voltage and current of the different inverters and plugs
• Check the isolation of the sensing lines.
• Check the protection against overcurrent and overvoltage
• Check its immunity to EMC, both conducted and radiated.

The control

• Add test routines to check the correct operation of the FPGA/DSP.

Interface

• Try the proper working of the differents modes of configuration.
• Debug the interface.
• Check the latency in communications.

Thank you!