CoCT excess reactive power at night?

[Elbow]

I came across this on the web:

CoCT EV tariff briefing note

This is some sort of briefing document for the council - not clear if it was commissioned by the council or if it was just an initiative of "Friedrich Naumann Foundation for Freedom".

At the end of section 3.6 there is an interesting paragraph about the City of Cape Town and their problem with "excess reactive power" in the network overnight.  The article references 200MVAR or 200 mega-VA or reactive power - that must be exported back to Eskom actually breaching contractual obligations.

There are 1.3million households in the city (source - page 15) so that's 153VAR per household.

I suppose the excess reactive power arises since load at night has a poor power factor?  No high demand heating etc leaving only switching power supplies and suchlike?  Otherwise I don't understand where it comes from?

The briefing note is actually saying that electric car chargers and SSEG systems can help consume that reactive power and reduce the amount that needs to be exported.   

Or maybe its saying such systems can absorb "real" power and turn it into "reactive" power.  Or, maybe they are saying that a customer with SSEG will be taking little "real" power from the grid but will still be consuming the out-of-phase "reactive" power?

Who understands it better than me?

Quote

Chargers have the potential to provide reactive power:

Voltage support by distributed control of reactive power provision is seen as one of the potential ways EVSE can help solve local voltage issues in distribution networks (Zecchino, et al., 2017). The loads in grid supported SSEG installations, for instance, consume reactive power disproportionally from the grid (the panels typically only supply active power through the inverters), creating such local voltage issues. Using power electronics based charging infrastructure EVSE’s can inject reactive power into the grid while consuming active power during charging to counter this but can also potentially either generate or consume reactive power during charging. Larger public charging is a more obvious application of this but modelling studies have shown the potential for leveraging co-ordinated charging of residential EV chargers (Paudyal, et al., 2017).

While Eskom applies an ancillary charge for supplying excess reactive power during the high demand season, the city currently rather has a problem with excess reactive power that can be of the order of 200 MVAR on the system at night which then has to be exported violating contractual obligations to Eskom. While Eskom is currently satisfied that the general situation is not excessive, the capital cost of equipment to correct this amount of excess reactive power has been estimated at R150 million. As an alternative Electricity Generation and Distribution at the City of cape Town has initiated a pilot with a private 1MW (SSEG) installation which will run for a week overnight in reaction mode to establish the efficiency of generating reactive power from grid active power consumption. There may therefore be an option to buy reactive power from both SSEG and EVSE installations at night and this might be considered as an option in a future tariff.

 

Edited November 15, 2019 by Elbow

[Stanley]

A common problem with grid-tied PV inverters is that they only inject real power into the grid, actually making the power factor worse since they don't decrease the reactive power. Some inverters do allow you to inject reactive power to improve power factor, I have done this with the Huawei inverters and I'm sure many others can too. Basically reactive power is caused by current that is not in phase with the grid voltage, but this could be either leading the voltage (capacitive loads) or lagging the voltage (inductive loads). To correct for reactive loads you need to inject current with the opposite phase error, i.e. to correct for current that is lagging the voltage you inject current that is leading the voltage. So inverters that run in parallel with the grid could help correct the power factor if they have spare capacity, and they can do this 'mostly' without using any power (I say mostly because the losses do increase slightly)

edit: To answer your question a bit better, it looks like the CoCT is thinking of paying private SSEG owners for reactive power at night (when the PV inverters would otherwise be doing nothing) to help improve the power factor since adding power factor correction equipment for 200 MVAR is quite expensive

Edited November 15, 2019 by Stanley

[The Bulldog]

I can understand that a PV inverter can create out of phase current injection during the day if solar is available. But at night ? What am I missing ? My Solis has such a setting but obviously that only works if it has something to give.

Leading or lagging load to charge my batteries - OK I can do that. But how do I know by how much ?

 

[GreenFields]

3 hours ago, The Bulldog said:

But at night ? What am I missing ?

Layman's 2c. Feel free to correct as appropriate. I think the article relates to electric vehicle charging equipment, and the idea to use their electric vehicles' batteries/chargers as an "embedded generator" to return power to the grid at night with current leading voltage (electronically controlled), thereby reducing Eskom's expenditure on power factor correcting capacitors. This is in response to the evening effect that real power consumption (heating/lighting) reduces more dramatically than inductive loads (fridges, industrial machines). Normal embedded generation equipment like PV inverters do not allow this PF-correction option, especially since the real power they generate is non-existent at night. Personally I don't like the thought of Eskom tapping juice out of my "petrol tank" at night.

Edited November 15, 2019 by GreenFields

[Stanley]

Actually, injecting reactive power doesn't consume any real power in theory, but in practice there is the power that the inverter itself needs to run and some losses that depend on the amount of current you want to inject. So the inverter would need to consume a small amount of real power to inject quite a lot of reactive power and this real power could come from the grid.