PVsyst's forum

Welcome on PVsyst's forum.
It is currently Wed Sep 23, 2020 6:23 pm

All times are UTC+02:00




Post new topic  Reply to topic  [ 9 posts ] 
Author Message
PostPosted: Wed Jun 12, 2013 1:29 am 

Joined: Thu Aug 02, 2012 3:38 am
Posts: 105
I'm working on some 500kW projects in Hawaii that are non-export grid tied systems. I am going to limit the real power production of the inverters to stay below the site load and I need to adjust the inverter power factor to offset the low power factor of the site load.

What would be the best approach to do this in version 5 of PVsyst so I can model this system? Can I just take an inverter model file and reduce the real power production number to be what I will set it at or is their an interaction with other parameters that I need to take into account?



Thanks,

Marvin Hamon


Top
   
PostPosted: Tue Jun 18, 2013 11:40 pm 

Joined: Thu Aug 02, 2012 3:38 am
Posts: 105
No help with this one, anyone? Someone's got to know if this is possible.


Top
   
PostPosted: Thu Jun 20, 2013 12:23 am 

Joined: Fri Jun 14, 2013 12:06 am
Posts: 6
Hey Marvin,

I've dealt with something similar in the past and haven't really come across a good way to do this either.

I have ended up applying this curtailment through some post PVSyst processing. I think is an acceptable solution since the real power component can be output and you can just clip the power to supply your load, in a post processing. This also is the most realistic approach, since the curtailment would actually reduce the input power drawn rather than reduce the inverter output. Further, this also gives you the ability to apply a curtailment distribution rather than a single load limit, which would be the case if you modified the inverter file.


Last edited by traj123 on Thu Jul 11, 2013 3:31 pm, edited 2 times in total.

Top
   
PostPosted: Thu Jun 20, 2013 6:59 pm 

Joined: Thu Aug 02, 2012 3:38 am
Posts: 105
If you do the curtailment on post processing how do you also adjust the annual energy production estimate?


Top
   
PostPosted: Thu Jun 20, 2013 11:44 pm 

Joined: Thu Aug 02, 2012 3:38 am
Posts: 105
After going through the documentation and the model file a few more times I feel pretty comfortable that if I want to put a limit on the inverter real power output I can simulate this by just creating a copy of the inverter model file and putting in a lower Nominal AC Power. PVSyst should use this correctly in the simulation. None of the other values in the model file are not used in the simulation as far as I can tell.

If anyone at PVSyst thinks this is incorrect please speak up.


Top
   
PostPosted: Fri Jun 21, 2013 10:14 pm 
User avatar

Joined: Mon Apr 16, 2012 7:29 pm
Posts: 1680
The power E_grid issued from the PVsyst simulation is the active power, expressed in [kWh].

Now we have to define exactly what we are speaking about.
When the grid manager requires a given cosPhi (reactive power), then this should be programmed in the inverter device, and there may be several possibilities:

- either the Pnom specified by the inverter's manufacturer corresponds to a real power, and the "apparent" output power will be E_OutApp = E_Grid[kWh]/CosPhi, expressed in [kVAh].
In this case the simulation adjusts correctly the overpower conditions.

- or the Pnom specified by the inverter's manufacturer is an apparent power. In this case the result of the whole simulation will be the same (E_OutApp[kVAh] = E_Grid[kWh]/CosPhi), but the power limitation should occur for PnomApp [kVA] = Pnom [kW]/cosPhi. Therefore we have to adjust the Pnom specified in the inverter's definition by dividing it by the required CosPhi.

PVsyst doesn't treat the reactive power in the present time.
For introducing it, we will require an additional parameter from the inverter's manufacturer, i.e. specify whether the inverter works with the first or second hypothesis. In the second case PVsyst will adjust the Pnom for the power factor CosPhi, what should be done manually in the present time (by modifying the Inverter's definition for each given CosPhi).
I don't have any information about this behaviour with real inverters in the present time: I will contact some manufacturers for precising this point.

NB: Another question may arise: is the power limit required by the grid manager a real or an apparent Power ? This will act on the choice of the PNom of the inverter.


Top
   
PostPosted: Mon Jun 24, 2013 7:11 pm 

Joined: Thu Aug 02, 2012 3:38 am
Posts: 105
It's not clear to me what you are doing in the second possibility.

"- or the Pnom specified by the inverter's manufacturer is an apparent power. In this case the result of the whole simulation will be the same (E_OutApp[kVAh] = E_Grid[kWh]/CosPhi), but the power limitation should occur for PnomApp [kVA] = Pnom [kW]/cosPhi. Therefore we have to adjust the Pnom specified in the inverter's definition by dividing it by the required CosPhi."

Maybe I am reading too much into this but it looks to me like you are saying that the energy simulation output will be in whatever units that are specified in the inverter definition for Pnom. That is, if I use PnomApp [kVA] I will get EoutApp [kVAh], if I use PnomReal [kW] I will get EoutReal [kWh], and if I use PnomReac [kVAR] I will get EoutReac [kVARh]. Is this correct? The remaining question then is what should the power limitation be based on, PnomReal or PnomApp. It looks like you are saying to use PnomApp but I am not sure.

In reality all inverters are already rated in kVA for the inverter output with a default power factor of 1 making PnomApp = PnomReal, so the second hypothesis is the correct one to use in all conditions. It has become customary for manufacturers to list inverter output in real power due to the assumption that the power factor everyone uses is 1. If you assume the second hypothesis and provide a variable in PVSyst for power factor to be entered you can cover both an inverter producing only real power and an inverter producing both real and reactive power.

Now that inverters are becoming more advanced there is much more that will need to be incorporated into PVSyst in the near future to keep up. Step one would be to allow for a fixed non-unity power factor or fixed VAR production, and also a fixed reduced real power production. After that and coming soon will be dynamic VAR control and dynamic real power production control, but since these will be variable it might prove to be very difficult to incorporate.


Top
   
PostPosted: Thu Jun 27, 2013 6:00 pm 

Joined: Thu Aug 02, 2012 3:38 am
Posts: 105
Since there is no reply to my last question I guess I will do some more speculation.

Let's restate the problem.

1. Some modern PV systems need to be operated at other than unity power factor (PF)
2. New inverters are coming on the market that allow PFs to be set as low as 0.80
3. PVSyst does not accept PF as an input and accepts the nominal power rating (Pnom) of an inverter in kW in the inverter definition file

Given this how can we simulate inverter operation at less than unity PF?

Options:

1.) Set Pnom to the real power that the inverter will be limited to based on the Papparent*PF. This will result in an output of the energy produced in kWh and the inverter will limit at the Pnom setting. The problem with this is that the simulation will not use the inverter efficiency curve correctly since PVSyst will be computing based on the real power output but the inverter will actually be producing more apparent power and therefore be on a different part of the curve. Power limiting will also occur at a lower power than the actual inverter power output.

2.) Set Pnom to the apparent power generated by the inverter. This will result in the energy output being in kVAh which would need to be converted to kWh and kVARh in post processing. PVSyst will use the correct points on the inverter efficiency curve for the simulation and power limiting will be correct. The problem then is the extra steps in post processing and that the PVSyst report becomes less useful since it is in kVAh.


Top
   
PostPosted: Fri Oct 04, 2013 1:44 pm 
User avatar

Joined: Mon Apr 16, 2012 7:29 pm
Posts: 1680
The Power factor definition has now been implemented in the version 6.11 of PVsyst.

See Can I define a Power Factor in PVsyst ? in our FAQ.


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 9 posts ] 

All times are UTC+02:00


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Limited