Confusing financial results on Time-controlled charging

[sunlightfuture]

Hi,

We have had some concerns with the accuracy of financial modelling when using the time-controlled charging but today we have inadvertently stumbled on a test case where it demonstrates that this aspect of the modelling does not appear to be working correctly.

We are modelling a site for a client and we had intended on suggesting the client uses off-peak charging to increase the battery utility.  When I reviewed the project I found we had made a mistake in the tariff configuration and had used a single rate tariff From-grid tariff (so there was no off-peak period).  The Feed-In tariff rate and the From-grid tariff were set at the same rate.

This means that if we activated the Time-control to set the SOC target value to charge overnight to 100% the battery would get used more (incuring increased losses) but not generate much of a change in the rate of return (savings) as the increase in power pulled from the grid would be offset by the increase in Feed-In Tariff, both at the same rate.

However, we instead saw that the more forced charging we applied the results improved considerably: 

No Time-control charging:

Time-controlled charge to 100% between 01:00-04:00: 

We set off-peak rates correctly with an off-peak rate 75% lower than the peak rate and had a result broadly similar result:

Digging down into the data it seemed that the difference suggested that PV Sol was not accounting for the cost of electricity used to charge the battery in the time-controlled period.

We checked the losses due to charging/discharging and these were higher in the Forced charge scenario, as expected.

The resultant difference in financial projections was troubling. In this example the amortization period dropped from 11.5 to approx 7 years (regardless of low we set the off-peak rate).

This cannot be correct.

For reference, this is the Time-controlled charging schedule we used is below.

We would be very grateful for some input into this.  I fear we have misled many customers with our reports when we applied these settings in previous projects.

Kind regards,

Oliver

 

[developer_mo]

Hi Oliver,

thank you for your post and sorry for the late reply.

I tried to recreate a scenario similar to yours to test the behavior of the time-controlled charging. In my project, I could not reproduce the behavior you described.

Could you send me your project file (.pvprj) via private message in this forum so that I can take a closer look at the problem?

Kind regards

Mikio