Energy storage at Day-Night time slice level

[Arne@IFE]

Hi

I'm trying to model a battery storage device, but I'm facing a problem with the relation between the energy storage level and the capacity of the battery. It works ok at seasonal and weekly time slice level, but when I try to use day-night time slice for the battery, the capacity (MW) of the battery becomes too low.

I'm using a storage device (Not NST or STK).

Any suggestions on how to model a battery at day-night times slice level?

[wnijs]

Hi,

A quick response.

A storage process has an activity (ACT) that represents the storage level.

The capacity (CAP) represents the maximum level (energy !) so some MWh for example.

If the storage process is a bottle of water, ACT is the level of the water and CAP is the fact that you can have 1.5 liter stored into it.

In a DAYNITE storage the capacity normally represents the max. daily amount that can be stored.

However, you need to take into acccount a conversion factor. 

For each timeslice you have following equation active:

ACT(STORAGE) < CAP(STORAGE).AF.G_YRFR.PRC(CAPACT) . Nr of days in the season.

(I hope this is correct)

I advice to get a very clear picture of all the units.
Do you still have insufficient capacity ?

Best,
Wouter

[Arne@IFE]

Hi

Thank you for the quick response. I have tried to use both MW and GWh/annum as the capacity unit (of course with different conversion factors), but the storage level (in GWh) is way too high.

What I actually get is that the ACT(Storage) > CAP(STORAGE).AF.G_YRFR.PRC(CAPACT) . Nr of days in the season.

This only happens when I change from seasonal or weekly to day-night time slice. So it works ok with season and week.

Best,

Arne

[wnijs]

Hi Arne,

If you could upload some information in Excel or upload an image with timeslice information, that would help.

Also, are you using the commodity-specific availability factor NCAP_AFC ?

In that case, the capacity represents the nominal maximum output flow (and not nominal maximum amount stored).

Ciao

Wouter

[Arne@IFE]

Hi



The figure shows the energy inflow (first column) and outflow (second column) to and from the battery for the first two weeks of the year.
The numbers are in GWh.
The capacity decided by TIMES was 15.39 GWh/year.
As you can see, there is a violation of the capacity already in W1-day2.

I'm not using the NCAP_AFC factor. Is that required?

Best,

Arne

[Antti-L]

I am sorry to interfere with the discussion, but I am not sure why you say that there is a violation. 

According to the documentation on EQ_CAPACT:

VAR_ACT(r,v,t,p,s) ≤ VAR_NCAP(r,v,t,p)×COEF_CPT(r,v,t,p)×COEF_AF(r,v,t,p,s,’UP’) ×PRC_CAPACT(r,p)×RS_STGPRD(r,s)

Therefore, if you have N days under the parent timeslice (apparently W01), the activity can be at most N times the capacity in each DAYNITE timeslice under W01, assuming NCAP_AF=1 (internally COEF_AF). If your capacity is 15.39 GWh, the activity can thus be at most N×15.39 GWh. As your input flow (84.53) occurs in a single timeslice, you would need to have at least 5.5 days under W01 to fit that amount in the capacity, as 5.5 × 15.39 = 84.6.

So, can you tell us what is the G_YRFR value of the parent timeslice ps, to see what the number of days is under it (N = G_YRFR(ps) ×365)?

[Arne@IFE]

Hi

Maybe I have got this all wrong, but my interpretation of the capacity value of 15.39 GWh/year, was that this is the maximum storage level at any time during the year. Is this assumption wrong?

The G_YRFR values are as follows:
G_YRFR     NO1     W01     -     0.0192
G_YRFR     NO1     W01-DAY1     -     0.0023
G_YRFR     NO1     W01-DAY2     -     0.0034
G_YRFR     NO1     W01-DAY3     -     0.0034
G_YRFR     NO1     W01-NIGHT     -     0.0045
G_YRFR     NO1     W01-WEEKEND     -     0.0056

Best,

Arne

[Antti-L]

Ok, your G_YRFR = 0.0192, and so you have 7 days under W01.

As you say, the capacity is the maximum storage level at any time during the year. You can thus have 15.39 stored on any individual day, and the daily input flow can thus also be at most 15.39 GWh in any timeslice (assuming there are no losses).  Consequently, in 7 days' time the input flow can be at  most 7 × 15.39  = 107.7 GWh, because the storage is discharged on a daily basis. The input flows you see in VEDA-BE for W01 represent the sum of the input flow in 7 days. And the same holds for the activity levels. Thus, I cannot see any violation.  But maybe I am all wrong?

[Arne@IFE]

Ok. Then I read the results from the VEDA-table in the wrong way.

Thank you for your valuable feedback

[Antti-L]

Ok, I am glad if you were able to understand and accept my explanation...

As far as I know, all the DAYNITE flow and activity results in VEDA-BE represent sums over the days under the parent timeslice; not just storage flows.

[wnijs]

Good this is cleared out.

Indeed, it is a bit confusing that ACT can be bigger than CAP (although the CAP is limiting the ACT like usual).

The ratio is determined by the nr of days/week or days/season or weeks/season.

(in a standard situation where COEF_CPT, COEF_AF as well as PRC_CAPACT are "1")

Wouter