Battery Case Study Questions

1 post / 0 new
Paul Gilman
Battery Case Study Questions

The following is from an email conversation about the following paper:

DiOrio, N.; Dobos, A.; Janzou, S. (2015). Economic Analysis Case Studies of Battery Energy Storage with SAM. 22 pp. NREL/TP-6A20-64987 (PDF 550 KB)

I am trying to reach the same results as your validation technical report "Economic Analysis Case Studies of Battery Energy Storage with SAM". 

For the electric load data, it is stated that it was taken from EnerNOC's free online database. As I checked your prepared reference for that

I am unable to find the electric load which is the same as yours for January 1st for Los-Angeles and Knoxville. I will really appreciate it if you can help me to have the load data. 

The data is:

  • Los-Angeles: EnerNoc site 384
  • Knoxville: EnerNoc site 101

I tried to solve the weather file issues with that work but I think the weather file that MIDC/Solarmap creates is for 1 minute and it is only for some parts of a day. For instance, from 7:00 AM to 17:00 PM. In the other side, the electric load data is for every 5 minutes. I think they are not consistent with each other, and I got some errors regarding the consistency of data from SAM. Is it the same weather file that you used for your simulation?

The data format that I created was a “SAM csv”.  The MIDC can create data down to 1 minute, but you should be able to get it for the full year at all times.  I believe I used the year 2012.  I’ve attached for your convenience in the 5-minute format.  1 minute is possible, but probably a bit overkill and will be very slow to run.  You could even up sample these to 15 minute data, and that would still be good.

1) PV meets load before charging or PV charges the battery before meeting load?

2) What is the percentage amount when we choose to charge from the grid?

3) What is the min time at charge state? Is it the same as our time steps (5 mins)?

4) What is the capacity of Tesla Powerwall (7 kWh or 10 kWh because of min state of charge)? I mean, in the desired bank capacity section, should I insert 7 or 10?

5) In Table 3 of your report, the voltage is mentioned 350 - 450 V. However, I should insert that as a solid number in the desired bank voltage section? what number did you use there?

6) What about Max C-rate of charge and discharge? I entered them 0.3 per/hour.

7) Where should we insert the base demand charge (mentioned in Table 6 of your work)?

1.) This refers to the ability to tell the controller how to prioritize the PV power production. You can either have the PV power meet the electric load first, or have it charge the battery first. Typically you will have it meet the electric load first, but the other option is available for specific use cases.

2.) The percentage amount refers to the battery capacity. For instance, if you set to 25% for one hour, it will only allow the battery to charge up to 25% of its capacity from the grid in that hour. This is a way to restrict how much grid power is used.

3.) This is a way to stop the battery from rapidly cycling. For instance, if you are running a simulation at 5 minutes, you may not want your battery to discharge one step, and then charge the next. Rather, you might want the battery to charge for one hour and discharge for one hour. The field allows you to restrict rapid cycling behavior, which can be damaging to a battery.

4.) This depends. You have two options, 1.) model the capacity as advertised by Tesla assuming it is the full cyclable capacity of the battery. For instance, if Tesla advertises 7 kWh, then you could model 7 kWh, being cycled from 0 – 100% capacity. 2.) Model the realistic capacity, assuming that Tesla has reserved 20-30% of the capacity. In this case, if Tesla has advertised a 7 kWh battery, they’ve probably reserved 3 kWh or so, so that the battery doesn’t cycle too deeply. In that case, you would enter 10, and then have the battery SOC limits from 30% - 100%.

5.) In the desired bank voltage section, you should enter the nominal voltage of the battery. I would say enter 400V, which is halfway between the diminished (350V) and full (450V) voltages of the battery.

6.) That sounds about right. You want to tailor the C-rate to achieve the power production advertised. In the case of Tesla, I think it was a bit over 3 kW for one power wall, which corresponds to a C-rate of about 0.3 for a 10 kWh battery.

7.) The base demand charge is entered here:

Screenshot of demand charge inputs


Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer