•   over 9 years ago

another question on PGE samples on DOE website

I am working on to do something with residential energy data samples provided by PGE. I understand these xml file are load profile (or sample files) that represent typical usage in a residential building. For my analysis, I have to compare energy usage with temperature data of that region.

Here is my question:
I would like to know what location temperature would make sense for this sample data. What region would make more sense? Fresno, Los Angeles, Sacramento, San Diego, San Francisco?

  • 8 comments

  • Manager   •   over 9 years ago

    The PG&E samples represent residential or commercial buildings, as indicated by the file names. I'll also point out that actual PG&E data would have the building address, not the region, in the data. To get from address to region, PG&E provides a zip to territory lookup table here: http://www.pge.com/tariffs/RESZIPS.XLS

    For your actual question, PG&E covers a large swath of northern CA, so San Francisco or Sacramento make the most sense. If you want to be more precise, you can use the building location to get temperature data from NOAA. http://appsforenergy.challenge.gov/forum_topics/942 has more details on how to do that.
    Matthew

  •   •   over 9 years ago

    I understood how to get temperature data from NOAA. But, unfortunately the sample PGE data on DOE website doesn't have any address on it to guess its region.

    If you by any chance can get the region information for the sample data provided, that would help us to be accurate in our estimation of building energy use.

  • Manager   •   over 9 years ago

    Ah, I understand the issue now. Since the actual PG&E data will include an address, would it work to just insert an arbitrary address for testing purposes? If not, let me know and I'll get some PG&E data with fake (but regionally appropriate zips, cities, and streets) addresses.
    Thanks,
    Matthew

  •   •   over 9 years ago

    I think we need to get this data from PG&E. It could be a fake address, but it should be representative of the region for which the sample xml data has been given.

    For example, if the sample data represents building energy use in Sacramento, I would go ahead and use the precise Sacramento temperature data. But, if it is Bakersfield or something, I have to use Bakersfield temperature data to better understand the energy use of the house.

    Without that I could end up correlating a house in Sacramento with Bakersfield temperature data, which doesn't make sense.

    - Mithun

  •   •   over 9 years ago

    Also, I am little skeptic about the date time stamps corresponding to kWh usage values provided on the PGE data. Because looking at E6 Average Usage data provided, it always look it for a residential house the average energy usage peaks up from 11:00 p.m to 3:00 a.m compared to any other time of the day, which doesn't make sense.

    I don't know if this could be real data or made up data. In case if this is made up data, it would help us only if the values represent the general trend. Please correct me if my understand is incorrect. Thanks!

  • Manager   •   over 9 years ago

    I've requested address data from PG&E, hopefully we'll get it today.

    I believe that the sample data provided is from actual customers, but I'm verifying that. One reason why usage might peak at night in this sample is that E6 is PG&E's time of use rate, so this customer might be trying to actively limit on peak usage.

    Matthew

  •   •   over 9 years ago

    It would be great if we can get representative addresses for different XML files from SDGE data too. It is all to pull up the right temperature data and do a comparative study of kWh usage with temperature data. Thanks!
    http://sdge.com/customer-service/billing-information/sample-regional-customer-data

  • Manager   •   over 9 years ago

    I just got the zip codes for the PG&E samples, and I've added them to the sample files. You can find them at http://energy.gov/downloads/green-button-sample-data-pge. As you might have guessed, they are the files that end with " zipcode".
    Matthew

Comments are closed.