Announcement

Collapse
No announcement yet.

Machine Learning’s Impact on Solar Energy

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Machine Learning’s Impact on Solar Energy

    In 2013, solar was the second-largest source of new electricity generating capacity in the U.S., exceeded only by natural gas. A USA SunShot Vision Study suggests solar power could provide as much as 14% of U.S. electricity demand by 2030, and 27% by 2050. Read more here: http://www.rdmag.com/article/2015/07/machine-learning%E2%80%99s-impact-solar-energy

  • #2
    How convenient that solar is 2nd without mentioning any numbers. Like, solar is so far behind you need a telescope to see it. Instead
    just throw around some optimistic predictions for a couple generations in the future. Bruce Roe

    Comment


    • #3
      Originally posted by Reworkdiane View Post
      In 2013, solar was the second-largest source of new electricity generating capacity in the U.S., exceeded only by natural gas. A USA SunShot Vision Study suggests solar power could provide as much as 14% of U.S. electricity demand by 2030, and 27% by 2050. Read more here: http://www.rdmag.com/article/2015/07/machine-learning%E2%80%99s-impact-solar-energy
      Not much else here besides playing with numbers and words in a fashion treehuggers and D.F.'s seem to like to gobble up and use as feed for their ignorance.

      Comment


      • #4
        Originally posted by J.P.M. View Post

        Not much else here besides playing with numbers and words in a fashion treehuggers and D.F.'s seem to like to gobble up and use as feed for their ignorance.
        You are correct. It is nothing but number playing. Even being able to supply 27% of the total US electrical demand by 2050 will still mean many hours and days that it will not be generating any power at all. And even with advancements in technology all the other forms of RE power generation will never fill in the time that solar can't provide power.

        You will still need a power source that is not dependent on the weather and be able to produce 24/7 to fill in the blank or the lights will go out for some.

        Comment


        • #5
          While machine learning and big data are real things, be very careful trusting reports of particular advances... especially those from very large marketing-driven organizations like IBM.
          I don't think the article is all that exciting. Yeah, better forecasting is nice and important. No, it's not a huge deal. That said, the road forward is paved with lots of small deals; better forecasting will happen, and will help a bit.

          Re the drive-by pessimism about solar and nighttime energy: storing energy for nighttime use is expensive and is just starting to be practical in limited areas for limited applications... but some of those are substantial. One PPA for 24 hour power from a pv + thermal solar plant came in at about 12 cents per kwh ( https://en.wikipedia.org/wiki/Copiap..._Solar_Project ).
          So it comes down to economics; if you're willing to pay extra for the storage, you can get nighttime solar. (You'd still need fossil fuel backup for cloudy/nonwindy days, of course.) Whether that is worth it is up to your needs.

          Comment


          • #6
            Originally posted by DanKegel View Post
            While machine learning and big data are real things, be very careful trusting reports of particular advances... especially those from very large marketing-driven organizations like IBM.
            I don't think the article is all that exciting. Yeah, better forecasting is nice and important. No, it's not a huge deal. That said, the road forward is paved with lots of small deals; better forecasting will happen, and will help a bit.

            Re the drive-by pessimism about solar and nighttime energy: storing energy for nighttime use is expensive and is just starting to be practical in limited areas for limited applications... but some of those are substantial. One PPA for 24 hour power from a pv + thermal solar plant came in at about 12 cents per kwh ( https://en.wikipedia.org/wiki/Copiap..._Solar_Project ).
            So it comes down to economics; if you're willing to pay extra for the storage, you can get nighttime solar. (You'd still need fossil fuel backup for cloudy/nonwindy days, of course.) Whether that is worth it is up to your needs.
            Something that might help with energy storage from RE besides batteries is the decision by SolarReserve to build additional plants similar to the Crescent Dunes solar in NV which heats molten salts to generate power at night through steam driven turbines.

            At least that's what this article mentioned.

            Comment


            • #7
              Originally posted by SunEagle View Post
              Something that might help with energy storage from RE besides batteries is the decision by SolarReserve to build additional plants similar to the Crescent Dunes solar in NV which heats molten salts to generate power at night through steam driven turbines.

              At least that's what this article mentioned.
              Yeah, now to see if any of the local power companies want to buy that power. I doubt they'll build without lining up customers first.

              Coincidentally, I'm heading off to watch LADWP's power IRP outreach presentation tonight; they'll be taking public comments on it for a couple weeks, and I'll put in my two bits for "more stored solar + wind, fewer new natural gas power stations". The city council told LADWP to start planning for 100% renewable energy, seems like locking us in to more fossil fuel use for decades with things like the Intermountain natural gas repowering project is the wrong way to start.

              Comment

              Working...
              X