Using Neural Networks To Better Forecast Renewables

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

Much has been made of the needs of the electricity grid as more and more renewables with their inherent fluctuations are plugged into the grid. No longer can a certain amount of energy be guaranteed, and with burgeoning populations and energy use the world over, it’s becoming increasingly difficult to predict how much energy is going to be needed as well.

Enter SENN — the Simulation Environment for Neural Networks — forecasting software that is currently used to forecast raw material prices, and the price of electricity over 20 day periods, and has now been turned to forecasting the amount of electricity that will be fed into the grid by renewables such as solar and wind.

The use of SENN in the field of renewables is not breaking news, however Siemens has afforded a lot of column space to the idea of better forecasting for renewables and neural networks in its latest ‘Pictures of the Future‘ magazine that is published twice a year. Pictures of the Future looks at the latest research in Siemens laboratories and investigates the major technology trends currently shaping the world.

Siemens has been using SENN modelling to determine the best time for a company to purchase electricity or key raw materials, a topic discussed in its Pictures of the Future magazine in the fall of 2011.

Chip in a few dollars a month to help support independent cleantech coverage that helps to accelerate the cleantech revolution!

Neural networks behave differently than analytical models, which require a massive amount of hard-to-come-by data. Neural networks, on the other hand, “don’t have to fully analyze and understand a problem in order to make a forecast,” says Ralph Grothmann, a researcher at Siemens Corporate Technology.

131025_sag_pof_infographics_micro_spring2013_Part01_01

Neural networks are trained using past data, and the more past data they incorporate they better their predictions become. Initial forecasts will often vary wildly from the actual output of a solar farm, for example. However, as it continues to learn and repeat its forecasting process over and over, SENN changes the weighting of individual parameters and provides increasingly more accurate predictions.

The neural networks magazine article can be read in full here, and it’s worth a read to understand the many and varied ways in which neural networks are helping move the integration of renewables into the energy grid forward.


Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.

Latest CleanTechnica TV Video


Advertisement
 
CleanTechnica uses affiliate links. See our policy here.

Joshua S Hill

I'm a Christian, a nerd, a geek, and I believe that we're pretty quickly directing planet-Earth into hell in a handbasket! I also write for Fantasy Book Review (.co.uk), and can be found writing articles for a variety of other sites. Check me out at about.me for more.

Joshua S Hill has 4403 posts and counting. See all posts by Joshua S Hill