Solar power is notorious for being one of the most fluctuating sources of renewable energy. Predicting how much solar energy will be available just for the next couple of hours is hard. However, having the ability to do so has several benefits, the major one being better stability on the power grid – including data centers, as well as every other appliance that relies on it.
A small team of engineers at the University of California in San Diego has now released software that is capable of easily modeling fluctuations in solar radiation caused by weather changes. The software only needs input from one sensor, called a pyranometer, and data from National Oceanic and Atmospheric Administration models to work. In other words, the software will be able to foresee fluctuations at low costs.
The software is based on the solar variability law that was developed by graduate student Matthew Lave at the Jacobs School of Engineering at UC San Diego.
The code is already in high demand for the development in solar power plants – especially those that operate under requirements set by the Puerto Rico Power Electric Power Authority – new utility-scale power plants has to commit to limiting changes in power output to 10 percent per minute.
This is potentially a problem for solar power plants where fluctuations might be significantly higher – a change in output of more than 70% in per second is possible.
Having the ability to predict solar radiation and the output of our solar panels will give us enough time to do something about incoming fluctuations and smoothen out rapid changes. It will be interesting to see how successful the software will be in making solar power less of a fluctuating source.
The capacity of renewable energy on the grid is increasing every single day. It is clear that something needs to be done about stability, as our old base-load energy sources are becoming more and more obsolete.