We are currently pursuing two projects at OCF, both are focused on Solar Photovoltaics (PV).

Our first project, Solar PV Nowcasting, investigates the possibilities of using Machine Learning to forecast the output of solar photovoltaic installations for the next few hours.

Solar PV power production is a hard to forecast: As clouds move over PV panels, the power output moves up and down rapidly. To keep the energy grid in balance, electricity grid operators need to have a readily available backup which usually comes from fossil sources.

Our assumption is the following: If we have more accurate predictions of how much electricity a PV installation will produce over the next few hours then we can reduce the amount of fossil fuel backup required.

PV Mapping is our second project. We would like to support the OpenStreetMap community to map the location of the world's PV panels in OpenStreetMap. OpenSteetMap is like the Wikipedia of maps: anyone can edit the database. We plan to use a combination of machine learning and the wisdom of the crowd to locate PV panels.

To be able to arrive at an accurate forecast for solar power generation, we need to know where all the PV panels are. You can support us in mapping the geographical location of PV Systems by using the tag:generator:source=solar.