talk on conference website
We are working to map all the solar panels (photovoltaic, "PV") in the world. Why? The data can be used directly to reduce carbon emissions from power generation. We will share our experiences of surveying, aerial mapping and machine vision to find all the hundreds of thousands of solar panels in our countries.
Together with a small group called OpenClimateFix, we are working to map all the solar panels (photovoltaic, "PV") in the world. Why? Because if we combine this with short-term forecasting of cloud and sunshine, we can directly predict the solar power electricity generation ahead of time. This means we don't need to burn as much coal or gas as backup. So, it can be valuable to know the exact location and characteristics of each solar PV installation - large solar farms, and small domestic installations.
In this talk we'll discuss how this all fits together. We'll talk about solar power tagging/mapping in OSM, for both large and small, to make it easy for people to map but also useful for the power network analysis. We'll also share our explorations of crowdsourcing and automation machine vision, and the use of other data sources such as government open data to guide the mapping process.
We have already mapped a significant portion of the UK's solar capacity, and trialled some crowdsourcing and machine vision tools. We will show visualisations and analyses of the work that's been done so far, and consider how to scale this worldwide.