MIT creates super accurate solar potential map of Cambridge
As solar power has become cheaper and more prevalent, there have been no shortage of solar potential maps for showing how much power could be generated in specific spots or throughout a city with the installation of solar arrays. Cities like San Francisco got mapped early on and there's even an app that lets anyone get data on the solar potential of their roof. But most of these maps so far have relied on Google maps data and pretty simple calculations.
That's where the smarties in MIT have come in to improve upon the solar mapping technology. A team of researchers there with the Mapdwell project has created a program called Solar System that could be the most accurate and impressive solar mapping tool to date. Not only is the system in 3D, but it factors in more variables that could affect solar energy output like roof angles and sky conditions for each hour of the year. In testing, the predictions have fallen within four to 10 percent of real-world results.
To show the Solar System tool off, the team created a full solar map of all 17,000 rooftops in Cambridge, Massachusetts, with the data accessible to anyone through a user-friendly website. The map shows that if photovoltaic panels were installed on all of the locations classified as either "good" or "excellent" on the map, the city of Cambridge could generate about one third of its energy needs from solar for a cost of about $2.8 billion.
Gizmag reports that "The Solar System takes data on photovoltaic panel potential and overlays it on top of Google satellite imagery. Selecting a building within the map will bring up a host of information on installing PV panels on that specific rooftop, including the system size in kilowatts, the projected cost to the owner (including deductions and local incentives), the carbon offset, and the projected yearly revenue generated by the array.
Users can also design their own system by either adjusting its size with a slider (in which case the software returns the best arrangement of the panels for the desired array size) or by drawing an area on the map (in which case the software will return information on the projected efficiency of the system)."
To data used to compile the map comes from a combination of Google maps satellite imagery with LiDAR (Light Detection and Ranging) data from a prior aerial survey that allowed a 3D model of the city. The 3D model let them see things like roof shapes and angles, trees and other possible obstructions to more accurately predict energy output. They then combined that model with weather information from local stations and ran it through a computer simulation package that calculated the potential solar energy output across a year.
The predictions could be on the conservative side since it's assumed that all panels are fixed parallel to the roof instead of at an angle to better capture sunlight, but the team plans to change that in the next iteration.
Check out the video below to see the map in action.