Almost a quarter of India's population—about 300 million people—are still without proper electricity. In the countryside, there are still swathes of villages with no grid connection. What power does exist is from private and uncoordinated sources, like solar panels, batteries, and diesel generators.
Microgrids are a highly promising idea for powering remote communities because they don't require a big government investment to hoist transmission and distribution lines and don't suffer the efficiency losses inherent in transporting power over long distances. There is a lot of interest in the concept of "islanding" the power grid into manageable blocks, which would allow villages to coordinate their electricity production and consumption independently and collectively.
Gridform, a project of four MIT PhD students, is a planning system to help establish microgrids more cheaply and professionally. It's sophisticated software that maps villages and identifies the best ways to link up houses and energy sources.
"The existing methods for electrifying these areas are off the table for hundreds of millions of citizens," says MIT's Brian Spatocco. He says Gridform's software helps people installing grids make smarter decisions. First, it identifies which villages are most suitable for the technology, and, second, the best way to actually wire up that village.
"One of the major challenges is the siting and the planning of these projects," says Spatocco. "People, for better and worse, will have good intentions and go into rural India or places in Africa, and they'll have a technology and they'll drop it on the ground. There isn't much diligence in planning because they don't know what it looks like on the ground."
The software takes satellite photos and identifies a village's buildings from the image. It then runs a "load analysis" to see how much electricity demand there is across the village and runs simulations showing optimal microgrid configuration for both cost and reliability. For example, one algorithm works out the best number of energy sources.
The MIT team believes its software, which it plans to distribute for free under license, will help save money and lead to more robust infrastructure development. "It tells the local entrepreneur where the most bang for the buck is because [they] don't have a lot of money to throw around," Spatocco says.
Spatacco and his fellow students, George Chen, Ling Xu, and Kendall Nowocin, went on a fact-finding trip to India this summer and are now planning a full pilot with a local partner, called SELCO, which has 30 years of experience in rural electrification.
"What separates us from other people is that we can do this over huge regions," Spatocco says. "We can run this code on a satellite image and do hundreds or thousands villages."