We wanted to tackle the disparity between access to opportunities in remote communities that are off the electricity grid. This is because they lack access to information and instant communication in our modern age, and these communities suffer from health effects due to toxic fuel usage and produce negative environmental effects as a result of not having electricity.

What it does

We've created a product that can visually represent strategic points of interest at which we can choose to build renewable energy infrastructures to optimally and efficiently distribute electricity in underserved regions. By choosing a location, we can see these key villages that can serve as the hosts for energy facilities and infrastructure, determine what type of renewable energy resource is most effective given that geographic location, and give various metrics on how many people are being affected by access to electricity.

How I built it

We first scraped several online data sets to create large data sets, including Google Maps, geoplugin, Accuweather, etc. These sets were necessary to create a graph of villages and a function for determining the cost of building electricity at a particular village. We used several algorithms on the graph, including Floyd-Warshall's and an approximate solution to the NP-hard problem of finding a minimum weighted vertex cover. In the end, the program minimizes a certain cost function and also determines the best renewable resource to use at each village where we build an electrical site. Our results were converted to an accessible form through HTML, CSS, and JavaScript.

Challenges I ran into

We ran into problems with finding the relevant data sets for our project. While it was possible to extract some data like temperature and wind speeds, we were unable to find a reliable source of population counts on a per village level. As a result, we used an approximation of each region that included a model that assumed that villages that are both clustered and closer to a city have more population. While this model may provide some inaccuracies, this is meant to demonstrate the proof of concept of our product.

Accomplishments that I'm proud of

We managed to tackle an NP-hard problem and came up with our own algorithm for finding a good solution to it. We applied these theoretical concepts to a real life problem and produced decisive results. We also picked up a variety of skills along the way in order to implement what we had in mind: both on front-end and back-end.

What I learned

We learned how energy resources are distributed in less developed regions and how our product can facilitate the process of bringing them onto the grid. Also, we learned how to incorporate Google's APIs into our programs and learned to manipulate large data sets in a way to visually display them nicely.

What's next for FlashPoint

We want to expand our product by figuring out better ways to model the distribution of energy within our system. While our current product gives great approximations, a lot of factors are assumed due to the lack of information that makes it hard to fully place confidence in developing energy systems in a village. Hopefully, we can work with larger data sets that are more thorough with governments and organizations that want to use our technology to expand access of renewable energy resources. Also, we would hope to make our map more user friendly and give much more detail. .

Share this project: