Inspiration

The inspiration for this project came from what we are learning with our main project, fenestra. We have learned that home efficiency can be greatly improved by opening the windows of your home. Knowing when to open windows is the key factor in order to achieve substantial energy savings. How do we know when? DATA.

What it does

Using data from NIST's Net-Zero Energy Residential Test Facility provided by the US Department of Commerce we use information such as:

  • Outdoor air temperature
  • Outdoor air humidity
  • Indoor air temperature
  • Outdoor air temperature
  • Energy consumption for a high-efficiency heat pump (HVAC)

to decide when to:

  • Open windows and close windows
  • Turn on and off and AC
  • Turn on and off a Whole House Fan
  • When to operate a dryer
  • When to put appliances in energy-saving mode

and then calculate the amount of energy savings we can bring the user all while keeping a comfortable indoor temperature. We did this by parsing through all the Subsystem data files provided by NIST which contain minute by minute timestamped values of more than 50 datapoints. We then parsed through one month of data and created an average reading of all these values which we then uploaded into a web API. This API is then accessed by a mobile app which is connected to a smart home device such as smart window and simulates a day worth of data in ~1 hour. This way we can see, at an accelerated rate, all the decisions which would be made along the day which would bring the energy savings we are looking for.

How I built it

This app is built upon Samsung's SmartThings platform given it is the most commonly adopted IoT smart home automation platform. With all this data we can then take all the decisions in order to improve a home's energy efficiency. The API is built using ruby on rails 5 on heroku and returns json data generated from the Subsystem data files provided by NIST.

Challenges I ran into

Reading through all the data which is provided by the US Department of Commerce and deciding which variables were most important was our first challenge. Next, we also had to make some assumptions regarding air-flow, how the amount of heat in the air varies with temperature and humidity levels, and how would our recommendations would reduce energy consumption. And finally, the technical challenges of creating the API which is accessed by the mobile app which talks to smart home devices. We can think of 100 ways to improve on what we have, mainly automating the generation of data... that's what's next.

Accomplishments that I'm proud of

Once we cracked the code, it was rewarding to experiment and calculate actual energy savings. Using the data from the Department of Commerce, we gained new insights and were able to quantify the impact fenestra can have on home energy consumption.

What I learned

We learned that by using our algorithm with generated data obtained by the Net Zero Energy Residential Test Facility house in Maryland we could achieve 19% savings on the energy used by the AC during the summer season. This is equal to approximately $50 USD a year.

What's next for Home Efficiency Genie

Create a home energy savings calculator which takes the homeowners zip code from the user and downloads datapoints such as dew bulb temperature, dry bulb temperature, and the most likely thermostat settings of this geolocation and defines how much energy can be saved by automating the opening and closing of home windows. With all this info, we'll be able to estimate home energy savings.

Share this project:
×

Updates