Currently, Homeowners have no way of knowing whether their strategies for reducing the cost to heat and cool their homes are working. This app enables Homeowners to improve their energy usage by displaying the impact of modifications to their homes or behavior. By integrating a modeled Baseline using outside temperatures into the existing Green Button system, users can see how their Actual Usage compares with the Baseline. This allows the Homeowner to graphically see how their actions change their energy usage.
Some examples of changes a Homeowner might consider include the following:
-opening and closing windows (manually or remotely)
-using fans instead of air conditioning
-sealing leaks in the house, with weather stripping and caulking
-providing overhangs at south-facing windows to eliminate summer solar gain
-insulating roofs (walls and floors)
-adding photovoltaic panels
Because Green Button is currently limited to showing hourly usage, it only allows for comparisons to immediate and regional neighbors. However, due to the huge number of variables (shade, trees, orientation, etc.) the most useful comparison is with one’s own house, which is the focus of this app. In addition, the app provides the Homeowner with the ability to define the unit of time, e.g. hour, day, week, month, offering an essential interactive feature.
Green Button is an important tool in this direction, because it provides Homeowners with hourly information. However, to provide an accurate assessment, one needs to have a baseline model that incorporates temperature, time of day and season. In order to demonstrate how this can work, we used real data available in the Green Button platform.
This app can easily be expanded in the near and more distant future in order to allow further comparisons between two specific houses. (This approach can easily be replicated for other climates. With enough data, an average can be generated, which is much more useful.)
The app has much to offer now and going forward. For this particular residence, located in San Diego, the correlation of 0.86 between the actual and modeled hourly usage shows how well approximately 50 parameters model the 8760 observations. Now the Homeowner can look at what happened during those clear instances when there’s deviation to see the impact of their changes.
Going forward, this approach can easily make use of near-future and upcoming technologies which can be incorporated into this platform.