Big building data visualization: a visualization tool for the ”Smart Living Lab”

Teachers: 
Denis Lalanne
Student: 
Johan Jobin
Project status: 
Finished
Year: 
2018

Collecting huge amount of data of building has gained importance in the last years. Indeed, with the growth of the Internet of Things, getting data using sensors has become always simpler and cheaper. All these raw data are usually stored in big databases, ready to be used for all kind of applications.

That’s exactly what was done in the ”Smart Living Lab”, a smart building in Fribourg, in which almost 3000 sensors were installed to measure temperature, carbon dioxide, power, energy, electric potential, etc. All these data are stored in a database which is accessible through a RESTful service that can be queried to get measures of specified sensors for a given interval of time.

Hence, this work consists in the creation of a web dashboard that brings a visual meaning to the data through various visualizations, allowing the user of the system to have a better understanding of the building. Six different visualizations with different objectives are imple- mented: a line chart that illustrates the variation of the measures, a scatter plot dedicated to binary values like lights, a box plot to identify the quartiles of the measures, a radial plot to highlight the periodicity of one week, a calendar heat map to compare the average value of each day during one year and finally a histogram to see the distribution of the data. All these visualizations are generated on-the-fly using a big table that is a visualization itself and that acts like a menu to choose which sensors to consider in order to display the visualizations. In addition to these visualizations a search engine is also available to explore the sensors and get information about them.

With the created dashboard it is possible to notice that approximately 15% of all the sensors of the building are activated. Furthermore, multiple interesting observations can be made concerning the measure of the sensors: it is possible to see that sometimes anomalies are reg- istered, which sensors of temperature are indoor and which are not, when it rained and when not, the differences between season, etc.