Skip to content

lucavehbiu/Time_series_forecast

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Optimizing energy consumption - 'Smart-home' perspective

With the rise of IoT, the precision and the real-time feature of the data that is retrieved, in this case submeters can really make a dent in the energy usage behavior of a household.

Machine learning algorithms for time series (Fourier, Holt-Winters and Arima) were i.mplemented and the results which have become interactive through a .Shiny application teach users how to consume less and more efficiently (less energy losses).

Predictions of future energy consumption and usage can be used from the home-builders to:

  • Better design their houses (thermo-efficient)
  • More knowledge to the household as to which appliance spends the most for e.g.

Moreover, such predictions can be sent to the energy suppliers which in turn can predict their energy demand more accurately. By doing so, they can make up to households for the data gifted to them, by providing accurate prices for the next 48 hours. Such symbiotic relationship between home-builder, energy supplier and household is what will happen frequently in the future, which we more commonly know as Smart-grid Cities.

Iot - Real benefits get visible as scale increases!

File explanation:

  • model.R -> code for modeling the time series and forecasting
  • viz.R -> code for visualizations of the dataset
  • app.R -> Shiny app for making the results interactive

About

Ubiqum Project 3 - Energy consumption

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages