DHW forecasts

In this vignette we build forecast models for forecasting of Domestic Hot Water (DHW) in different settings.

We will do it in two settings: first forecast the hourly average DHW flow to a multi-family building in Copenhagen, second, forecast the hourly heat load for DHW in single family buildings from Sønderborg.

DHW flow to a multi-family building

The DHW flow was recorded at the supply to a multi-family building in Copenhagen. The data has been resampled to hourly data. It originates from the project “EnergyLab Nordhavn-New Urban Energy Infrastructures and Smart Components” project grant by the Danish Energy Technology Development and Demonstration Program (No. 64015-0055).

Explore and prepare data

First, load the package and data:

# For storing data and cache files
cachedir <- tempdir()
# Load the package and data
library(onlineforecast)
path <- paste0(cachedir,"/DdhwMultiFamily.RDS")
download.file("https://onlineforecasting.org/examples/data/DdhwMultiFamily.RDS", path)
D <- as.data.list(readRDS(path))
# See the variables
str(D)
##     List of 2
##      $ t  : POSIXct[1:2159], format: "2019-01-01 02:00:00" "2019-01-01 03:00:00" ...
##      $ dhw: num [1:2159] 412.29 537.45 23.87 227.33 7.99 ...
##      - attr(*, "class")= chr "data.list"

Plot of all data:

plot_ts(D)