Part III — Data Visualizations#
In the following chapters, visualizations are presented that are frequently used in the field of building energy monitoring. Sometimes not the visualization itself is challenging, but the preceding data preparation. Therefore, all recipes show the complete workflow from raw data to final plot.
Plotly and pyedautils#
All visualizations in this book use Plotly for interactive plots — every chart is zoomable, hoverable, and exportable. Many recipes use pyedautils, a Python package that provides ready-to-use visualization functions for energy data analysis.
Recipe Pattern#
Each visualization notebook follows a consistent structure:
Goal
What the visualization shows and when to use it.
Data Basis
Which data is needed — format, columns, and resolution.
Data Preparation — Load and transform the raw data.
Visualization — Create the plot using pyedautils or manual Plotly code.
Customization
Optional parameters for colors, labels, and layout.
Categories#
The visualizations are organized into the following categories:
Statistical — Boxplots, histograms, density by season, correlation, outliers, autocorrelation, cross-correlation
Decomposition — Long-term and short-term time series decomposition
Seasonal — Overlapping years, miniplots, polar charts, before/after comparisons
Heat Maps — Calendar heat maps and median week patterns
Daily Profiles — Overview, overlaid, mean, and decomposed profiles
Sum & Frequency — Daily and hourly duration curves
Comfort — Temperature and humidity over time, temperature vs. humidity, Mollier h-x, comfort donuts, comfort compass, SIA 180
Miscellaneous — Energy signatures, household electricity, multi-axis plots