Analysis of Accuracy of Flexibility Baseline Prediction Methods for Office Building at Different Measuring Points

Challenge and objective

  • Customer baseline load (CBL) prediction plays an important role in calculating the volume and value of the flexibility provided by end-users.​
  • In this paper, two different CBL methods are applied to investigate their prediction accuracy for a given load with high resolution metered data.

Work performed

  • Analyzed input data at office-building level and at EV parking-lot level, compared their accuracy​.
  • Implemented existing CBL-calculation methods, both before/after measurements, and historical data points

Significant results

  • Building-level measurement point performed best with use of load patterns before/after activation, due to a fixed 200-kW load import level​.
  • EV parking lot had much variation due to being an internal demand response for the 200-kW limit.​
  • Historical data was influenced by recent seasonal changed in load patterns

Impact for distribution system innovation

  • The methods can be applied to predict the baseline load from a flexible user during flexibility activation​.
  • Important to be aware of internal flexible measures that could affect historical data accuracy.


Magnus Korpås

WP3 Lead
+47 970 42 009
Magnus Korpås
WP3 Lead


Reference in CINELDI