Paper Image

Managing building energy use with data-driven flexibility predictions

Published on:

9 November 2023

Primary Category:

Systems and Control

Paper Authors:

Paul Scharnhorst,

Baptiste Schubnel,

Rafael E. Carrillo,

Pierre-Jean Alet,

Colin N. Jones


Key Details

Characterizes building flexibility with data-driven, uncertainty-aware 'flexibility envelopes'

Formulates optimization problems to schedule buildings and dispatch requests

In simulation coordinates up to 500 buildings for self-consumption or peak reduction

Achieves good performance while maintaining temperature comfort constraints

Demonstrates computational scalability to thousands of buildings

AI generated summary

Managing building energy use with data-driven flexibility predictions

This paper proposes a method to coordinate the flexible energy consumption of buildings for grid services like self-consumption or peak reduction. It characterizes building flexibility through data-driven 'flexibility envelopes' that predict the possible increase or decrease in consumption over time. These envelopes incorporate uncertainty estimates to be risk-aware. The paper then formulates optimization problems to schedule building activation times and dispatch flexibility requests between active buildings. Through simulation of up to 500 buildings, it shows the approach can substantially increase self-consumption or reduce peak loads while maintaining comfort.

Answers from this paper


No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up