Adaptive Learning Based Building Load Prediction for Microgrid Economic Dispatch

Rumia Masburah1,a, Rajib Lochan Jana1,b, Ainuddin Khan1,c, Shichao Xu2,a, Shuyue Lan2,b, Soumyajit Dey1,d and Qi Zhu2,c
11Indian Institute of Technology Kharagpur
arumiamasburah@iitkgp.ac.in
brajib.jana@iitkgp.ac.in
cainuddin@iitkgp.ac.in
dsoumya@cse.iitkgp.ac.in
2Northwestern University
ashichaoxu2023@u.northwestern.edu
bslan@u.northwestern.edu
cqzhu@northwestern.edu

ABSTRACT


Given that building loads consume roughly 40% of the energy produced in developed countries, smart buildings with local renewable resources offer a viable alternative towards achieving a greener future. Building temperature control strategies typically employ detailed physical models which require a significant amount of time, information and finesse. Even then, due to unknown building parameters and related inaccuracies, future power demands by the building loads are difficult to estimate. This creates unique challenges in the domain of microgrid economic power dispatch for satisfying building power demands through efficient control and scheduling of renewable and non-renewable local resources in conjunction with supply from the main grid. In this work, we estimate the real-time uncertainties in building loads using Gaussian Process (GP) learning and establish the effectiveness of run time model correction in the context of microgrid economic dispatch.

Keywords: GAussian Process Learning, Deep Reinforcement Learning, Predictive Control, Economic Dispatch, Building Thermal Model.



Full Text (PDF)