Premium
Non‐linear mixed‐effects models for time series forecasting of smart meter demand
Author(s) -
Roach Cameron,
Hyndman Rob,
Ben Taieb Souhaib
Publication year - 2021
Publication title -
journal of forecasting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.543
H-Index - 59
eISSN - 1099-131X
pISSN - 0277-6693
DOI - 10.1002/for.2750
Subject(s) - smart meter , demand response , computer science , benchmark (surveying) , time series , electricity , random forest , building model , linear model , building automation , mixed model , econometrics , simulation , machine learning , engineering , mathematics , physics , geodesy , electrical engineering , thermodynamics , geography
Abstract Buildings are typically equipped with smart meters to measure electricity demand at regular intervals. Smart meter data for a single building have many uses, such as forecasting and assessing overall building performance. However, when data are available from multiple buildings, there are additional applications that are rarely explored. For instance, we can explore how different building characteristics influence energy demand. If each building is treated as a random effect and building characteristics are handled as fixed effects, a mixed‐effects model can be used to estimate how characteristics affect energy usage. In this paper, we demonstrate that producing 1‐day‐ahead demand predictions for 123 commercial office buildings using mixed models can improve forecasting accuracy. We experiment with random intercept, random intercept and slope and non‐linear mixed models. The predictive performance of the mixed‐effects models are tested against naive, linear and non‐linear benchmark models fitted to each building separately. This research justifies using mixed models to improve forecasting accuracy and to quantify changes in energy consumption under different building configuration scenarios.