Short-Term Ambient Temperature Forecasting for Smart Heaters - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2021

Short-Term Ambient Temperature Forecasting for Smart Heaters

Abstract

Maintaining Cloud data centers is a worrying challenge in terms of energy efficiency. This challenge leads to solutions such as deploying Edge nodes that operate inside buildings without massive cooling systems. Edge nodes can act as smart heaters by recycling their consumed energy to heat these buildings. We propose a novel technique to perform temperature forecasting for Edge Computing smart heater environments. Our approach uses time series algorithms to exploit historical air temperature data with smart heaters’ power consumption and heat-sink temperatures to create models to predict short-term ambient temperatures. We implemented our approach on top of Facebook’s Prophet time series forecasting framework, and we used the real-time logs from Qarnot Computing as a usecase of a smart heater Edge platform. Our best trained model yields ambient temperature forecasts with less than 2.66% Mean Absolute Percentage Error showing the feasibility of near realtime forecasting.
Fichier principal
Vignette du fichier
Short_Term_Forecasting_Smart_Heaters_Time_Seris_Temperatures(42).pdf (751.14 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03364728 , version 1 (04-10-2021)

Licence

Identifiers

  • HAL Id : hal-03364728 , version 1

Cite

Danilo Carastan-Santos, Anderson Andrei da Silva, Denis Trystram, Alfredo Goldman, Angan Mitra, et al.. Short-Term Ambient Temperature Forecasting for Smart Heaters. 26th IEEE Symposium on Computers and Communications (ISCC 2021), Sep 2021, Athens, Greece. ⟨hal-03364728⟩
160 View
185 Download

Share

Gmail Mastodon Facebook X LinkedIn More