Building Electrical Load Forecasting through Neural Network Models with Exogenous Inputs

Cristina Nichiforov, Grigore Stamatescu, Iulia Stamatescu, Ioana Fagarasan, Sergiu Stelian Iliescu

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandForschungBegutachtung

Abstract

As buildings become key actors in the economic and sustainable operation of future electrical grids and smart cities, reliable models which capture the underlying electrical energy consumption become an important factor for robust control algorithms. Current ubiquitous field devices supported by complex data infrastructures allow generation, storage and online analysis of large quantities of data for deriving usable black-box models of building energy patterns. The paper presents an approach to model the energy consumption of medium and large sized buildings using Non-linear Autoregressive Neural Networks with eXogenous Input (NARX). We show that the chosen network architectures offers good performance for time series prediction from historical values and external input signals such as outdoor temperature in comparison to a baseline approach. Model evaluation and validation are carried out on public dataset for replicable research outcomes.
Originalspracheenglisch
Titel2019 23nd International Conference on System Theory, Control and Computing (ICSTCC)
Herausgeber (Verlag)Institute of Electrical and Electronics Engineers
Seiten474-479
Seitenumfang6
ISBN (elektronisch)978-1-7281-0699-1
DOIs
PublikationsstatusVeröffentlicht - 2019
VeranstaltungICSTCC 2019: 23rd International Conference on System Theory, Control and Computing, - Sinaia, Rumänien
Dauer: 9 Okt 201911 Okt 2019

Konferenz

KonferenzICSTCC 2019
LandRumänien
OrtSinaia
Zeitraum9/10/1911/10/19

Fingerprint

Neural networks
Energy utilization
Robust control
Network architecture
Time series
Economics
Temperature
Smart city

Dies zitieren

Nichiforov, C., Stamatescu, G., Stamatescu, I., Fagarasan, I., & Iliescu, S. S. (2019). Building Electrical Load Forecasting through Neural Network Models with Exogenous Inputs. in 2019 23nd International Conference on System Theory, Control and Computing (ICSTCC) (S. 474-479). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICSTCC.2019.8885964

Building Electrical Load Forecasting through Neural Network Models with Exogenous Inputs. / Nichiforov, Cristina; Stamatescu, Grigore; Stamatescu, Iulia; Fagarasan, Ioana; Iliescu, Sergiu Stelian.

2019 23nd International Conference on System Theory, Control and Computing (ICSTCC). Institute of Electrical and Electronics Engineers, 2019. S. 474-479.

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandForschungBegutachtung

Nichiforov, C, Stamatescu, G, Stamatescu, I, Fagarasan, I & Iliescu, SS 2019, Building Electrical Load Forecasting through Neural Network Models with Exogenous Inputs. in 2019 23nd International Conference on System Theory, Control and Computing (ICSTCC). Institute of Electrical and Electronics Engineers, S. 474-479, Sinaia, Rumänien, 9/10/19. https://doi.org/10.1109/ICSTCC.2019.8885964
Nichiforov C, Stamatescu G, Stamatescu I, Fagarasan I, Iliescu SS. Building Electrical Load Forecasting through Neural Network Models with Exogenous Inputs. in 2019 23nd International Conference on System Theory, Control and Computing (ICSTCC). Institute of Electrical and Electronics Engineers. 2019. S. 474-479 https://doi.org/10.1109/ICSTCC.2019.8885964
Nichiforov, Cristina ; Stamatescu, Grigore ; Stamatescu, Iulia ; Fagarasan, Ioana ; Iliescu, Sergiu Stelian. / Building Electrical Load Forecasting through Neural Network Models with Exogenous Inputs. 2019 23nd International Conference on System Theory, Control and Computing (ICSTCC). Institute of Electrical and Electronics Engineers, 2019. S. 474-479
@inproceedings{8d06216e0fec4eb08453f46903e1b04b,
title = "Building Electrical Load Forecasting through Neural Network Models with Exogenous Inputs",
abstract = "As buildings become key actors in the economic and sustainable operation of future electrical grids and smart cities, reliable models which capture the underlying electrical energy consumption become an important factor for robust control algorithms. Current ubiquitous field devices supported by complex data infrastructures allow generation, storage and online analysis of large quantities of data for deriving usable black-box models of building energy patterns. The paper presents an approach to model the energy consumption of medium and large sized buildings using Non-linear Autoregressive Neural Networks with eXogenous Input (NARX). We show that the chosen network architectures offers good performance for time series prediction from historical values and external input signals such as outdoor temperature in comparison to a baseline approach. Model evaluation and validation are carried out on public dataset for replicable research outcomes.",
author = "Cristina Nichiforov and Grigore Stamatescu and Iulia Stamatescu and Ioana Fagarasan and Iliescu, {Sergiu Stelian}",
year = "2019",
doi = "10.1109/ICSTCC.2019.8885964",
language = "English",
pages = "474--479",
booktitle = "2019 23nd International Conference on System Theory, Control and Computing (ICSTCC)",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

TY - GEN

T1 - Building Electrical Load Forecasting through Neural Network Models with Exogenous Inputs

AU - Nichiforov, Cristina

AU - Stamatescu, Grigore

AU - Stamatescu, Iulia

AU - Fagarasan, Ioana

AU - Iliescu, Sergiu Stelian

PY - 2019

Y1 - 2019

N2 - As buildings become key actors in the economic and sustainable operation of future electrical grids and smart cities, reliable models which capture the underlying electrical energy consumption become an important factor for robust control algorithms. Current ubiquitous field devices supported by complex data infrastructures allow generation, storage and online analysis of large quantities of data for deriving usable black-box models of building energy patterns. The paper presents an approach to model the energy consumption of medium and large sized buildings using Non-linear Autoregressive Neural Networks with eXogenous Input (NARX). We show that the chosen network architectures offers good performance for time series prediction from historical values and external input signals such as outdoor temperature in comparison to a baseline approach. Model evaluation and validation are carried out on public dataset for replicable research outcomes.

AB - As buildings become key actors in the economic and sustainable operation of future electrical grids and smart cities, reliable models which capture the underlying electrical energy consumption become an important factor for robust control algorithms. Current ubiquitous field devices supported by complex data infrastructures allow generation, storage and online analysis of large quantities of data for deriving usable black-box models of building energy patterns. The paper presents an approach to model the energy consumption of medium and large sized buildings using Non-linear Autoregressive Neural Networks with eXogenous Input (NARX). We show that the chosen network architectures offers good performance for time series prediction from historical values and external input signals such as outdoor temperature in comparison to a baseline approach. Model evaluation and validation are carried out on public dataset for replicable research outcomes.

U2 - 10.1109/ICSTCC.2019.8885964

DO - 10.1109/ICSTCC.2019.8885964

M3 - Conference contribution

SP - 474

EP - 479

BT - 2019 23nd International Conference on System Theory, Control and Computing (ICSTCC)

PB - Institute of Electrical and Electronics Engineers

ER -