Hybrid model for short-term forecasting of loads and load control responses

Pekka Koponen, Harri Niska

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

2 Citations (Scopus)

Abstract

A hybrid model for short-term forecasting of aggregated thermal loads and their load control responses is studied in this paper using field test data. Inputs include temperature measurement and forecast, measured power and control signals. The hybrid model comprises 1) partly physically based forecasting of the responses of the controlled thermal loads and the non-controlled power, and 2) forecasting the residual using Support Vector Machine (SVM). Their summation gives the hourly interval power forecast. Here partly physically based means that the model structure models thermal dynamics of houses. The response forecasting needs as inputs the daily heating energy demand and controllable power range, which are forecast by partly physically based models. Weekly load pattern is used to forecast the non-controllable power. SVM forecasts the residual. Alternative configurations are compared. The response forecasting model improved the forecasting accuracy most and adding the SVM for the residual further improved the accuracy.
Original languageEnglish
Title of host publicationPES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), 2016 IEEE
PublisherInstitute of Electrical and Electronic Engineers IEEE
Pages1-6
ISBN (Electronic)978-1-5090-3358-4, 978-1-5090-3357-7
ISBN (Print)978-1-5090-3359-1
DOIs
Publication statusPublished - 2 Jul 2016
MoE publication typeA4 Article in a conference publication
EventIEEE PES Innovative Smart Grid Technologies, Europe - Ljubljana, Slovenia
Duration: 9 Oct 201612 Oct 2016

Conference

ConferenceIEEE PES Innovative Smart Grid Technologies, Europe
Abbreviated titleISGT 2016
CountrySlovenia
CityLjubljana
Period9/10/1612/10/16

Fingerprint

Support vector machines
Thermal load
Model structures
Temperature measurement
Heating
Hot Temperature

Keywords

  • demand response
  • forecasting
  • hybrid models
  • machine learning
  • physical models

Cite this

Koponen, P., & Niska, H. (2016). Hybrid model for short-term forecasting of loads and load control responses. In PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), 2016 IEEE (pp. 1-6). [7856320] Institute of Electrical and Electronic Engineers IEEE. https://doi.org/10.1109/ISGTEurope.2016.7856320
Koponen, Pekka ; Niska, Harri. / Hybrid model for short-term forecasting of loads and load control responses. PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), 2016 IEEE . Institute of Electrical and Electronic Engineers IEEE, 2016. pp. 1-6
@inproceedings{7eae71c11c0e4fa08791a51721157f4d,
title = "Hybrid model for short-term forecasting of loads and load control responses",
abstract = "A hybrid model for short-term forecasting of aggregated thermal loads and their load control responses is studied in this paper using field test data. Inputs include temperature measurement and forecast, measured power and control signals. The hybrid model comprises 1) partly physically based forecasting of the responses of the controlled thermal loads and the non-controlled power, and 2) forecasting the residual using Support Vector Machine (SVM). Their summation gives the hourly interval power forecast. Here partly physically based means that the model structure models thermal dynamics of houses. The response forecasting needs as inputs the daily heating energy demand and controllable power range, which are forecast by partly physically based models. Weekly load pattern is used to forecast the non-controllable power. SVM forecasts the residual. Alternative configurations are compared. The response forecasting model improved the forecasting accuracy most and adding the SVM for the residual further improved the accuracy.",
keywords = "demand response, forecasting, hybrid models, machine learning, physical models",
author = "Pekka Koponen and Harri Niska",
note = "SDA: MIP: Intelligent Energy Grids",
year = "2016",
month = "7",
day = "2",
doi = "10.1109/ISGTEurope.2016.7856320",
language = "English",
isbn = "978-1-5090-3359-1",
pages = "1--6",
booktitle = "PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), 2016 IEEE",
publisher = "Institute of Electrical and Electronic Engineers IEEE",
address = "United States",

}

Koponen, P & Niska, H 2016, Hybrid model for short-term forecasting of loads and load control responses. in PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), 2016 IEEE ., 7856320, Institute of Electrical and Electronic Engineers IEEE, pp. 1-6, IEEE PES Innovative Smart Grid Technologies, Europe, Ljubljana, Slovenia, 9/10/16. https://doi.org/10.1109/ISGTEurope.2016.7856320

Hybrid model for short-term forecasting of loads and load control responses. / Koponen, Pekka; Niska, Harri.

PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), 2016 IEEE . Institute of Electrical and Electronic Engineers IEEE, 2016. p. 1-6 7856320.

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - Hybrid model for short-term forecasting of loads and load control responses

AU - Koponen, Pekka

AU - Niska, Harri

N1 - SDA: MIP: Intelligent Energy Grids

PY - 2016/7/2

Y1 - 2016/7/2

N2 - A hybrid model for short-term forecasting of aggregated thermal loads and their load control responses is studied in this paper using field test data. Inputs include temperature measurement and forecast, measured power and control signals. The hybrid model comprises 1) partly physically based forecasting of the responses of the controlled thermal loads and the non-controlled power, and 2) forecasting the residual using Support Vector Machine (SVM). Their summation gives the hourly interval power forecast. Here partly physically based means that the model structure models thermal dynamics of houses. The response forecasting needs as inputs the daily heating energy demand and controllable power range, which are forecast by partly physically based models. Weekly load pattern is used to forecast the non-controllable power. SVM forecasts the residual. Alternative configurations are compared. The response forecasting model improved the forecasting accuracy most and adding the SVM for the residual further improved the accuracy.

AB - A hybrid model for short-term forecasting of aggregated thermal loads and their load control responses is studied in this paper using field test data. Inputs include temperature measurement and forecast, measured power and control signals. The hybrid model comprises 1) partly physically based forecasting of the responses of the controlled thermal loads and the non-controlled power, and 2) forecasting the residual using Support Vector Machine (SVM). Their summation gives the hourly interval power forecast. Here partly physically based means that the model structure models thermal dynamics of houses. The response forecasting needs as inputs the daily heating energy demand and controllable power range, which are forecast by partly physically based models. Weekly load pattern is used to forecast the non-controllable power. SVM forecasts the residual. Alternative configurations are compared. The response forecasting model improved the forecasting accuracy most and adding the SVM for the residual further improved the accuracy.

KW - demand response

KW - forecasting

KW - hybrid models

KW - machine learning

KW - physical models

UR - http://www.scopus.com/inward/record.url?scp=85017561442&partnerID=8YFLogxK

U2 - 10.1109/ISGTEurope.2016.7856320

DO - 10.1109/ISGTEurope.2016.7856320

M3 - Conference article in proceedings

SN - 978-1-5090-3359-1

SP - 1

EP - 6

BT - PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), 2016 IEEE

PB - Institute of Electrical and Electronic Engineers IEEE

ER -

Koponen P, Niska H. Hybrid model for short-term forecasting of loads and load control responses. In PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), 2016 IEEE . Institute of Electrical and Electronic Engineers IEEE. 2016. p. 1-6. 7856320 https://doi.org/10.1109/ISGTEurope.2016.7856320