Assessment of some methods for short-term load forecasting

Pekka Koponen; A. Mutanen; Harri Niska

doi:10.1109/ISGTEurope.2014.7028901

Assessment of some methods for short-term load forecasting

Pekka Koponen, A. Mutanen, Harri Niska

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

4 Citations (Scopus)

Abstract

Accurate forecasting of loads is essential for smart grids and energy markets. This paper compares the performance of the following models in short-term load forecasting: 1) smart metering data based profile models, 2) a neural network (NN) model, and 3) a Kalman-filter based predictor with input nonlinearities and a physically based main structure. The comparison helps method selection for the development of hybrid models for forecasting the load control responses. According to the results all these three modeling approaches show much better performance than 4) the traditional load profiles and 5) a static outdoor temperature dependency model applied with a lag. The neural network model was the most accurate in the comparison, but the differences of the three methods developed were rather small and also other aspects and other methods must be considered and compared when selecting the method for a specific purpose.

Original language	English
Title of host publication	Proceedings
Subtitle of host publication	IEEE PES Innovative Smart Grid Technologies, Europe, ISGT-Europe, 2014
Publisher	IEEE Institute of Electrical and Electronic Engineers
Number of pages	6
ISBN (Electronic)	978-1-4799-7720-8
DOIs	https://doi.org/10.1109/ISGTEurope.2014.7028901
Publication status	Published - 2014
MoE publication type	A4 Article in a conference publication
Event	5th IEEE PES Innovative Smart Grid Technologies Conference Europe, ISGT-Europe 2014 - Istanbul, Turkey Duration: 12 Oct 2014 → 15 Oct 2014

Conference

Conference	5th IEEE PES Innovative Smart Grid Technologies Conference Europe, ISGT-Europe 2014
Abbreviated title	ISGT-Europe 2014
Country	Turkey
City	Istanbul
Period	12/10/14 → 15/10/14

Keywords

power demand
demand forecasting
load modeling
prediction algorithms
artificial neural networks

UN SDGs

This output contributes to the following Sustainable Development Goal(s)

Access to Document

10.1109/ISGTEurope.2014.7028901

Fingerprint Dive into the research topics of 'Assessment of some methods for short-term load forecasting'. Together they form a unique fingerprint.

Cite this

@inproceedings{3cff76b1545045f39ec7a20d702861d9,

title = "Assessment of some methods for short-term load forecasting",

abstract = "Accurate forecasting of loads is essential for smart grids and energy markets. This paper compares the performance of the following models in short-term load forecasting: 1) smart metering data based profile models, 2) a neural network (NN) model, and 3) a Kalman-filter based predictor with input nonlinearities and a physically based main structure. The comparison helps method selection for the development of hybrid models for forecasting the load control responses. According to the results all these three modeling approaches show much better performance than 4) the traditional load profiles and 5) a static outdoor temperature dependency model applied with a lag. The neural network model was the most accurate in the comparison, but the differences of the three methods developed were rather small and also other aspects and other methods must be considered and compared when selecting the method for a specific purpose.",

keywords = "power demand, demand forecasting, load modeling, prediction algorithms, artificial neural networks",

author = "Pekka Koponen and A. Mutanen and Harri Niska",

year = "2014",

doi = "10.1109/ISGTEurope.2014.7028901",

language = "English",

booktitle = "Proceedings",

publisher = "IEEE Institute of Electrical and Electronic Engineers",

address = "United States",

note = "5th IEEE PES Innovative Smart Grid Technologies Conference Europe, ISGT-Europe 2014, ISGT-Europe 2014 ; Conference date: 12-10-2014 Through 15-10-2014",

}

Koponen, P, Mutanen, A & Niska, H 2014, Assessment of some methods for short-term load forecasting. in Proceedings: IEEE PES Innovative Smart Grid Technologies, Europe, ISGT-Europe, 2014. IEEE Institute of Electrical and Electronic Engineers, 5th IEEE PES Innovative Smart Grid Technologies Conference Europe, ISGT-Europe 2014, Istanbul, Turkey, 12/10/14. https://doi.org/10.1109/ISGTEurope.2014.7028901

Assessment of some methods for short-term load forecasting. / Koponen, Pekka; Mutanen, A.; Niska, Harri.

Proceedings: IEEE PES Innovative Smart Grid Technologies, Europe, ISGT-Europe, 2014. IEEE Institute of Electrical and Electronic Engineers, 2014.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Assessment of some methods for short-term load forecasting

AU - Koponen, Pekka

AU - Mutanen, A.

AU - Niska, Harri

PY - 2014

Y1 - 2014

N2 - Accurate forecasting of loads is essential for smart grids and energy markets. This paper compares the performance of the following models in short-term load forecasting: 1) smart metering data based profile models, 2) a neural network (NN) model, and 3) a Kalman-filter based predictor with input nonlinearities and a physically based main structure. The comparison helps method selection for the development of hybrid models for forecasting the load control responses. According to the results all these three modeling approaches show much better performance than 4) the traditional load profiles and 5) a static outdoor temperature dependency model applied with a lag. The neural network model was the most accurate in the comparison, but the differences of the three methods developed were rather small and also other aspects and other methods must be considered and compared when selecting the method for a specific purpose.

AB - Accurate forecasting of loads is essential for smart grids and energy markets. This paper compares the performance of the following models in short-term load forecasting: 1) smart metering data based profile models, 2) a neural network (NN) model, and 3) a Kalman-filter based predictor with input nonlinearities and a physically based main structure. The comparison helps method selection for the development of hybrid models for forecasting the load control responses. According to the results all these three modeling approaches show much better performance than 4) the traditional load profiles and 5) a static outdoor temperature dependency model applied with a lag. The neural network model was the most accurate in the comparison, but the differences of the three methods developed were rather small and also other aspects and other methods must be considered and compared when selecting the method for a specific purpose.

KW - power demand

KW - demand forecasting

KW - load modeling

KW - prediction algorithms

KW - artificial neural networks

U2 - 10.1109/ISGTEurope.2014.7028901

DO - 10.1109/ISGTEurope.2014.7028901

M3 - Conference article in proceedings

BT - Proceedings

PB - IEEE Institute of Electrical and Electronic Engineers

T2 - 5th IEEE PES Innovative Smart Grid Technologies Conference Europe, ISGT-Europe 2014

Y2 - 12 October 2014 through 15 October 2014

ER -