Short-term load forecasting model based on smart metering data: Daily energy prediction using physically based component model structure

Pekka Koponen (Corresponding author)

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

1 Citation (Scopus)

Abstract

Performance of smart grids and energy markets depends on the accuracy of forecasted power balances and power flows. This document describes the following approach to predict daily energy consumption of large groups of small customers that have electrical heating and cooling. The model is divided into parallel submodels, such as transfer function models, for differently behaving load types. Each linear transfer function has also physically based input nonlinearities such as saturation defining the heating and cooling ranges, or heat pump coefficient of performance. The submodels and their input nonlinearities were identified one after another in decreasing size order. 13 months of hourly metered data from about 6672 houses were used in the model development and verification. The model was identified from 2664 randomly selected houses. The model is described and its simulations are compared with measured loads. Future verification and development steps are briefly discussed.
Original languageEnglish
Title of host publicationIEEE 2012 International Conference on Smart Grid Technology, Economics and Policies
PublisherIEEE Institute of Electrical and Electronic Engineers
Number of pages4
ISBN (Electronic)978-1-4673-5932-0 , 978-1-4673-5930-6
DOIs
Publication statusPublished - 3 Dec 2012
MoE publication typeA4 Article in a conference publication
EventIEEE International Conference on Smart Grid Technology, Economics and Policies, SG-TEP 2012 - Nuremberg, Germany
Duration: 3 Jul 20124 Jul 2012

Conference

ConferenceIEEE International Conference on Smart Grid Technology, Economics and Policies, SG-TEP 2012
Abbreviated titleSG-TEP 2012
CountryGermany
CityNuremberg
Period3/07/124/07/12

Fingerprint

Model structures
Transfer functions
Cooling
Heating
Energy utilization
Pumps

Cite this

Koponen, P. (2012). Short-term load forecasting model based on smart metering data: Daily energy prediction using physically based component model structure. In IEEE 2012 International Conference on Smart Grid Technology, Economics and Policies IEEE Institute of Electrical and Electronic Engineers . https://doi.org/10.1109/SG-TEP.2012.6642386
Koponen, Pekka. / Short-term load forecasting model based on smart metering data : Daily energy prediction using physically based component model structure. IEEE 2012 International Conference on Smart Grid Technology, Economics and Policies. IEEE Institute of Electrical and Electronic Engineers , 2012.
@inproceedings{4f427dd9b7ac42439060134f918f8888,
title = "Short-term load forecasting model based on smart metering data: Daily energy prediction using physically based component model structure",
abstract = "Performance of smart grids and energy markets depends on the accuracy of forecasted power balances and power flows. This document describes the following approach to predict daily energy consumption of large groups of small customers that have electrical heating and cooling. The model is divided into parallel submodels, such as transfer function models, for differently behaving load types. Each linear transfer function has also physically based input nonlinearities such as saturation defining the heating and cooling ranges, or heat pump coefficient of performance. The submodels and their input nonlinearities were identified one after another in decreasing size order. 13 months of hourly metered data from about 6672 houses were used in the model development and verification. The model was identified from 2664 randomly selected houses. The model is described and its simulations are compared with measured loads. Future verification and development steps are briefly discussed.",
author = "Pekka Koponen",
year = "2012",
month = "12",
day = "3",
doi = "10.1109/SG-TEP.2012.6642386",
language = "English",
booktitle = "IEEE 2012 International Conference on Smart Grid Technology, Economics and Policies",
publisher = "IEEE Institute of Electrical and Electronic Engineers",
address = "United States",

}

Koponen, P 2012, Short-term load forecasting model based on smart metering data: Daily energy prediction using physically based component model structure. in IEEE 2012 International Conference on Smart Grid Technology, Economics and Policies. IEEE Institute of Electrical and Electronic Engineers , IEEE International Conference on Smart Grid Technology, Economics and Policies, SG-TEP 2012, Nuremberg, Germany, 3/07/12. https://doi.org/10.1109/SG-TEP.2012.6642386

Short-term load forecasting model based on smart metering data : Daily energy prediction using physically based component model structure. / Koponen, Pekka (Corresponding author).

IEEE 2012 International Conference on Smart Grid Technology, Economics and Policies. IEEE Institute of Electrical and Electronic Engineers , 2012.

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

TY - GEN

T1 - Short-term load forecasting model based on smart metering data

T2 - Daily energy prediction using physically based component model structure

AU - Koponen, Pekka

PY - 2012/12/3

Y1 - 2012/12/3

N2 - Performance of smart grids and energy markets depends on the accuracy of forecasted power balances and power flows. This document describes the following approach to predict daily energy consumption of large groups of small customers that have electrical heating and cooling. The model is divided into parallel submodels, such as transfer function models, for differently behaving load types. Each linear transfer function has also physically based input nonlinearities such as saturation defining the heating and cooling ranges, or heat pump coefficient of performance. The submodels and their input nonlinearities were identified one after another in decreasing size order. 13 months of hourly metered data from about 6672 houses were used in the model development and verification. The model was identified from 2664 randomly selected houses. The model is described and its simulations are compared with measured loads. Future verification and development steps are briefly discussed.

AB - Performance of smart grids and energy markets depends on the accuracy of forecasted power balances and power flows. This document describes the following approach to predict daily energy consumption of large groups of small customers that have electrical heating and cooling. The model is divided into parallel submodels, such as transfer function models, for differently behaving load types. Each linear transfer function has also physically based input nonlinearities such as saturation defining the heating and cooling ranges, or heat pump coefficient of performance. The submodels and their input nonlinearities were identified one after another in decreasing size order. 13 months of hourly metered data from about 6672 houses were used in the model development and verification. The model was identified from 2664 randomly selected houses. The model is described and its simulations are compared with measured loads. Future verification and development steps are briefly discussed.

U2 - 10.1109/SG-TEP.2012.6642386

DO - 10.1109/SG-TEP.2012.6642386

M3 - Conference article in proceedings

BT - IEEE 2012 International Conference on Smart Grid Technology, Economics and Policies

PB - IEEE Institute of Electrical and Electronic Engineers

ER -

Koponen P. Short-term load forecasting model based on smart metering data: Daily energy prediction using physically based component model structure. In IEEE 2012 International Conference on Smart Grid Technology, Economics and Policies. IEEE Institute of Electrical and Electronic Engineers . 2012 https://doi.org/10.1109/SG-TEP.2012.6642386