LSTM neural network model for ultra-short-term distribution zone substation peak demand prediction

Ibrahim, IA; Hossain, MJ

LSTM neural network model for ultra-short-term distribution zone substation peak demand prediction

Ibrahim, IA Hossain, MJ

Permalink

Publisher:: IEEE
Publication Type:: Conference Proceeding
Citation:: IEEE Power and Energy Society General Meeting, 2020, 2020-August, pp. 1-5
Issue Date:: 2020-08-02

In Progress

	Filename	Description	Size
	PESGM-19.pdf	Published version	267.08 kB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is being processed and is not currently available.

Full metadata record

Field	Value	Language
dc.contributor.author	Ibrahim, IA
dc.contributor.author	Hossain, MJ
dc.date	2020-08-02
dc.date.accessioned	2021-06-06T01:02:29Z
dc.date.available	2021-06-06T01:02:29Z
dc.date.issued	2020-08-02
dc.identifier.citation	IEEE Power and Energy Society General Meeting, 2020, 2020-August, pp. 1-5
dc.identifier.isbn	9781728155081
dc.identifier.issn	1944-9925
dc.identifier.issn	1944-9933
dc.identifier.uri	http://hdl.handle.net/10453/149429
dc.description.abstract	The accurate prediction of the distribution load demand data is the corner stone of future planning of the power system networks and energy management strategies and policies. This paper presents a long short-term memory (LSTM) neural networks model to predict the distribution zone substation peak demand data in New South Wales state, Australia for 14 years and based on 15-minute intervals. The obtained results are compared with those obtained by feed-forward neural networks (FFNNs) and recurrent neural networks (RNNs) models. Three statistical performance evaluation, namely, the root-mean-square error (RMSE), mean bias error (MBE) and mean absolute percentage error (MAPE) are used to verify the effectiveness of the proposed model. The RMSE, MBE and MAPE of the LSTM neural network model are 1.2556%, 1.2201% and 2.2250%, respectively. In addition, the computational time is 12.3309 second which is faster than FFNNs and RNNs models. The results show the effectiveness of the proposed model over the aforementioned models in terms of accuracy and computational speed.
dc.language	en
dc.publisher	IEEE
dc.relation.ispartof	IEEE Power and Energy Society General Meeting
dc.relation.ispartof	2020 IEEE Power & Energy Society General Meeting (PESGM)
dc.relation.isbasedon	10.1109/PESGM41954.2020.9281973
dc.rights	info:eu-repo/semantics/restrictedAccess
dc.title	LSTM neural network model for ultra-short-term distribution zone substation peak demand prediction
dc.type	Conference Proceeding
utslib.citation.volume	2020-August
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Electrical and Data Engineering
utslib.copyright.status	in_progress	*
dc.date.updated	2021-06-06T01:02:28Z
pubs.finish-date	2020-08-06
pubs.publication-status	Published
pubs.start-date	2020-08-02
pubs.volume	2020-August

Abstract:

The accurate prediction of the distribution load demand data is the corner stone of future planning of the power system networks and energy management strategies and policies. This paper presents a long short-term memory (LSTM) neural networks model to predict the distribution zone substation peak demand data in New South Wales state, Australia for 14 years and based on 15-minute intervals. The obtained results are compared with those obtained by feed-forward neural networks (FFNNs) and recurrent neural networks (RNNs) models. Three statistical performance evaluation, namely, the root-mean-square error (RMSE), mean bias error (MBE) and mean absolute percentage error (MAPE) are used to verify the effectiveness of the proposed model. The RMSE, MBE and MAPE of the LSTM neural network model are 1.2556%, 1.2201% and 2.2250%, respectively. In addition, the computational time is 12.3309 second which is faster than FFNNs and RNNs models. The results show the effectiveness of the proposed model over the aforementioned models in terms of accuracy and computational speed.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/149429