Performance Evaluation of Deep Recurrent Neural Networks Architectures: Application to PV Power Forecasting

Mohamed Massaoudi; Ines Chihi; Lilia Sidhom; Mohamed Trabelsi; Shady S. Refaat; Fakhreddine S. Oueslati

doi:10.1109/SGRE46976.2019.9020965

Performance Evaluation of Deep Recurrent Neural Networks Architectures: Application to PV Power Forecasting

Mohamed Massaoudi, Ines Chihi, Lilia Sidhom, Mohamed Trabelsi, Shady S. Refaat, Fakhreddine S. Oueslati

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

10 Citations (Scopus)

Abstract

Smart grid systems require an accurate energy prediction from renewable sources to ensure high sustainability and power quality. For PV plants, a precise estimation of the generated PV power is crucial for the reduction of the production/demand unbalance. This essential need comes from the high variability of weather parameters during the PV electricity generation. Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) recurrent neural networks proved their high efficiency in forecasting applications. Thus, this paper proposes a comprehensive evaluation of the LSTM and GRU techniques for PV power estimation in the medium/long horizon. The evaluation is based on a fair assessment of the aforementioned architectures for one week and more than three months (98 days) periods.

Original language	English
Title of host publication	2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728129600
DOIs	https://doi.org/10.1109/SGRE46976.2019.9020965
Publication status	Published - Nov 2019
Externally published	Yes
Event	2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Doha, Qatar Duration: 19 Nov 2019 → 21 Nov 2019

Publication series

Name	2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings

Conference

Conference	2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019
Country/Territory	Qatar
City	Doha
Period	19/11/19 → 21/11/19

Keywords

Gated Recurrent Unit
GRU
Long Short Term Memory
LSTM
PV Power Forecasting
Smart Grids

Access to Document

10.1109/SGRE46976.2019.9020965

Cite this

Massaoudi, M., Chihi, I., Sidhom, L., Trabelsi, M., Refaat, S. S., & Oueslati, F. S. (2019). Performance Evaluation of Deep Recurrent Neural Networks Architectures: Application to PV Power Forecasting. In 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings [9020965] (2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SGRE46976.2019.9020965

Massaoudi, Mohamed ; Chihi, Ines ; Sidhom, Lilia et al. / Performance Evaluation of Deep Recurrent Neural Networks Architectures : Application to PV Power Forecasting. 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. (2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings).

@inproceedings{c6af4fddcc184edfa48e078322234a04,

title = "Performance Evaluation of Deep Recurrent Neural Networks Architectures: Application to PV Power Forecasting",

abstract = "Smart grid systems require an accurate energy prediction from renewable sources to ensure high sustainability and power quality. For PV plants, a precise estimation of the generated PV power is crucial for the reduction of the production/demand unbalance. This essential need comes from the high variability of weather parameters during the PV electricity generation. Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) recurrent neural networks proved their high efficiency in forecasting applications. Thus, this paper proposes a comprehensive evaluation of the LSTM and GRU techniques for PV power estimation in the medium/long horizon. The evaluation is based on a fair assessment of the aforementioned architectures for one week and more than three months (98 days) periods.",

keywords = "Gated Recurrent Unit, GRU, Long Short Term Memory, LSTM, PV Power Forecasting, Smart Grids",

author = "Mohamed Massaoudi and Ines Chihi and Lilia Sidhom and Mohamed Trabelsi and Refaat, {Shady S.} and Oueslati, {Fakhreddine S.}",

note = "Funding Information: This publication was made possible by NPRP grant [NPRP10-0101-170082] from the Qatar National Research Fund (a member of Qatar Foundation) and the co-funding by IBERDROLA QSTP LLC. The statements made herein are solely the responsibility of the authors. Publisher Copyright: {\textcopyright} 2019 IEEE.; 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 ; Conference date: 19-11-2019 Through 21-11-2019",

year = "2019",

month = nov,

doi = "10.1109/SGRE46976.2019.9020965",

language = "English",

series = "2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings",

address = "United States",

}

Massaoudi, M, Chihi, I, Sidhom, L, Trabelsi, M, Refaat, SS & Oueslati, FS 2019, Performance Evaluation of Deep Recurrent Neural Networks Architectures: Application to PV Power Forecasting. in 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings., 9020965, 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019, Doha, Qatar, 19/11/19. https://doi.org/10.1109/SGRE46976.2019.9020965

Performance Evaluation of Deep Recurrent Neural Networks Architectures : Application to PV Power Forecasting. / Massaoudi, Mohamed; Chihi, Ines; Sidhom, Lilia et al.

2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. 9020965 (2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Performance Evaluation of Deep Recurrent Neural Networks Architectures

T2 - 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019

AU - Massaoudi, Mohamed

AU - Chihi, Ines

AU - Sidhom, Lilia

AU - Trabelsi, Mohamed

AU - Refaat, Shady S.

AU - Oueslati, Fakhreddine S.

N1 - Funding Information: This publication was made possible by NPRP grant [NPRP10-0101-170082] from the Qatar National Research Fund (a member of Qatar Foundation) and the co-funding by IBERDROLA QSTP LLC. The statements made herein are solely the responsibility of the authors. Publisher Copyright: © 2019 IEEE.

PY - 2019/11

Y1 - 2019/11

N2 - Smart grid systems require an accurate energy prediction from renewable sources to ensure high sustainability and power quality. For PV plants, a precise estimation of the generated PV power is crucial for the reduction of the production/demand unbalance. This essential need comes from the high variability of weather parameters during the PV electricity generation. Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) recurrent neural networks proved their high efficiency in forecasting applications. Thus, this paper proposes a comprehensive evaluation of the LSTM and GRU techniques for PV power estimation in the medium/long horizon. The evaluation is based on a fair assessment of the aforementioned architectures for one week and more than three months (98 days) periods.

AB - Smart grid systems require an accurate energy prediction from renewable sources to ensure high sustainability and power quality. For PV plants, a precise estimation of the generated PV power is crucial for the reduction of the production/demand unbalance. This essential need comes from the high variability of weather parameters during the PV electricity generation. Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) recurrent neural networks proved their high efficiency in forecasting applications. Thus, this paper proposes a comprehensive evaluation of the LSTM and GRU techniques for PV power estimation in the medium/long horizon. The evaluation is based on a fair assessment of the aforementioned architectures for one week and more than three months (98 days) periods.

KW - Gated Recurrent Unit

KW - GRU

KW - Long Short Term Memory

KW - LSTM

KW - PV Power Forecasting

KW - Smart Grids

UR - http://www.scopus.com/inward/record.url?scp=85082302943&partnerID=8YFLogxK

U2 - 10.1109/SGRE46976.2019.9020965

DO - 10.1109/SGRE46976.2019.9020965

M3 - Conference contribution

AN - SCOPUS:85082302943

T3 - 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings

BT - 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 19 November 2019 through 21 November 2019

ER -

Massaoudi M, Chihi I, Sidhom L, Trabelsi M, Refaat SS, Oueslati FS. Performance Evaluation of Deep Recurrent Neural Networks Architectures: Application to PV Power Forecasting. In 2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. 9020965. (2nd International Conference on Smart Grid and Renewable Energy, SGRE 2019 - Proceedings). doi: 10.1109/SGRE46976.2019.9020965

Performance Evaluation of Deep Recurrent Neural Networks Architectures: Application to PV Power Forecasting

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this