Home > Research > Publications & Outputs > Big data analytics for short and medium-term el...

Links

Text available via DOI:

View graph of relations

Big data analytics for short and medium-term electricity load forecasting using an AI techniques ensembler

Research output: Contribution to journalJournal articlepeer-review

Published

Standard

Big data analytics for short and medium-term electricity load forecasting using an AI techniques ensembler. / Ayub, N.; Irfan, M.; Awais, M.; Ali, U.; Ali, T.; Hamdi, M.; Alghamdi, A.; Muhammad, F.

In: Energies, Vol. 13, No. 19, 05.10.2020.

Research output: Contribution to journalJournal articlepeer-review

Harvard

Ayub, N, Irfan, M, Awais, M, Ali, U, Ali, T, Hamdi, M, Alghamdi, A & Muhammad, F 2020, 'Big data analytics for short and medium-term electricity load forecasting using an AI techniques ensembler', Energies, vol. 13, no. 19. https://doi.org/10.3390/en13195193

APA

Ayub, N., Irfan, M., Awais, M., Ali, U., Ali, T., Hamdi, M., Alghamdi, A., & Muhammad, F. (2020). Big data analytics for short and medium-term electricity load forecasting using an AI techniques ensembler. Energies, 13(19). https://doi.org/10.3390/en13195193

Vancouver

Author

Ayub, N. ; Irfan, M. ; Awais, M. ; Ali, U. ; Ali, T. ; Hamdi, M. ; Alghamdi, A. ; Muhammad, F. / Big data analytics for short and medium-term electricity load forecasting using an AI techniques ensembler. In: Energies. 2020 ; Vol. 13, No. 19.

Bibtex

@article{1c8abbcb13934495be74747a169fe482,
title = "Big data analytics for short and medium-term electricity load forecasting using an AI techniques ensembler",
abstract = "Electrical load forecasting provides knowledge about future consumption and generation of electricity. There is a high level of fluctuation behavior between energy generation and consumption. Sometimes, the energy demand of the consumer becomes higher than the energy already generated, and vice versa. Electricity load forecasting provides a monitoring framework for future energy generation, consumption, and making a balance between them. In this paper, we propose a framework, in which deep learning and supervised machine learning techniques are implemented for electricity-load forecasting. A three-step model is proposed, which includes: feature selection, extraction, and classification. The hybrid of Random Forest (RF) and Extreme Gradient Boosting (XGB) is used to calculate features{\textquoteright} importance. The average feature importance of hybrid techniques selects the most relevant and high importance features in the feature selection method. The Recursive Feature Elimination (RFE) method is used to eliminate the irrelevant features in the feature extraction method. The load forecasting is performed with Support Vector Machines (SVM) and a hybrid of Gated Recurrent Units (GRU) and Convolutional Neural Networks (CNN). The meta-heuristic algorithms, i.e., Grey Wolf Optimization (GWO) and Earth Worm Optimization (EWO) are applied to tune the hyper-parameters of SVM and CNN-GRU, respectively. The accuracy of our enhanced techniques CNN-GRU-EWO and SVM-GWO is 96.33% and 90.67%, respectively. Our proposed techniques CNN-GRU-EWO and SVM-GWO perform 7% and 3% better than the State-Of-The-Art (SOTA). In the end, a comparison with SOTA techniques is performed to show the improvement of the proposed techniques. This comparison showed that the proposed technique performs well and results in the lowest performance error rates and highest accuracy rates as compared to other techniques. {\textcopyright} 2020 by the authors. Licensee MDPI, Basel, Switzerland.",
keywords = "Big data analytics, Deep learning, Load forecasting, Optimization techniques, Advanced Analytics, Big data, Convolutional neural networks, Data Analytics, Decision trees, Electric power plant loads, Extraction, Feature extraction, Forecasting, Heuristic algorithms, Learning systems, Optimization, Recurrent neural networks, Support vector machines, Electrical load forecasting, Electricity load forecasting, Feature extraction methods, Feature selection methods, Meta heuristic algorithm, Monitoring frameworks, Recursive feature elimination, Supervised machine learning, Electric load forecasting",
author = "N. Ayub and M. Irfan and M. Awais and U. Ali and T. Ali and M. Hamdi and A. Alghamdi and F. Muhammad",
year = "2020",
month = oct,
day = "5",
doi = "10.3390/en13195193",
language = "English",
volume = "13",
journal = "Energies",
issn = "1996-1073",
publisher = "MDPI AG",
number = "19",

}

RIS

TY - JOUR

T1 - Big data analytics for short and medium-term electricity load forecasting using an AI techniques ensembler

AU - Ayub, N.

AU - Irfan, M.

AU - Awais, M.

AU - Ali, U.

AU - Ali, T.

AU - Hamdi, M.

AU - Alghamdi, A.

AU - Muhammad, F.

PY - 2020/10/5

Y1 - 2020/10/5

N2 - Electrical load forecasting provides knowledge about future consumption and generation of electricity. There is a high level of fluctuation behavior between energy generation and consumption. Sometimes, the energy demand of the consumer becomes higher than the energy already generated, and vice versa. Electricity load forecasting provides a monitoring framework for future energy generation, consumption, and making a balance between them. In this paper, we propose a framework, in which deep learning and supervised machine learning techniques are implemented for electricity-load forecasting. A three-step model is proposed, which includes: feature selection, extraction, and classification. The hybrid of Random Forest (RF) and Extreme Gradient Boosting (XGB) is used to calculate features’ importance. The average feature importance of hybrid techniques selects the most relevant and high importance features in the feature selection method. The Recursive Feature Elimination (RFE) method is used to eliminate the irrelevant features in the feature extraction method. The load forecasting is performed with Support Vector Machines (SVM) and a hybrid of Gated Recurrent Units (GRU) and Convolutional Neural Networks (CNN). The meta-heuristic algorithms, i.e., Grey Wolf Optimization (GWO) and Earth Worm Optimization (EWO) are applied to tune the hyper-parameters of SVM and CNN-GRU, respectively. The accuracy of our enhanced techniques CNN-GRU-EWO and SVM-GWO is 96.33% and 90.67%, respectively. Our proposed techniques CNN-GRU-EWO and SVM-GWO perform 7% and 3% better than the State-Of-The-Art (SOTA). In the end, a comparison with SOTA techniques is performed to show the improvement of the proposed techniques. This comparison showed that the proposed technique performs well and results in the lowest performance error rates and highest accuracy rates as compared to other techniques. © 2020 by the authors. Licensee MDPI, Basel, Switzerland.

AB - Electrical load forecasting provides knowledge about future consumption and generation of electricity. There is a high level of fluctuation behavior between energy generation and consumption. Sometimes, the energy demand of the consumer becomes higher than the energy already generated, and vice versa. Electricity load forecasting provides a monitoring framework for future energy generation, consumption, and making a balance between them. In this paper, we propose a framework, in which deep learning and supervised machine learning techniques are implemented for electricity-load forecasting. A three-step model is proposed, which includes: feature selection, extraction, and classification. The hybrid of Random Forest (RF) and Extreme Gradient Boosting (XGB) is used to calculate features’ importance. The average feature importance of hybrid techniques selects the most relevant and high importance features in the feature selection method. The Recursive Feature Elimination (RFE) method is used to eliminate the irrelevant features in the feature extraction method. The load forecasting is performed with Support Vector Machines (SVM) and a hybrid of Gated Recurrent Units (GRU) and Convolutional Neural Networks (CNN). The meta-heuristic algorithms, i.e., Grey Wolf Optimization (GWO) and Earth Worm Optimization (EWO) are applied to tune the hyper-parameters of SVM and CNN-GRU, respectively. The accuracy of our enhanced techniques CNN-GRU-EWO and SVM-GWO is 96.33% and 90.67%, respectively. Our proposed techniques CNN-GRU-EWO and SVM-GWO perform 7% and 3% better than the State-Of-The-Art (SOTA). In the end, a comparison with SOTA techniques is performed to show the improvement of the proposed techniques. This comparison showed that the proposed technique performs well and results in the lowest performance error rates and highest accuracy rates as compared to other techniques. © 2020 by the authors. Licensee MDPI, Basel, Switzerland.

KW - Big data analytics

KW - Deep learning

KW - Load forecasting

KW - Optimization techniques

KW - Advanced Analytics

KW - Big data

KW - Convolutional neural networks

KW - Data Analytics

KW - Decision trees

KW - Electric power plant loads

KW - Extraction

KW - Feature extraction

KW - Forecasting

KW - Heuristic algorithms

KW - Learning systems

KW - Optimization

KW - Recurrent neural networks

KW - Support vector machines

KW - Electrical load forecasting

KW - Electricity load forecasting

KW - Feature extraction methods

KW - Feature selection methods

KW - Meta heuristic algorithm

KW - Monitoring frameworks

KW - Recursive feature elimination

KW - Supervised machine learning

KW - Electric load forecasting

U2 - 10.3390/en13195193

DO - 10.3390/en13195193

M3 - Journal article

VL - 13

JO - Energies

JF - Energies

SN - 1996-1073

IS - 19

ER -