Demand response for efficient power generation in smart grids using hyper–local weather prediction

Shaikh, F; Khattak, HA; Alavi, HS; Farman, H; Nasralla, MM

Demand response for efficient power generation in smart grids using hyper–local weather prediction

Shaikh, F Khattak, HA Alavi, HS Farman, H Nasralla, MM

Permalink

Publisher:: TAYLOR & FRANCIS LTD
Publication Type:: Journal Article
Citation:: Cogent Engineering, 2025, 12, (1)
Issue Date:: 2025-01-01

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download Published versionAdobe PDF (5.71 MB)

View on publisher's site

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Shaikh, F
dc.contributor.author	Khattak, HA
dc.contributor.author	Alavi, HS
dc.contributor.author	Farman, H
dc.contributor.author	Nasralla, MM
dc.date.accessioned	2026-07-02T03:25:52Z
dc.date.available	2026-07-02T03:25:52Z
dc.date.issued	2025-01-01
dc.identifier.citation	Cogent Engineering, 2025, 12, (1)
dc.identifier.issn	2331-1916
dc.identifier.issn	2331-1916
dc.identifier.uri	http://hdl.handle.net/10453/195528
dc.description.abstract	Optimizing power generation is crucial in the evolving smart grid and urban ecosystem landscape. Efficient energy use is vital in today’s tech-driven world, with smart grids leading the way in managing energy consumption and distribution. Given the link between climate and energy use, integrating weather data is now essential. This research presents a method combining demand response in smart grids with hyperlocal weather predictions using machine learning to enhance power generation forecasts. Five models—SARIMAX, Prophet, XGBoost, LSTM, and Holt–Winters are trained on the smart grid and weather data, including precipitation, humidity, temperature, and cloud cover. Extensive feature engineering, such as lag terms and cyclic encoding, improved accuracy. SARIMAX outperformed others with an MAE of 3.54 and an R-squared of 0.97, while the hybrid model, combining SARIMAX, Prophet, and Holt–Winters, achieved an R-squared of 0.96. The hybrid model’s dynamic recalibration improved accuracy with new data. A real-time dashboard is also developed, offering dynamic, user-specific forecasting insights. This research enhances smart grid management by integrating machine learning and demand response for more efficient energy use.
dc.language	English
dc.publisher	TAYLOR & FRANCIS LTD
dc.relation.ispartof	Cogent Engineering
dc.relation.isbasedon	10.1080/23311916.2025.2555341
dc.rights	info:eu-repo/semantics/openAccess
dc.subject.classification	40 Engineering
dc.title	Demand response for efficient power generation in smart grids using hyper–local weather prediction
dc.type	Journal Article
utslib.citation.volume	12
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	open_access	*
dc.rights.license	This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0). To view a copy of this license, visit https://creativecommons.org/licenses/by/4.0/
dc.date.updated	2026-07-02T03:25:48Z
pubs.issue	1
pubs.publication-status	Published
pubs.volume	12
utslib.citation.issue	1

Abstract:

Optimizing power generation is crucial in the evolving smart grid and urban ecosystem landscape. Efficient energy use is vital in today’s tech-driven world, with smart grids leading the way in managing energy consumption and distribution. Given the link between climate and energy use, integrating weather data is now essential. This research presents a method combining demand response in smart grids with hyperlocal weather predictions using machine learning to enhance power generation forecasts. Five models—SARIMAX, Prophet, XGBoost, LSTM, and Holt–Winters are trained on the smart grid and weather data, including precipitation, humidity, temperature, and cloud cover. Extensive feature engineering, such as lag terms and cyclic encoding, improved accuracy. SARIMAX outperformed others with an MAE of 3.54 and an R-squared of 0.97, while the hybrid model, combining SARIMAX, Prophet, and Holt–Winters, achieved an R-squared of 0.96. The hybrid model’s dynamic recalibration improved accuracy with new data. A real-time dashboard is also developed, offering dynamic, user-specific forecasting insights. This research enhances smart grid management by integrating machine learning and demand response for more efficient energy use.

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

http://hdl.handle.net/10453/195528