The accurate forecasting of electric load in smart grid applications and the prediction of synthesizability for crystalline materials are crucial aspects in decision-making and the advancement of power systems and materials science. This study proposes an integrated framework that combines deep learning techniques with heuristic algorithms to address these challenges. A novel hybrid model is introduced for electric load forecasting, encompassing a data pre-processing and feature selection module based on the modified mutual information (MMI) technique. The training and forecasting module employs a factored conditional restricted Boltzmann machine (FCRBM), designed to capture the non-linear and stochastic behavior inherent in electric load profiles. To optimize the model, a genetic wind-driven (GWDO) optimization algorithm is introduced, facilitating the fine-tuning of adjustable parameters to enhance accuracy. Shifting the focus to materials science, a distinctive approach is taken to predict the synthesizability of crystalline materials by representing atomic structures through three-dimensional pixel-wise images. These images, color-coded by chemical attributes, enable the use of a convolutional encoder to extract latent features related to synthesizability. The model accurately classifies materials into synthesizable crystals and anomalies across diverse crystal structures and chemical compositions. The proposed integrated framework is evaluated using historical hourly load data from three power grids and hypothetical crystals for battery electrode and thermoelectric applications. Comparative analyses with existing forecasting models and synthesizability benchmarks demonstrate the effectiveness of this approach in providing accurate predictions for both electric load and crystalline materials synthesizability. This research contributes to advancing the fields of smart grid management and materials science by offering a unified, data-driven methodology for forecasting and predicting synthesizability.
Choose the style your journal or department asks for, then copy it.
Every version below is generated from this paper's own record.
IJRTI — journal style
"Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning", IJRTI - International Journal for Research Trends and Innovation (www.ijrti.org), ISSN:2456-3315, Vol.9, Issue 1, page no.787 - 795, January-2024, Available :https://ijrti.org/papers/IJRTI2401118.pdf
APA — 7th edition
Dhanush, A., Shinde, N., Nayeemulla, S., & Biradar, P. (2024). Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning. International Journal for Research Trends and Innovation, 9(1), 787 - 795. https://doi.org/10.5281/zenodo.10580719
MLA — 9th edition
Dhanush, Akkari, et al. "Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning." International Journal for Research Trends and Innovation, vol. 9, no. 1, 2024, pp. 787 - 795, https://doi.org/10.5281/zenodo.10580719.
Chicago — 17th, bibliography
Dhanush, Akkari, et al. "Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning." International Journal for Research Trends and Innovation 9, no. 1 (2024): 787 - 795. https://doi.org/10.5281/zenodo.10580719.
Harvard — author–date
Dhanush, A. et al. (2024) 'Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning', International Journal for Research Trends and Innovation, 9(1), pp. 787 - 795. Available at: https://doi.org/10.5281/zenodo.10580719
IEEE — numbered reference
A. Dhanush, N. Shinde, S. Nayeemulla and P. Biradar, "Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning," IJRTI, vol. 9, no. 1, pp. 787 - 795, Jan. 2024, doi: 10.5281/zenodo.10580719.
Vancouver — biomedical
Dhanush A, Shinde N, Nayeemulla S, Biradar P. Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning. IJRTI. 2024 Jan;9(1):787 - 795. doi: 10.5281/zenodo.10580719
AMA — 11th edition
Dhanush A, Shinde N, Nayeemulla S, Biradar P. Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning. IJRTI. 2024;9(1):787 - 795. doi:10.5281/zenodo.10580719
BibTeX — Zotero, Mendeley, LaTeX
@article{IJRTI2401118,
author = {Akkari Dhanush and Nikhil Shinde and Syed Nayeemulla and Prajakta Biradar},
title = {Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning},
journal = {International Journal for Research Trends and Innovation},
volume = {9},
number = {1},
pages = {787 - 795},
year = {2024},
month = {January},
issn = {2456-3315},
doi = {10.5281/zenodo.10580719},
url = {https://doi.org/10.5281/zenodo.10580719}
}
RIS — EndNote, RefWorks
TY - JOUR
AU - Dhanush, Akkari
AU - Shinde, Nikhil
AU - Nayeemulla, Syed
AU - Biradar, Prajakta
TI - Towards Sustainable Smart Systems: A Deep Learning Approach to Electric Load Forecasting, Materials Synthesizability, and Machine Learning
T2 - International Journal for Research Trends and Innovation
JA - IJRTI
VL - 9
IS - 1
PY - 2024
SN - 2456-3315
DO - 10.5281/zenodo.10580719
UR - https://doi.org/10.5281/zenodo.10580719
SP - 787
EP - 795
ER -
Check for updates. Crossmark shows whether this article has been
corrected, updated or retracted since publication, and links to the
current version of record.
About Publisher
International Journal for Research Trends and InnovationPublished by IJRTI (JW Publication)
2456-3315ISSN
10.57Impact Factor
2016ESTD Year
OpenAccess
Impact Factor 10.57 calculated by Google Scholar and Semantic Scholar.
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 10.57 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
UGC CARE Approved Journal — Transparent Peer-reviewed journal aligned with UGC’s 2025 suggestive parameters, with CrossRef DOI registration and Scopus Standard metadata on every published paper.
Disclaimer:
The content, data and findings in this article are based on the authors’ research and have been
peer-reviewed for academic purposes only. Readers are advised to verify all information before practical
or commercial use. The journal and its editorial board are not liable for any errors, losses or
consequences arising from its use.
Declarations
Funding
No external funding was received for this study.
Conflict of Interest
The authors declare that they have no conflict of interest.
Acknowledgements
The authors would like to thank the reviewers and the editorial board of International Journal for Research Trends and Innovation for their careful reading and constructive comments, and all colleagues who supported the preparation of this manuscript.