Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68695
Volume:
10 |
11 |
Nov 2026
Pages:
1-4
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Computer Engineering Department
SVPM’s College of Engineering, Malegaon (Bk.), Baramati
Computer Engineering Department
SVPM’s College of Engineering, Malegaon (Bk.), Baramati
Computer Engineering Department
SVPM’s College of Engineering, Malegaon (Bk.), Baramati
Computer Engineering Department
SVPM’s College of Engineering, Malegaon (Bk.), Baramati
Computer Engineering Department
SVPM’s College of Engineering, Malegaon (Bk.), Baramati
Accurate forecasting of electricity demand and renewable energy generation is central to reliable grid planning, because solar and wind output varies with the weather while demand follows season, time of day and human behavior. This paper surveys five representative studies on deep learning for energy forecasting: a CNN-LSTM model for residential consumption, an LSTM model tuned with feature selection and a genetic algorithm for load forecasting, a CNN with multi-layer bidirectional LSTM for household power, a U-shaped LSTM model for solar irradiance, and a multi-branch CNN, LSTM and dense network for renewable energy consumption. For each study, the core idea, processing flow, advantages, limitations and reported outcome are summarized. A comparative analysis shows that existing work usually forecasts either demand or generation, rarely connects the two, and seldom explains its predictions. This gap motivates a proposed smart energy demand and renewable energy prediction system that forecasts both quantities, computes the energy gap or excess, and produces a planning recommendation.
KEYWORDS
This paper reviewed five studies on deep learning for energy forecasting. Taken together, they show that CNN-LSTM style hybrids, bidirectional layers, systematic feature and hyperparameter selection, and multi-branch designs each improve forecasting accuracy. At the same time, the studies treat demand and generation separately, stop at the forecast, and offer limited explanation. The proposed system addresses these limitations by forecasting both demand and renewable generation, computing the energy gap or excess, and producing a recommendation for energy planning. Future work will implement and evaluate the system on real data, add explainability, and extend it with live weather forecasts and storage-aware planning.
1. Kim, T.-Y., Cho, S.-B.: Predicting residential energy consumption using CNN-LSTM neural networks. Energy 182 (2019) 72–81. doi:10.1016/j.energy.2019.05.230 2. Bouktif, S., Fiaz, A., Ouni, A., Serhani, M.A.: Optimal deep learning LSTM model for electric load forecasting using feature selection and genetic algorithm: Comparison with machine learning approaches. Energies 11(7) (2018) 1636. doi:10.3390/en11071636 3. Ullah, F.U.M., Ullah, A., Haq, I.U., Rho, S., Baik, S.W.: Short-term prediction of residential power energy consumption via CNN and multi-layer bi-directional LSTM networks. IEEE Access (2020). doi:10.1109/ACCESS.2019.2963045 4. Zhu, L. et al.: A novel U-LSTM-AFT model for hourly solar irradiance forecasting. Renewable Energy 238 (2025) 121955 5. Li, Z., Zhang, Y., Liu, Y.: Research on U.S. renewable energy consumption prediction based on multi-branch deep learning architecture. In: Proc. 8th Int. Conf. Renewable Energy and Power Engineering (REPE). IEEE (2025). doi:10.1109/REPE66317.2025.11340927 6. Suganthi, L., Samuel, A.A.: Energy models for demand forecasting—A review. Renewable and Sustainable Energy Reviews 16(2) (2012) 1223–1240 7. Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Computation 9(8) (1997) 1735–1780
Publication Date
11/5/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
5
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...