EUVIMEDEUVIMED entdecken
Crossref

Hybrid XGBoost-BiLSTM Net-Load Forecasting Framework With Holographic Integration for High-PV IoT Environments

Ronak Doshi, Rajendrasinh Jadeja, Amit Ved

IEEE Transactions on Consumer Electronics · 2026-02 · 72 (1) · 1410-1417

DOI: 10.1109/tce.2025.3647579 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Consumer Electronics, 2026-02. Verfasst von Ronak Doshi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 41 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Net-load forecasting is essential for reliable and economic operation of power systems grid, especially in high photovoltaic (PV) penetration environments. By incorporating Holographic Counterparts (HCs) into Internet of Things (IoT) ecosystems, a new paradigm for enhancing consumer electronics autonomous decision-making, real-time data interaction, and system scalability is introduced. This study proposes novel net load forecasting method that combines automated feature extraction based on historical load, solar generation, and weather data, and a hyperparameter optimization method to improve the forecasting accuracy. The hybrid model is constructed by taking Extreme Gradient Boosting (XGBoost), which captures complicated interactions between predictors, and combining it with a Bidirectional Long Short-Term Memory (BiLSTM) network, which captures temporal dynamics in models. The forecasts from both models will be fused, to develop point forecasts and a probabilistic forecast. The new proposed method will be tested on real-world datasets collected from New York and Austria and compared to standard direct and indirect forecasting models. The different models will be evaluated by Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). The results show that the novel method performed better than the baseline methods, achieving the lowest RMSE of 26.57 for New York and 159.4 for Austria under PV penetration scenario.

Zitieren

APA

Doshi, R., Jadeja, R., & Ved, A. (2026). Hybrid XGBoost-BiLSTM Net-Load Forecasting Framework With Holographic Integration for High-PV IoT Environments. IEEE Transactions on Consumer Electronics, 72(1), 1410-1417. https://doi.org/10.1109/tce.2025.3647579

Vancouver

Doshi R, Jadeja R, Ved A. Hybrid XGBoost-BiLSTM Net-Load Forecasting Framework With Holographic Integration for High-PV IoT Environments. IEEE Transactions on Consumer Electronics. 2026;72(1):1410-1417. doi:10.1109/tce.2025.3647579

BibTeX
@article{Doshi2026,
  author = {Doshi, Ronak and Jadeja, Rajendrasinh and Ved, Amit},
  title = {Hybrid XGBoost-BiLSTM Net-Load Forecasting Framework With Holographic Integration for High-PV IoT Environments},
  journal = {IEEE Transactions on Consumer Electronics},
  year = {2026},
  volume = {72},
  number = {1},
  pages = {1410--1417},
  doi = {10.1109/tce.2025.3647579}
}
Kostenlos starten →