EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Addressing Data Sparsity in EV Charging Load Forecasting: A Novel Zero-Inflated Neural Network Approach

Huiya Xiang, Zhe Li, Lisha Liu, Yujin Yang, Lin Lu, Binxin Zhu

Energies · 24.04.2026 · 19 (9) · 2068

DOI: 10.3390/en19092068 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Energies, online erschienen am 24.04.2026. Verfasst von Huiya Xiang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hubei Provincial Natural Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate electric vehicle (EV) charging load forecasting is essential for grid planning and resource allocation, yet existing approaches struggle with the inherent sparsity of charging data—a phenomenon characterized by excessive zeros representing periods of no charging activity. This paper addresses this challenge through a novel framework combining a Zero-Inflated Neural Network (ZINN) architecture with an Evolutionary Neural Architecture Search (ENAS) algorithm. ZINN explicitly decomposes the forecasting problem into binary classification (predicting charging occurrence) and regression (estimating energy magnitude conditioned on occurrence), enabling the model to learn distinct patterns for the absence and presence of charging events. Rather than relying on manually designed architectures, ENAS automatically discovers optimal encoder and decoder configurations from a comprehensive search space encompassing modern architectures (LSTM, GRU, Transformer, and iTransformer), layer configurations, activation functions, and hyperparameters. The evolutionary algorithm balances prediction accuracy with computational efficiency through multi-objective optimization. Extensive experiments on real-world EV charging data from 30 stations in Wuhan demonstrate that the ZINN+ENAS framework achieves the lowest prediction error compared to conventional baselines, with the discovered optimal configuration substantially outperforming hand-crafted designs. Comprehensive ablation studies reveal that the asymmetric dual-head architecture and adaptive regularization strategies are critical for handling data sparsity. These findings highlight the importance of explicit zero-inflation modeling and automated architecture discovery for specialized forecasting tasks, providing practitioners with an open-source framework for practical EV charging load prediction.

Förderung

  • Hubei Provincial Natural Science Foundation

Zitieren

APA

Xiang, H., Li, Z., Liu, L., Yang, Y., Lu, L., & Zhu, B. (2026). Addressing Data Sparsity in EV Charging Load Forecasting: A Novel Zero-Inflated Neural Network Approach. Energies, 19(9), 2068. https://doi.org/10.3390/en19092068

Vancouver

Xiang H, Li Z, Liu L, Yang Y, Lu L, Zhu B. Addressing Data Sparsity in EV Charging Load Forecasting: A Novel Zero-Inflated Neural Network Approach. Energies. 2026;19(9):2068. doi:10.3390/en19092068

BibTeX
@article{Xiang2026,
  author = {Xiang, Huiya and Li, Zhe and Liu, Lisha and Yang, Yujin and Lu, Lin and Zhu, Binxin},
  title = {Addressing Data Sparsity in EV Charging Load Forecasting: A Novel Zero-Inflated Neural Network Approach},
  journal = {Energies},
  year = {2026},
  volume = {19},
  number = {9},
  pages = {2068},
  doi = {10.3390/en19092068}
}
Kostenlos starten →