TCN ‐ GWO ‐Based Photovoltaic Power Forecasting and Intelligent Hosting Capacity Assessment for Distributed PV Power Grids
Engineering Reports · 2026-11 · 8 (11)
DOI: 10.1002/eng2.71058 ↗Kurz gefasst
Fachartikel in der Zeitschrift Engineering Reports, online erschienen am 13.11.2026. Verfasst von Yijie Huang und 8 weiteren Autor:innen. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The rapid integration of distributed photovoltaic (PV) systems into modern power grids creates challenges related to PV intermittency, dynamic hosting capacity, and real‐time grid operation. Conventional static capacity assessment approaches may not adequately capture temporal variations in solar generation and grid conditions, limiting efficient PV integration. This study develops an Intelligent Open Capacity Publishing Platform based on a hybrid Temporal Convolutional Network–Grey Wolf Optimizer (TCN‐GWO) framework for accurate PV generation forecasting and dynamic hosting‐capacity allocation. PV generation, solar irradiance, temperature, humidity, voltage, and load data are preprocessed using Min‐Max normalization and arranged as sequential inputs for TCN‐based forecasting. The TCN employs causal and dilated convolutions to capture short‐ and long‐term temporal dependencies, while GWO optimizes feasible PV hosting capacity under voltage, thermal‐loading, and current constraints. The forecasting performance is evaluated using MSE, MAE, MAPE, nMSE, and R 2 , together with computational‐time analysis. Experimental results show that the proposed TCN‐GWO achieves an MSE of 0.00103, MAE of 0.0250, MAPE of 1.80%, nMSE of 0.00173, and R 2 of 0.9939, outperforming the evaluated LSTM‐GWO, GRU‐PSO, CNN‐LSTM, and TCN‐PSO models under the same experimental conditions. The framework further enables node‐level PV hosting‐capacity assessment while maintaining grid operating constraints. The proposed platform therefore provides an accurate, computationally efficient, and data‐driven approach for supporting transparent and adaptive distributed PV integration.
Zitieren
APA
Huang, Y., Wu, G., Kang, X., Wang, W., Li, W., Lin, Y., Cai, Q., Wu, Q., & Lan, Z. (2026). TCN ‐ GWO ‐Based Photovoltaic Power Forecasting and Intelligent Hosting Capacity Assessment for Distributed PV Power Grids. Engineering Reports, 8(11). https://doi.org/10.1002/eng2.71058
Vancouver
Huang Y, Wu G, Kang X, Wang W, Li W, Lin Y, et al. TCN ‐ GWO ‐Based Photovoltaic Power Forecasting and Intelligent Hosting Capacity Assessment for Distributed PV Power Grids. Engineering Reports. 2026;8(11). doi:10.1002/eng2.71058
BibTeX
@article{Huang2026,
author = {Huang, Yijie and Wu, Guoyao and Kang, Xiaoqing and Wang, Wei and Li, Wanping and Lin, Yayun and Cai, Qi and Wu, Qiyou and Lan, Zhiqiang},
title = {TCN ‐ GWO ‐Based Photovoltaic Power Forecasting and Intelligent Hosting Capacity Assessment for Distributed PV Power Grids},
journal = {Engineering Reports},
year = {2026},
volume = {8},
number = {11},
doi = {10.1002/eng2.71058}
}
EUVIMED