Title: Day-ahead power prediction of photovoltaic power generation based on WOA-BiGRU-AT model

Authors: Yi Zhang; Wenzhen Meng

Addresses: School of Electrical and Computer Science, Jilin Jianzhu University, Changchun, Jilin, China ' School of Electrical and Computer Science, Jilin Jianzhu University, Changchun, Jilin, China

Abstract: At present, photovoltaic power generation is developing rapidly worldwide, and more accurate photovoltaic power prediction technology plays a significant role in developing photovoltaic power generation. To improve the accuracy of the current traditional photovoltaic power prediction model, a photovoltaic power prediction algorithm based on the WOA-BiGRU-AT model is proposed in this paper. The attention mechanism is added to accelerate the convergence rate of the model. Then, by simulating the phenomenon of whale predation in nature, the super parameters of the BiGRU-AT model are optimised in the forward propagation process. The experimental results show that the model predicts the data set better than the existing models.

Keywords: day-ahead generation power prediction; wireless sensor; BiGRU; bi-directional gated recurrent unit; WOA; whale optimisation algorithm; attention mechanism.

DOI: 10.1504/IJWMC.2026.155714

International Journal of Wireless and Mobile Computing, 2026 Vol.31 No.2, pp.120 - 130

Received: 31 Jul 2023
Accepted: 27 Dec 2023
Published online: 11 Aug 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article