DG-LSTM-SA model: A deep gated LSTM network with self-attention mechanism for power generation and load forecasting
Guoqiang Sun, Yang Zhao, Jianglong Li, Junfu Cui, Xiaoyan Qi · 2026
Read DG-LSTM-SA model: A deep gated LSTM network with self-attention mechanism for power generation and load forecasting by Guoqiang Sun, Yang Zhao, Jianglong Li, Junfu Cui, Xiaoyan Qi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0350071.txt.
Description
Guoqiang Sun and co-authors built a forecasting model for power generation and load in 2026. Grid operators must predict demand closely to keep supply steady, yet power data has long patterns that simple recurrent networks lose, and big attention models cost too much.
Their model stacks gated memory layers with attention placed inside the stack, letting it stress the time steps that matter without ballooning in size. On three real energy data sets it beat ten rival models, cutting errors sharply against plain recurrent networks. It matched leading attention models on accuracy while training faster, a practical gain for daily grid planning.
Catalog record
| CREATOR | Guoqiang Sun, Yang Zhao, Jianglong Li, Junfu Cui, Xiaoyan Qi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46390 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |