Power load forecasting combining deep learning models and improved CLPO algorithm

Jianguang Song, Xiang Wei, Zhipeng Li, Xia Zhang, Guoheng Zhang, Lu Wang · 2026

Read Power load forecasting combining deep learning models and improved CLPO algorithm by Jianguang Song, Xiang Wei, Zhipeng Li, Xia Zhang, Guoheng Zhang, Lu Wang (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-0351428.txt.

Description

Jianguang Song and five colleagues build a hybrid forecasting system for electricity demand in this 2026 study. Accurate load forecasts keep power grids stable and cheap to run. Demand curves are knotted and nonlinear, and standard optimisers get stuck in local traps. The team combined three deep learning networks that read time patterns differently. Their outputs were merged by weight. An improved parrot-inspired optimisation algorithm then tuned the system, with an error-correction step on top.

The combined model converged far faster than gradient descent training. It reached strong accuracy in about two hundred rounds, where the standard route needed well over three hundred. Final accuracy was higher too. On a British distribution network dataset, its long-term forecast errors were the lowest among rival methods. Extra machinery guarded against drifting weights and short-term bias. Grid operators gain a forecaster that learns quickly and stays honest over long horizons.

Catalog record

CREATORJianguang Song, Xiang Wei, Zhipeng Li, Xia Zhang, Guoheng Zhang, Lu Wang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE77170
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0351428

Browse the full library · About the Tower of Babel