Multi-Gradient-Descent Federated Learning With Parity for Cooperative Short-Term Load Forecasting

Short-term load forecasting (STLF) is essential in power system research. In scenarios where multiple distribution system operators are involved in cooperative forecasting, it is important to uphold the fairness of multi-party cooperation and drive the parity of model gains among parties under the p...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on power systems Jg. 40; H. 2; S. 1199 - 1213
Hauptverfasser: Wang, Haijin, Xing, Shuangshuang, Si, Caomingzhe, Pan, Zibin, Zhao, Junhua, Qiu, Jing, Dong, Zhaoyang
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 01.03.2025
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Accuracy
Algorithms
Cooperative STLF
Data models
Federated learning
Forecasting
Kuhn-Tucker method
Load forecasting
Load modeling
Long short term memory
multi-gradient-descent
Multiple objective analysis
Parity
performance parity
Privacy
privacy-preserving
ISSN:0885-8950, 1558-0679
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Short-term load forecasting (STLF) is essential in power system research. In scenarios where multiple distribution system operators are involved in cooperative forecasting, it is important to uphold the fairness of multi-party cooperation and drive the parity of model gains among parties under the premise of privacy protection. In this context, this paper proposes a short-term load forecasting method namely Multi-Gradient-Descent Federated Learning with Parity Long-short Term Memory (MGD-FLP-LSTM) under a federated learning framework. First, MGD-FLP-LSTM transforms cooperative STLF into a Multi-Objective Optimization (MOO) problem with parity-driven objective. The designed algorithm optimizes multiple objectives simultaneously and incorporates Karush-Kuhn-Tucker (KKT) conditions in solving the dual form of the problem, thus accelerating the global model convergence. Second, MGD-FLP-LSTM constitutes a parity-driven objective with cosine similarity in the MOO iteration. The objective allows all local loss functions to have similar descents on the common gradient descending direction. Third, MGD-FLP-LSTM constructs a Performance Parity Control (PPC) scheme that enables an active moderate trade-off between forecasting accuracy and parity. Experiments on real-measured load dataset highlight the method's enhanced accuracy, efficiency, and performance while maintaining consumer privacy.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0885-8950
1558-0679
DOI:10.1109/TPWRS.2024.3420811