Asynchronous Distributed Optimal Load Scheduling Algorithm

Distributed optimization strategies are promising solutions for direct load control (DLC), since the centralized ones have privacy and heavy communication issues. However, conventional synchronous distributed methods suffer from communication delay or failures. In this paper, we present an asynchron...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE Power & Energy Society General Meeting S. 1 - 5
Hauptverfasser: Wang, Qi, Wu, Wenchuan, Lin, Chenhui, Li, Li, Yang, Yinguo
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 02.08.2020
Schlagworte:
Asynchronous alternating direction method of multipliers
demand response
home energy management systems
three-phase power flow model
ISSN:1944-9933
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Distributed optimization strategies are promising solutions for direct load control (DLC), since the centralized ones have privacy and heavy communication issues. However, conventional synchronous distributed methods suffer from communication delay or failures. In this paper, we present an asynchronous distributed approach based on a Fast Alternative Direction Method of Multipliers (ADMM) in which a synthetic asynchronous strategy is adopted to tackle with the time-delay into the upload and download link. To accelerate the developed algorithm's convergence further, methods including adaptive updating of penalty parameter, warm start and over-relaxation are introduced. Since the residential loads are connected on low voltage level networks, three-phase unbalanced distribution network is formulated in the DLC model. The objective considers the operational cost of LSE and the utility function of consumers. Numerical tests on IEEE 33-bus and IEEE 123-bus systems show that the proposed method can realize reliable and fast convergence both with and without communication delay, attaining the optimal solution consistent with the centralized algorithm simultaneously, and requires much less computing time compared with the synchronous algorithm.
ISSN:1944-9933
DOI:10.1109/PESGM41954.2020.9281471