We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Electrical Engineering and Systems Science > Systems and Control

Title: Extremum-Seeking Adaptive-Droop for Model-free and Localized Volt-VAR Optimization

Abstract: In an active power distribution system, Volt-VAR optimization (VVO) methods are employed to achieve network-level objectives such as minimization of network power losses. The commonly used model-based centralized and distributed VVO algorithms perform poorly in the absence of a communication system and with model and measurement uncertainties. In this paper, we proposed a model-free local Volt-VAR control approach for network-level optimization that does not require communication with other decision-making agents. The proposed algorithm is based on extremum-seeking approach that uses only local measurements to minimize the network power losses. To prove that the proposed extremum-seeking controller converges to the optimum solution, we also derive mathematical conditions for which the loss minimization problem is convex with respect to the control variables. Local controllers pose stability concerns during highly variable scenarios. Thus, the proposed extremum-seeking controller is integrated with an adaptive-droop control module to provide a stable local control response. The proposed approach is validated using IEEE 4-bus and IEEE 123-bus systems and achieves the loss minimization objective while maintaining the voltage within the pre-specific limits even during highly variable DER generation scenarios.
Comments: 11 pages, 15 figures
Subjects: Systems and Control (eess.SY)
DOI: 10.1109/TPWRS.2021.3093831
Cite as: arXiv:2107.04515 [eess.SY]
  (or arXiv:2107.04515v1 [eess.SY] for this version)

Submission history

From: Hongda Ren [view email]
[v1] Fri, 9 Jul 2021 16:02:25 GMT (3015kb)

Link back to: arXiv, form interface, contact.