Restoration of High-Renewable-Penetrated Distribution Systems Considering Uncertain Repair Workloads

In distribution systems with high penetration of renewable energy, fast restoration after extreme events faces a variety of uncertainties from the output of distributed renewable generations, as well as incomplete knowledge on accurate repair workloads of fault components. Considering these uncertai...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaotian Sun, Haipeng Xie, Zhaohong Bie, Gengfeng Li
Format: Article
Language:English
Published: China electric power research institute 2025-01-01
Series:CSEE Journal of Power and Energy Systems
Subjects:
Online Access:https://ieeexplore.ieee.org/document/9862576/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In distribution systems with high penetration of renewable energy, fast restoration after extreme events faces a variety of uncertainties from the output of distributed renewable generations, as well as incomplete knowledge on accurate repair workloads of fault components. Considering these uncertainties, this paper proposes an implicit decision rule based restoration-oriented combined unit-crew dispatch (RCUCD) method to boost the ex-post restoration. First, a “wait and see” uncertainty model is presented to depict the non-anticipativity of repair workloads of fault components. Thereby, a dynamic crew grouping dispatch algorithm is proposed to address the uncertainties in repair workloads based on the improved time-space network theory. Further, a multi-stage stochastic RCUCD model is set up for post-disaster restoration. Based on implicit decision rules, the proposed model is reformulated into a mixed integer linear programming (MILP) problem by introducing axillary safety region variables. To boost computational efficiency, the simplified RCUCD model, as well as the rolling look-ahead RCUCD model is proposed. The effectiveness and efficiency of the proposed methods are tested on IEEE 34-bus and 69-bus distribution systems.
ISSN:2096-0042