Physical-Model-Aided Data-Driven Linear Power Flow Model: An Approach to Address Missing Training Data

Zhentong Shao, Qiaozhu Zhai, Xiaohong Guan

Research output: Contribution to journalArticlepeer-review

34 Scopus citations

Abstract

Data-driven linear power flow (D-LPF) models are prevalent due to their excellent accuracy. Typically, D-LPF models rely on sufficient training data. However, in practice, the training data may be insufficient due to recording errors or limited measurement conditions. To address this practical and important issue, this letter presents a physical-model-aided data-driven linear power flow (PD-LPF) model, in which, physical model parameters are introduced to assist the data-driven training process, thereby avoiding unreasonable training results, and guaranteeing linearization accuracy for critical operating points with the maximum probability. The proposed method is applicable for both transmission and distribution systems. Compared to current LPF models, the PD-LPF model exhibits excellent accuracy and robustness under severe missing-data conditions.

Original languageEnglish
Pages (from-to)2970-2973
Number of pages4
JournalIEEE Transactions on Power Systems
Volume38
Issue number3
DOIs
StatePublished - 1 May 2023

Keywords

  • Data-driven
  • chance constraints
  • distributionally robust
  • linear power flow
  • missing data

Fingerprint

Dive into the research topics of 'Physical-Model-Aided Data-Driven Linear Power Flow Model: An Approach to Address Missing Training Data'. Together they form a unique fingerprint.

Cite this