Reconstructing Missing Data in State Estimation With Autoencoders

被引：81

作者：

Miranda, Vladimiro ^{[1
,2
]}

Krstulovic, Jakov ^{[1
]}

Keko, Hrvoje ^{[1
,2
]}

Moreira, Cristiano ^{[1
,2
]}

Pereira, Jorge ^{[1
,2
]}

机构：

[1] INESC TEC INESC Technol & Sci, Oporto, Portugal

[2] Univ Porto, FEUP, Fac Engn, Oporto, Portugal

来源：

IEEE TRANSACTIONS ON POWER SYSTEMS | 2012年 / 27卷 / 02期

关键词：

Autoencoders; distribution management systems; energy management systems; neural networks; state estimation;

D O I：

10.1109/TPWRS.2011.2174810

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

This paper presents the proof of concept for a new solution to the problem of recomposing missing information at the SCADA of energy/distribution management systems (EMS/DMS), through the use of offline trained autoencoders. These are neural networks with a special architecture, which allows them to store knowledge about a system in a nonlinear manifold characterized by their weights. Suitable algorithms may then recompose missing inputs (measurements). The paper shows that, trained with adequate information, autoencoders perform well in recomposing missing voltage and power values, and focuses on the particularly important application of inferring the topology of the network when information about switch status is absent. Examples with the IEEE RTS 24-bus network are presented to illustrate the concept and technique.

引用

页码：604 / 611

页数：8

共 22 条

[1]

Abdella M, 2005, COMPUT INFORM, V24, P577

[2]

[Anonymous], 1979, IEEE T POWER AP SYST, V98, P2047, DOI 10.1109/TPAS.1979.319398

[3]

BILLINTON R, 1985, IEEE T POWER AP SYST, V104, P2649, DOI 10.1109/TPAS.1985.319105

[4] Critical pseudo-measurement selection for unreduced external system modeling [J].