TY  - JOUR
AU  - Ferdowsi, Mohsen
AU  - Benigni, Andrea
AU  - Monti, Antonello
AU  - Ponci, Ferdinanda
TI  - Measurement Selection for Data-Driven Monitoring of Distribution Systems
JO  - IEEE systems journal
VL  - 13
IS  - 4
SN  - 1932-8184
CY  - New York, NY
PB  - IEEE
M1  - FZJ-2019-05331
SP  - 4260 - 4268
PY  - 2019
AB  - This article investigates the problem of measurement selection for data-driven monitoring approaches. Several approaches to input variable selection (IVS) are analyzed, and a general procedure for finding the optimal order for the selection of candidate measurements is presented. The method is based on the extensions of partial correlation and minimal redundancy maximum relevance criteria to support IVS problems involving multiple outputs. This method can be used to find the minimal set of measurements for achieving a target estimation accuracy. The results demonstrate the advantages and limits of the introduced method in comparison to the other approaches discussed in this article.
LB  - PUB:(DE-HGF)16
UR  - <Go to ISI:>//WOS:000503182300064
DO  - DOI:10.1109/JSYST.2019.2939500
UR  - https://juser.fz-juelich.de/record/866112
ER  -