Using kNN Algorithm for classification of Distribution transformers Health index

Distribution transformers are significant components in a grid utility system. Therefore, knowing the health status of a transformer becomes vital for the safe operation of a grid. Health Index (HI) defined the health status based on laboratory tests and interpreted by the standards specified by IEE...

Full description

Saved in:
Bibliographic Details
Published in2021 International Conference on Innovative Computing (ICIC) pp. 1 - 6
Main Authors Javid, Jahanzaib, Mughal, Muhammad Ali, Karim, Mustansir
Format Conference Proceeding
LanguageEnglish
Published IEEE 09.11.2021
Subjects
Classification algorithms
Distribution transformers
Health index
IEC Standards
Indexes
kNN algorithm
Machine learning
Machine learning algorithms
Prediction algorithms
Transformers
Online AccessGet full text

Cover

More Information
Summary:Distribution transformers are significant components in a grid utility system. Therefore, knowing the health status of a transformer becomes vital for the safe operation of a grid. Health Index (HI) defined the health status based on laboratory tests and interpreted by the standards specified by IEEE, IEC. The obtained HI number represents whether a transformer is healthy and able to operate further or unhealthy and needs a replacement. In this paper, the k Nearest Neighbors (kNN) algorithm is employed to predict the health status based on the laboratory test data. A total of six tests performed on a class of distribution transformers are used as predictors. The kNN algorithm has proved to be a good classifier for the transformer health status prediction problem by achieving good accuracy.
DOI:10.1109/ICIC53490.2021.9693013