RSSPN:Robust Semi-Supervised Prototypical Network for Fault Root Cause Classification in Power Distribution Systems

The power distribution system's fault root cause classification is an important but challenging problem. Traditional classifiers fail to achieve high accuracy and good generalization performance due to data insufficiency. A large volume of unlabeled data is available, which can be utilized to i...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on power delivery 2022-08, Vol.37 (4), p.3282-3290
Main Authors: Zheng, Tianqing, Liu, Yadong, Yan, Yingjie, Xiong, Siheng, Lin, Tao, Chen, Yanxia, Wang, Zhiyong, Jiang, Xiuchen
Format: Article
Language:English
Subjects:
Classification
Classifiers
Computer architecture
data insufficiency
Electric power distribution
fault cause classification
Feature extraction
Few-shot learning
Lightning
Power distribution
power distribution systems
Prototypes
Robustness
Semi-supervised learning
Semisupervised learning
Task analysis
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The power distribution system's fault root cause classification is an important but challenging problem. Traditional classifiers fail to achieve high accuracy and good generalization performance due to data insufficiency. A large volume of unlabeled data is available, which can be utilized to improve classification performance. This paper proposes a novel classifier called Robust Semi-Supervised Prototypical Network (RSSPN) based on Prototypical Network architecture and semi-supervised learning to address this issue. The proposed method can mine information from unlabeled data to improve the generalization ability and classification accuracy. Furthermore, RSSPN adopts the idea of meta-learning to obtain the "few-shot learning" ability for identifying new fault classes using very few samples encountered during the operation and update online. Experiments have been conducted on a dataset consisting of 1152 labeled samples belonging to 12 different classes and 10000 unlabeled samples. The accuracy of the proposed method is significantly better than the traditional classifiers.
ISSN:0885-8977
1937-4208
DOI:10.1109/TPWRD.2021.3125704