Anomaly detection of online monitoring data of power equipment based on association rules and clustering algorithm

DOI

10.2991/eeeis-16.2017.38How to use a DOI?

Keywords

Big data, Anomaly detection, Data cleaning, Association rules, FCM.

Abstract

With the continuous research and development of smart grid and energy Internet, as well as the rapid construction of power transmission and transformation equipment in various places, the amount of data collected from the equipment is also increasing. To dig out the effective information must be to ensure the accuracy of the data. However, large data set must contain erroneous or abnormal data. The traditional method cannot handle the big data anomaly detection well. Therefore, this paper presents anomaly detection based on association rules and clustering algorithms. The association rules are used to find out the sequences with relevance in the dataset. Then the FCM algorithm are used to separate the abnormal data into a sensor abnormal that can be cleaned and a device abnormality that cannot be cleaned. For the correlation sequence, the sensor anomaly and the device abnormality are found by the method of association and clustering, then early warning and maintenance advice are given.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yu-Xiang Cai
AU  - Li-Jun Cai
AU  - Zhou Lu
PY  - 2016/12
DA  - 2016/12
TI  - Anomaly detection of online monitoring data of power equipment based on association rules and clustering algorithm
BT  - Proceedings of the 2nd Annual International Conference on Electronics, Electrical Engineering and Information Science (EEEIS 2016)
PB  - Atlantis Press
SP  - 289
EP  - 298
SN  - 2352-5401
UR  - https://doi.org/10.2991/eeeis-16.2017.38
DO  - 10.2991/eeeis-16.2017.38
ID  - Cai2016/12
ER  -