Presentation | 2017-10-20 A Fundamental Study of Training Data Selection Method for Wind Turbine Health Management Using SCADA Data Akihisa Yasuda, Jun Ogata, Yoko Furusawa, Masahiro Murakawa, Hiroyuki Morikawa, Makoto Iida, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Wind turbines need to be stopped for a long period if the internal equipment breaks down. Therefore, it is important for the wind power business to detect the anomaly related to breakdown of the wind turbine quickly and to implement repair work for extending the life of the equipment. In this paper, assuming a system health monitoring method which uses data collected by SCADA (Supervisory Control And Data Acquisition) which is installed as a wind turbine standard equipment, we propose a method of extracting normal data from SCADA data using ideal power curve and evaluating the normal behavior of the data with change point detection. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Wind Turbine / SCADA / Normal Behavior / Anomaly Detection / Machine Learning / Training Data |
Paper # | R2017-47 |
Date of Issue | 2017-10-13 (R) |
Conference Information | |
Committee | R |
---|---|
Conference Date | 2017/10/20(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Tetsushi Yuge(National Defense Academy) |
Vice Chair | Akira Asato(Fujitsu) |
Secretary | Akira Asato(Hosei Univ.) |
Assistant | Shinji Inoue(Kansai Univ.) / Hiroyuki Okamura(Hiroshima Univ.) |
Paper Information | |
Registration To | Technical Committee on Reliability |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | A Fundamental Study of Training Data Selection Method for Wind Turbine Health Management Using SCADA Data |
Sub Title (in English) | |
Keyword(1) | Wind Turbine |
Keyword(2) | SCADA |
Keyword(3) | Normal Behavior |
Keyword(4) | Anomaly Detection |
Keyword(5) | Machine Learning |
Keyword(6) | Training Data |
1st Author's Name | Akihisa Yasuda |
1st Author's Affiliation | The University of Tokyo(UT) |
2nd Author's Name | Jun Ogata |
2nd Author's Affiliation | National Institute of Advanced Industrial Science and Technology(AIST) |
3rd Author's Name | Yoko Furusawa |
3rd Author's Affiliation | The University of Tokyo(UT) |
4th Author's Name | Masahiro Murakawa |
4th Author's Affiliation | National Institute of Advanced Industrial Science and Technology(AIST) |
5th Author's Name | Hiroyuki Morikawa |
5th Author's Affiliation | The University of Tokyo(UT) |
6th Author's Name | Makoto Iida |
6th Author's Affiliation | The University of Tokyo(UT) |
Date | 2017-10-20 |
Paper # | R2017-47 |
Volume (vol) | vol.117 |
Number (no) | R-253 |
Page | pp.pp.17-22(R), |
#Pages | 6 |
Date of Issue | 2017-10-13 (R) |