Title:
設備機器の異常診断システム
Document Type and Number:
Japanese Patent JP6778132
Kind Code:
B2
Abstract:
When abnormality diagnosis of an equipment machine is performed using data clustering processing, the abnormality diagnosis can be realized even when the normal data for the learning process with respect to the data of the equipment machine is unclear and cannot be defined. An abnormality diagnosis system for an equipment machine is configured such that abnormality diagnosis is performed using ananalysis method including learning processing using normal data and diagnosis processing using a model created in a learning process based on a plurality of data from the equipment machine; with respect to a plurality of data from the equipment machine, data of a normal operation of the equipment machine is not known; and the system includes: a first processing unit that performs clustering processing on a plurality of data from the equipment machine, a second processing unit defining a normal group based on the number of data conform to each group, a third processing unit extracting data conform to the normal group as normal data, and a fourth processing unit using the extracted normal data for learning processing .
Inventors:
Kiji Hayashi
Tatsuya Maeda
Shuntaro Hitomi
Koichiro Miyamoto
Tatsuya Maeda
Shuntaro Hitomi
Koichiro Miyamoto
Application Number:
JP2017047170A
Publication Date:
October 28, 2020
Filing Date:
March 13, 2017
Export Citation:
Assignee:
株式会社日立製作所
International Classes:
G05B23/02
Domestic Patent References:
| JP5369246B1 | ||||
| JP2009070071A | ||||
| JP2015114967A | ||||
| JP2016081482A | ||||
| JP2010092355A |
Foreign References:
| US20070088550 | ||||
| KR1020150015989A |
Attorney, Agent or Firm:
Polaire Patent Business Corporation