US 11,709,474 B2
Method and apparatus for detecting abnormality of manufacturing facility
Hyun Chui Kang, Daejeon (KR); Ho Jin Park, Daejeon (KR); Ji Yeon Son, Daejeon (KR); and Eun Seo Lee, Daejeon (KR)
Assigned to ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE, Daejeon (KR)
Filed by Electronics and Telecommunications Research Institute, Daejeon (KR)
Filed on Dec. 29, 2020, as Appl. No. 17/136,654.
Claims priority of application No. 10-2020-0013089 (KR), filed on Feb. 4, 2020.
Prior Publication US 2021/0240167 A1, Aug. 5, 2021
Int. Cl. G06N 20/00 (2019.01); G06F 18/214 (2023.01); G05B 19/4063 (2006.01); G05B 19/418 (2006.01)
CPC G05B 19/4063 (2013.01) [G05B 19/4183 (2013.01); G05B 19/4184 (2013.01); G05B 19/4185 (2013.01); G05B 19/41865 (2013.01); G06F 18/214 (2023.01); G06N 20/00 (2019.01)] 13 Claims
OG exemplary drawing
 
1. A computer program product embodied on a non-transitory computer-readable medium, the computer program product comprising software code portions being configured, when run on a processor, to perform a method of a learning model generating for detecting an abnormality of a manufacturing facility, the method comprising:
receiving, via the processor connected to a multi-sensor, a measured value for a normal state of a manufacturing facility collected through the multi-sensor on a time-by-time basis;
generating, via the processor, a learning model including a predetermined weight set and training the learning model using the measured value;
determining, via the processor, using the learning model, a threshold corresponding to a boundary between the normal state and an abnormal state of the manufacturing facility and a criterion for determining the abnormal state in a local window representing a predetermined time interval; and
removing a direction component of the received measured value to use the measured value for training of the learning model.