US 12,247,752 B2
Abnormality cause estimation device, abnormality cause estimation method, and medium
Yasuhiro Toyama, Tokyo (JP)
Assigned to MITSUBISHI ELECTRIC CORPORATION, Tokyo (JP)
Filed by MITSUBISHI ELECTRIC CORPORATION, Tokyo (JP)
Filed on Jun. 15, 2021, as Appl. No. 17/347,876.
Application 17/347,876 is a continuation of application No. PCT/JP2019/001638, filed on Jan. 21, 2019.
Prior Publication US 2021/0310679 A1, Oct. 7, 2021
Int. Cl. F24F 11/38 (2018.01); F24F 11/63 (2018.01)
CPC F24F 11/38 (2018.01) [F24F 11/63 (2018.01)] 6 Claims
OG exemplary drawing
 
1. An abnormality cause estimation device, comprising:
processing circuitry configured to
acquire sensor data from a sensor installed in a target device;
acquire additional data related to the target device, the additional data including at least one or more data of a set value related to an operation of the target device, an environmental value around the target device, and information about a product manufactured by the target device;
classify the acquired sensor data for each group according to the acquired additional data;
detect an abnormal tendency of the sensor data by comparing the acquired sensor data with a normal model in accordance with a group of the sensor data; and
estimate an abnormality cause of the target device by comparing the group of the sensor data and the detected abnormal tendency with an abnormality cause model in which a group being identical to the group of the sensor data, the abnormal tendency of past sensor data, and the abnormality cause of the target device are associated with each other, to identify the abnormality cause of the target device without estimating the abnormality cause by threshold value determination to streamline maintenance work of the target device and predict and prevent abnormality occurrence of the target device.