Abnormality diagnosis device and abnormality diagnosis method

Patent metadata

Field	Value
Publication number	US-12103169-B2
Application number	US-201817266402-A
Country	US
Kind code	B2
Filing date	Aug 6, 2018
Priority date	Aug 6, 2018
Publication date	Oct 1, 2024
Grant date	Oct 1, 2024

Abstract

An abnormality diagnosis device diagnoses an abnormality of a plurality of speed reducers included in a robot in accordance with disturbance torque regarding a state of the respective speed reducers acquired from a sensor installed in the robot, and outputs a result of the diagnosis to a display unit, the abnormality diagnosis device including a maintenance history DB configured to store maintenance data on maintenance made for the respective speed reducers, and a control unit configured to detect an abnormality in the respective speed reducers in accordance with the disturbance torque. The control unit, when detecting an abnormality in one speed reducer in accordance with the disturbance torque, predicts an abnormality in another speed reducer caused in association with the abnormality in the one speed reducer in accordance with the maintenance data, and outputs information on the predicted abnormality to the display unit.

First claim

The invention claimed is:

1. An abnormality diagnosis device for diagnosing an abnormality of a plurality of movable parts included in an apparatus comprising:
a sensor installed in the apparatus, wherein movable-part data regarding a state of the plurality of movable parts is acquired from the sensor;
a maintenance history database configured to store maintenance data on maintenance made for the respective movable parts; and
an information processing circuit configured to detect the abnormality in the respective movable parts in accordance with the movable-part data,
the information processing circuit further configured to:
determine, in response to detecting the abnormality in one movable part of the plurality of movable parts in accordance with the movable part data, whether the abnormality has occurred in any of other movable parts of the plurality of movable parts caused in association with the abnormality in the one movable part in accordance with the maintenance data,
execute a contradiction analysis between the abnormality detected in the one moveable part and the maintenance data,
determine, in response to the abnormality in the one moveable part not being contradictory to the maintenance data, a maintenance command for the one moveable part,
determine, in response to a contradiction between the abnormality detected in the one moveable and the maintenance data, whether the maintenance data indicates the occurrence of the abnormality in any of the other moveable parts by referring to the maintenance data stored in the maintenance history database,
determine, in response to the maintenance data indicating the occurrence of the abnormality in any of the other moveable parts, a maintenance command for the one moveable part and any of the other moveable parts; and
output, when determining the abnormality in any of the other moveable parts, information on the abnormality detected in the one moveable part and information on the abnormality determined in any of the other movable parts to a display, the information including maintenance commands for the one movable part and for any of the other movable parts, respectively,

wherein the sensor comprises an encoder and a torque sensor,
wherein the movable-part data comprises a position of the one movable part and a measured torque at the one movable part, the position and the measured torque determined with the encoder and torque sensor, respectively.

2. The abnormality diagnosis device according to claim 1 , wherein the information processing circuit determines the information on the abnormality in any of the other movable parts in accordance with a time of the maintenance and a content of the maintenance made for the respective movable parts acquired from the maintenance data stored in the maintenance history database.

3. The abnormality diagnosis device according to claim 1 , wherein the information processing circuit, when detecting the abnormality in the one movable part, refers to the maintenance data regarding the abnormality of at least one of the other movable parts to determine the information on the abnormality in any of the other movable parts.

4. The abnormality diagnosis device according to claim 1 , further comprising a correlation storage unit configured to store a correlation between the abnormality caused in the one movable part and the abnormality caused in any of the other movable parts,
wherein the information processing circuit, when detecting the abnormality in the one movable part, refers to the correlation stored in the correlation storage unit to predict the abnormality in any of the other movable parts having the correlation with the abnormality detected in the one movable part.

5. The abnormality diagnosis device according to claim 1 , wherein the information processing circuit executes machine learning for learning a pattern of the maintenance data that has a high probability of an occurrence of an abnormality based on the maintenance data regarding the respective movable parts at least within a part of periods stored in the maintenance history database to predict the abnormality in any of the other movable parts in accordance with a result of machine learning.

6. The abnormality diagnosis device according to claim 1 , wherein the information processing circuit is configured to:
when detecting the abnormality in the one movable part, generate a tree image indicating, with a tree structure, information on the one movable part and information on the other movable parts; and
output the information on the other movable parts to the display while changing a displaying configuration of the information depending on a content upon predicting the abnormality in the other movable parts.

7. The abnormality diagnosis device according to claim 6 , wherein the information processing circuit is configured to:
generate a maintenance history image indicating the maintenance data on the one movable part and at least one of the other movable parts in a time-series manner in addition to the tree image; and
when predicting the abnormality in any of the other movable parts, cause the maintenance history image to be accompanied with a sign indicating the maintenance data that is a basis upon predicting, and cause the information on the other movable parts in the tree image to be accompanied with the same sign to output the tree image and the maintenance history image to the display.

8. The abnormality diagnosis device according to claim 7 , wherein the information processing circuit is configured to:
generate, as the maintenance history image, a first maintenance history image and a second maintenance history image having a different scale in a time axis from the first maintenance history image; and
emphasize or enlarge and output, to the display, either the first maintenance history image or the second maintenance history image including the maintenance data that is the basis upon predicting the abnormality in any of the other movable parts.

9. The abnormality diagnosis device according to claim 6 , wherein the information processing circuit is configured to:
calculate an abnormality level indicating a degree of abnormality caused in the movable-part data;
generate an abnormality level display image including the movable-part data and the abnormality level in addition to the tree image; and
when predicting the abnormality in any of the other movable parts, cause the abnormality level display image to be accompanied with a sign indicating the abnormality level that is a basis upon predicting, and cause the information on the other movable parts in the tree image to be accompanied with the same sign to output the tree image and the maintenance history image to the display.

10. The abnormality diagnosis device according to claim 6 , wherein the tree image includes a maintenance command for the one movable part and any of the other movable parts in which an occurrence of the abnormality is predicted, and is associated with a relation between the occurrence of the abnormality and the maintenance command.

11. The abnormality diagnosis device according to claim 1 , wherein the information processing circuit outputs the information on the abnormality to the display only when diagnosing at least one of the movable parts as having the abnormality.

12. The abnormality diagnosis device according to claim 1 ,
wherein when detecting the abnormality in one movable part, determining whether the abnormality has occurred in any of the other movable parts of the plurality of movable parts comprises:
determining, based on the maintenance data, whether maintenance correlated with the abnormality detected in the one movable

Assignees

• Nissan Motor

Inventors

• Tanaka Yasuhiro

Classifications

• G05B23/0283 Primary

  Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] (preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06) · CPC title

• G05B23/0224

  Process history based detection method, e.g. whereby history implies the availability of large amounts of data · CPC title

• G05B13/0265

  the criterion being a learning criterion · CPC title

• G01L5/0061

  Force sensors associated with industrial machines or actuators (for the specific machine or actuator involved see relevant class, e.g. F01, F04, F16, B66, E21) · CPC title

• G06N20/00

  Machine learning · CPC title