IP Library › Granted Patent US 11,353,860
Granted Patent B2
US 11,353,860 · App. 15/733,898 · Granted Jun 7, 2022

Data analysis device, system, method, and recording medium storing program

Inventor: Osamu Nasu (Tokyo, JP)
Assignee: MITSUBISHI ELECTRIC CORPORATION
G05B23/0229G06N5/04G06N20/00
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,353,860
App. No.
15/733,898
Granted
Jun 7, 2022
Kind
B2
Abstract

A data analysis device ( 1 ) according to the present disclosure includes a data collector ( 131 ) and a data organizer ( 135 ). The data collector ( 131 ) collects an input signal that is input from an apparatus connected to the data analysis device ( 1 ). The data organizer ( 135 ) removes, from among waveforms included in the input signal collected by the data collector ( 131 ), a waveform that has an appearance frequency lower than a reference frequency. The data organizer ( 135 ) generates learning data by collecting, from among the waveforms included in the input signal, waveforms that have an appearance frequency higher than the reference frequency.

Claims (58)

1. A data analysis device comprising:

a data collector to collect signals supplied from a plurality of apparatuses connected to the data analysis device, the signals including a signal supplied when at least one of the apparatuses is in a normal operating state and a signal supplied when at least one of the apparatuses is in an abnormal operating state;

a data learner to create waveforms including signals cut out at predetermined intervals from the signals collected by the data collector, the data learner being further configured to classify the waveforms into groups of mutually-similar waveforms in accordance with a predetermined classification basis; and

a data organizer to generate learning data by (i) removing waveforms, determined by an external inspection device to be rejected, associated with at least one of the plurality of apparatuses and (ii) collecting waveforms remaining after the removing as the learning data,

wherein the external inspection device determines whether to accept or reject a workpiece processed by the at least one of the plurality of apparatuses, and

the data organizer removes the waveforms associated with a workpiece processed by the at least one of the plurality of apparatuses that the external inspection device determines to reject.

2. The data analysis device according to claim 1 , wherein the data organizer (i) allocates, for each of the groups, a number of waveforms included in the corresponding group and (ii) when the allocated number of waveforms included in a group is equal to or smaller than a predetermined threshold value, removes the group from the learning data.

3. The data analysis device according to claim 2 , further comprising:

an auxiliary storage, wherein

the data learner determines a representative waveform for each group based on the waveforms included in the learning data and stores in the auxiliary storage the representative waveforms as a learning result.

4. The data analysis device according to claim 3 , wherein

the data collector collects a monitoring signal supplied from the apparatus that is a diagnosis target, and

the data analysis device further includes:

a data determiner to read the learning result from the auxiliary storage and to determine whether a waveform of the monitoring signal matches the waveforms included in the learning data; and

a response implementer to transmit, to the diagnosis target apparatus, based on a result of the determination by the data determiner, a signal to cause performance of predetermined processing.

5. The data analysis device according to claim 2 , wherein each of the apparatuses connected to the data analysis device is an apparatus that repeatedly performs processing, supplies the signal to the data collector repeatedly, and maintains a normal operating state thereof longer than an abnormal operating state thereof.

6. The data analysis device according to claim 1 , wherein the data organizer (i) allocates, for each of the groups, a number of waveforms included in the corresponding group and (ii) when the allocated number of waveforms included in a group is equal to or larger than a predetermined threshold value, generates the learning data by collecting the waveforms included in the group.

7. The data analysis device according to claim 6 , further comprising:

an auxiliary storage, wherein

the data learner determines a representative waveform for each group based on the waveforms included in the learning data and stores in the auxiliary storage the representative waveforms as a learning result.

8. The data analysis device according to claim 7 , wherein

the data collector collects a monitoring signal supplied from the apparatus that is a diagnosis target, and

the data analysis device further includes:

a data determiner to read the learning result from the auxiliary storage and to determine whether a waveform of the monitoring signal matches the waveforms included in the learning data; and

a response implementer to transmit, to the diagnosis target apparatus, based on a result of the determination by the data determiner, a signal to cause performance of predetermined processing.

9. The data analysis device according to claim 6 , wherein each of the apparatuses connected to the data analysis device is an apparatus that repeatedly performs processing, supplies the signal to the data collector repeatedly, and maintains a normal operating state thereof longer than an abnormal operating state thereof.

10. The data analysis device according to claim 1 , further comprising:

an auxiliary storage, wherein

the data learner determines a representative waveform for each group based on the waveforms included in the learning data and stores in the auxiliary storage the representative waveforms as a learning result.

11. The data analysis device according to claim 10 , wherein

the data collector collects a monitoring signal supplied from the apparatus that is a diagnosis target, and

the data analysis device further includes:

a data determiner to read the learning result from the auxiliary storage and to determine whether a waveform of the monitoring signal matches the waveforms included in the learning data; and

a response implementer to transmit, to the diagnosis target apparatus, based on a result of the determination by the data determiner, a signal to cause performance of predetermined processing.

12. The data analysis device according to claim 11 , wherein each of the apparatuses connected to the data analysis device is an apparatus that repeatedly performs processing, supplies the signal to the data collector repeatedly, and maintains a normal operating state thereof longer than an abnormal operating state thereof.

13. The data analysis device according to claim 10 , wherein each of the apparatuses connected to the data analysis device is an apparatus that repeatedly performs processing, supplies the signal to the data collector repeatedly, and maintains a normal operating state thereof longer than an abnormal operating state thereof.

14. The data analysis device according to claim 1 , wherein the apparatus connected to the data analysis device is an apparatus that repeatedly performs processing, supplies the signal to the data collector repeatedly, and maintains a normal operating state thereof longer than an abnormal operating state thereof.

15. The data analysis device according to claim 1 , wherein each of the apparatuses connected to the data analysis device is an apparatus that repeatedly performs processing, supplies the signal to the data collector repeatedly, and maintains a normal operating state thereof longer than an abnormal operating state thereof.

16. A system comprising:

a data learner to create waveforms including signals collected at predetermined intervals from a plurality of apparatuses and classified into groups of mutually-similar waveforms in accordance with a predetermined classification basis;

a data organizer to generate learning data by (i) removing waveforms, determined by an external inspection device to be rejected, associated with at least one of the plurality of apparatuses, and (ii) collecting waveforms remaining after the removing as the learning data; and

the data learner is further configured to determine a representative waveform based on the learning data generated by the data organizer and to store the representative waveform as a learning result,

wherein the external inspection device determines whether to accept or reject a workpiece processed by the at least one of the plurality of apparatuses, and

the data organizer removes the waveforms associated with a workpiece processed by the at least one of the plurality of apparatuses that the external inspection device determines to reject.

17. A method implemented by a computer, the method comprising:

collecting input signals from a plurality of apparatuses including a signal supplied when at least one of the apparatuses is in a normal operating state and a signal supplied when at least one of the apparatuses is in an abnormal operating state;

creating waveforms including signals cut out at predetermined intervals from the signals collected by the collecting;

classifying the waveforms into groups of mutually-similar waveforms in accordance with a predetermined classification basis;

generating learning data by (i) removing waveforms, determined by an external inspection device to be rejected, associated with at least one of the plurality of apparatuses and (ii) collecting waveforms remaining after the removing as the learning data; and

determining a representative waveform based on the learning data and storing the representative waveform as a learning result,

wherein the external inspection device determines whether to accept or reject a workpiece processed by the at least one of the plurality of apparatuses, and

the generating learning data removes the waveforms associated with a workpiece processed by the at least one of the plurality of apparatuses that the external inspection device determines to reject.

18. A non-transitory recording medium storing a program causing a computer to function as:

a data collector to collect signals supplied from a plurality of apparatuses, the signals including a signal supplied when at least one of the apparatuses is in a normal operating state and a signal supplied when at least one of the apparatuses is in an abnormal operating state;

a data learner to create waveforms including signals cut out at predetermined intervals from the signals collected by the data collector, the data learner being further configured to classify the waveforms into groups of mutually-similar waveforms in accordance with a predetermined classification basis; and

a data organizer to generate learning data by (i) removing waveforms, determined by an external inspection device to be rejected, associated with at least one of the plurality of apparatuses and (ii) collecting waveforms remaining after the removing as the learning data,

wherein the external inspection device determines whether to accept or reject a workpiece processed by the at least one of the plurality of apparatuses, and

the data organizer removes the waveforms associated with a workpiece processed by the at least one of the plurality of apparatuses that the external inspection device determines to reject.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 1, 2020
From: NASU, OSAMU
To: MITSUBISHI ELECTRIC CORPORATION
Reel/Frame 054498/0768 →
Continuity (1)
Related Publication 20210223764A1 · Jul 22, 2021
Cited By (1)
US 12,288,429