IP Library Granted Patent US 12,422,839
Granted Patent B2
US 12,422,839 · App. 18/015,846 · Granted Sep 23, 2025

Cause estimation apparatus and cause estimation method

Inventors: Shunsaku Matsumoto (Tokyo, JP); Kuniaki Aoyama (Hyogo, JP); Yasuyuki Ogawa (Tokyo, JP); Takashi Sonoda (Tokyo, JP); Daisuke Kudo (Tokyo, JP)
Assignee: MITSUBISHI HEAVY INDUSTRIES, LTD.
G05B23/0275
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 12,422,839
App. No.
18/015,846
Granted
Sep 23, 2025
Kind
B2
Abstract

A cause estimation device includes: an acquisition unit that acquires a measurement value of a target apparatus; a likelihood calculation unit that calculates, in a case where it is assumed that events occur due to a common cause, a likelihood of an occurrence of each of the events based on the measurement value; a table storage unit that stores a cause table in which a cause of occurrence of the event and a frequency of the cause of occurrence are associated with each other for each of the plurality of events; and an estimation unit that rewrites the frequency registered in the cause table into a frequency in the case where it is assumed that the plurality of events occur due to the common cause, and estimates the cause of occurrence based on the rewritten cause table and the likelihood.

Claims (41)

1. A cause estimation device comprising:

an input/output interface;

a storage device;

a central processing unit (CPU) that:

acquires a measurement value of a target apparatus;

calculates, in a case where it is assumed that a plurality of events that occur in the target apparatus occur due to a common cause, a likelihood of an occurrence of each of the events based on the measurement value that has been acquired;

stores, in the storage device, a cause table in which a cause of occurrence of the event and a frequency of the cause of occurrence are associated with each other for each of the events;

rewrites the frequency registered in the cause table into a frequency in the case where it is assumed that the events occur due to the common cause, and estimates the cause of occurrence based on the rewritten cause table and the likelihood;

receives setting of a predetermined noise value for an item having a frequency of 0 set in the cause table, and updates the frequency registered in the cause table based on the setting; and

outputs, via the input/output interface, the cause of occurrence that has been estimated in order of the likelihood.

2. The cause estimation device according to claim 1 , wherein

the CPU rearranges the events in descending order of the likelihood, and calculates the likelihood in the case where it is assumed that the events occur due to the common cause by setting a value obtained by subtracting, from the likelihood corresponding to a first event, the likelihood corresponding to the event one rank lower than the first event in the rearranged order, as the likelihood corresponding to the first event.

3. The cause estimation device according to claim 1 , wherein

the CPU rearranges the cause table for the events in the same order as when the events are rearranged in descending order of the likelihood, and

the cause table is rewritten by setting a value obtained by subtracting, from the frequency for each cause of occurrence corresponding to a first event, the frequency for each cause of occurrence corresponding to the event one rank lower than the first event in the rearranged order, as the frequency for each cause of occurrence corresponding to the first event.

4. The cause estimation device according to claim 1 , wherein

in the cause table, the cause of occurrence, the event, and the number of times the event is confirmed when an anomaly related to the cause of occurrence occurs are associated with each other, and

the CPU calculates a weighted sum of the likelihood and the number of times for each cause of occurrence, and estimates the cause of occurrence based on the weighted sum.

5. The cause estimation device according to claim 1 , wherein

in the cause table, the cause of occurrence, the event, and a probability that the event is confirmed when an anomaly related to the cause of occurrence occurs are associated with each other, and

the CPU calculates a weighted sum of the likelihood and the probability for each cause of occurrence, and estimates the cause of occurrence based on the weighted sum.

6. The cause estimation device according to claim 1 , wherein

the case where it is assumed that the events occur due to the common cause is a case in which, in a case where the event is represented by a closed space having an area corresponding to a magnitude of the likelihood that the event occurs, for each of the events, a relationship in which the closed space of the event is included in a closed space of the event having the next smallest likelihood after the event is established.

7. The cause estimation device according to claim 1 , wherein

the CPU receives information for deleting the cause set in the cause table and sets the frequency related to the cause to be deleted to 0 based on the information.

8. The cause estimation device according to claim 1 , wherein

the CPU determines, based on the measurement value, to delete the cause set in the cause table related to the measurement value in a case where the measurement value does not change by a predetermined threshold or more, and sets the frequency related to the cause to be deleted to 0.

9. A cause estimation method using a cause estimation device that comprises an input/output interface, a storage device, and a central processing unit (CPU), the method comprising:

acquiring a measurement value of a target apparatus;

calculating, in a case where it is assumed that a plurality of events that occur in the target apparatus occur due to a common cause, a likelihood of an occurrence of each of the events based on the acquired measurement value; and

storing, in the storage device, a cause table in which a cause of occurrence of the event and a frequency of the cause of occurrence are associated with each other for each of the events;

rewriting the frequency registered in the cause table into a frequency in the case where it is assumed that the events occur due to the common cause, and estimating the cause of occurrence based on the rewritten cause table and the likelihood;

receiving setting of a predetermined noise value for an item having a frequency of 0 set in the cause table, and updates the frequency registered in the cause table based on the setting; and

outputting, via the input/output interface, the cause of occurrence that has been estimated in order of the likelihood.

10. A non-transitory computer-readable storage medium storing a program that causes a cause estimation device that comprises an input/output interface, a storage device, and a central processing unit (CPU) to execute:

acquiring a measurement value of a target apparatus;

calculating, in a case where it is assumed that a plurality of events that may occur in the target apparatus occur due to a common cause, a likelihood of an occurrence of each of the events based on the acquired measurement value; and

storing, in the storage device, a cause table in which a cause of occurrence of the event and a frequency of the cause of occurrence are associated with each other for each of the events;

rewriting the frequency registered in the cause table into a frequency in the case where it is assumed that the events occur due to the common cause, and estimating the cause of occurrence based on the rewritten cause table and the likelihood;

receiving setting of a predetermined noise value for an item having a frequency of 0 set in the cause table, and updating the frequency registered in the cause table based on the setting; and

outputting, via the input/output interface, the cause of occurrence that has been estimated in order of the likelihood.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 25, 2024
From: MATSUMOTO, SHUNSAKU; AOYAMA, KUNIAKI; OGAWA, YASUYUKI; SONODA, TAKASHI; KUDO, DAISUKE
To: MITSUBISHI HEAVY INDUSTRIES, LTD.
Reel/Frame 066251/0314 →
Priority Claims (1)
JP 2020-176531 · Oct 21, 2020 · national
Continuity (1)
Related Publication 20230259116A1 · Aug 17, 2023
References Cited (7)
US 20050114743A1 · Moorhouse · 2005 [cited by examiner]
US 20100198555A1 · Takahama · 2010 [cited by examiner]
US 20190018402A1 · Enomoto · 2019 [cited by examiner]
US 20190187676A1 · Nagano et al. · 2019 [cited by applicant]
JP 2013199804A · 2013 [cited by applicant]
WO 2017209167A1 · 2017 [cited by applicant]
International Search Report issued in Application No. PCT/JP2021/037894 mailed on Dec. 21, 2021 (4 pages). [cited by applicant]