EP-4741794-A1 - INFORMATION PROCESSING METHOD AND INFORMATION PROCESSING DEVICE

EP4741794A1EP 4741794 A1EP4741794 A1EP 4741794A1EP-4741794-A1

Abstract

To provide a technique capable of easily monitoring the state of an object. For example, the information processing apparatus 40 according to one embodiment of the present disclosure acquires a time-series measurement signal of the acoustic emission of the mechanical seal 52D, calculates the energy of the acoustic emission based on the measurement signal, extracts a time-series signal of the periodicity-removed component obtained by removing the periodicity component from the time-series signal of the calculated energy or a time-series signal of the periodicity component, and calculates the standard deviation σ2 of the extracted time-series signal of the periodicity-removed component or the standard deviation σ1 of the time-series signal of the periodicity component.

Inventors

SUDO, Katsutaka
SUMINAKA, Yukiko
HAMASAKI, Junichi
OKAYAMA, Hideto
SUGIMOTO, Iori
NISHIMOTO, SHIGETO

Assignees

Daikin Industries, Ltd.

Dates

Publication Date: 20260513
Application Date: 20240702

Claims (15)

An information processing method comprising: an acquisition step of acquiring, by an information processing apparatus, a time-series measurement signal of acoustic emission of an object; an energy calculation step of calculating, by the information processing apparatus, energy of the acoustic emission of the object based on the measurement signal; a first extraction step of extracting, by the information processing apparatus, either a time-series signal of a first component obtained by removing a second component having periodicity from a time-series signal of the energy, or a time-series signal of the second component; and a variation calculation step of calculating, by the information processing apparatus, an index value relating to variation of the time-series signal of the first component or the time-series signal of the second component.
The information processing method according to claim 1, further comprising: a second extraction step of extracting, by the information processing apparatus, the time-series signal of the second component from the time-series signal of the energy, wherein in the first extraction step, the time-series signal of the first component is extracted, and wherein in the variation calculation step, index values relating to variation of both the time-series signal of the first component and the time-series signal of the second component are calculated.
The information processing method according to claim 1, further comprising: a diagnosis step of diagnosing, by the information processing apparatus, presence or absence of a sign of failure of the object mounted on a predetermined apparatus, based on a trend of a time-dependent change in history of the index value relating to variation of either the time-series signal of the first component or the time-series signal of the second component.
The information processing method according to claim 2, further comprising: a diagnosis step of diagnosing, by the information processing apparatus, presence or absence of a sign of failure of the object mounted on a predetermined apparatus, based on trends of time-dependent changes in history of the index values relating to variation of both the time-series signal of the first component and the time-series signal of the second component.
The information processing method according to claim 3, wherein in the diagnosis step, the object mounted on the predetermined apparatus is diagnosed as exhibiting a sign of failure when the history of the index value relating to variation of the time-series signal of the second component transitions from an increasing trend with respect to elapsed time to a decreasing trend with respect to elapsed time.
The information processing method according to claim 3, wherein in the diagnosis step, the object mounted on the predetermined apparatus is diagnosed as exhibiting a sign of failure when the history of the index value relating to variation of the time-series signal of the first component transitions from a state in which change with respect to elapsed time is relatively small to a state in which increase with respect to elapsed time is relatively large.
The information processing method according to claim 4, wherein in the diagnosis step, the object mounted on the predetermined apparatus is diagnosed as exhibiting a sign of failure when the history of the index value relating to variation of the time-series signal of the second component transitions from an increasing trend with respect to elapsed time to a decreasing trend with respect to elapsed time, and subsequently the history of the index value relating to variation of the time-series signal of the first component transitions from a state in which change with respect to elapsed time is relatively small to a state in which increase with respect to elapsed time is relatively large.
The information processing method according to claim 1, further comprising: a setting step of setting, by the information processing apparatus, a reference value for determining presence or absence of a sign of failure of the object mounted on a predetermined apparatus, wherein in the acquisition step, a plurality of measurement signals of acoustic emission of the object mounted on the predetermined apparatus are acquired, the plurality of measurement signals including both measurement signals in a situation before occurrence of failure of the object and measurement signals in a situation after occurrence of failure of the object, wherein in the energy calculation step, the energy corresponding to each of the plurality of measurement signals acquired in the acquisition step is calculated, wherein in the first extraction step, the time-series signal of the first component is extracted from the time-series signal of the energy corresponding to each of the plurality of measurement signals, wherein in the variation calculation step, the index value relating to variation of the time-series signal of the first component corresponding to each of the plurality of measurement signals is calculated, and wherein in the setting step, a first reference value for the index value relating to variation of the time-series signal of the first component is set, based on a distribution of the index values relating to variation of the time-series signal of the first component corresponding to the measurement signals in the situation before occurrence of failure of the object and a distribution of the index values relating to variation of the time-series signal of the first component corresponding to the measurement signals in the situation after occurrence of failure of the object, among the index values relating to variation of the time-series signal of the first component corresponding to each of the plurality of measurement signals.
The information processing method according to claim 2, further comprising: a setting step of setting, by the information processing apparatus, reference values for determining presence or absence of a sign of failure of the object mounted on a predetermined apparatus, wherein in the acquisition step, a plurality of measurement signals of acoustic emission of the object mounted on the predetermined apparatus are acquired, the plurality of measurement signals including both measurement signals in a situation before occurrence of failure of the object and measurement signals in a situation after occurrence of failure of the object, wherein in the energy calculation step, the energy corresponding to each of the plurality of measurement signals acquired in the acquisition step is calculated, wherein in the first extraction step, the time-series signal of the first component is extracted from the time-series signal of the energy corresponding to each of the plurality of measurement signals, wherein in the second extraction step, the time-series signal of the second component obtained by removing the first component from the time-series signal of the energy corresponding to each of the plurality of measurement signals is extracted, wherein in the variation calculation step, the index value relating to variation of the time-series signal of the first component corresponding to each of the plurality of measurement signals is calculated, and the index value relating to variation of the time-series signal of the second component corresponding to each of the plurality of measurement signals is calculated, and wherein in the setting step, a first reference value for the index value relating to variation of the time-series signal of the first component and a second reference value for the index value relating to variation of the time-series signal of the second component are set, based on distributions of the combinations of the index values relating to variation of the time-series signals of the first component and the second component corresponding to the measurement signals in the situation before occurrence of failure of the object and distributions of the combinations of the index values relating to variation of the time-series signals of the first component and the second component corresponding to the measurement signals in the situation after occurrence of failure of the object, among combinations of the index values of the time-series signals of both the first component and the second component corresponding to each of the plurality of measurement signals.
The information processing method according to claim 8, wherein the plurality of measurement signals are a collection of measurement signals at different timings for a same predetermined apparatus, and wherein in the setting step, the first reference value is set, based on the index value relating to variation of a signal of the first component corresponding to the measurement signal at a timing obtained by tracing back a predetermined time from a timing of a first measurement signal recorded after occurrence of failure of the object.
The information processing method according to claim 9, wherein the plurality of measurement signals are a collection of measurement signals at different timings for a same predetermined apparatus, and wherein in the setting step, the first reference value is set, based on the index value relating to variation of a signal of the first component corresponding to the measurement signal at a timing obtained by tracing back a predetermined time from a timing of a first measurement signal recorded after occurrence of failure of the object, and the second reference value is set, based on the index value relating to variation of a signal of the second component corresponding to the first measurement signal recorded after occurrence of failure of the object.
The information processing method according to any one of claims 8 to 11, wherein in the setting step, the reference value for determining presence or absence of a sign of failure of the object is set for each group of a plurality of predetermined apparatuses, the groups being classified according to operating conditions of the object.
The information processing method according to any one of claims 3 to 12, wherein the predetermined apparatus is a polymerization reactor, and the object is a mechanical seal for shaft sealing.
The information processing method according to claim 13, wherein in the acquisition step, a measurement signal recorded at a timing before a process reaction in the polymerization reactor is acquired.
An information processing apparatus comprising: an acquisition unit configured to acquire a time-series measurement signal of acoustic emission of an object; an energy calculation unit configured to calculate energy of the acoustic emission based on the measurement signal; an extraction unit configured to extract either a time-series signal of a first component obtained by removing a second component having periodicity from a time-series energy signal, or a time-series signal of the second component; and a variation calculation unit configured to calculate an index value relating to variation of the time-series signal of the first component or the time-series signal of the second component.

Description

Technical Field The present disclosure relates to an information processing method, and the like. Background Art For example, there is a known technique for measuring high-frequency vibrations of an object and monitoring the state of the object based on the measurement results (see, for example, Patent Document 1). Citation List Patent Document [Patent Document 1] Japanese Examined Patent Publication No. H7-11466 Summary of Invention Technical Problem However, Patent Document 1 requires the use of the maximum value, minimum value, average value, and variance of the effective value of high-frequency vibration per unit time. Accordingly, the number of parameters required for determining the state of the object is relatively large, which may complicate the setting of determination criteria. The present disclosure is intended to provide a technique that enables easy monitoring of the state of an object. Solution to Problem According to a first aspect of the present disclosure, an information processing method is provided. The information processing method includes: an acquisition step of acquiring, by an information processing apparatus, a time-series measurement signal of acoustic emission of an object;an energy calculation step of calculating, by the information processing apparatus, energy of the acoustic emission of the object based on the measurement signal;a first extraction step of extracting, by the information processing apparatus, either a time-series signal of a first component obtained by removing a second component having periodicity from a time-series signal of the energy, or a time-series signal of the second component; anda variation calculation step of calculating, by the information processing apparatus, an index value relating to variation of the time-series signal of the first component or the time-series signal of the second component. According to this embodiment, the information processing apparatus can calculate an index value relating to variation of either a time-series signal of a first component or a time-series signal of a second component of the energy of the acoustic emission of the object. Therefore, for example, the information processing apparatus can relatively easily monitor the state of the object based on the index value. In addition, for example, the information processing apparatus can set a reference value for the index value, thereby enabling easier monitoring of the state of the object based on a comparison between the index value and the reference value. Further, in a second aspect of the present disclosure, based on the first aspect described above, the information processing method may further include: a second extraction step of extracting, by the information processing apparatus, the time-series signal of the second component from the time-series signal of the energy,wherein in the first extraction step, the time-series signal of the first component is extracted, andwherein in the variation calculation step, index values relating to variation of both the time-series signal of the first component and the time-series signal of the second component are calculated. Further, in a third aspect of the present disclosure, based on the first aspect described above, the information processing method may further include: a diagnosis step of diagnosing, by the information processing apparatus, presence or absence of a sign of failure of the object mounted on a predetermined apparatus, based on a trend of a time-dependent change in history of the index value relating to variation of either the time-series signal of the first component or the time-series signal of the second component. Further, in a fourth aspect of the present disclosure, based on the second aspect described above, the information processing method may further include: a diagnosis step of diagnosing, by the information processing apparatus, presence or absence of a sign of failure of the object mounted on a predetermined apparatus, based on trends of time-dependent changes in history of the index values relating to variation of both the time-series signal of the first component and the time-series signal of the second component. Further, in a fifth aspect of the present disclosure, based on the third aspect described above, in the diagnosis step, the object mounted on the predetermined apparatus may be diagnosed as exhibiting a sign of failure when the history of the index value relating to variation of the time-series signal of the second component transitions from an increasing trend with respect to elapsed time to a decreasing trend with respect to elapsed time. Further, in a sixth aspect of the present disclosure, based on the third aspect described above, in the diagnosis step, the object mounted on the predetermined apparatus is diagnosed as exhibiting a sign of failure when the history of the index value relating to variation of the time-series signal of the first component transitions from a state in w