Granted Patent B2

US 12,553,870 · App. 17/688,228 · Granted Feb 17, 2026

Inventors: Kazuki Masuo (Tokyo, JP); Ryota Ichimura (Tokyo, JP); Hideyuki Mochizuki (Tokyo, JP)

Assignee: YOKOGAWA ELECTRIC CORPORATION

G01N30/86G01N30/62G06N20/00G01N2030/025

Abstract

Claims (29)

1 . An analysis apparatus comprising:

an extraction unit that extracts a baseline portion whose change is much smaller than that in a peak portion from chromatographic data that is output from a target apparatus of an analysis process; and

a determination unit that determines a normality in the target apparatus of the analysis process based on normal chromatographic data of the target apparatus of the analysis process and a feature of the baseline portion of the chromatographic data extracted by the extraction unit, wherein

the target apparatus is a Thermal Conductivity Detector,

the determination unit determines the target apparatus is abnormal when a difference of an inclination of the baseline portion between the normal chromatographic data and the chromatographic data acquired from the target apparatus exceeds a predetermined threshold value, and

when the determination unit determines the target apparatus is abnormal, the determination unit outputs an alarm or an error message.

2 . The analysis apparatus according to claim 1 , wherein

the determination unit determines the normality in the target apparatus based on the normal chromatographic data and an intensity of noise in the baseline portion as the feature of the baseline portion.

3 . The analysis apparatus according to claim 1 , wherein

the determination unit uses, as input data, the chromatographic data acquired from the target apparatus to determine the normality in the target apparatus by using a model on which learning is executed by using the normal chromatographic data as teacher data.

4 . The analysis apparatus according to claim 3 , wherein

the determination unit uses, as the input data, the chromatographic data acquired from the target apparatus to determine the normality in the target apparatus by using the model on which the learning is executed by using, as the teacher data, abnormal chromatographic data in addition to the normal chromatographic data or instead of the normal chromatographic data.

5 . The analysis apparatus according to claim 3 further comprising:

a learning unit that uses, as the input data, the chromatographic data acquired from the target apparatus to learn the model that outputs an evaluation value reflecting the normality in the target apparatus.

6 . The analysis apparatus according to claim 1 , wherein

the extraction unit extracts the baseline portion from the chromatographic data that is acquired from the target apparatus in operation by dividing a measurement sample of a multicomponent mixed gas into single components, and

the determination unit determines the normality in the target apparatus in operation based on the normal chromatographic data.

7 . An analysis method comprising:

extracting a baseline portion whose change is much smaller than that in a peak portion from chromatographic data output from a target apparatus of an analysis process; and

determining a normality in the target apparatus of the analysis process based on normal chromatographic data of the target apparatus of the analysis process and a feature of the baseline portion of the chromatographic data, wherein

the target apparatus is a Thermal Conductivity Detector,

the determining determines the target apparatus is abnormal when a difference of an inclination of the baseline portion between the normal chromatographic data and the chromatographic data acquired from the target apparatus exceeds a predetermined threshold value, and

when the determining determines the target apparatus is abnormal, the determining outputs an alarm or an error message.

8 . A non-transitory computer-readable recording medium having stored therein a control program configured to cause a computer to execute:

extracting a baseline portion whose change is much smaller than that in a peak portion from chromatographic data output from a target apparatus of an analysis process; and

determining a normality in the target apparatus of the analysis process based on normal chromatographic data of the target apparatus of the analysis process and a feature of the baseline portion of the chromatographic data, wherein

the target apparatus is a Thermal Conductivity Detector,

the determining determines the target apparatus is abnormal when a difference of an inclination of the baseline portion between the normal chromatographic data and the chromatographic data acquired from the target apparatus exceeds a predetermined threshold value, and

when the determining determines the target apparatus is abnormal, the determining outputs an alarm or an error message.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (21)