US-12619210-B2 - State-monitoring device and state-monitoring method

US12619210B2US 12619210 B2US12619210 B2US 12619210B2US-12619210-B2

Abstract

A diagnosis-parameter calculation unit calculates a diagnosis parameter in a second time period, based on sensing data obtained in the second time period. A characteristic-data input unit obtains characteristic data representing characteristics of a monitoring target device in a first time period and the second time period, from a logger included in the monitoring target device. Using the characteristic data in the second time period and the diagnosis parameter in the second time period, a regression-model generation unit) generates a regression model representing a relationship between the characteristic data and the diagnosis parameter. Using the regression model, a diagnosis-parameter estimation unit estimates a diagnosis parameter in the first time period, which corresponds to the characteristic data in the first time period. Using the diagnosis parameters in the first time period and the second time period, the diagnosis-criteria setting unit sets diagnosis criteria.

Inventors

Yota SHIMOYAMA

Assignees

NTN CORPORATION

Dates

Publication Date: 20260505
Application Date: 20210927
Priority Date: 20200930

Claims (8)

1 . A state-monitoring device connectable to a monitoring target device, wherein a first time period, a second time period, and a third time period are set, the first time period being a time period before the monitoring target device is connected to the state-monitoring device, the second time period being an adjustment period, after the monitoring target device is connected to the state-monitoring device, for diagnosing abnormality of the monitoring target device, the third time period being a time period, after the monitoring target device is connected to the state-monitoring device, during which abnormality of the monitoring target device is diagnosed, the state-monitoring device comprising: a sensing-data input unit that obtains, during the second time period and the third time period, sensing data representing a state of the monitoring target device from a sensor included in the monitoring target device; a diagnosis-parameter calculation unit that calculates a diagnosis parameter in the second time period and a diagnosis parameter in the third time period based respectively on the sensing data obtained during the second time period and on the sensing data obtained during the third time period; a characteristic-data input unit that obtains characteristic data representing characteristics of the monitoring target device in the first time period and the second time period from a logger included in the monitoring target device; a regression-model generation unit that generates, using the characteristic data in the second time period and the diagnosis parameter in the second time period, a regression model representing a relationship between the characteristic data and the diagnosis parameter; a diagnosis-parameter estimation unit that estimates, using the regression model, a diagnosis parameter in the first time period, which corresponds to the characteristic data in the first time period; a diagnosis-criteria setting unit that sets diagnosis criteria, using the diagnosis parameter in the first time period and the diagnosis parameter in the second time period; and a diagnostic unit that diagnoses abnormality of the monitoring target device, based on the diagnosis parameter in the third time period and the diagnosis criteria, wherein the monitoring target device is wind power generation equipment, and the sensor is a vibration sensor.
2 . The state-monitoring device according to claim 1 , further comprising a characteristic data selection unit that selects characteristic data whose correlation coefficient with the diagnosis parameter in the second time period is greater than or equal to a threshold, from among the characteristic data in a plurality of the second time periods, wherein the regression-model generation unit generates the regression model, using the selected characteristic data.
3 . The state-monitoring device according to claim 1 , further comprising a regression model evaluation unit that determines a score of the regression model, based on predetermined metrics, wherein acquisition of the sensing data, calculation of the diagnosis parameter, acquisition of the characteristic data, and generation of the regression model are repeated until the score of the regression model reaches a predetermined value.
4 . The state-monitoring device according to claim 1 , wherein the diagnosis parameter is a vibration root mean square, and the characteristic data is a rotational speed of a shaft.
5 . The state-monitoring device according to claim 1 , wherein the diagnosis-criteria setting unit creates diagnosis criteria which include a first threshold Th1, based on an average μ and a standard deviation σ of the diagnosis parameters in the first time period and the diagnosis parameters in the second time period, according to Equation: Th ⁢ 1 = µ + a × σ ( 1 ) where a is an any real number, and wherein the diagnostic unit determines that the monitoring target device is normal when the diagnosis parameter in the third time period is less than or equal to the first threshold Th1.
6 . The state-monitoring device according to claim 5 , wherein the diagnosis-criteria setting unit creates diagnosis criteria which include a second threshold Th2, based on the average μ and the standard deviation σ of the diagnosis parameters in the first time period and the diagnosis parameters in the second time period, according to Equation: Th ⁢ 2 = µ + a × σ × n ( 2 ) where a is an any real number and n is any constant greater than one, and wherein the diagnostic unit determines that the monitoring target device is in an inspection required state when the diagnosis parameter in the third time period is greater than the first threshold and less than or equal to the second threshold.
7 . The state-monitoring device according to claim 6 , wherein the diagnostic unit determines that the monitoring target device is in a pause required state when the diagnosis parameter in the third time period is greater than the second threshold.
8 . A state-monitoring method in a state-monitoring device connectable to a monitoring target device, wherein a first time period, a second time period, and a third time period are set, the first time period being a time period before the monitoring target device is connected to the state-monitoring device, the second time period being a preparation period, after the monitoring target device is connected to the state-monitoring device, for diagnosing abnormality of the monitoring target device, the third time period being a time period, after the monitoring target device is connected to the state-monitoring device, during which abnormality of the monitoring target device is diagnosed, the state-monitoring method comprising: obtaining, by the state-monitoring device, during the second time period, sensing data representing a state of the monitoring target device from a sensor included in the monitoring target device; calculating, by the state-monitoring device, a diagnosis parameter in the second time period based on the sensing data obtained during the second time period; obtaining, by the state-monitoring device, characteristic data representing characteristics of the monitoring target device in the first time period and the second time period from a logger included in the monitoring target device; generating, by the state-monitoring device, using the characteristic data in the second time period and the diagnosis parameter in the second time period, a regression model representing a relationship between the characteristic data and the diagnosis parameter; estimating, by the state-monitoring device, using the regression model, a diagnosis parameter in the first time period, which corresponds to the characteristic data in the first time period; setting, by the state-monitoring device, diagnosis criteria, using the diagnosis parameter in the first time period and the diagnosis parameter in the second time period; and obtaining, by the state-monitoring device, during the third time period, the sensing data representing the state of the monitoring target device from the sensor included in the monitoring target device; calculating, by the state-monitoring device, a diagnosis parameter in the third time period, based on the sensing data obtained in the third time period; and diagnosing, by the state-monitoring device, abnormality of the monitoring target device, based on the diagnosis parameter in the third time period and the diagnosis criteria, wherein the monitoring target device is wind power generation equipment, and the sensor is a vibration sensor.

Description

CROSS-REFERENCE OF RELATED APPLICATIONS This application is the U.S. National Phase under 35 U.S.C. § 371 of International Patent Application No. PCT/JP2021/035274, filed on Sep. 27, 2021, which in turn claims the benefit of Japanese Patent Application No. 2020-165518, filed on Sep. 30, 2020, the entire disclosures of which Applications are incorporated by reference herein. TECHNICAL FIELD The present invention relates to a state-monitoring device and a state-monitoring method, and, more particularly, to a state-monitoring device and a state-monitoring method for monitoring the state of rotating equipment, for example. BACKGROUND ART Conventionally, an apparatus is known which sets diagnosis criteria based on measurement data representing a state of equipment, and diagnoses the state of the equipment according to the diagnosis criteria. For example, the state-monitoring system disclosed in PTL 1 diagnoses abnormality of equipment provided in a wind power generator, the state-monitoring system including: a monitor device which includes multiple sensors for measuring measurement data indicating the state of the equipment and operating conditions data indicating operating conditions of the wind power generator and calculates a diagnosis parameter from the measurement data; a monitoring-side control device which includes a storage unit storing information, sets a first threshold for diagnosing the abnormality of the equipment, and diagnoses the abnormality of the equipment based on the first threshold; and a monitoring terminal device which includes a display unit for displaying information, and monitors the state of the equipment. The monitoring-side control device is coupled to the monitor device and the monitoring terminal device by communications lines. The monitoring terminal device sequentially sets a first time period for collecting underlying data, a second time period for setting the first threshold, and a third time period for diagnosing the abnormality of the equipment. During the first time period, the measurement data, operating conditions data, and diagnosis parameter measured and calculated by the monitor device are stored into the storage unit and displayed on the display unit, and diagnostic operating conditions are set using the monitoring terminal device. During the second time period, the diagnosis parameter, calculated from the measurement data measured when the operating conditions for the wind power generator meet the diagnostic operating conditions, is stored into the storage unit, and the monitoring-side control device generates the first threshold based on the diagnosis parameter stored in the storage unit. During the third time period, the diagnosis parameter, calculated from the measurement data measured when the operating conditions for the wind power generator meet the diagnostic operating conditions, is stored into the storage unit, and the monitoring-side control device compares the diagnosis parameter stored in the storage unit with the first threshold, determines that the equipment is subject to abnormality if the diagnosis parameter is greater than the first threshold, and a result of the determination is stored into the storage unit and displayed on the display unit. CITATION LIST Patent Literature PTL 1: Japanese Patent Laying-Open No. 2013-185507 SUMMARY OF INVENTION Technical Problem PTL 1 requires data in order to set the diagnosis criteria. The abnormality of the equipment cannot be diagnosed during an adjustment period in which the state-monitoring device is set up and the diagnosis criteria are set based on the data. Accordingly, a short adjustment period is preferable. However, if the state of the monitoring target device in the operational period differs from that in the adjustment period, the state-monitoring system may determine that the monitoring target device is subject to abnormality even though the monitoring target device is in a normal state. Thus, it is necessary to extend the adjustment period so that the diagnosis criteria can be set base on as many states as possible of the monitoring target device. For the reasons stated above, conventionally, the adjustment period and the diagnostic accuracy are in trade off. Therefore, an object of the present invention is to provide a state-monitoring device and a state-monitoring method which require a short adjustment period and are capable of accurate diagnosis. Solution to Problem The present invention is a state-monitoring device connectable to a monitoring target device. A first time period, a second time period, and a third time period are set, the first time period being a time period before the monitoring target device is connected to the state-monitoring device, the second time period being a preparation period, after the monitoring target device is connected to the state-monitoring device, for diagnosing abnormality of the monitoring target device, the third time period being a time period, after the monit