KR20160006281A - Monitoring method and system - Google Patents

Monitoring method and system Download PDF

Info

Publication number
KR20160006281A
KR20160006281A KR1020140084941A KR20140084941A KR20160006281A KR 20160006281 A KR20160006281 A KR 20160006281A KR 1020140084941 A KR1020140084941 A KR 1020140084941A KR 20140084941 A KR20140084941 A KR 20140084941A KR 20160006281 A KR20160006281 A KR 20160006281A
Authority
KR
South Korea
Prior art keywords
database
signal
target system
dimensional
value
Prior art date
Application number
KR1020140084941A
Other languages
Korean (ko)
Other versions
KR101616072B1 (en
Inventor
임종순
Original Assignee
주식회사 글로비즈
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 주식회사 글로비즈 filed Critical 주식회사 글로비즈
Priority to KR1020140084941A priority Critical patent/KR101616072B1/en
Priority to PCT/KR2014/006139 priority patent/WO2015005663A1/en
Priority to US14/897,006 priority patent/US10281909B2/en
Publication of KR20160006281A publication Critical patent/KR20160006281A/en
Application granted granted Critical
Publication of KR101616072B1 publication Critical patent/KR101616072B1/en

Links

Images

Abstract

The present invention relates to a monitoring method and system capable of accurately monitoring and diagnosing a state of a target mechanical system in various use environments by appropriately classifying signals measured in a target system operated under complex operating conditions according to operating conditions. According to an embodiment of the present invention, there is provided a method of monitoring a target system, the method comprising: classifying a signal measured by the target system into an analysis value and a frequency for an n-dimensional (n? And verifying the behavior of the target system by expressing the analysis value and the frequency of the signal measured in the target system on an n-dimensional (n? 3 integer) parameter using a bubble chart .

Figure P1020140084941

Description

Monitoring method and system

The present invention relates to a method and system for monitoring a mechanical system, for example, for performing a predetermined purpose, and more particularly to a method and system for monitoring a mechanical system, And more particularly, to a monitoring method and system capable of accurately monitoring and diagnosing a state of a target system in various use environments.

In general, mechanical systems such as wind turbines operate under complex operating conditions.

Accordingly, in the case of a wind turbine, which is an example of a mechanical system, it is necessary to classify the measurement (measurement) signal simultaneously considering the rotational speed of the driving system, external wind speed,

Conventionally, in the case of a wind turbine generator, most of the power generation power is considered in order to monitor the driving condition of the wind turbine generator, so that a signal that is measured (measured) by a large factor of the external wind speed is sometimes misdiagnosed as a failure.

In order to prevent the above-described false diagnosis and improve the reliability of the diagnosis, multidimensional signal monitoring capable of simultaneously considering various operating states and building a database therefor are required.

The types of the target system and the parameters related to the measurement signal or the measurement signal according to the multidimensional operation state may be as follows.

First, the parameters affecting the behavior of the gear box and the bearing of the wind turbine as described above include rotor speed, external wind speed and generated power; Second, the number of revolutions, radial load (acceleration) and axial load (acceleration) of the hub bearings of automobiles and railway cars; Third, there are rotor revolutions of large generators (thermal and nuclear), steam temperature / pressure and generated power; Fourth, there may be the number of revolutions and the radial load (acceleration) of the rolling mill.

In such a target system, it is an important issue to improve the accuracy of the diagnosis of the driveline system in which a relatively large number of current status monitoring devices are attached.

In the above-described object system, in the conventional embodiment, once the parameter for the operation condition is selected, a knowledge database is first constructed as an expert system for diagnosis in general.

That is, a signal for the operating state of the target system in a healthy state without a failure is measured, and a database is constructed for a parameter selected as a physical quantity having a physical meaning.

However, according to the existing signal classification method, since the measurement signal value is classified based on the one-dimensional parameter, it is often difficult to accurately identify the monitoring stability and the cause of the diagnosis.

For example, in the case of a conventional wind turbine, the root-mean-square (RMS) value of the vibration signal of the drivetrain measured for monitoring the condition of the gear teeth and the bearing is divided by a predetermined stage . However, in this case, when the external wind speed suddenly becomes high or a gust of wind blows locally, but the generated power is not high, a relatively large value is measured in the monitoring device, which can be mistaken for the damage to the target system.

If this false diagnosis is repeated, the operator may doubt the reliability of the monitoring system itself, and in a severe case, the diagnosis result may be ignored.

If users, operators and administrators disbelieve about the monitoring system, they may not be prepared even if the actual target system fails, or the monitoring system may be turned off and not used.

Therefore, it is important to maximize the reliability of the diagnosis of the monitoring system.

1. Korean Registered Patent No. 10-0522342 (Registration date 2005.10.11)

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a monitoring method and system capable of accurately diagnosing a fault by monitoring the state of a target system (e.g., a wind turbine generator, etc.) in various use environments.

Another problem to be solved by the present invention is to analyze a signal waveform in a healthy state measured in a target system, to build a database by classifying the operating state of the target system in a multidimensional space, And to provide a monitoring method and system capable of selecting a judgment criterion.

Another problem to be solved by the present invention is to present a signal waveform analysis value and a distribution in a space in the case of a multidimensional (e.g., three-dimensional) operation parameter, And a method of upgrading a database to provide a precise selection of a reference value for diagnosis and a monitoring method and system that can accurately diagnose the diagnosis based on the selection.

According to an aspect of the present invention, there is provided a method of monitoring a target system, the method comprising: classifying signals measured by the target system into analysis values and frequency of n-dimensional parameters ; And verifying the behavior of the target system by expressing the analysis value and the frequency of the signal measured in the target system on an n-dimensional (n? 3 integer) parameter using a bubble chart .

Expressing the analysis value and frequency in a three-dimensional bubble chart with a multi-layered two-dimensional bubble chart or a three-dimensional surface color chart; Forming a two-dimensional bubble chart or a three-dimensional surface color chart by shortening an operation parameter less than a set value in the three-dimensional bubble chart to a set value; And N (N > = 3 integers) reference planes, and reducing the analysis values and frequencies to at least one parameter.

Constructing the database by classifying the analyzed value and the frequency of the measured signal for the first set period in the steady state condition with the n-dimensional (n? 3 integer) parameter; Setting an alarm or alarm signal value of n-dimensional (n? 3 integer) classification type requiring warning or alarm based on data of the database classified and constructed during the set period in the steady state condition; And diagnosing the target system by comparing the currently measured signal with the warning or alarm signal value.

Constructing the database by classifying the analyzed values and frequencies of the measured signals for the second set period in the steady state condition with the n dimension (n? 3 integer) parameters; And comparing the database established during the first setting period with the database constructed during the second setting period to generate a warning when the magnitude or rate of change of the analysis value established in each database exceeds the set size .

According to another aspect of the present invention, there is provided a system for monitoring a target system, comprising: a sensor for measuring a behavior signal of the target system; A database configured to classify signals in the normal and abnormal states of the target system; An inference engine for analyzing a signal measured based on the data of the database to determine a normal or abnormal signal; A description module for describing the result of the inference engine; And a database building module for building the database, wherein the database building module is operable by a set of programs for performing a signal monitoring method of a target system according to an embodiment of the present invention.

As described above, according to the embodiment of the present invention, signals measured in a target system operated under complex operating conditions are appropriately classified according to various operating conditions and converted into a database, thereby monitoring measurement (measurement) You can accurately monitor and diagnose the status of the target system in your environment.

1 is a block diagram showing a monitoring system of a target system according to an embodiment of the present invention.
2 is a flowchart of a method of monitoring a target system according to an embodiment of the present invention.
FIGS. 3 to 9 are graphs for explaining the operation of the monitoring method and system of the target system according to the embodiment of the present invention.
10 is a schematic block diagram of a signal measurement and diagnosis system to which a monitoring method and system according to an embodiment of the present invention can be applied.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. However, the present invention is not limited to the embodiments described herein but may be embodied in other forms.

Throughout the specification, when a section includes a constituent element, it is understood that it can include other constituent elements, not excluding other constituent elements unless specifically stated otherwise. Like numbers refer to like elements throughout the specification.

1 is a block diagram schematically illustrating a signal monitoring system of a target system according to an embodiment of the present invention.

A signal monitoring system of a target system according to an embodiment of the present invention includes a sensor 710 for measuring a behavior signal of the target system 10; A database 730 constructed by classifying signals in the normal and abnormal states of the target system 10; An inference engine 740 for analyzing the measured signals based on the data of the database 730 and determining normal or abnormal signals; A description module 750 for explaining the result of the inference engine 740; And a database building module 720 for building the database 730.

The sensor 710 may be any sensor capable of measuring a behavior signal of the target system 10. For example, the sensor 710 may include a temperature sensor, a rotation speed sensor, a speed sensor, and the like.

In addition, the sensor 710 may be configured to sense the signals disclosed in Application No. 10-2013-0081045 (name: Signal Measurement and Diagnosis System and Method) filed by the applicant of the present invention. That is, the sensor 710 may detect the signals as shown in FIGS. 10 and 2 and input the signals to the analog / digital converter 100 and the pre-signal processing device 200. 10 is a diagram disclosed in the above application No. 10-2013-0081045 (entitled "Signal Measurement and Diagnosis System and Method").

Although not shown in FIG. 1, the analog / digital converter 100 and the pre-signal processing device 200 shown in FIGS. 2 and 10 can be included in the monitoring system 700 or the database building module 720 have.

The database 730 can be constructed by classifying signals in normal and abnormal states. That is, the database 730 is constructed by dividing a signal in a safe state or a signal in a case where a problem occurs, with respect to a signal measured or measured by the sensor 710. The reason why the database 730 is constructed as described above is that the signals measured in the target system 10 operated under a complex operating condition are appropriately classified according to the operating conditions so that the state of the target system 10 can be accurately Monitoring and diagnosing faults.

The reasoning engine 740 and the description module 750 may be composed of a combination of hardware and software, and may be integrated into a database building module 720 described later.

The database building module 720 may include at least one microprocessor and a microprocessor operating according to a set program. The set program may include a series of commands for performing a monitoring method according to an embodiment of the present invention As shown in FIG.

The database building module 720 may be included in the embedded local server 300 shown in FIG. 10 or may include the embedded local server 300, but the protection scope of the present invention is not limited thereto It should not be construed as necessarily limited thereto. The technical idea of the present invention can be applied to a configuration in which a substantial database 730 can be constructed even if the configuration is different.

The database building module 720 according to the embodiment of the present invention analyzes the signal waveforms of the healthy state measured by the target system 10 and classifies the signal waveforms in a space representing the operation state of the target system 10 in a multidimensional manner, (730), and a criterion for determining a failure can be selected on the basis thereof.

In particular, for example, in the case of a three-dimensional operation parameter, the database building module 720 expresses signal waveform analysis values and distributions in space and uses them to analyze trends in spatial values, trends (slopes), and By upgrading the database, it is possible to precisely select the reference value for diagnosis and make an accurate diagnosis when diagnosing it.

In addition, the monitoring system 700 can degenerate the database in two dimensions based on three-dimensional parameters.

Hereinafter, a monitoring method of a target system according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

2 is a flowchart illustrating a method of monitoring a target system according to an embodiment of the present invention.

Referring to FIG. 2, the monitoring system 700 including the database building module 720 detects the behavior of the target system 10 through the sensor 710. That is, the monitoring system 700 receives the signal of the target system 10 measured or measured by the sensor 710.

In the embodiment of the present invention, the monitoring system 700 can receive signals processed by the A / D converter 100 and the pre-signal processing device 200, and in this embodiment, 10-2013-0081045.

The database building module 720 of the monitoring system 700 may be configured to initially build the database 730 in the steady state conditions, i.e., in an initial period (during the first set time) The time T is set in steps S101, S110, S112, and S121 shown in FIG. 2 in order to construct the database 730 according to the frequency or the frequency. When the corresponding conditions by the time setting related steps are satisfied, the monitoring system 700 constructs the database 730 according to the initial section multi-dimensional measurement signal analysis value and / or frequency (S113) And a trend change rate reference value Tvref are set (S114).

The trend analysis reference value Tref and the trend change rate reference value Tvref set in S114 are used to calculate the tendency Tr of the measurement signal value and the tendency change rate Tv performed in S122.

The monitoring system 700 receives the data of the database 730 classified and constructed during the first setting period under the steady state condition through the database building module 720 in an n dimension (n? 3 The current measurement signal analysis value Cv of the multidimensional form is set to the alarm signal value Wv or the alarm signal value Av To diagnose the target system (S140, S131).

If the multivariate current measurement signal analysis value Cv is greater than the warning signal value Wv or the alarm signal value Av as a result of the diagnosis, a warning or alarm report is generated at step S132. The current measured signal Cv may be defined as a multi-dimensional current measurement signal analysis value.

The multidimensional current measurement signal analysis value Cv that is diagnosed may be only the analysis value within the set time range as shown in S101, for example.

In addition to the first setting period, the monitoring system 700 classifies the analyzed value and the frequency of the measured signal as n-dimensional (n? 3 integer) parameters during the other setting period (during the second setting period) The database 730 can be constructed (S120).

When the database 730 related to the second set period is established in S120, the monitoring system 700 calculates the signal trend (Tr) and the trend change rate Tv (S122), and calculates the calculated trend (Tr) (Tv) is compared with the reference values Tref and Tvref set in S114, and attention is generated when the set value is out of the set size (S123, S124).

The monitoring system 700 classifies the measured signals into analytic values and frequencies for multidimensional (e.g., n? 3 integer) parameters so as to be applied to the monitoring method according to the embodiment of the present invention through the database building module 720 The data is converted into a database, which will be described in detail below.

The database building module 720 may use the bubble chart to express the analysis value and the frequency of the signal measured in the target system 10 on the n-dimensional (n? 3 integer) ).

As shown in the graph of Fig. 3, for example, the database construction module 720 selects three-dimensional cells as three operation state parameters, and calculates the average value, the minimum value, the maximum value, the standard deviation You can build a database for. In the graph of Fig. 3, the number of measurement signals corresponds to the bubble size, and the analysis value can be expressed in color.

The monitoring system 700 according to the embodiment of the present invention can superimpose signal analysis values using a three-dimensional bubble chart as shown in FIG. 4, and can confirm the behavior of the target system 10 at a glance have.

The monitoring system 700 can display the analysis value and frequency by the database building module 720 in a three-dimensional bubble chart as a multi-layered two-dimensional chart or a three-dimensional surface color chart as shown in FIG. 5 . The two-dimensional charts shown in Fig. 5 each refer to one parameter value.

The monitoring system 700 can form a two-dimensional chart by shortening the operation parameters below the set value in the three-dimensional bubble chart shown in Figs. 3 and 4 to the set value as shown in Fig. 7, Also, a three-dimensional surface color chart can be formed as shown in Figs. 6A to 6C. In Fig. 6, the color can be made to correspond to the frequency, and the bubble size can be made to correspond to the frequency. In addition, the monitoring system 700 can reduce the analysis value and frequency to N (N > = 3 integers) reference planes to at least one parameter.

The monitoring system 700 can display the projection of three sides of the three-dimensional bubble chart on a parameter basis, as shown in Figs. 8A to 8C.

The monitoring system 700 can set the warning signal value to about 1.3 to 1.8 times the healthy signal value, and the alarm signal value can be set to about 1.5 to 3 times the healthy signal.

In addition, the monitoring system 700 can display alarms and warnings on a three-dimensional graph as bubble size and various bubble shapes, as shown in Fig. For example, a bubble corresponding to a signal for which a warning and an alarm are required may be displayed larger than a bubble corresponding to a healthy signal, or a bubble corresponding to a signal for which a warning and an alarm are required may be made semi-transparent.

The monitoring system 700 can database the trends and trends of the measured signals in a three-dimensional and / or two-dimensional manner through graphs related to the bubble charts shown in FIGS. 3 to 9, Can be analyzed and diagnosed. In addition, the monitoring system 700 can display trends and trends of the measured signals in the color of the bubble chart.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

10: Target system 100: Analog / digital (A / D) converter
200: pre-signal processing device 700: monitoring system
710: Sensor (s) 720: Database building module
730: Database 740: Inference Engine
750: Description module

Claims (6)

A method for monitoring a target system,
Classifying the signals measured by the target system into an analysis value and frequency for an n-dimensional (n? 3 integer) parameter and converting the signal into a database; And
(B) confirming the behavior of the target system by expressing the analysis value and the frequency of the signal measured in the target system on an n-dimensional (n? 3 integer) parameter using a bubble chart;
≪ / RTI >
The method of claim 1,
Expressing the analysis value and the frequency with a three-dimensional bubble chart or a two-dimensional bubble chart or a three-dimensional surface color chart;
Forming a two-dimensional bubble chart or a three-dimensional surface color chart by shortening an operation parameter less than a set value in the three-dimensional bubble chart to a set value;
Superimposing the analysis values and frequencies on N (N > = 3 integers) reference planes and reducing them to at least one parameter;
≪ / RTI >
The method of claim 1,
Constructing the database by classifying the analyzed value and the frequency of the measured signal for the first set period in the steady state condition with the n-dimensional (n? 3 integer) parameter;
Setting an alarm or alarm signal value of n-dimensional (n? 3 integer) classification type requiring warning or alarm based on data of the database classified and constructed during the set period in the steady state condition;
Diagnosing the target system by comparing the currently measured signal with the warning or alarm signal value;
≪ / RTI >
4. The method of claim 3,
Constructing the database by classifying the analyzed values and frequencies of the measured signals for the second set period in the steady state condition with the n dimension (n? 3 integer) parameters;
Comparing the database constructed during the first setting period with the database constructed during the second setting period, and generating attention when the magnitude or rate of change of the analytic value built in each database is out of the set size;
≪ / RTI >
A system for monitoring a target system,
A sensor for measuring a behavior signal of the target system;
A database configured to classify signals in the normal and abnormal states of the target system;
And a database building module for building the database,
Wherein the database building module is operated by a set program for performing the method of any one of claims 1 to 4.
The method of claim 5,
An inference engine for analyzing a signal measured based on the data of the database to determine a normal or abnormal signal; And
A description module for describing the result of the inference engine;
Further comprising:
KR1020140084941A 2013-07-10 2014-07-08 Monitoring method and system KR101616072B1 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
KR1020140084941A KR101616072B1 (en) 2014-07-08 2014-07-08 Monitoring method and system
PCT/KR2014/006139 WO2015005663A1 (en) 2013-07-10 2014-07-09 Signal measurement diagnosis monitoring system and method therefor, and method and system for applying same to individual device
US14/897,006 US10281909B2 (en) 2013-07-10 2014-07-09 Signal measuring/diagnosing system, and method for applying the same to individual devices

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020140084941A KR101616072B1 (en) 2014-07-08 2014-07-08 Monitoring method and system

Publications (2)

Publication Number Publication Date
KR20160006281A true KR20160006281A (en) 2016-01-19
KR101616072B1 KR101616072B1 (en) 2016-04-28

Family

ID=55305901

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020140084941A KR101616072B1 (en) 2013-07-10 2014-07-08 Monitoring method and system

Country Status (1)

Country Link
KR (1) KR101616072B1 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10273920A (en) * 1997-03-31 1998-10-13 Komatsu Ltd Abnormality monitoring method for machine
KR100522342B1 (en) 2000-07-04 2005-10-19 아사히 가세이 엔지니어링 가부시키가이샤 System for diagnosing facility apparatus, managing apparatus and diagnostic apparatus
JP5484591B2 (en) * 2010-12-02 2014-05-07 株式会社日立製作所 PLANT DIAGNOSIS DEVICE AND PLANT DIAGNOSIS METHOD

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10273920A (en) * 1997-03-31 1998-10-13 Komatsu Ltd Abnormality monitoring method for machine
KR100522342B1 (en) 2000-07-04 2005-10-19 아사히 가세이 엔지니어링 가부시키가이샤 System for diagnosing facility apparatus, managing apparatus and diagnostic apparatus
JP5484591B2 (en) * 2010-12-02 2014-05-07 株式会社日立製作所 PLANT DIAGNOSIS DEVICE AND PLANT DIAGNOSIS METHOD

Also Published As

Publication number Publication date
KR101616072B1 (en) 2016-04-28

Similar Documents

Publication Publication Date Title
US10725439B2 (en) Apparatus and method for monitoring a device having a movable part
Villa et al. Statistical fault diagnosis based on vibration analysis for gear test-bench under non-stationary conditions of speed and load
JP4995134B2 (en) Wind turbine monitoring device, method and program
US20110106510A1 (en) Methods, apparatus and computer readable storage mediums for model-based diagnosis
EP2940449A1 (en) Bearing device vibration analysis method, bearing device vibration analysis device, and rolling bearing status monitoring device
KR101748559B1 (en) Apparatus for inspecting rotating device and method thereof
Saidi et al. An integrated wind turbine failures prognostic approach implementing Kalman smoother with confidence bounds
JP5101396B2 (en) Soundness diagnosis method and program, and windmill soundness diagnosis device
JP2019128704A (en) Facility state monitoring device and facility state monitoring method
KR20140084159A (en) A method and a system for the purpose of condition monitoring of gearboxes
US20160371137A1 (en) Preprocessor of Abnormality Sign Diagnosing Device and Processing Method of the Same
WO2011024304A1 (en) Device and method for monitoring wind turbine, and program
EP2746885A1 (en) Method of monitoring the condition of a wind turbine
KR102040179B1 (en) Method for sensing and diagnosing abnormality of manufacture equipment
EP3367198A1 (en) Method for monitoring and diagnosing rotating machines based on signature analysis of electrical signals
CN109416023B (en) Wind turbine monitoring device, wind turbine monitoring method, wind turbine monitoring program, and storage medium
KR20140033944A (en) System and method for monitoring-diagnose wind power transformer
CN113383215A (en) System and process for mode-matched bearing vibration diagnostics
KR102226971B1 (en) Method for fault diagnosis based on multiple variables and apparatus using the method
US20220260985A1 (en) System and method for automated failure mode detection of rotating machinery
KR101616072B1 (en) Monitoring method and system
CN112985578B (en) Wind turbine generator set resonance diagnosis method and device
EP2560059B1 (en) Computing device and methods of presenting data to identify faults within power systems
Koukoura et al. Wind turbine intelligent gear fault identification
KR102230463B1 (en) Diagnosis system and method of defect of equipment component

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
AMND Amendment
E601 Decision to refuse application
AMND Amendment
X701 Decision to grant (after re-examination)
GRNT Written decision to grant