KR101268474B1 - Predictive diagnostic method and system on mechanical integrity of generator stator - Google Patents

Abstract

A method of predicting the integrity of a generator stator and a system therefor are disclosed. Generator Stator For the purpose of predicting soundness, a sensor can be installed at a plurality of windings of a stator terminal section to monitor the vibration amplitude and frequency of the windings.

Description

Technical Field [0001] The present invention relates to a method of predicting the integrity of a generator stator,

The present invention relates to a power generation system, and more particularly, to a method and system for predicting the health of a stator of a generator in real time.

The most reliable diagnosis method among the existing techniques is a method of diagnosing during stoppage, and there is a method of predicting and diagnosing the damage of the stator winding of the generator in mechanical aspect. This technology is a technique to test the inherent vibration and mode by using an impact hammer for each winding of the stator of the target generator after stopping the power plant. This method is a technology to determine the soundness by analyzing the frequency and amplitude around 120Hz by obtaining the natural vibration spectrum by the response rate of the acceleration vibration signal of the winding with respect to the excitation force appearing on the force sensor when the winding is hit by the impact hammer. There is a weak point that it is impossible to know the abnormality occurring in the actual operation state such as the characteristic information and the mode of the winding can be analyzed precisely, In addition, this technology can not be performed in the normal operation state and can be performed only under the condition that the power plant is stopped, resulting in economic loss.

In order to diagnose the abnormality of the generator stator in terms of electric power, a technique of analyzing the partial discharge waveform and spectrum by installing a partial discharge sensor on each core of the stator winding at the core boundary, and testing the impedance or resistance of the stator by flowing high- Thereby diagnosing the overall insulation state. The electrical method during stoppage has the advantage of precisely diagnosing the stator as a whole or different or abnormal condition, but it is difficult to diagnose the characteristic of each winding and the damaged winding, In the case of discharging, it is only possible to determine whether there is a tendency to switch to an online system or not.

In recent years, a state monitoring system for the stator terminal of the generator stator is being built around domestic nuclear power plants during operation. In this system, the system cost is high among the terminal windings (standard thermal power generator: 42 or 48 windings, gas turbine generator: 64 windings, etc.) of the generator stator and vibration sensor is installed only on a part of windings of the stator terminal part, It is not possible to know the information that can detect the abnormality of the entire section such as the stator terminal section. When the abnormality occurs in the winding where the sensor is not installed, the position or the damage of the abnormal winding can be diagnosed or predicted by the system itself In case of an accident, it takes a lot of time to identify the winding, which can increase the economic loss. Failure of generator stator windings requires a maintenance period of at least 15 days. In particular, the life span of the generator is 25-30 years, but the life span of the state monitoring system is about 10 years, which causes the performance deterioration and failure. However, it is difficult to maintain the equipment due to the purchase of the related accessories and the deterioration of the equipment. Therefore, not only existing state monitoring system technology, but also fault diagnosis diagnosis technology and system development are urgent.

The present invention provides a system and a method for predicting and diagnosing the integrity of a generator stator by monitoring the vibration amplitude and frequency of the winding with a vibration sensor installed in a winding within 3 to 4 points of the stator terminal section will be.

According to an aspect of the present invention, there is provided a method of generating a vibration signal, comprising: collecting a vibration signal from a sensor installed in a winding of a same section of a generator stator; Performing a Fourier transform on the vibration signal to calculate phase and amplitude information on a rotational frequency and an electromagnetic frequency corresponding to the generator main vibration source or a characteristic frequency of the external force; Extracting a response component for the rotational frequency and the electromagnetic frequency at a frequency for the Fourier transformed vibration signal; In order to synchronize the inverse-converted waveform phase information, the start position of the rotational frequency and the electromagnetic frequency waveform information corresponding to the sensor installed in the reference winding is shifted to the reference position to perform the inverse Fourier transform on the extracted response component, Deriving an angle; And estimating the presence or absence of the abnormality of the generator stator by synchronizing the inverse converted waveform phase information corresponding to the other sensor using the movement phase angle.

The sensor is located in the same section of the generator stator, and three or more windings are selected in consideration of the natural vibration mode and are installed in the radial direction and the normal direction of the selected winding.

The frequencies extracted corresponding to the response vibration waveform may be 60 Hz, 120 Hz, and 240 Hz.

And generating the behavior pattern data according to the change of the extraction waveform of the response vibration to the rotational frequency and the electromagnetic force frequency, wherein the behavior pattern data is generated by combining the position coordinates of the sensor in the radial direction and the tangential direction .

And a step of deriving a variation amount according to the coordinate movement in the same section of the generator stator to recognize a change to the behavior pattern data.

And learning the behavior pattern data according to the change of the torque frequency and the electromagnetic force frequency waveform information when the generator stator is in a steady state in order to predict the failure of the same section of the generator stator by a back propagation method.

According to another aspect of the present invention, there is provided a generator stator soundness predicting system in which a sensor is installed at a plurality of windings of a stator terminal section to monitor the vibration amplitude and frequency of the windings.

A plurality of sensors installed in a plurality of windings of the same section of the generator stator; An information acquisition device for acquiring a vibration signal from the sensor and converting the vibration signal into a digital form; And a diagnostic predicting device for extracting the characteristic frequency from the digitally converted vibration signal and for synchronizing the phases to predict the abnormality of the standing generator stator.

The sensor is located in the same section of the generator stator, and three or more windings are selected in consideration of the natural vibration mode and installed in the radial direction and the normal direction of the selected winding.

The diagnostic predicting device performs a Fourier transform on the vibration signal to calculate phase and amplitude information for a characteristic frequency of a rotational force frequency, an electromagnetic force frequency or an external force corresponding to a main excitation source of the generator, And a controller for performing inverse Fourier transform on the extracted response component and outputting the inverse Fourier transform to the inverse transformed waveform phase information, It is possible to shift the starting position of the rotational frequency and the electromagnetic force frequency waveform information to the reference position to derive the moving phase angle and synchronize the inversely converted waveform phase information corresponding to the other sensor using the moving phase angle

The diagnostic predictor generates behavior pattern data in accordance with a change in the extracted waveform of the response vibration to the rotational frequency and the electromagnetic force frequency to recognize the change, and the behavior pattern data includes position coordinates of the sensor in the radial direction and tangential direction Can be derived.

The diagnostic predicting device can recognize a change in the behavior pattern data by deriving a variation amount according to the coordinate movement in the same section of the generator stator.

The diagnostic predictor can learn the behavior pattern data according to the change of the torque frequency and the electromagnetic force frequency waveform information when the generator stator is in a normal state for predicting the failure of the same section of the generator stator by the back propagation method.

The present invention provides a method and system for predicting generator stator integrity according to an embodiment of the present invention to estimate the behavior of an abnormal rotor by using synchronization phase information of vibration of a weak section of a generator stator in real- It is possible to secure the integrity of the generator and prevent a large-scale accident.

Also, the present invention can monitor and analyze the response characteristics of the region of interest even under the normal operation conditions, thereby reducing the periodic inspection cost such as vibration test during stoppage.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram schematically illustrating a system for predicting and diagnosing the health of a generator stator in real time according to an embodiment of the present invention. FIG.
Figures 2 and 3 are diagrams for illustrating the installation of a sensor in a stator.
4 is a block diagram schematically showing an internal configuration of the information obtaining apparatus of FIG.
5 is a block diagram schematically showing an internal configuration of a diagnostic prediction apparatus according to an embodiment of the present invention;
6 is a flowchart illustrating a process of synchronizing an output signal of a sensor by a diagnostic prediction apparatus according to an embodiment of the present invention.
7 illustrates synchronized waveform information including phase information obtained through a sensor installed in a reference winding through an inverse FFT transform for a 60 Hz spectral component according to an embodiment of the present invention;
FIG. 8 is a diagram illustrating coordinate movement corresponding to behavior pattern data for each waveform of a generator stator winding according to an embodiment of the present invention; FIG.
FIG. 9 is a diagram illustrating coordinate shifts to a sensor installed in a radial direction of a generator stator winding according to an embodiment of the present invention; FIG.
10 is a diagram illustrating coordinate shifts for a sensor installed in a tangential direction of a generator stator winding according to an embodiment of the present invention;
11 is a diagram showing a behavior pattern for each waveform of a generator stator winding.
12 is a diagram illustrating the phase change of sensors for a 60 Hz vibration waveform according to an embodiment of the present invention.
Fig. 13 is a polar coordinate representation of a behavior pattern for a specific phase in the waveform data of Fig. 12; Fig.
FIG. 14 is a diagram illustrating a phase change of sensors with respect to a 120 Hz vibration waveform according to an embodiment of the present invention; FIG.
Fig. 15 is a polar coordinate representation of a behavior pattern for a specific phase in the waveform data of Fig. 14; Fig.
16 is a view showing an example of a behavior pattern for 120 Hz applied to a failure prediction system of a stator of a generator stator according to an embodiment of the present invention.
17 is a diagram showing a screen for learning a behavior pattern according to an embodiment of the present invention and predicting whether a failure has occurred.

BRIEF DESCRIPTION OF THE DRAWINGS The present invention is capable of various modifications and various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. It is to be understood, however, that the invention is not to be limited to the specific embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[Description of Figures 1 to 3]

FIG. 1 is a block diagram schematically showing a configuration of a system for predicting and diagnosing the integrity of a generator stator in real time according to an embodiment of the present invention. FIGS. 2 and 3 are views FIG.

Referring to FIG. 1, the predictive diagnostic system includes a plurality of sensors 110, an information acquisition device 120, and a diagnostic prediction device 130.

The sensor 110 is installed in the winding of the generator stator terminal section. At this time, the sensor 110 is installed at three or more points on the target winding of the end portion of the generator stator. Here, the sensor 110 may be a vibration sensor.

2 is a view showing a method of installing the sensor 110 in the same section of the generator stator winding. As shown in FIG. 2, in the radial direction 210 and the connection direction 220 of the stator winding (terminal portion or core winding portion) positioned at a predetermined angle among the windings of the same section of the generator stator, Are provided in each of the windings.

At this time, when the spacing angle between the windings is three around the reference winding 230 on which the sensor is installed, the windings positioned at 120 degrees are selected as the windings having the sensors 110, The adjacent windings are selected in consideration of the mode of the natural vibration frequency adjacent to the frequency component so that the sensor is installed in the tangential direction and in the radial direction of the selected winding so as to be sensitive to the state monitoring and the failure prediction of the stator.

FIG. 3 is a view illustrating that a sensor 110 is installed in an actual generator stator.

The information obtaining apparatus 120 is a means for receiving a vibration signal through the sensor 110 and outputting the vibration signal to the diagnostic prediction apparatus 130. [

The diagnostic predicting device 130 may receive the vibration signal directly from each sensor 110 according to the implementation method. However, since the distances of outputable distances are limited due to the characteristics of the sensor 110 installed at the distal end of the generator stator, the information acquiring device 120 is provided in each local unit, It is assumed that all of the vibration signals are collected.

The diagnostic predicting device 130 is a device for diagnosing the integrity of the generator stator by analyzing the vibration signal obtained through the information obtaining device 120.

For example, the diagnostic predicting device 130 may parameterize the phase of the vibration signal for the same section windings of each generator stator, the behavior pattern by the synchronization waveform, and the related operation information, One or more of the phase information, the displacement information, and the behavior pattern of the radial direction and tangential vibration data of the stator winding can be learned to diagnose the dryness of the stator winding.

4)

4 is a block diagram schematically showing an internal configuration of the information acquiring apparatus of FIG.

4, the information obtaining apparatus 120 includes a communication unit 410, an A / D conversion unit 415, a collecting unit 420, and a control unit 425.

The communication unit 410 is a means for transmitting and receiving data with other devices.

For example, the communication unit 410 may output the vibration signal converted into the digital signal to the diagnostic prediction unit 130 through the A / D conversion unit 415. [ Of course, the communication unit 410 is means for outputting operation data and state data for the generators collected through the collecting unit 420 to the diagnostic prediction apparatus 130. [

The A / D conversion section 415 is means for converting an analog signal into a digital signal.

For example, the A / D converter 415 can convert a vibration signal of an analog form obtained through each sensor 110 into a vibration signal of a digital form.

The collecting unit 420 is a means for collecting the operation of the generator and various status data. For example, the collecting unit 420 may collect various operation data such as the output load of the generator, the winding temperature, and the like, and output it to the diagnostic predicting apparatus 130 through the communication unit 410.

The control unit 425 is a means for controlling the internal components (e.g., the communication unit 410, the collection unit 420, etc.) of the information obtaining apparatus 120 according to an embodiment of the present invention.

5)

5 is a block diagram schematically illustrating an internal configuration of a diagnostic prediction apparatus according to an embodiment of the present invention.

5, the diagnostic prediction apparatus 130 includes a communication unit 510, a data processing unit 515, a database 520, a display unit 525, and a control unit 530.

The communication unit 510 is a means for transmitting and receiving data with other devices.

The data processing unit 515 is a means for diagnosing the dryness of the stator winding by analyzing the digital type vibration signal received through the communication unit 510. [ Detailed functions of the data processing unit 515 will be described in detail below with reference to related drawings.

The database 520 is a means for storing the vibration signal of the stator, the operation and status data of the generator, etc., received through the communication unit 510.

The display unit 525 is a means for outputting data input / received through the diagnostic predicting device 130, stored data, etc. in the form of time information. For example, the display unit 525 may be a liquid crystal display (LCD).

The control unit 530 may include internal components (e.g., the communication unit 510, the data processing unit 515, the database 520, the display unit 525, and the like) of the diagnostic prediction apparatus 130 according to an embodiment of the present invention. Etc.).

[Description of Figures 6 and 7]

FIG. 6 is a flowchart illustrating a process of synchronizing output signals of a sensor by a diagnostic predicting apparatus according to an embodiment of the present invention. FIG. FIG. 8 is a view illustrating coordinate shift corresponding to behavior pattern data of waveforms of a generator stator winding according to an embodiment of the present invention. FIG. 8 is a view showing synchronized waveform information including phase information obtained through a sensor installed in a winding. And FIG. 9 is a diagram illustrating coordinate shifts of a sensor installed in a radial direction of a generator stator winding according to an embodiment of the present invention. FIG. 10 is a graph illustrating the coordinate shifts of sensors installed in a tangential direction of a generator stator winding according to an embodiment of the present invention. And the coordinate movement with respect to the sensor.

In step 610, the diagnostic predicting device 130 collects the vibration signals obtained through the respective sensors 110 and stores them in the database 520.

That is, the diagnostic predicting device 130 may collect the vibration signals in real time through the sensors 110 through the information obtaining device 120 and store them in the database 520.

In step 615, the diagnostic predicting device 130 calculates the phase and amplitude information of the characteristic frequencies of the rotational frequency and the electromagnetic frequency (for example, 60 Hz, 120 Hz) or the external force corresponding to the main oscillator affecting the generator Performs Fast Fourier Transform (FFT) conversion on the vibration signal.

In step 620, the diagnostic predicting device 130 extracts response components such as the rotational frequency and the electromagnetic frequency from among the FFT-transformed spectral data.

In step 625, the diagnostic predicting device 130 performs an inverse FFT transform on the extracted response component.

In step 630, the diagnostic predicting device 130 moves the start position of the rotational frequency of the sensor and the frequency information of the electromagnetic force frequency provided in the reference winding to the position of zero amplitude to synchronize the inverse FFT transformed waveform phase information, . The waveform information including the phase information obtained through the sensor installed in the reference winding through the inverse FFT transform for the 60 Hz spectrum component is shown as an example of the waveform synchronized in Fig.

In step 635, the diagnostic predicting device 130 moves the derived moving phase angle by the moving phase angle obtained from the extracted waveform of the reference sensor with respect to the extracted waveforms corresponding to the other sensors except for the reference sensor, .

FIG. 8 is a view for explaining the coordinate movement coupled with the radial direction and the tangential direction of the windings of the sensor attached to the position B of the stator in FIG.

In the condition where the generator is stopped, as shown in FIG. 2, when the generator is operated, the external force such as the rotational force and the electromagnetic force acts on the stator and changes from the initial coordinates to the new position. Accordingly, the diagnostic predicting device 130 can derive a new position by combining a radial sensor and a tangential sensor to calculate a new position for the sensor.

9 is a case of a sensor installed in the radial direction of the stator winding. The sensor coordinates are set to + and the opposite direction to the stator center direction with respect to the tangential direction, and the coordinate movement is determined as the maximum amplitude. Fig. 10 shows a case where the sensor is installed in the tangential direction of the stator winding. The coordinates of the sensor are set in the counterclockwise direction (CCW) in the tangential direction as + and the clockwise direction (CW) in the tangential direction. The movement of the coordinates is determined by the maximum amplitude as in the radial direction. When vibration occurs due to electromagnetic force or rotational force, the radial coordinate moves with respect to the radial axis and the tangential coordinate moves with respect to the tangential axis.

Figures 9 and 10 show examples of coordinate shifts for radial / tangential sensors attached at the "B" position of the stator. In Figs. 9 and 10, the amount of change of the coordinate movement changes according to the synchronized phase 55. Fig. When the phase is 0 ° and 270 ° - the maximum amplitude and 90 ° and 180 ° are shifted to the + maximum amplitude. Therefore, a method of obtaining a new coordinate movement by combining the radial direction and the tangential coordinate is as shown in Equation (1).

을 각각 나타낸다.
here,

Respectively.

11)

FIG. 11 is a view showing a behavior pattern of a waveform of a generator stator winding according to a waveform, FIG. 12 is a diagram showing a phase change of sensors with respect to a 60 Hz vibration waveform according to an embodiment of the present invention, FIG. 14 is a diagram showing phase shifts of sensors with respect to a 120 Hz vibration waveform according to an embodiment of the present invention, FIG. 15 is a graph showing a phase shift of a specific phase FIG. 16 is a diagram showing an example of a behavior pattern for a 120 Hz applied to a failure prediction system of a stator of a generator stator according to an embodiment of the present invention. FIG. FIG. 5 is a diagram illustrating a screen for learning a behavior pattern according to an embodiment of the present invention and predicting whether a failure has occurred.

Fig. 11 shows the movement change of the behavior pattern when the phase changes from 0 deg. To 90 deg. It shows the behavior pattern that is obtained by extracting the torque frequency and the frequency of the electromagnetic force from the output signal of each sensor installed in the same section winding of the generator stator. Each behavior pattern has a certain shape when it is in the normal state, but when the specific vibration is generated or changed according to the load, the magnitude and phase of the vibration of the extraction frequency changes, so the behavior pattern changes to another form. By learning the behavioral pattern change of the excitation frequency in the steady state by the back propagation method, the state of any change of the stator can be monitored. The learning pattern for predicting the failure of the stator section is modeled with the coordinate shift variation of the behavior pattern according to the phase change (0 ° → 90 ° → 180 ° → 270 ° → 0 °). The variation of the coordinate movement at one position is shown in Equation 2.

Here, (x1, y1), (x2, y2) are the coordinates of each point, and? Represents the angle between two straight lines.

FIGS. 12 and 13 show the behavior patterns of the extraction waveforms of the generator stator windings, respectively, and show the movement patterns of the behavior patterns when the phase changes from 0 ° to 90 °.

FIGS. 12 and 14 illustrate the behavior patterns obtained by synchronizing the extracted 60 Hz and 120 Hz waveforms of the vibration signals in the radial direction and the tangential direction of the four stator windings through the output results of the respective reference sensors. And examples of behavior patterns for learning the operation pattern, the phase information, and the displacement information. The behavior pattern can be obtained by synchronizing the phase signals of other sensors based on the phase signal of the reference sensor. FIGS. 12 and 14 show phase shifts of the sensors for the 60 Hz and 120 Hz vibration waveforms, in which the reference sensor is phase-synchronized with respect to the first sensor.

FIGS. 13 and 15 show polar patterns of the behavior pattern for specific phases (0 °, 90 °, 180 °, and 270 °) among the phase change data of FIGS. 12 and 14. From the movement change of the behavior pattern with respect to these specific phases, a learning pattern is created using Equation (2).

FIG. 16 shows an example of a behavior pattern for a 120 Hz applied to a failure prediction system of a generator stator stator. FIG. 17 shows an example of a behavior pattern learning and prediction of a failure, The pattern of each load and operation mode occurring during normal operation of the inverter can be stored and learned by the back propagation technique, and it can be monitored and diagnosed periodically or in real time through the fault. Also, when the fault occurs, the position of the winding at the abnormal point can be calculated using the behavior pattern.

Meanwhile, a method for authenticating an application program executed in a user terminal according to an exemplary embodiment of the present invention may be implemented in a form of a program command that can be executed through a variety of means for processing information electronically and recorded in a storage medium. The storage medium may include program instructions, data files, data structures, and the like, alone or in combination.

Program instructions to be recorded on the storage medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of software. Examples of storage media include magnetic media such as hard disks, floppy disks and magnetic tape, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, magneto-optical media and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. The above-mentioned medium may also be a transmission medium such as a light or metal wire, wave guide, etc., including a carrier wave for transmitting a signal designating a program command, a data structure and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as devices for processing information electronically using an interpreter or the like, for example, a high-level language code that can be executed by a computer.

The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention as defined in the appended claims. It will be understood that the invention may be varied and varied without departing from the scope of the invention.

Claims (15)

(1) Generator Stator Collecting a vibration signal from a sensor installed in a winding in the same section:
(2) performing a Fourier transform on the vibration signal to calculate phase and amplitude information for a torque frequency and an electromagnetic frequency or a characteristic frequency of an external force corresponding to a main vibration source of the generator;
(3) extracting a response component for the rotational frequency and the electromagnetic frequency at a frequency with respect to the Fourier transformed vibration signal;
(4) The inverse Fourier transform is performed on the extracted response component, and the start position of the torque frequency and the electromagnetic frequency waveform information corresponding to the sensor installed in the reference winding is shifted to the reference position in order to synchronize the inversely converted waveform phase information Deriving a movement phase angle; And
(5) estimating the presence or absence of abnormality of the generator stator by synchronizing the inversely converted waveform phase information corresponding to the other sensors using the movement phase angle.
The method according to claim 1,
Wherein the sensor is located in the same section of the generator stator and is configured to select three or more windings in consideration of the natural vibration mode and to be installed in a radial direction and a normal direction of the selected windings.
The method according to claim 1,
Wherein frequencies to be extracted corresponding to the response components are 60 Hz, 120 Hz, and 240 Hz. The method according to claim 1,
After the step (1)
(1-1) generating the behavior pattern data according to the change of the extracted waveform of the response vibration to the torque frequency and the electromagnetic force frequency.
5. The method of claim 4,
The behavior pattern data includes:
And the position coordinates of the sensor in the radial direction and the tangential direction are combined with each other.
6. The method of claim 5,
After the step (1-1)
(1-2) A method for estimating a generator stator soundness predicting method, further comprising: determining a change in the behavior pattern data by deriving a variation amount according to a coordinate movement in the same section of the generator stator.
The method according to claim 6,
After the step (1-2)
(1-3) The method further includes learning the behavior pattern data according to the change of the rotational frequency and the electromagnetic frequency waveform information when the generator stator is in a steady state in order to predict the failure of the same section of the generator stator, by a back propagation method Wherein the stator stator comprises a plurality of stator poles.
delete In a generator stator soundness predicting system,
A plurality of sensors installed in a plurality of windings of the same section of the generator stator;
An information acquiring device for acquiring vibration signals from the plurality of sensors and converting them into a digital form; And
And a diagnostic predicting device for extracting a characteristic frequency from the digitally converted vibration signal and for predicting the abnormality of the generator stator by synchronizing the phases,
Wherein the diagnostic predicting device performs Fourier transform on the vibration signal to calculate phase and amplitude information for a characteristic frequency of a rotational force frequency, an electromagnetic force frequency, or an external force corresponding to a main virtual origin of the generator,
Extracting a response component for the rotational frequency and the electromagnetic force frequency at a frequency with respect to the Fourier transformed vibration signal, performing an inverse Fourier transform on the extracted response component,
The starting position of the waveform information of the rotational frequency and the electromagnetic frequency corresponding to the sensor provided in the reference winding is moved to the reference position to derive the moving phase angle so as to synchronize the inversely converted waveform phase information,
And synchronizes the inversely converted waveform phase information corresponding to the other sensors using the moving phase angle.
10. The method of claim 9,
Characterized in that the plurality of sensors are located in the same section of the generator stator and are arranged in the radial direction and the normal direction of the selected winding in consideration of three or more windings in consideration of the natural vibration mode, system.
delete 10. The method of claim 9,
The diagnostic predicting device comprises:
And generates the behavior pattern data according to the change of the extracted waveform of the response vibration to the rotational frequency and the electromagnetic force frequency to recognize the change.
13. The method of claim 12,
The behavior pattern data includes:
And the position coordinates of the sensor in the radial direction and the tangential direction are combined with each other.
14. The method of claim 13,
The diagnostic predicting device comprises:
Wherein a change in the behavior pattern data is recognized by deriving a change amount according to coordinate movement in the same section of the generator stator.
10. The method of claim 9,
The diagnostic predicting device comprises:
Characterized in that the behavior pattern data according to the change of the waveform information of the rotational frequency and the electromagnetic frequency when the generator stator is in the normal state is predicted by the back propagation method in order to predict the failure of the same section of the generator stator Prediction system.

Images