KR101688303B1 - Apparatus and method for estimating parameter of oscilation mode - Google Patents

Abstract

An apparatus and a method for estimating a parameter of a vibration mode are disclosed. The apparatus for estimating a parameter of a vibration mode according to an embodiment of the present invention includes a sampling unit that receives a time series signal and samples it according to a predetermined sampling period, accumulates the sampled data, acquires a Fourier spectrum for each accumulated number of data samples A spectrum acquisition unit for acquiring a peak value, a peak frequency, and a phase corresponding to the peak frequency of the Fourier amplitude spectrum for each of the accumulated data sample counts; and a control unit for controlling, based on the peak frequency change according to the accumulated number of data samples, And a parameter estimator for estimating a parameter of the vibration mode included in the time series signal.

Description

[0001] APPARATUS AND METHOD FOR ESTIMATING PARAMETER OF OSCILATION MODE [0002]

Embodiments of the present invention relate to techniques for estimating critical parameters included in signals measured in signal processing and low frequency vibrations in a power system.

Due to the rapid development of industrialized information technology in today's world, the power system is characterized by large-scale and heavy-load operation. Such characteristics of the power system cause various stability problems and make safe driving more difficult. In particular, low frequency vibration generated in the power system limits the operation of the power system and threatens safe operation. Therefore, accurately estimating the vibration mode of the power system provides important information in the operation of the power system.

Significant vibration modes occurring in the power system occur mainly in the low frequency region of 2.5 Hz or less. Also, due to the large scale of the power system, there are many signals to estimate the vibration mode, and there are various kinds, and all the signals are not independent but are dependent on each other. Therefore, the estimation of the vibration mode in the power system can be important information that can be used for safe operation by quickly estimating the vibration mode characteristics of the power system.

Generally, when estimating the frequency of a vibration mode included in a time series signal by using FFT (Fast Fourier Transform), the peak frequency of the Fourier amplitude spectrum is estimated as a vibration mode frequency, and theoretically, In infinity, the peak frequency of the Fourier amplitude spectrum is equal to the frequency of the real vibration mode. However, in order to estimate the parameters of the vibration mode within a short time, the parameters of the vibration mode should be estimated using a limited number of data samples.

Korean Patent Publication No. 2002-0080789 (October 23, 2002)

Embodiments of the present invention are intended to provide an apparatus and method for finding a sampling interval that performs the discrete Fourier transform on a given data repeatedly for the most accurate time in the shortest time and estimating the parameters accurately using the sampling interval.

The apparatus for estimating a parameter of a vibration mode according to an embodiment of the present invention includes a sampling unit for receiving a time series signal and sampling the time series signal according to a predetermined sampling period, accumulating the sampled data, acquiring a Fourier spectrum for each accumulated number of data samples A spectrum acquisition unit for acquiring a peak value, a peak frequency and a phase corresponding to the peak frequency of the Fourier amplitude spectrum for each of the accumulated data sample counts, And a parameter estimator for estimating a parameter of the vibration mode included in the time series signal.

The parameter estimator may estimate the number of data samples corresponding to the i-th period of the vibration mode based on the change in the peak frequency and estimate the parameter of the vibration mode using the estimated number of data samples.

The parameter estimator may obtain a singular point of the peak frequency variation and estimate the number of data samples corresponding to the i < th > period of the vibration mode using the singular point.

The parameter estimator may obtain the singular point from a difference between a peak frequency corresponding to the number of n accumulated data samples and a peak frequency corresponding to n-1 accumulated data samples.

The parameter estimator may estimate the number of data samples corresponding to the i-th period of the oscillation mode using the peak frequency of the singular point and the average value of the peak frequency for the accumulated number of data samples immediately before the singular point.

The parameter estimator may estimate the number of data samples corresponding to the i-th period of the oscillation mode using the peak value of the singularity and the intermediate value of the peak frequencies for the accumulated number of data samples before the next singularity appears have.

Wherein the parameter estimator calculates an average value of the peak frequency of the peak point of the singular point and a peak value of the peak value of the peak value of the peak value of the peak value of the peak value of the peak value of the peak value, It is possible to estimate the number of data samples corresponding to the i-th period of the vibration mode.

The parameter estimator may estimate a peak frequency corresponding to the number of data samples corresponding to the i-th cycle of the vibration mode among the peak frequencies of the Fourier amplitude spectrum as the frequency of the vibration mode.

The parameter estimator may estimate a phase corresponding to the number of data samples corresponding to the i-th period of the vibration mode among the phases corresponding to the peak frequency as the phase of the vibration mode.

The parameter estimator may estimate a damping factor of the vibration mode using a peak value corresponding to the number of data samples corresponding to the i-th cycle of the vibration mode among the peak values of the Fourier amplitude spectrum.

The parameter estimator may estimate the magnitude of the vibration mode using the estimated vibration coefficient.

A method of estimating a parameter of a vibration mode according to an embodiment of the present invention includes receiving a time series signal, sampling the time series signal according to a predetermined sampling period, accumulating the sampled data, Acquiring a Fourier spectrum, obtaining a peak value, a peak frequency, and a phase corresponding to the peak frequency of a Fourier amplitude spectrum for each of the accumulated number of data samples, and obtaining a peak frequency change according to the accumulated number of data samples And estimating a parameter of the vibration mode included in the time series signal.

Wherein estimating the parameter of the vibration mode comprises: estimating a number of data samples corresponding to an i < th > period of the vibration mode based on the peak frequency change; and estimating a parameter of the vibration mode And a step of estimating.

Estimating the number of data samples may further include obtaining a singular point of the peak frequency variation and estimating the number of data samples corresponding to the i < th > period of the vibration mode using the singular point .

Obtaining the singular point of the peak frequency variation may obtain the singular point from a difference between a peak frequency corresponding to the number of accumulated n data samples and a peak frequency corresponding to n-1 accumulated data sample counts.

The step of estimating the number of data samples may include estimating the number of data samples corresponding to the i-th period of the oscillation mode using the peak frequency of the singular point and the average value of the peak frequency for the accumulated number of data samples immediately before the singular point, can do.

Wherein the step of estimating the number of data samples comprises the step of estimating the number of data samples corresponding to the i < th > period of the oscillation mode by using the peak value of the singularity and the intermediate value of the peak frequencies for the number of accumulated data samples before the next singularity appears, It is possible to estimate the number.

Wherein the step of estimating the number of data samples comprises calculating a peak frequency of the singular point, an average value of the peak frequency with respect to the accumulated number of data samples immediately before the singular point, and a peak frequency of the singular point with respect to the cumulative number of data samples before the next singularity appears It is possible to estimate the number of data samples corresponding to the i < th > period of the vibration mode using the average value of the median values of the peak frequencies.

The step of estimating the parameter of the vibration mode may estimate a peak frequency corresponding to the number of data samples corresponding to the i-th cycle of the vibration mode among the peak frequencies of the Fourier amplitude spectrum as the frequency of the vibration mode.

The step of estimating the parameter of the vibration mode may estimate a phase corresponding to the number of data samples corresponding to the i-th period of the vibration mode among the phases corresponding to the peak frequency as the phase of the vibration mode.

Wherein the step of estimating the parameter of the vibration mode includes calculating a damping factor of the vibration mode by using a peak value corresponding to the number of data samples corresponding to the i-th period of the vibration mode among the peak values of the Fourier amplitude spectrum Can be estimated.

The step of estimating the parameter of the vibration mode may estimate the magnitude of the vibration mode using the estimated vibration coefficient.

A computer program stored in a recording medium according to an exemplary embodiment of the present invention includes a step of sampling a time series signal according to a preset sampling period, accumulating sampled data to obtain a Fourier spectrum for each accumulated number of data samples, Obtaining a peak value, a peak frequency and a phase corresponding to the peak frequency of a Fourier amplitude spectrum for each of the accumulated data sample counts, and based on the peak frequency variation according to the accumulated number of data samples, And estimating a parameter of the vibration mode.

According to embodiments of the present invention, by estimating the parameters of the vibration mode using the change of the peak frequency according to the accumulated number of data samples, the speed and accuracy of parameter estimation can be improved with only a small number of sampling data.

1 is a block diagram of a vibration mode estimating apparatus according to an embodiment of the present invention;
2 shows an example of the peak frequency variation of the amplitude spectrum according to the number of accumulated data samples
FIGS. 3 and 4 are diagrams showing the results of estimation of parameters of a vibration mode, according to an embodiment of the present invention; FIG.
5 is a flowchart of a method of estimating a parameter of a vibration mode according to an embodiment of the present invention
6 is a flowchart showing a detailed procedure for estimating a parameter of a vibration mode based on a change in peak frequency

Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. The following detailed description is provided to provide a comprehensive understanding of the methods, apparatus, and / or systems described herein. However, this is merely an example and the present invention is not limited thereto.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. The following terms are defined in consideration of the functions of the present invention, and may be changed according to the intention or custom of the user, the operator, and the like. Therefore, the definition should be based on the contents throughout this specification. The terms used in the detailed description are intended only to describe embodiments of the invention and should in no way be limiting. Unless specifically stated otherwise, the singular form of a term includes plural forms of meaning. In this description, the expressions "comprising" or "comprising" are intended to indicate certain features, numbers, steps, operations, elements, parts or combinations thereof, Should not be construed to preclude the presence or possibility of other features, numbers, steps, operations, elements, portions or combinations thereof.

1 is a configuration diagram of an apparatus 100 for estimating a vibration mode parameter according to an embodiment of the present invention.

Referring to FIG. 1, an apparatus 100 for estimating a vibration mode parameter according to an exemplary embodiment of the present invention includes a sampling unit 110, a spectrum acquisition unit 120, and a parameter estimation unit 130.

Meanwhile, the respective components 110, 120, and 130 included in the vibration mode estimation apparatus 100 shown in FIG. 1 are not necessarily implemented by separate devices that are physically separated from each other, . For example, the vibration mode estimation apparatus 100 shown in FIG. 1 may be implemented by hardware such as, for example, a DSP (Digital Signal Processor), and each of the configurations 110, 120, The function may be classified according to the function.

The sampling unit 110 may receive the time series signal and may sample the time series signal according to a predetermined sampling interval. At this time, the time series signal may be, for example, a signal measured in the power system or a signal arbitrarily generated by the user.

Meanwhile, the sampling interval of the sampling unit 110 can be set to a predetermined value, and can be set and changed to an appropriate value by the user in consideration of the accuracy of parameter estimation, calculation load, and the like.

The spectrum acquisition unit 120 accumulates data sampled by the sampling unit 110 and acquires a Fourier spectrum for each accumulated number of data samples. Specifically, the spectrum acquisition unit 120 sequentially accumulates data sampled by the sampling unit 110, and Fourier-transforms the accumulated data to acquire a Fourier spectrum for each accumulated number of data samples.

On the other hand, the spectrum acquisition unit 120 may acquire a phase corresponding to the peak value, the peak frequency, and the peak frequency of the Fourier amplitude spectrum for each accumulated number of data samples. At this time, the peak frequency means a frequency corresponding to the peak value in the Fourier amplitude spectrum.

Specifically, the peak value of the Fourier amplitude spectrum can be obtained using Equation (1).

[Equation 1]

Here, A (?) Means the Fourier amplitude spectrum.

Therefore, the peak value of the amplitude spectrum for each of the accumulated number of data samples is expressed by Equation (2).

&Quot; (2) "

In this case, n denotes the number of accumulated data samples, and An (?) Denotes a Fourier amplitude spectrum for data obtained by accumulating n data samples, respectively.

On the other hand, the peak frequency of each Fourier amplitude spectrum can be obtained using Equation (3).

&Quot; (3) "

Thus, the peak frequency of the Fourier amplitude spectrum for each of the accumulated number of data samples is given by Equation (4).

&Quot; (4) "

On the other hand, the phase corresponding to the peak frequency can be obtained from the Fourier phase spectrum, which is shown in Equation (5).

&Quot; (5) "

는 퓨리에 위상 스펙트럼을 의미한다.At this time,

Means the Fourier phase spectrum.

Thus, the phase corresponding to the peak frequency of the Fourier amplitude spectrum for each of the accumulated number of data samples is shown in Equation (6).

&Quot; (6) "

은 n개의 데이터 샘플을 누적한 데이터에 대한 퓨리에 위상 스펙트럼을 의미한다.At this time,

Refers to the Fourier phase spectrum for the data in which n data samples are accumulated.

The parameter estimator 130 may estimate the parameter of the vibration mode included in the time-series signal based on the change of the peak frequency according to the accumulated number of data samples.

Specifically, the peak frequency of the amplitude spectrum obtained by the spectrum acquisition unit 120 changes according to the accumulated number of data samples. The parameter estimator 130 estimates the number of data samples corresponding to the i-th period of the oscillation mode using the peak frequency variation of the amplitude spectrum, and estimates the number of data samples corresponding to the oscillation mode included in the time series signal Parameters can be estimated.

 A change in the peak frequency of the amplitude spectrum according to the accumulated number of data samples will be described below.

The frequency resolution of the discrete Fourier spectrum in the discrete Fourier transform is given by Equation (7).

&Quot; (7) "

는 주파수 해상도, n은 데이터 샘플 개수, t는 샘플링 간격, T는 총 샘플링 시간을 의미한다.At this time,

N is the number of data samples, t is the sampling interval, and T is the total sampling time.

On the other hand, assuming that there is no spectral leakage, the peak frequency of the amplitude spectrum obtained through the discrete Fourier transform is expressed by Equation (8).

&Quot; (8) "

In this case, ω ^* denotes the actual vibration mode frequency, and round denotes a function of rounding the value in parentheses at the first decimal place.

As a result, the error between the peak frequency of the amplitude spectrum obtained through the discrete Fourier transform and the actual vibration mode frequency is expressed by Equation (9).

 &Quot; (9) "

From Equation (9), it can be seen that the peak frequency of the amplitude spectrum obtained through discrete Fourier transform is the same as the frequency of the actual vibration mode when the remainder of? ^* Nt is zero. For example, when ω ^* is 1.04 [Hz] and Δω is 0.03 [Hz], the peak frequency of the amplitude spectrum obtained through discrete Fourier transform is 1.05 [Hz]. On the other hand, when ω ^* is 1.04 [Hz] and Δω is 0.04 [Hz], the peak frequency of the amplitude spectrum obtained through the discrete Fourier transform becomes 1.04 [Hz] and becomes equal to the actual vibration mode frequency.

On the other hand, by summarizing Equation (8) using Equation (7), Equation (8) can be expressed as Equation (10).

&Quot; (10) "

From Equations 9 and 10, the peak frequency of the amplitude spectrum obtained through the discrete Fourier transform is varied by the number of accumulated data samples, so that the difference between the peak frequency of the amplitude spectrum obtained through the Fourier transform and the actual oscillation mode frequency Is also varied according to the number of accumulated data samples. This will be described in more detail with reference to FIG.

Figure 2 is an illustration of the peak frequency variation of the amplitude spectrum according to the number of accumulated data samples.

More specifically, FIG. 2 shows the case where the actual vibration mode frequency? ^* Is 0.7854 [Hz], the sampling interval t is 0.1 [s], and the total sampling time T is 6.4 [s] It shows the change of the peak frequency of the amplitude spectrum according to the number of samples. The horizontal axis represents the number of accumulated data samples, the vertical axis represents the peak frequency, and the vertical axis represents Hz. A portion indicated by a solid line in the vicinity of 0.8 [Hz] means the actual vibration mode frequency.

In the example shown in FIG. 2, it can be seen that the peak frequency of the amplitude spectrum gradually approaches the actual vibration mode frequency while vibrating around the actual vibration mode frequency as the number of accumulated data samples increases.

Therefore, theoretically, when the number of accumulated data samples is infinite, the peak frequency of the amplitude spectrum becomes equal to the actual vibration mode frequency. However, it is impossible to obtain infinite data samples in actual implementations. Also, as the number of accumulated data samples increases, the vibration mode frequency estimated through the peak frequency becomes closer to the actual vibration mode frequency, but the time required for the vibration frequency estimation also increases.

On the other hand, in the example shown in Fig. 2, when the number of accumulated data samples is a specific value (i.e., the number of accumulated data samples close to the i-th cycle of the actual vibration mode), the peak frequency of the amplitude spectrum As shown in Fig. Therefore, in order to estimate the accurate vibration mode frequency using a limited data sample, it is necessary to find the number of data samples capable of obtaining a peak frequency close to the actual vibration mode frequency.

According to one embodiment of the present invention, the parameter estimator 130 obtains the singularity of the peak frequency variation according to the increase in the number of accumulated data samples, and calculates the i < th > The number of accumulated data samples corresponding to the period can be estimated.

More specifically, In the example shown in Figure 2, the peak frequency of the amplitude spectrum is a specific part (210, 220, 230 as the rest of the while gradually decreases with increasing the number of the accumulated data samples, ω ^* nt is changed, and 240). Hereinafter, the point at which the peak frequency of the amplitude spectrum increases as the number of accumulated data samples increases is defined as the singular point of the peak frequency change. This singularity of the peak frequency change occurs theoretically in the (2m + 1) / 2 period of the oscillation mode, where m is a positive integer.

The parameter estimator 130 determines the singularities 210, 220, 230, and 240 of the peak frequency change from the difference between the peak frequency corresponding to the n accumulated data sample counts and the peak frequency corresponding to n-1 accumulated data sample counts Can be obtained. That is, as described above, the singularities (210, 220, 230, and 240) of the peak frequency change are the points at which the peak frequency decreases with an increase in the number of accumulated data samples, but increases sharply. Therefore, the parameter estimator 130 can obtain the singular point of the peak frequency change by finding the point where the value of the following Equation 11 becomes positive (+).

&Quot; (11) "

Here, sign is a function that extracts the sign of the value in parentheses.

On the other hand, the number of data samples corresponding to the i-th period of the vibration mode can be estimated, for example, by using the peak frequency corresponding to the singularity of the peak frequency change and the average value of the peak frequency for the number of accumulated data samples just before the singularity appears can do.

Specifically, the peak frequency corresponding to the singular point of the peak frequency variation and the average value of the peak frequency with respect to the accumulated number of data samples immediately before the singular point appears can be obtained using the equation (12).

 &Quot; (12) "

In this case, k _i denotes the number of data samples of the i-th singular point.

Further, the number of data samples corresponding to the i-th period of the vibration mode can be estimated from the result of Expression (12) using Expression (13).

&Quot; (13) "

As another example, the number of data samples corresponding to the i-th period of the oscillation mode can be estimated using the peak frequency of the singularity and the intermediate value of the peak frequencies for the accumulated number of data samples before the next singularity appears.

Specifically, the peak value of the singularity and the intermediate value of the peak frequencies with respect to the cumulative number of data samples before the next singularity appears can be obtained using Equation (14).

&Quot; (14) "

Further, from the result of Expression (14), the number of data samples corresponding to the i-th period of the vibration mode can be estimated using Expression (15).

&Quot; (15) "

 As another example, the number of data samples corresponding to the i-th cycle of the vibration mode may be estimated by using an average of the values obtained by Equations (12) and (14) in order to reduce errors due to spectral leakage.

Specifically, the average of the values obtained through the expressions (12) and (14) can be obtained by using the expression (16).

&Quot; (16) "

Further, from the result of Expression (16), the number of data samples corresponding to the i-th cycle of the vibration mode can be estimated using Expression (17).

&Quot; (17) "

According to an embodiment of the present invention, the parameter estimator 130 may calculate the frequency, phase, damping factor, and frequency of the vibration mode included in the time-series signal from the number of data samples corresponding to the i- The size can be estimated.

Specifically, the frequency of the vibration mode included in the time-series signal can be estimated through Equation (18).

&Quot; (18) "

Further, the phase of the vibration mode included in the time-series signal can be estimated through Equation (19).

&Quot; (19) "

In addition, the vibration coefficient of the vibration mode included in the time-series signal can be estimated through Equation (20).

&Quot; (20) "

On the other hand, the parameter estimator 130 can estimate the magnitude of the vibration mode included in the time-series signal by using the estimated vibration coefficient. Specifically, the magnitude of the vibration mode included in the time-series signal can be estimated through Equation (21).

&Quot; (21) "

는 추정된 진동 모드 주파수에서 진폭 스펙트럼의 크기를 의미한다.At this time,

Means the magnitude of the amplitude spectrum at the estimated vibration mode frequency.

FIGS. 3 and 4 are views showing the results of estimation of parameters of a vibration mode according to an embodiment of the present invention. FIG.

Fig. 3 shows a test function used for parameter estimation of the vibration mode. The horizontal axis represents time (unit: sec), and the vertical axis represents a test function value. Specifically, the test function uses an exponential decay cosine function of the form of Equation (22).

&Quot; (22) "

The parameters of the test function are as follows.

Vibration mode 1: A ₁ = 1, α ₁ = 0.02, ω ₁ = 0.5323 [Hz], φ ₁ = 60 [degree]

Vibration mode 2: A ₂ = 1, α ₂ = -0.05, ω ₂ = 1.2135 [Hz], φ ₂ = 0 [degree]

On the other hand, the sampling frequency is set to 0.01 [sec], and in the illustrated example, the portion indicated by circles represents the sampled data.

Fig. 4 shows the results of estimating the parameters of the vibration mode using the test function of Fig.

In Fig. 4, Frequency is the frequency of the estimated vibration mode [Hz], Damp is the vibration coefficient of the estimated vibration mode, Phase is the phase of the estimated vibration mode [Degree], and Amp is the magnitude of the estimated vibration mode . On the other hand, Sec means a singularity occurrence time of a peak frequency change used for parameter estimation. That is, Sec corresponds to (2m + 1) / 2 cycles of the vibration mode (where m is a positive integer). Also, Cycles means a cycle of the vibration mode corresponding to the number of data samples estimated by Equation (13), (15) or (17).

Referring to FIG. 4, in the case of the vibration mode 1, an approximate parameter is estimated at about 6.5 cycles (12.26 seconds) of the vibration mode frequency. Also, in the case of the vibration mode 2, it can be seen that it is estimated at about 6.5 cycles (5.37 seconds) of the vibration mode frequency.

Thus, it can be seen that the approximate vibration mode parameter can be estimated through a very small number of data samples, and the calculation time of the parameter estimation can be used for on-line parameter estimation less than 1 second.

5 is a flowchart of a method of estimating a parameter of a vibration mode according to an embodiment of the present invention.

5, the parameter estimation apparatus 100 of the vibration mode receives a time series signal (510) and samples the input time series signal according to a predetermined sampling period (520).

Thereafter, the parameter estimation apparatus 100 of the vibration mode accumulates the sampled data and acquires a Fourier spectrum for each of the accumulated data samples (530).

Thereafter, the parameter estimation apparatus 100 of the vibration mode acquires 540 the phase corresponding to the peak value, the peak frequency and the peak frequency of the Fourier amplitude spectrum for each of the accumulated data sample counts.

Thereafter, the parameter estimating apparatus 100 of the vibration mode estimates a parameter of the vibration mode included in the time-series signal based on the peak frequency change according to the accumulated number of data samples (550).

Specifically, the parameter estimating apparatus 100 for a vibration mode estimates the number of data samples corresponding to the i-th cycle of the vibration mode based on the peak frequency change according to the accumulated number of data samples, The parameters of the vibration mode included in the time series signal can be estimated.

6 is a flowchart showing a detailed procedure for estimating a parameter of a vibration mode based on a change in peak frequency.

Referring to FIG. 6, the parameter estimation apparatus 100 of the vibration mode acquires the singularity of the peak frequency variation according to the accumulated number of data samples (610). At this time, the singularity of the peak frequency change can be obtained from the difference between the peak frequency corresponding to the number of accumulated data samples n and the peak frequency corresponding to the accumulated sample number of n-1 data samples.

Then, the parameter estimating apparatus 100 of the vibration mode estimates the number of data samples corresponding to the i-th cycle of the vibration mode included in the time-series signal using the singularity of the peak-to-peak frequency variation (620).

At this time, the number of data samples corresponding to the i-th period of the vibration mode can be estimated, for example, by using the peak frequency of the singular point and the average value of the peak frequency with respect to the accumulated number of data samples just before the singular point.

As another example, the number of data samples corresponding to the i-th period of the oscillation mode can be estimated using the peak frequency of the singularity and the intermediate value of the peak frequencies for the accumulated number of data samples before the next singularity appears.

As another example, the number of data samples corresponding to the i-th period of the oscillation mode may be calculated by multiplying the peak frequency of the singular point, the average value of the peak frequency with respect to the accumulated number of data samples just before the singular point, and the peak value of the singularity, The average value of the intermediate values of the peak frequencies with respect to the number of data samples.

Thereafter, the parameter estimating apparatus 100 of the vibration mode estimates the parameters of the vibration mode using the estimated number of data samples (630). At this time, the parameter of the estimated vibration mode may include the frequency, phase, vibration coefficient and magnitude of the vibration mode.

For example, the parameter estimation apparatus 100 of the vibration mode can estimate the peak frequency corresponding to the estimated number of data samples out of the peak frequencies of the Fourier amplitude spectrum for each of the accumulated data sample frequencies as the frequency of the vibration mode.

Further, the parameter estimation apparatus 100 of the vibration mode can estimate the phase corresponding to the estimated number of data samples out of the phases corresponding to the peak frequency of the Fourier amplitude spectrum for each of the accumulated number of data samples, as the phase of the vibration mode .

The parameter estimation apparatus 100 of the vibration mode estimates a damping factor of the vibration mode using a peak value corresponding to the estimated number of data samples among the peak values of the Fourier amplitude spectrum for each of the accumulated data sample counts. can do.

In addition, the parameter estimating apparatus 100 of the vibration mode can estimate the magnitude of the vibration mode using the estimated vibration coefficient.

In the flowcharts shown in FIGS. 5 and 6, the method is described by dividing the method into a plurality of steps. However, at least some of the steps may be performed in reverse order, or may be performed in combination with other steps, omitted, Or one or more steps not shown may be added.

On the other hand, an embodiment of the present invention may include a computer-readable recording medium including a program for performing the methods described herein on a computer. The computer-readable recording medium may include a program command, a local data file, a local data structure, or the like, alone or in combination. The media may be those specially designed and constructed for the present invention, or may be those that are commonly used in the field of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floppy disks, and magnetic media such as ROMs, And hardware devices specifically configured to store and execute program instructions. Examples of program instructions may include machine language code such as those generated by a compiler, as well as high-level language code that may be executed by a computer using an interpreter or the like.

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, but, on the contrary, . Therefore, the scope of the present invention should not be limited to the above-described embodiments, but should be determined by equivalents to the appended claims, as well as the appended claims.

100: Parameter estimating device
110: Sampling unit
120: Spectrum acquisition unit
130: Parameter estimation unit

Claims (23)

A sampling unit for receiving a time series signal and sampling the time series signal according to a predetermined sampling period;
Accumulating the sampled data to acquire a Fourier spectrum for each of the accumulated number of data samples, and obtaining a spectrum corresponding to the peak value, the peak frequency, and the phase corresponding to the peak frequency of the Fourier amplitude spectrum for each of the accumulated data samples, An acquisition unit; And
Estimating a number of data samples corresponding to an i-th period of a vibration mode included in the time-series signal by using a singular point of a change in the peak frequency according to a change in the number of accumulated data samples, And a parameter estimator for estimating a parameter of the vibration mode based on a peak value of a Fourier amplitude spectrum for each of the accumulated number of data samples and a phase corresponding to the peak frequency. delete delete The method according to claim 1,
Wherein the parameter estimator obtains the singular point from a difference between a peak frequency corresponding to the number of n accumulated data samples and a peak frequency corresponding to n-1 accumulated data sample counts. The method according to claim 1,
Wherein the parameter estimator calculates a parameter of a vibration mode for estimating the number of data samples corresponding to the i-th cycle of the vibration mode, using the peak frequency of the singular point and the average value of the peak frequency for the accumulated number of data samples immediately before the singular point, Estimating device. The method according to claim 1,
Wherein the parameter estimator estimates the number of data samples corresponding to the i < th > period of the oscillation mode using the peak value of the singularity and the intermediate value of the peak frequencies with respect to the accumulated number of data samples before the next singularity appears, Mode. The method according to claim 1,
Wherein the parameter estimator calculates an average value of the peak frequency of the peak point of the singular point and a peak value of the peak value of the peak value of the peak value of the peak value of the peak value of the peak value of the peak value, And estimates the number of data samples corresponding to the i-th period of the vibration mode using an average value of the values of the data samples. The method according to claim 1,
Wherein the parameter estimating unit estimates a peak frequency corresponding to the number of data samples corresponding to the i-th cycle of the vibration mode among the peak frequencies of the Fourier amplitude spectrum as the frequency of the vibration mode. The method according to claim 1,
Wherein the parameter estimating unit estimates a phase corresponding to the number of data samples corresponding to the i-th period of the vibration mode among the phases corresponding to the peak frequency as the phase of the vibration mode. The method according to claim 1,
Wherein the parameter estimator calculates a parameter of a vibration mode that estimates a damping factor of the vibration mode using a peak value corresponding to the number of data samples corresponding to the i-th cycle of the vibration mode among the peak values of the Fourier amplitude spectrum Estimating device. The method of claim 10,
Wherein the parameter estimating unit estimates the magnitude of the vibration mode using the estimated vibration coefficient. Receiving a time series signal;
Sampling the time series signal according to a predetermined sampling period;
Accumulating the sampled data to obtain a Fourier spectrum for each of the accumulated data samples;
Obtaining a peak value, a peak frequency, and a phase corresponding to the peak frequency of the Fourier amplitude spectrum for each of the accumulated number of data samples;
Obtaining a singularity of a change in the peak frequency according to a change in the number of accumulated data samples;
Estimating a number of data samples corresponding to an i-th period of the vibration mode included in the time series signal using the singular point; And
Estimating a parameter of the vibration mode based on the estimated number of data samples, a peak value of a Fourier amplitude spectrum for each of the accumulated number of data samples, and a phase corresponding to the peak frequency, Estimation method. delete delete The method of claim 12,
The step of obtaining the singularity of the peak frequency variation comprises the steps of obtaining the singularity from the difference between the peak frequency corresponding to the number of accumulated n data samples and the peak frequency corresponding to the accumulated number of n-1 data samples Parameter estimation method. The method of claim 12,
The step of estimating the number of data samples may include estimating the number of data samples corresponding to the i-th period of the oscillation mode using the peak frequency of the singular point and the average value of the peak frequency for the accumulated number of data samples immediately before the singular point, Wherein the method comprises: The method of claim 12,
Wherein the step of estimating the number of data samples comprises the step of estimating the number of data samples corresponding to the i < th > period of the oscillation mode by using the peak value of the singularity and the intermediate value of the peak frequencies for the number of accumulated data samples before the next singularity appears, And estimating the number of vibration modes. The method of claim 12,
Wherein the step of estimating the number of data samples comprises calculating a peak frequency of the singular point, an average value of the peak frequency with respect to the accumulated number of data samples immediately before the singular point, and a peak frequency of the singular point with respect to the cumulative number of data samples before the next singularity appears And estimating the number of data samples corresponding to the i-th period of the vibration mode using an average value of the median values of the peak frequencies. The method of claim 12,
Wherein the step of estimating the parameter of the vibration mode includes the step of estimating a parameter of a vibration mode that estimates a peak frequency corresponding to the number of data samples corresponding to the i < th > period of the vibration mode among the peak frequencies of the Fourier amplitude spectrum, Estimation method. The method of claim 12,
Wherein the step of estimating the parameter of the vibration mode comprises: estimating a phase corresponding to the number of data samples corresponding to the i-th cycle of the vibration mode among the phases corresponding to the peak frequency as a phase of the vibration mode, Way. The method of claim 12,
Wherein the step of estimating the parameter of the vibration mode includes calculating a damping factor of the vibration mode by using a peak value corresponding to the number of data samples corresponding to the i-th period of the vibration mode among the peak values of the Fourier amplitude spectrum And estimating the parameter of the vibration mode. 23. The method of claim 21,
Wherein the step of estimating the parameter of the vibration mode estimates the magnitude of the vibration mode using the estimated vibration coefficient. Combined with hardware,
Sampling the time series signal according to a predetermined sampling period;
Accumulating the sampled data to obtain a Fourier spectrum for each of the accumulated data samples;
Obtaining a peak value, a peak frequency, and a phase corresponding to the peak frequency of the Fourier amplitude spectrum for each of the accumulated number of data samples;
Obtaining a singularity of a change in the peak frequency according to a change in the number of accumulated data samples;
Estimating a number of data samples corresponding to an i-th period of the vibration mode included in the time series signal using the singular point; And
Estimating a parameter of the vibration mode based on the estimated number of data samples, a peak value of a Fourier amplitude spectrum for each of the accumulated number of data samples, and a phase corresponding to the peak frequency, A stored computer program.

Images