CN114659791B - Turbine fault detection method, system, device and storage medium - Google Patents
Turbine fault detection method, system, device and storage medium Download PDFInfo
- Publication number
- CN114659791B CN114659791B CN202210189833.1A CN202210189833A CN114659791B CN 114659791 B CN114659791 B CN 114659791B CN 202210189833 A CN202210189833 A CN 202210189833A CN 114659791 B CN114659791 B CN 114659791B
- Authority
- CN
- China
- Prior art keywords
- displacement signal
- signal
- amplitude
- power frequency
- frequency noise
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
- G06F17/156—Correlation function computation including computation of convolution operations using a domain transform, e.g. Fourier transform, polynomial transform, number theoretic transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a method, a system, a device and a storage medium for detecting a turbine fault, which can be applied to the technical field of fault detection. According to the method, a Nuttall window function is added to an original vibration displacement signal to improve subsequent processing precision, then a second displacement signal obtained after the Nuttall window function is added is used as Fourier transform to generate an amplitude spectrum, a power frequency noise signal is calculated according to the amplitude spectrum, a third displacement signal is obtained after the power frequency noise signal is removed from the original vibration displacement signal, and after singular value decomposition is carried out on the third displacement signal, a random noise signal in the third displacement signal is restrained through a priority differential spectrum peak value algorithm, so that the power frequency noise signal and the random noise signal in the original vibration displacement signal are effectively eliminated, the loss of the useful signal is reduced, and the result of turbine fault analysis of a target displacement signal obtained after the power frequency noise signal and the random noise signal are eliminated is more accurate.
Description
Technical Field
The invention relates to the technical field of fault detection, in particular to a method, a system, a device and a storage medium for detecting the fault of a steam turbine.
Background
In the related art, the vibration displacement signal of the steam turbine contains power frequency noise of a power supply and random noise, and if the signals are directly utilized without treatment, erroneous judgment on the actual running state of the steam turbine can be caused. Currently, separate methods are generally used to suppress the power frequency noise and the random noise, respectively, for this type of signal. One of the most common power frequency noise suppression methods at present is a wave trap method, and the premise of applying the wave trap method is to assume that the power frequency noise spectrum is fixed, but the actual power frequency noise has a certain range of fluctuation, so that the wave trap spectrum needs to have a certain width in order to effectively filter the actual power frequency noise, and even if the power frequency interference is filtered, the useful signal is damaged at the same time, and the accuracy is poor.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides a method, a system, a device and a storage medium for detecting the fault of the steam turbine, which can effectively inhibit power frequency noise signals and random noise signals and reduce the loss of useful signals.
In one aspect, an embodiment of the present invention provides a method for detecting a turbine fault, including the steps of:
collecting an original vibration displacement signal of a steam turbine as a first displacement signal;
adding a Nuttall window function to the first displacement signal to obtain a second displacement signal;
generating an amplitude spectrum by taking the second displacement signal as Fourier transform;
calculating a power frequency noise signal according to the amplitude spectrum;
removing the power frequency noise signal from the first displacement signal to obtain a third displacement signal;
after singular value decomposition is carried out on the third displacement signal, a random noise signal in the third displacement signal is restrained through a preferential differential spectrum peak algorithm, and a target displacement signal is obtained;
and carrying out turbine fault analysis according to the target displacement signal.
In some embodiments, the singular value decomposition of the third displacement signal comprises:
constructing a Hankel matrix with preset dimensions according to the third displacement signal;
and carrying out singular value decomposition on the Hankel matrix to obtain a singular value sequence of the Hankel matrix.
In some embodiments, the suppressing the random noise signal in the third displacement signal by using a preferential differential spectrum peak algorithm, to obtain a target displacement signal includes:
generating a singular value differential spectrum according to the singular value sequence;
determining a reconstruction order according to the singular value difference spectrum;
and reconstructing the time domain signal according to the reconstruction order to obtain a target displacement signal.
In some embodiments, the calculating the power frequency noise signal from the amplitude spectrum includes:
determining the corresponding amplitude of a preset three spectral lines in the amplitude spectrum;
calculating an interpolation coefficient according to the amplitude value;
calculating power frequency noise signal parameters according to the interpolation coefficient and the amplitude corresponding to the preset three spectral lines;
and determining the power frequency noise signal according to the power frequency noise signal parameter.
In some embodiments, the determining the corresponding amplitude of the preset triplets in the amplitude spectrum includes:
extracting the maximum amplitude spectral line in the preset range of the peak frequency point of the amplitude spectrum as a target spectral line;
extracting a left spectral line and a right spectral line of the target spectral line from the amplitude spectrum respectively;
and determining the amplitude values of the target spectral line, the left spectral line and the right spectral line on the amplitude spectrum.
In some embodiments, the calculating the power frequency noise signal parameter according to the interpolation coefficient and the amplitude corresponding to the preset triplex line includes:
and calculating the power frequency noise signal parameters according to the interpolation coefficient and the amplitude of the target spectral line on the amplitude spectrum.
In some embodiments, the acquiring the raw vibration displacement signal of the steam turbine comprises:
and acquiring original vibration displacement signals of a plurality of data points according to a preset sampling frequency.
In another aspect, an embodiment of the present invention provides a turbine fault detection system, including:
the acquisition module is used for acquiring an original vibration displacement signal of the steam turbine as a first displacement signal;
the first processing module is used for adding a Nuttall window function to the first displacement signal to obtain a second displacement signal;
the second processing module is used for generating an amplitude spectrum by taking the second displacement signal as Fourier transform;
the calculation module is used for calculating a power frequency noise signal according to the amplitude spectrum;
the third processing module is used for removing the power frequency noise signal from the first displacement signal to obtain a third displacement signal;
the fourth processing module is used for restraining random noise signals in the third displacement signals through a preferential differential spectrum peak algorithm after singular value decomposition is carried out on the third displacement signals, so as to obtain target displacement signals;
and the analysis module is used for carrying out turbine fault analysis according to the target displacement signal.
In another aspect, an embodiment of the present invention provides a turbine fault detection apparatus, including:
at least one memory for storing a program;
and the at least one processor is used for loading the program to execute the turbine fault detection method.
In another aspect, an embodiment of the present invention provides a storage medium in which a computer-executable program is stored, the computer-executable program being for implementing the turbine fault detection method when executed by a processor.
The method for detecting the fault of the steam turbine has the following beneficial effects:
according to the embodiment, a Nuttall window function is added to an original vibration displacement signal acquired in real time to improve subsequent processing precision, then a second displacement signal obtained after the Nuttall window function is added is used as Fourier transform to generate an amplitude spectrum, a power frequency noise signal is calculated according to the amplitude spectrum, the power frequency noise signal is removed from the original vibration displacement signal to obtain a third displacement signal, after singular value decomposition is carried out on the third displacement signal, a random noise signal in the third displacement signal is restrained through a priority differential spectrum peak value algorithm, so that the power frequency noise signal and the random noise signal in the original vibration displacement signal are effectively eliminated, loss of a useful signal is reduced, and a result of turbine fault analysis of a target displacement signal obtained after the power frequency noise signal and the random noise signal are eliminated is more accurate.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The invention is further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flow chart of a method for detecting a turbine failure according to an embodiment of the present invention;
FIG. 2 is a time domain waveform diagram corresponding to an original vibration displacement signal according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an amplitude spectrum corresponding to the time domain waveform shown in FIG. 3 according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of an amplitude spectrum of a windowed vibration displacement signal after Fourier transform according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of amplitude spectrum after Fourier transform of vibration displacement signals with power frequency noise signals removed according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of singular value spectrum of a vibration displacement signal after singular value decomposition according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a singular value difference spectrum corresponding to the singular value spectrum shown in FIG. 6 according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of a time domain signal reconstructed from the singular value difference spectrum of FIG. 7, according to an embodiment of the present invention;
fig. 9 is a schematic diagram of an amplitude spectrum corresponding to fig. 8 according to an embodiment of the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention.
In the description of the present invention, it should be understood that references to orientation descriptions such as upper, lower, front, rear, left, right, etc. are based on the orientation or positional relationship shown in the drawings, are merely for convenience of description of the present invention and to simplify the description, and do not indicate or imply that the apparatus or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus should not be construed as limiting the present invention.
In the description of the present invention, the meaning of a number is one or more, the meaning of a number is two or more, and greater than, less than, exceeding, etc. are understood to exclude the present number, and the meaning of a number is understood to include the present number. The description of the first and second is for the purpose of distinguishing between technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
In the description of the present invention, unless explicitly defined otherwise, terms such as arrangement, installation, connection, etc. should be construed broadly and the specific meaning of the terms in the present invention can be reasonably determined by a person skilled in the art in combination with the specific contents of the technical scheme.
In the description of the present invention, the descriptions of the terms "one embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The structure of the main machine equipment of the steam turbine is complex, the parameters are numerous and mutually influence, and the effective monitoring of the state of the steam turbine is always a focus of attention. Because of the sensitivity of vibration to the state of the turbine unit, when the turbine unit is abnormal or fails, the vibration of the unit always reacts immediately, so that the analysis of vibration signals is an important means for monitoring the running state of the turbine, and is beneficial to early analysis and diagnosis of the failure and timely taking corresponding measures. The non-contact type eddy current displacement sensor is widely used for monitoring the state of low-frequency vibration objects such as a steam turbine, and the like, and the two mutually perpendicular eddy current displacement sensors can be used for generating an axle center track reflecting the running condition of a rotor, so that whether the rotor has faults such as unbalance, misalignment, axle crack, oil film whirl, oil film instability and the like can be intuitively judged through the axle center track. However, because the working environment of the steam turbine is complex and severe, the vibration displacement signal of the main engine of the steam turbine unit is very easy to be interfered by power frequency noise of a power supply in the acquisition process, and the acquired vibration displacement signal also contains a large amount of random noise, if the signals are directly utilized without treatment, the erroneous judgment on the actual running state of the steam turbine can be caused. Therefore, an effective signal purification method is needed to suppress power frequency noise and random noise of a power supply in the vibration displacement signal as much as possible, and a signal capable of reflecting the actual working state of the steam turbine is extracted.
The turbine vibration displacement signal contains power supply power frequency noise and random noise at the same time, and the power frequency noise and the random noise are usually restrained by adopting an independent method for the type of signal. One of the most common power frequency noise suppression methods at present is a wave trap method, and the premise of applying the wave trap method is to assume that the power frequency noise spectrum is fixed, but because the actual power frequency noise has a certain range of fluctuation, in order to effectively filter the actual power frequency noise, the wave trap spectrum needs to have a certain width, so that even if the power frequency interference is filtered, the useful signal is damaged at the same time, and the accuracy is poor.
Based on the above, the embodiment of the invention provides a turbine fault detection method, which adds a nutall window function to an original vibration displacement signal acquired in real time to improve the subsequent processing precision, then uses a second displacement signal obtained after the nutall window function is added as fourier transform to generate an amplitude spectrum, calculates a power frequency noise signal according to the amplitude spectrum, removes the power frequency noise signal from the original vibration displacement signal to obtain a third displacement signal, and then carries out singular value decomposition on the third displacement signal, and then suppresses random noise signals in the third displacement signal by a preferential differential spectrum peak algorithm, thereby effectively eliminating the power frequency noise signal and the random noise signal in the original vibration displacement signal, reducing the loss of the useful signal, and enabling the result of turbine fault analysis based on the target displacement signal obtained after the power frequency noise signal and the random noise signal are eliminated to be more accurate.
Specifically, referring to fig. 1, an embodiment of the present invention provides a method for detecting a turbine fault, including steps 110 to 170:
In the embodiment of the present application, a plurality of data points may be acquired according to a preset sampling frequencyAnd starting vibration displacement signals. Specifically, when the preset sampling frequency is f s Then can be based on the sampling frequency f s Data of N data points are collected to form vibration displacement signal x 0 (N), (n=0, 1, …, N-1). The power frequency noise signal ζ (n) in the vibration displacement signal may be expressed as shown in formula (1):
wherein A in formula (1) 1 、f 1 、
The amplitude, the frequency and the initial phase of the power frequency noise parameters are respectively.
For example, the original vibration displacement signal x is obtained after collecting 1024 points of data at a sampling frequency of 1024Hz 0 (t) the original vibration displacement signal x 0 The time domain waveform of (t) is shown in fig. 2, and the corresponding amplitude spectrum is shown in fig. 3.
In the embodiment of the present application, the time domain expression of the nutall window function may be as shown in formula (2):
in the time domain expression of equation (2), a m Representing a preset value, n=0, 1, …, N-1, N representing the number of points. For example, a 0 ＝10/32，a 1 ＝15/32，a 2 ＝6/32，a 3 ＝1/32。
The first displacement signal x is then subjected to a nutall window function w (n) as shown in equation (2) 0 (n) performing a plurality of point truncations to obtain a discrete signal sequence as a second displacement signal x 1 (n) wherein x 1 (n)＝x 0 (n)·w(n)。
And 130, generating an amplitude spectrum by taking the second displacement signal as Fourier transform.
In the embodiment of the application, the windowed vibration displacement signal x 1 (n) performing a fast Fourier transform to generate an amplitude spectrum as shown in FIG. 4, in which the spectral line k with the largest amplitude is a =50, corresponding amplitude |x (k a ) |= 0.002465, left line k a-1 =49, corresponding amplitude |x (k a-1 ) |= 0.002001, right spectral line k a+1 =51, corresponding amplitude |x (k a+1 ) |= 0.001651. The phase arg [ X (k) a )]＝-0.213052rad。
And 140, calculating a power frequency noise signal according to the amplitude spectrum.
In the embodiment of the application, the power frequency noise signal parameter can be calculated by firstly determining the amplitude corresponding to the preset triplex line in the amplitude spectrum, then calculating the interpolation coefficient according to the amplitude, then calculating the power frequency noise signal parameter according to the interpolation coefficient and the amplitude corresponding to the preset triplex line, and then determining the power frequency noise signal according to the power frequency noise signal parameter. Specifically, for determining the amplitude corresponding to the preset three spectral lines in the amplitude spectrum, the maximum amplitude spectral line in the preset range of the peak frequency point of the amplitude spectrum can be extracted as a target spectral line, then the left spectral line and the right spectral line of the target spectral line are extracted from the amplitude spectrum respectively, and then the amplitude of the target spectral line, the left spectral line and the right spectral line on the amplitude spectrum is determined. And for the determination of the power frequency noise parameters, calculating the power frequency noise signal parameters according to the interpolation coefficient and the amplitude of the target spectral line on the amplitude spectrum.
In this embodiment, when the vibration displacement signal is sampled asynchronously, the frequency point f of the power frequency noise signal is due to the fence effect 1 Is difficult to be located exactly at the sampling frequency point, i.e. f 1 Typically not an integer. Let the interpolation coefficient δ=k-k a Then there is-0.5<δ<0.5. In the amplitude spectrum, the maximum spectral line of the amplitude obtained by sampling near the peak frequency point is set as k a Taking the maximum amplitude spectrum as a target spectrum, wherein the left spectrum of the target spectrum isk a-1 The right spectral line is k a+1 ，k a-1 、k a And k a+1 The three preset spectral lines correspond to the amplitudes of |X (k) a-1 )|、|X(k a ) I and X (k) a+1 ) | a. The invention relates to a method for producing a fibre-reinforced plastic composite. And the three spectral line interpolation is to accurately solve the delta value by utilizing the information of the three preset spectral lines. Specifically, assume that
The interpolation coefficient δ is as shown in formula (3):
delta=4 (α -1)/(α+1) equation (3)
The interpolation coefficient delta can be used for calculating the power frequency noise signal parameter. Specifically, amplitude A 1 The calculation formula of (2) is shown as formula (4), frequency f 1 The calculation formula of (2) is shown as formula (5) and the initial phase
The calculation formula of (2) is shown as formula (6):
f 1 ＝(k α +delta. Δf equation (5)
The power frequency noise parameters in the fault characteristic signals can be accurately calculated by using the formula, and then the expression of the power frequency noise signals zeta (n) as shown in the formula (7) is obtained:
wherein n=0, 1, …,1023 in formula (7).
For example, the amplitude of the power frequency noise signal calculated according to equation (4), equation (5) and equation (6) is 0.007950, the frequency is 49.836749Hz, and the phase is 0.299313rad. The mathematical expression of the power frequency noise signal obtained according to the power frequency noise signal parameters is shown as a formula (8):
ζ (n) = 0.007950 ·sin (2pi· 49.836749 ·n/1024+0.299313) formula (8)
Wherein n=0, 1, …,1023 in formula (8).
And 150, removing the power frequency noise signal from the first displacement signal to obtain a third displacement signal.
In this embodiment of the present application, the third displacement signal may be obtained by performing a difference between the first displacement signal and the power frequency noise signal. For example, from a first displacement signal x 0 And (n) subtracting the power frequency component zeta (n) from the (n) to obtain a turbine vibration displacement signal with the power frequency noise being suppressed as a third displacement signal.
For example, the third displacement signal is x 2 (n) from the first displacement signal x 0 Subtracting the power frequency noise signal ζ (n) from (n) may be expressed as shown in equation (9):
x 2 (n)＝x 0 (n) - ζ (n) formula (9)
Specifically, for x 2 (n) performing a fast fourier transform to generate an amplitude spectrum as shown in fig. 5. As can be seen from fig. 5, the power frequency noise in the first displacement signal is effectively eliminated, and the influence of the adjacent components of the power frequency noise is small. Therefore, the power frequency noise reduction method of the embodiment has obvious advantages compared with a wave trap filter method.
And 160, after performing singular value decomposition on the third displacement signal, suppressing a random noise signal in the third displacement signal by using a preferential differential spectrum peak algorithm to obtain a target displacement signal.
In this embodiment of the present application, singular value decomposition is performed on the third displacement signal, and after a Hankel matrix with a preset dimension is constructed according to the third displacement signal, singular value decomposition is performed on the Hankel matrix, so as to obtain a singular value sequence of the Hankel matrix. Specifically, by the third displacement signal x 2 (n) (n=0, 1, …, 1023) constructing a Hankel matrix of 512 x 513 dimensions in a predetermined dimension as shown in formula (10)Wherein, the Hankel matrix refers to a square matrix with equal elements on each secondary diagonal:
then, singular value decomposition is carried out on the matrix A to generate an orthogonal matrix U epsilon R 512×512 And V.epsilon.R 513×513 So that the formula (11) holds:
A＝UΛV T formula (11)
In equation (11), Λ is a diagonal matrix, R is the rank of Λ, and non-zero diagonal element σ i (n=0, 1, …, 511) is called the singular value of a, i.e., the singular value spectrum shown in fig. 6.
In this embodiment, after the third displacement signal is decomposed by singular values, all singular values are divided into smaller singular values reflecting noise and larger singular values reflecting useful signal features, the smaller singular values are partially set to zero, and the larger singular values are partially reserved, so that the purpose of random noise suppression is achieved. The singular value decomposition noise reduction method is to select proper order k value for reconstruction, the selection of the reconstruction order k is important, the signal distortion is serious due to the fact that the order value is too small, and more noise is caused due to the fact that the value is too large. Thus, the present embodiment determines the k value using the preferred differential spectral peak method.
Specifically, for the target displacement signal obtained by suppressing the random noise signal in the third displacement signal by the priority differential spectrum peak value algorithm, the target displacement signal can be obtained by generating a singular value differential spectrum according to the singular value sequence, determining a reconstruction order according to the singular value differential spectrum, and then reconstructing the time domain signal according to the reconstruction order. Wherein the singular value sequence sigma i Generating a singular value difference spectrum, singular value difference spectrum beta i The definition is as shown in formula (12):
β i ＝σ i -σ i+1 formula (12)
In formula (12), i=0, 1,2, …,510.
The reconstruction matrix a' is performed with the preferred differential spectral peak method, i.e. selecting the most suitable reconstruction order k value from the differential spectral peaks, and then using equation (13):
then, the first row and the last column in the matrix A' are extracted again to generate a target displacement signal x 3 (n)，x 3 And (n) is the vibration displacement signal with the random noise suppressed. X is x 3 (n) is also the turbine current vortex vibration displacement signal which is finally purified.
For example, a singular value difference spectrum as shown in fig. 7 is generated from the singular value spectrum shown in fig. 6. As can be seen from fig. 7, when the singular value number is greater than 20, the peak value of the differential spectrum is small, and the small peak value indicates that the subsequent singular values are not changed basically, and k=20 is the optimal reconstruction order. Reconstruction is performed with a reconstruction order k=20, and a time domain signal x as shown in fig. 8 is obtained through reconstruction 3 (n) the amplitude spectrum of the time domain signal is shown in fig. 9. Comparing fig. 8 with fig. 2 and fig. 9 with fig. 3, it can be found that the suppression effect on random noise by the preferred differential spectrum peak method is very remarkable, the noise due to random noise is much reduced in the waveform of fig. 8, and the high frequency spectrum corresponding to random noise is very small in fig. 9. The time domain waveforms and spectra shown in fig. 8 and 9 are also the result of the refinement of the final vibration displacement signal.
And 170, performing turbine fault analysis according to the target displacement signal.
In the embodiment of the application, the power frequency noise is suppressed by using a windowing interpolation method for the original vibration displacement signal, the random noise is suppressed by using a preferential differential spectrum peak value method, and finally the purified turbine vibration displacement signal is obtained as a target displacement signal which can truly reflect the actual working condition of the turbine. According to the method and the device, the fault analysis of the state of the steam turbine is carried out based on the purified target displacement signal, and the accuracy of the fault analysis result can be effectively improved.
The embodiment of the invention provides a turbine fault detection system, which comprises:
the acquisition module is used for acquiring an original vibration displacement signal of the steam turbine as a first displacement signal;
the first processing module is used for adding a Nuttall window function to the first displacement signal to obtain a second displacement signal;
the second processing module is used for generating an amplitude spectrum by taking the second displacement signal as Fourier transform;
the calculation module is used for calculating a power frequency noise signal according to the amplitude spectrum;
the third processing module is used for removing the power frequency noise signal from the first displacement signal to obtain a third displacement signal;
the fourth processing module is used for restraining random noise signals in the third displacement signals through a preferential differential spectrum peak algorithm after singular value decomposition is carried out on the third displacement signals, so as to obtain target displacement signals;
and the analysis module is used for carrying out turbine fault analysis according to the target displacement signal.
The content of the method embodiment of the invention is suitable for the system embodiment, the specific function of the system embodiment is the same as that of the method embodiment, and the achieved beneficial effects are the same as those of the method.
The embodiment of the invention provides a turbine fault detection device, which comprises:
at least one memory for storing a program;
at least one processor for loading the program to perform the turbine fault detection method of FIG. 1.
The content of the method embodiment of the invention is suitable for the device embodiment, the specific function of the device embodiment is the same as that of the method embodiment, and the achieved beneficial effects are the same as those of the method.
An embodiment of the present invention provides a storage medium in which a computer-executable program for implementing the turbine fault detection method shown in fig. 1 is stored, when the computer-executable program is executed by a processor.
Embodiments of the present invention also provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device may read the computer instructions from the computer-readable storage medium, and execute the computer instructions to cause the computer device to perform the turbine fault detection method shown in fig. 1.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of one of ordinary skill in the art without departing from the spirit of the present invention. Furthermore, embodiments of the invention and features of the embodiments may be combined with each other without conflict.
Claims (7)
1. A method for detecting a turbine fault, comprising the steps of:
collecting an original vibration displacement signal of a steam turbine as a first displacement signal;
adding a Nuttall window function to the first displacement signal to obtain a second displacement signal;
generating an amplitude spectrum by taking the second displacement signal as Fourier transform;
determining the corresponding amplitude of a preset three spectral lines in the amplitude spectrum;
calculating an interpolation coefficient according to the amplitude value;
calculating power frequency noise signal parameters according to the interpolation coefficient and the amplitude corresponding to the preset three spectral lines;
determining a power frequency noise signal according to the power frequency noise signal parameter;
removing the power frequency noise signal from the first displacement signal to obtain a third displacement signal;
after singular value decomposition is carried out on the third displacement signal, a random noise signal in the third displacement signal is restrained through a preferential differential spectrum peak algorithm, and a target displacement signal is obtained;
performing turbine fault analysis according to the target displacement signal;
wherein said performing a singular value decomposition on said third displacement signal comprises:
constructing a Hankel matrix with preset dimensions according to the third displacement signal;
performing singular value decomposition on the Hankel matrix to obtain a singular value sequence of the Hankel matrix;
the step of suppressing the random noise signal in the third displacement signal by using a preferential differential spectrum peak algorithm to obtain a target displacement signal comprises the following steps:
generating a singular value differential spectrum according to the singular value sequence;
determining a reconstruction order according to the singular value difference spectrum;
and reconstructing the time domain signal according to the reconstruction order to obtain a target displacement signal.
2. The method for detecting a turbine fault according to claim 1, wherein said determining the corresponding magnitudes of the preset triplets in the magnitude spectrum comprises:
extracting the maximum amplitude spectral line in the preset range of the peak frequency point of the amplitude spectrum as a target spectral line;
extracting a left spectral line and a right spectral line of the target spectral line from the amplitude spectrum respectively;
and determining the amplitude values of the target spectral line, the left spectral line and the right spectral line on the amplitude spectrum.
3. The method for detecting a turbine fault according to claim 2, wherein calculating the power frequency noise signal parameter according to the interpolation coefficient and the amplitude corresponding to the preset triplets comprises:
and calculating the power frequency noise signal parameters according to the interpolation coefficient and the amplitude of the target spectral line on the amplitude spectrum.
4. The method for detecting the failure of the steam turbine according to claim 1, wherein the step of acquiring the original vibration displacement signal of the steam turbine comprises the steps of:
and acquiring original vibration displacement signals of a plurality of data points according to a preset sampling frequency.
5. A turbine fault detection system, comprising:
the acquisition module is used for acquiring an original vibration displacement signal of the steam turbine as a first displacement signal;
the first processing module is used for adding a Nuttall window function to the first displacement signal to obtain a second displacement signal;
the second processing module is used for generating an amplitude spectrum by taking the second displacement signal as Fourier transform;
the calculation module is used for determining the corresponding amplitude value of the preset three spectral lines in the amplitude value spectrum; calculating an interpolation coefficient according to the amplitude value; calculating power frequency noise signal parameters according to the interpolation coefficient and the amplitude corresponding to the preset three spectral lines; determining a power frequency noise signal according to the power frequency noise signal parameter;
the third processing module is used for removing the power frequency noise signal from the first displacement signal to obtain a third displacement signal;
the fourth processing module is used for restraining random noise signals in the third displacement signals through a preferential differential spectrum peak algorithm after singular value decomposition is carried out on the third displacement signals, so as to obtain target displacement signals;
the analysis module is used for carrying out turbine fault analysis according to the target displacement signal;
wherein said performing a singular value decomposition on said third displacement signal comprises:
constructing a Hankel matrix with preset dimensions according to the third displacement signal;
performing singular value decomposition on the Hankel matrix to obtain a singular value sequence of the Hankel matrix;
the step of suppressing the random noise signal in the third displacement signal by using a preferential differential spectrum peak algorithm to obtain a target displacement signal comprises the following steps:
generating a singular value differential spectrum according to the singular value sequence;
determining a reconstruction order according to the singular value difference spectrum;
and reconstructing the time domain signal according to the reconstruction order to obtain a target displacement signal.
6. A turbine fault detection apparatus, comprising:
at least one memory for storing a program;
at least one processor for loading the program to perform the turbine fault detection method according to any one of claims 1-4.
7. A storage medium having stored therein a computer executable program for implementing the turbine fault detection method according to any one of claims 1-4 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210189833.1A CN114659791B (en) | 2022-02-28 | 2022-02-28 | Turbine fault detection method, system, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210189833.1A CN114659791B (en) | 2022-02-28 | 2022-02-28 | Turbine fault detection method, system, device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114659791A CN114659791A (en) | 2022-06-24 |
| CN114659791B true CN114659791B (en) | 2023-07-04 |
Family
ID=82028378
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210189833.1A Active CN114659791B (en) | 2022-02-28 | 2022-02-28 | Turbine fault detection method, system, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114659791B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03199939A (en) * | 1989-12-27 | 1991-08-30 | I M V Kk | Vibration controlling apparatus |
| US6226321B1 (en) * | 1998-05-08 | 2001-05-01 | The United States Of America As Represented By The Secretary Of The Air Force | Multichannel parametric adaptive matched filter receiver |
| US6621860B1 (en) * | 1999-02-08 | 2003-09-16 | Advantest Corp | Apparatus for and method of measuring a jitter |
| CN101701984A (en) * | 2009-11-23 | 2010-05-05 | 浙江大学 | Fundamental wave and harmonic wave detecting method based on three-coefficient Nuttall windowed interpolation FFT |
| CN102253282A (en) * | 2011-06-10 | 2011-11-23 | 中国矿业大学 | Method for obtaining continuous frequency spectrum interpolation power harmonic parameter of Nuttall window function |
| CN106895985A (en) * | 2017-03-10 | 2017-06-27 | 汉威广园(广州)机械设备有限公司 | The fault-signal noise reduction reconstruct characteristic recognition method of high-speed rod-rolling mill |
| CN109782063A (en) * | 2018-10-23 | 2019-05-21 | 国网安徽省电力有限公司芜湖供电公司 | A kind of dynamic m-Acetyl chlorophosphonazo analysis method based on three spectral line interpolation FFT of Nuttall self-convolution window |
| CN111611870A (en) * | 2020-04-26 | 2020-09-01 | 广东机电职业技术学院 | Fault feature identification method, system, device and medium for tandem cold mill |
| CN111999635A (en) * | 2020-08-23 | 2020-11-27 | 国网山东省电力公司莱芜供电公司 | Board card fault signal analysis method and terminal based on 4-item 5-order Nuttall window |
| CN113032716A (en) * | 2019-12-24 | 2021-06-25 | 南京理工大学 | Harmonic and inter-harmonic analysis method based on windowed interpolation and Prony algorithm |
| CN113361331A (en) * | 2021-05-17 | 2021-09-07 | 广东机电职业技术学院 | Power frequency interference elimination method, system and medium based on windowing interpolation FFT |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7929375B2 (en) * | 2007-06-26 | 2011-04-19 | The United States Of America As Represented By The Secretary Of The Navy | Method and apparatus for improved active sonar using singular value decomposition filtering |
-
2022
- 2022-02-28 CN CN202210189833.1A patent/CN114659791B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03199939A (en) * | 1989-12-27 | 1991-08-30 | I M V Kk | Vibration controlling apparatus |
| US6226321B1 (en) * | 1998-05-08 | 2001-05-01 | The United States Of America As Represented By The Secretary Of The Air Force | Multichannel parametric adaptive matched filter receiver |
| US6621860B1 (en) * | 1999-02-08 | 2003-09-16 | Advantest Corp | Apparatus for and method of measuring a jitter |
| CN101701984A (en) * | 2009-11-23 | 2010-05-05 | 浙江大学 | Fundamental wave and harmonic wave detecting method based on three-coefficient Nuttall windowed interpolation FFT |
| CN102253282A (en) * | 2011-06-10 | 2011-11-23 | 中国矿业大学 | Method for obtaining continuous frequency spectrum interpolation power harmonic parameter of Nuttall window function |
| CN106895985A (en) * | 2017-03-10 | 2017-06-27 | 汉威广园(广州)机械设备有限公司 | The fault-signal noise reduction reconstruct characteristic recognition method of high-speed rod-rolling mill |
| CN109782063A (en) * | 2018-10-23 | 2019-05-21 | 国网安徽省电力有限公司芜湖供电公司 | A kind of dynamic m-Acetyl chlorophosphonazo analysis method based on three spectral line interpolation FFT of Nuttall self-convolution window |
| CN113032716A (en) * | 2019-12-24 | 2021-06-25 | 南京理工大学 | Harmonic and inter-harmonic analysis method based on windowed interpolation and Prony algorithm |
| CN111611870A (en) * | 2020-04-26 | 2020-09-01 | 广东机电职业技术学院 | Fault feature identification method, system, device and medium for tandem cold mill |
| CN111999635A (en) * | 2020-08-23 | 2020-11-27 | 国网山东省电力公司莱芜供电公司 | Board card fault signal analysis method and terminal based on 4-item 5-order Nuttall window |
| CN113361331A (en) * | 2021-05-17 | 2021-09-07 | 广东机电职业技术学院 | Power frequency interference elimination method, system and medium based on windowing interpolation FFT |
Non-Patent Citations (5)
| Title |
|---|
| MT时间序列的FIR数字滤波分析;李建华;林品荣;何畏;郑采君;;物探与化探(第S1期);全文 * |
| 基于加四项Nuttall窗递推DFT插值算法的高精度测频方法研究;许珉;刘晓辉;王玺;许辉;王奕;曹苑芊;;郑州大学学报(工学版)(第01期);全文 * |
| 基于非同步采样电力稳态谐波及间谐波检测方法分析;许晨昱;《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》;全文 * |
| 球磨机振动信号特征提取与负荷建模方法研究;卿宗胜;中国优秀硕士学位论文全文数据库 (工程科技Ⅰ辑);全文 * |
| 结合短时傅里叶变换和谱峭度的电力系统谐波检测方法;黄建明;李晓明;;电力系统保护与控制(第07期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114659791A (en) | 2022-06-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108195587B (en) | Motor rolling bearing fault diagnosis method and system | |
| Li et al. | A new strategy of instantaneous angular speed extraction and its application to multistage gearbox fault diagnosis | |
| CN110991564B (en) | Variable working condition bearing fault diagnosis method based on multiscale dispersion entropy deviation mean value and nonlinear mode decomposition | |
| CN107689059B (en) | Method and device for identifying abnormal variable pitch of wind generating set | |
| CN113865867B (en) | Bearing fault diagnosis method based on singular value decomposition of amplitude characteristic | |
| CN109540560B (en) | Absolute anti-aliasing multi-scale filtering method for complex harmonic dynamic process of rotating mechanical structure | |
| CN113361331B (en) | Power Frequency Interference Elimination Method, System and Medium Based on Windowed Interpolation FFT | |
| CN117836599A (en) | Method for detecting bearing defects in a rotating system and monitoring system for implementing said method | |
| Jiang et al. | A tacholess order tracking method based on spectral amplitude modulation for variable speed bearing fault diagnosis | |
| CN114659791B (en) | Turbine fault detection method, system, device and storage medium | |
| CN111854930B (en) | Vibration signal power frequency interference suppression method based on priori estimation | |
| CN106980722B (en) | Method for detecting and removing harmonic component in impulse response | |
| CN117686232A (en) | Method, device and storage medium for extracting vibration fundamental frequency of gas turbine in real time | |
| Xu et al. | A novel method for extracting maximum kurtosis component and its applications in rolling bearing fault diagnosis | |
| CN110147637B (en) | Rub-impact fault diagnosis method based on wavelet and harmonic component greedy sparse identification | |
| LU502454B1 (en) | A Bearing Fault Identification Method and System Based on EEMD Sparse Decomposition | |
| CN117109923A (en) | Rolling bearing fault diagnosis method and system | |
| CN116304648A (en) | Gear fault identification method based on optimized pulse enhancement and envelope synchronous averaging | |
| CN112781723B (en) | Harmonic component detection method based on frequency spectrum variance | |
| Damine et al. | Enhancement of the AR-MED deconvolution process using ensemble empirical mode decomposition in bearing fault detection | |
| Ma et al. | Envelope demodulation method based on SET for fault diagnosis of rolling bearings under variable speed | |
| CN111650436B (en) | Subsynchronous oscillation identification method for high-proportion renewable energy power system | |
| CN111611870B (en) | Fault feature identification method, system, device and medium for cold tandem mill | |
| Feng et al. | Filter Realization of the Time‐Domain Average Denoising Method for a Mechanical Signal | |
| CN114487589A (en) | Power grid broadband signal self-adaptive measurement method, device and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |