CN108596215B - Multi-modal signal analysis and separation method, device, equipment and storage medium - Google Patents
Multi-modal signal analysis and separation method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN108596215B CN108596215B CN201810297110.7A CN201810297110A CN108596215B CN 108596215 B CN108596215 B CN 108596215B CN 201810297110 A CN201810297110 A CN 201810297110A CN 108596215 B CN108596215 B CN 108596215B
- Authority
- CN
- China
- Prior art keywords
- signal
- modal
- time domain
- decomposed
- domain signal
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2134—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on separation criteria, e.g. independent component analysis
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a method, a device, equipment and a storage medium for multi-modal signal analysis and separation, wherein the method comprises the following steps: decomposing the multi-mode frequency measurement signal to obtain a first modal component signal, wherein the multi-mode frequency measurement signal is a first time domain signal; judging whether the first modal component signal is a single modal component signal, if so, judging that the first modal component signal is correctly decomposed, and generating a time domain signal to be decomposed for next decomposition based on the first time domain signal and the first modal component signal; if not, performing band-pass filtering processing on the first modal component signal with modal aliasing to obtain an updated single-modal component signal, and generating a time domain signal to be decomposed for next decomposition; and judging whether the time domain signal to be decomposed meets the termination condition of modal component decomposition, if not, repeating the steps to continue the decomposition until the complete decomposition of each modal component signal is completed. The method solves the problem of accurate separation of single-frequency modal components in the multi-mode frequency measurement signal.
Description
Technical Field
The invention relates to the field of recognition of modal parameters of parts of an aircraft engine, in particular to a multi-modal signal analysis and separation method, device, equipment and storage medium.
Background
The modal parameter identification is a necessary means for knowing the vibration characteristics of the mechanical parts, and mainly comprises the identification of frequency, damping and vibration mode, wherein the damping is difficult to identify accurately. The damping of the mechanical parts has important significance for inhibiting mechanical vibration and noise, enhancing the reliability and stability of the operation of the mechanical parts and prolonging the service life of the mechanical parts.
The traditional modal parameter identification mainly adopts a modal test method, and depends on the existing modal test software to directly acquire the modal parameters of mechanical parts. The method firstly needs to measure free damping vibration signals of mechanical parts, then converts the free damping vibration signals into frequency domain signals by utilizing Fourier transform, further reads modal frequency, and obtains damping ratio by utilizing a half-power bandwidth method. The method is easily influenced by factors such as measurement noise, sampling frequency, frequency resolution, sampling point number and the like, and particularly, although the half-power bandwidth method is widely used for identifying damping, the accuracy and the repeatability are poor. The time-frequency signal decomposition method can process nonlinear and non-stationary signals, and is more suitable for analyzing free damping vibration signals of mechanical parts under pulse excitation compared with the Fourier transform which is only suitable for analyzing stationary signals. The method is characterized in that a time-frequency signal decomposition method is utilized to process multi-modal free damping vibration signals to identify each modal parameter, the key is to decompose the multi-modal free damping vibration signals into a plurality of single-modal component signals, and then the modal frequency and the damping ratio can be obtained by utilizing a numerical method according to an expression of displacement response under the pulse excitation of a single-degree-of-freedom system.
Time-frequency signal decomposition methods such as EMD (empirical mode decomposition), LMD (local mean decomposition) can decompose a complex signal of multiple components into a sum of a series of single-component signals. Taking the LMD method as an example, it can decompose the signal x (t) into k PF components and a residual signal uk(t) sum of. Then there is
From PFp(t) and its Hilbert transform H [ PF ]p(t)]Can construct an analytic signal zp(t)：
Modal Parameter Identification (MPI) is an important research content in mechanical vibration analysis, and is significant for understanding the vibration characteristics of a mechanical system. For a mechanical system with multiple degrees of freedom, the displacement response under pulse excitation can be expressed as the superposition of the displacement responses of a plurality of single-degree-of-freedom systems, namely:
in the formula, Ap、ζp、ωnpAnd phipDisplacement amplitude coefficient, modal damping ratio, natural angular frequency and initial phase of the p-th order mode respectivelyA bit.
For small damping situations, comparing equations (2) and (3), the instantaneous amplitude and instantaneous phase can be expressed as
Taking logarithm of two sides of the formula (4) to obtain:
lnap(t)＝-ζpωnpt+ln Ap (6)
and (5) obtaining an instantaneous phase curve and a logarithmic amplitude curve by the formulas (5) and (6), performing linear fitting on the instantaneous phase curve and the logarithmic amplitude curve, and identifying the modal natural frequency and the damping ratio of the structure according to the slope of a straight line after fitting.
When the existing time-frequency signal decomposition method, such as LMD or EMD, decomposes a free damping vibration signal under pulse excitation to obtain a PF component or an IMF component, two modal aliasing phenomena, which are incomplete decomposition (different modes are mixed up in one PF component or IMF component) or excessive decomposition (the same mode is decomposed into a plurality of PF components or IMF components), often exist. Due to the generation of modal aliasing, the correct single-modal components cannot be obtained without correctly obtaining the modal frequencies and damping ratios from the use of equations (5) and (6).
Since the slope of the phase curve and the logarithmic decay curve is used for modal parameter identification, only single-frequency modal components can be targeted. For the frequency measurement signal of the multi-mode component, in order to identify the mode frequency and the damping ratio, the multi-mode component must be accurately decomposed into a plurality of single-frequency mode components, and then the single-frequency mode components are subjected to mode parameter identification one by one. The frequency measurement signal is a free vibration attenuation signal, which shows certain nonlinearity and non-stationarity, and it is urgently needed to design a method for analyzing and separating each mode of a multi-mode signal, so as to solve the technical problem that the modal aliasing phenomenon can occur in the existing multi-mode signal decomposition, which causes that each modal component in the frequency measurement signal cannot be accurately separated, and thus single-frequency modal component modal parameter identification cannot be satisfied.
Disclosure of Invention
The invention provides a multi-modal signal analyzing and separating method, a multi-modal signal analyzing and separating device, multi-modal signal analyzing and separating equipment and a multi-modal signal storing medium, and aims to solve the technical problem that modal parameters of a single-frequency modal component cannot be accurately identified due to the fact that modal signals in existing multi-modal signals cannot be accurately separated.
The technical scheme adopted by the invention is as follows:
according to an aspect of the present invention, there is provided a multi-modal signal analysis and separation method, the method comprising:
decomposing the multi-mode frequency measurement signal by using a time-frequency signal decomposition method to obtain a first modal component signal, wherein the multi-mode frequency measurement signal is a first time domain signal;
judging whether the first modal component signal is a single modal component signal, if so, judging that the first modal component signal is correctly decomposed, and generating a time domain signal to be decomposed for next decomposition based on the first time domain signal and the first modal component signal; if not, performing band-pass filtering processing on the first modal component signal with modal aliasing to obtain an updated single-modal component signal, and generating a time domain signal to be decomposed for next decomposition;
and judging whether the time domain signal to be decomposed meets the termination condition of modal component decomposition, if not, repeating the steps to continue the decomposition until the complete decomposition of each modal component signal is completed.
Further, determining whether the first modal component signal is a single modal component signal comprises:
performing Fourier transform on the first modal component signal to obtain a first frequency domain signal, wherein q resonance peaks exist in the first frequency domain signal, and q is more than or equal to 1;
if q is 1, judging that the first modal component signal is decomposed correctly, and directly generating a time domain signal to be decomposed for next decomposition;
if q is>1, calculating the q and q-1 resonant peak frequencies f1,q(omega) and f1,q-1Mean frequency f of (omega)1，BPF(ω)＝(f1，q(ω)+f1，q-1(ω))/2, the mean frequency being the cut-off frequency of the band-pass filter.
Further, the step of performing band-pass filtering processing on the first modal component signal with modal aliasing to obtain an updated single-modal component signal, and generating a time domain signal to be decomposed for next decomposition includes:
carrying out first band-pass filtering on the first frequency domain signal to obtain a second frequency domain signal;
performing inverse Fourier transform on the second frequency domain signal to obtain a second time domain signal, wherein the second time domain signal is used as an updated single-mode component signal;
subtracting the second time domain signal from the first time domain signal to obtain a quasi residual signal;
performing Fourier transform on the residual signal to obtain a third frequency domain signal;
carrying out second band-pass filtering on the third frequency domain signal to obtain a fourth frequency domain signal;
and performing inverse Fourier transform on the fourth frequency domain signal to obtain a time domain signal to be decomposed for next decomposition.
Further, the first band-pass filter has a band-pass filtering frequency range of [ f [ ]P,BPF(ω),fP,end(ω)]P is the number of modal components, fP,end(ω) is the maximum frequency of the first time domain signal or the time domain signal to be decomposed when the p-th modal component is decomposed;
the second band-pass filter has a band-pass filter frequency range of [ fP,0(ω),fP,BPF(ω)]，fP,0And (omega) is the minimum frequency of the first time domain signal or the time domain signal to be decomposed when the p-th modal component is decomposed.
Further, the termination condition is that the time domain signal to be decomposed is approximately a monotonic function or has at most one extreme point.
Further, the method of the invention also comprises the following steps:
and carrying out modal parameter identification on each decomposed single-modal component signal.
Furthermore, the time-frequency signal decomposition method is a local mean decomposition method, an empirical mode decomposition method, an intrinsic time scale decomposition method or a local characteristic scale decomposition method.
According to another aspect of the present invention, there is also provided a multi-modal signal analysis and separation apparatus, the apparatus comprising:
the modal signal decomposition module is used for decomposing the multi-modal frequency measurement signal by using a time-frequency signal decomposition method to obtain a first modal component signal, and the multi-modal frequency measurement signal is a first time domain signal;
the modal aliasing detection module is used for judging whether the first modal component signal is a single modal component signal or not, judging that the first modal component signal is correctly decomposed if the first modal component signal is a single modal component signal, and generating a time domain signal to be decomposed for next decomposition based on the first time domain signal and the first modal component signal;
the aliasing mode separation module is used for performing band-pass filtering processing on the first mode component signal with mode aliasing to obtain an updated single mode component signal and generating a time domain signal to be decomposed for next decomposition;
and the decomposition judging module is used for judging whether the time domain signal to be decomposed meets the termination condition of modal component decomposition until all the modal component signals are decomposed.
According to another aspect of the present invention, there is also provided a multi-modal signal analysis and separation apparatus, including a processor, where the processor is configured to run a program, and the program is configured to execute the multi-modal signal analysis and separation method of the present invention when the program is run.
According to another aspect of the present invention, there is also provided a storage medium, which includes a stored program, and the program controls a device on which the storage medium is executed to perform the multi-modal signal analysis and separation method of the present invention.
The invention has the following beneficial effects:
the multi-mode signal analyzing and separating method, the device, the equipment and the storage medium decompose the multi-mode frequency measurement signals by the time-frequency signal decomposing method, not only utilize the characteristics of self-adaptive decomposition, but also effectively solve the problem of mode aliasing generated in the decomposing process, decompose the multi-mode frequency measurement signals into single-frequency mode component signals, and facilitate the next step of mode parameter identification.
In addition to the objects, features and advantages described above, other objects, features and advantages of the present invention are also provided. The present invention will be described in further detail below with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic flow diagram of a multi-modal signal analysis and separation method according to a preferred embodiment of the present invention;
FIG. 2 is a diagram illustrating the steps of the multi-modal signal analysis and separation method according to the preferred embodiment of the present invention;
FIG. 3 is a flow chart illustrating modal parameter identification in a preferred embodiment of the present invention;
FIG. 4 is a schematic diagram of a time domain waveform of each PF component of a displacement simulation signal decomposed by a method according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a magnitude spectrum of each PF component of FIG. 4;
FIG. 6 is a time domain waveform of each PF component obtained by decomposing a compressor guide vane frequency measurement signal by a method of an embodiment of the invention;
FIG. 7 is a schematic diagram of a magnitude spectrum of each PF component of FIG. 6;
fig. 8 is a schematic block diagram of the principle of multi-modal signal analytic separation in accordance with the preferred embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
The preferred embodiment of the invention provides a multi-modal signal analysis and separation method, which aims to solve the problem that multi-modal components are difficult to accurately separate in modal parameter identification. Referring to fig. 1, the method of the present embodiment includes:
s100, decomposing the multi-mode frequency measurement signal by using a time-frequency signal decomposition method to obtain a first modal component signal, wherein the multi-mode frequency measurement signal is a first time domain signal;
step S200, judging whether the first modal component signal is a single modal component signal, if so, judging that the first modal component signal is decomposed correctly, and generating a time domain signal to be decomposed for next decomposition based on the first time domain signal and the first modal component signal; if not, performing band-pass filtering processing on the first modal component signal with modal aliasing to obtain an updated single-modal component signal, and generating a time domain signal to be decomposed for next decomposition;
and step S300, judging whether the time domain signal to be decomposed meets the termination condition of modal component decomposition, if not, repeating the steps S100 and S200 to continue the decomposition until the complete decomposition of each modal component signal is completed.
In this embodiment, the determining whether the first modal component signal is a single modal component signal includes:
performing Fourier transform on the first modal component signal to obtain a first frequency domain signal, wherein q resonance peaks exist in the first frequency domain signal, and q is more than or equal to 1;
if q is 1, judging that the first modal component signal is decomposed correctly, and directly generating a time domain signal to be decomposed for next decomposition;
if q is>1, calculating the q and q-1 resonant peak frequencies f1,q(omega) and f1,q-1Mean frequency f of (omega)1，BPF(ω)＝(f1，q(ω)+f1，q-1(ω))/2, the mean frequency being the cut-off frequency of the band-pass filter.
In this embodiment, performing band-pass filtering on the first modal component signal with modal aliasing to obtain an updated single-modal component signal, and generating a time domain signal to be decomposed for next decomposition includes:
carrying out first band-pass filtering on the first frequency domain signal to obtain a second frequency domain signal;
performing inverse Fourier transform on the second frequency domain signal to obtain a second time domain signal, wherein the second time domain signal is used as an updated single-mode component signal;
subtracting the second time domain signal from the first time domain signal to obtain a quasi residual signal;
performing Fourier transform on the residual signal to obtain a third frequency domain signal;
carrying out second band-pass filtering on the third frequency domain signal to obtain a fourth frequency domain signal;
and performing inverse Fourier transform on the fourth frequency domain signal to obtain a time domain signal to be decomposed for next decomposition.
Referring to fig. 2, in the preferred embodiment, the implementation process of each modal analysis separation method of the multi-modal signal is as follows: firstly, a time frequency signal decomposition method such as an LMD method is utilized to decompose the frequency measurement signal, and a first PF component is decomposed. The LMD adaptively decomposes a complex non-stationary multi-component signal into a sum of product functions PF where several instantaneous frequencies have physical significance. Since the LMD decomposition has modal aliasing, the PF component obtained by the decomposition is not the PF component in the strict sense and is called as a pre-PF component. In order to detect whether the decomposed pre-PF component is a single-frequency component, modal aliasing detection needs to be performed on the pre-PF component. Firstly, Fourier transform (FFT) is carried out on the pre-PF component to obtain the pre-PF1(t) spectrum pre-PF of component1(ω), i.e. the first frequency domain signal. If pre-PF1(ω) there are q (q.gtoreq.1) formants, when q is>1, calculating the q and q-1 resonant peak frequencies f1,q(omega) and f1，q-1Mean frequency f of (omega)1，BPF(ω)＝(f1，q(ω)+f1，q-1(ω))/2, said mean frequency f1，BPFAnd (ω) is the Band Pass Filter (BPF) cutoff frequency. Performing first band-pass filtering (BPF1) on the first frequency domain signal to obtain a second frequency domain signal, performing inverse Fourier transform (IFFT) on the second frequency domain signal to obtain a second time domain signal, and subtracting the second time domain signal from the first time domain signal to obtain a second time domain signalTo the quasi-residual signal pre-u1(t) of (d). Performing Fourier transform (FFT) on the quasi-residual signal to obtain a third frequency domain signal pre-u1(ω). Further to said third frequency domain signal pre-u1(ω) second bandpass filtering (BPF2) to obtain a fourth frequency domain signal u of the residual signal1(ω) and the residual signal u is obtained by an inverse Fourier transform (IFFT)1(t) of (d). The remaining signal, i.e. the time domain signal to be decomposed, is taken as the first time domain signal to continue the decomposition of the next modal component. Similarly, pre-PF can be obtained from LMD decomposition2(t) obtaining a time domain signal PF of a second mode according to the specific flow of each mode analysis and separation module of the multi-mode signal2(t) of (d). Similarly, the separation of all the remaining modes can be completed by repeating the steps to obtain a series of PFsp(t) component and residual signal ukAnd (t) completing the whole process of analyzing and separating each mode of the frequency measurement signal.
In the modal aliasing detection module, if q is 1, it is described that the PF component obtained by LMD decomposition is a normal modal component, and there is no modal aliasing phenomenon, that is, the PF component is not subjected to band-pass filtering, and there is a PF componentp(t)＝pre-PFp(t)。
In this embodiment, the first band-pass filter (BPF1) has a band-pass filter frequency range of [ f [ ]P,BPF(ω),fP,end(ω)]P is the number of modal components, fP,end(ω) is the maximum frequency of the first time domain signal or the time domain signal to be decomposed when the p-th modal component is decomposed;
the second band-pass filter (BPF2) has a band-pass filter frequency range of [ f [ ]P,0(ω),fP,BPF(ω)]，fP,0And (omega) is the minimum frequency of the first time domain signal or the time domain signal to be decomposed when the p-th modal component is decomposed.
Preferably, the termination condition is that the time domain signal to be decomposed is approximately a monotonic function or has at most one extreme point.
Preferably, the method of this embodiment further includes:
and carrying out modal parameter identification on each decomposed single-modal component signal.
As shown in fig. 3, forPF after correct separation of modalitiespAnd (t) component, wherein the identification of the modal parameters can be further realized through a modal parameter identification module. In particular to PFp(t) the component is Hilbert transformed to obtain H [ PFp(t)]Then constructing an analytic signal
Reuse of single degree of freedom system displacement response
Expressions for instantaneous amplitude and instantaneous phase can be derived. And further taking logarithms of two sides of the instantaneous amplitude expression, performing linear fitting on the instantaneous phase curve and the logarithmic amplitude curve, and obtaining the modal natural frequency and the damping ratio of the structure according to the slope of the fitted straight line.
The method utilizes the self-adaptive time-frequency signal decomposition method to decompose the frequency measurement signal, not only utilizes the characteristics of the self-adaptive decomposition, but also effectively solves the problem of modal aliasing generated in the decomposition process, can decompose the multi-modal frequency measurement signal into a single-frequency modal component signal, and is convenient for the next modal parameter identification. In addition, in the embodiment, the frequency measurement signal is decomposed by using the self-adaptive time-frequency signal decomposition method, and the steps of modal aliasing detection and aliasing modal separation are integrated, so that compared with the step of directly performing band-pass filtering processing on the frequency measurement signal, the advantages of the self-adaptive time-frequency signal decomposition method in signal decomposition and signal noise reduction are fully utilized.
FIG. 4 is a schematic diagram showing a time domain waveform of each PF component obtained by decomposing a displacement simulation signal by a method according to an embodiment of the present invention; FIG. 5 is a schematic diagram of a magnitude spectrum of each PF component of FIG. 4; therefore, modal aliasing phenomenon does not exist in each PF component, separation is accurate, and subsequent modal parameter identification is facilitated.
The method is applied and verified in the mode separation and damping identification of the frequency measurement signal of the guide blade of a certain gas compressor. Application and verification results show that, referring to fig. 6 and 7, the method successfully decomposes the multi-modal compressor guide vane frequency measurement signal into a plurality of single-frequency modal component signals, and the decomposition effect is much better than that when the self-adaptive time-frequency signal decomposition method is directly applied to decompose the frequency measurement signal.
It should be noted that, the above method for analyzing and separating each mode of a multi-mode signal is only an example of Local Mean Decomposition (LMD) and a compressor guide vane frequency measurement signal, and the specific implementation process of the present invention is not limited to what time-frequency signal decomposition method and what frequency measurement signal of a component are used, and for those skilled in the art, the present invention may be modified and changed variously (the time-frequency signal decomposition method may be freely selected, and the frequency measurement signals of various mechanical components are also generally applicable), for example, an Empirical mode decomposition method (EMD), an Intrinsic time-scale decomposition (ITD), a Local feature-scale decomposition (LCD), and an improvement method thereof may also be used.
According to another aspect of the present invention, there is also provided a multi-modal signal analysis and separation apparatus, referring to fig. 8, the apparatus of this embodiment includes:
the modal signal decomposition module 100 is configured to decompose the multi-modal frequency measurement signal by using a time-frequency signal decomposition method to obtain a first modal component signal, where the multi-modal frequency measurement signal is a first time domain signal;
the modal aliasing detection module 200 is configured to determine whether the first modal component signal is a single modal component signal, determine that the first modal component signal is correctly decomposed if the first modal component signal is a single modal component signal, and generate a time domain signal to be decomposed for next decomposition based on the first time domain signal and the first modal component signal;
an aliasing mode separation module 300, configured to perform band-pass filtering on the first mode component signal with mode aliasing to obtain an updated single-mode component signal, and generate a time domain signal to be decomposed for next decomposition;
and the decomposition judging module 400 is configured to judge whether the time domain signal to be decomposed meets a termination condition of modal component decomposition until all the modal component signals are decomposed.
Each module of this embodiment corresponds to a step of the above method embodiment, and each step specifically refers to an execution process of the above method embodiment.
Preferably, the multi-modal signal analyzing and separating apparatus of this embodiment further includes a modal parameter identification module 500, configured to perform modal parameter identification on each decomposed single-modal component signal. The execution process refers to the above method embodiments, and is not described herein.
According to another aspect of the present invention, there is also provided a multi-modal signal analyzing and separating apparatus, including a processor, where the processor is configured to execute a program, and the program executes the multi-modal signal analyzing and separating method according to an embodiment of the present invention.
According to another aspect of the present invention, a storage medium is further provided, where the storage medium includes a stored program, and the program is executed to control a device in which the storage medium is located to execute the multi-modal signal analysis and separation method according to the embodiment of the present invention.
It should be noted that the steps illustrated in the flow charts of the figures may be performed in a computer system such as a set of executable instructions and that, although a logical order is illustrated in the flow charts, in some cases, the steps illustrated or described may be performed in an order different than presented herein.
The functions described in the method of the present embodiment, if implemented in the form of software functional units and sold or used as independent products, may be stored in one or more storage media readable by a computing device. Based on such understanding, part of the contribution of the embodiments of the present invention to the prior art or part of the technical solution may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computing device (which may be a personal computer, a server, a mobile computing device, a network device, or the like) to execute all or part of the steps of the method described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (4)
1. A multi-modal signal analysis separation method is characterized by comprising the following steps:
decomposing the multi-mode frequency measurement signal by using a time-frequency signal decomposition method to obtain a first modal component signal, wherein the multi-mode frequency measurement signal is a first time domain signal;
judging whether the first modal component signal is a single modal component signal, if so, judging that the first modal component signal is decomposed correctly, and generating a time domain signal to be decomposed for next decomposition based on the first time domain signal and the first modal component signal; if not, performing band-pass filtering processing on the first modal component signal with modal aliasing to obtain an updated single-modal component signal, and generating a time domain signal to be decomposed for next decomposition;
judging whether the time domain signal to be decomposed meets the termination condition of modal component decomposition, if not, repeating the steps to continue the decomposition until the complete decomposition of each modal component signal is completed;
performing modal parameter identification on each decomposed single-modal component signal;
determining whether the first modal component signal is a single modal component signal comprises:
performing Fourier transform on the first modal component signal to obtain a first frequency domain signal, wherein q resonance peaks exist in the first frequency domain signal, and q is more than or equal to 1;
if q is 1, judging that the first modal component signal is decomposed correctly, and directly generating a time domain signal to be decomposed for next decomposition;
if q is>1, calculating the q and q-1 resonant peak frequencies f1,q(omega) and f1,q-1Mean frequency f of (omega)1,BPF(ω)＝(f1,q(ω)+f1,q-1(ω))/2, the mean frequency being the cut-off frequency of the band-pass filtering;
the method comprises the following steps of performing band-pass filtering processing on a first modal component signal with modal aliasing to obtain an updated single modal component signal, and generating a time domain signal to be decomposed for next decomposition, wherein the time domain signal to be decomposed comprises:
performing first band-pass filtering on the first frequency domain signal to obtain a second frequency domain signal;
performing inverse Fourier transform on the second frequency domain signal to obtain a second time domain signal, wherein the second time domain signal is used as an updated single-mode component signal;
subtracting the second time domain signal from the first time domain signal to obtain a quasi-residual signal;
carrying out Fourier transform on the quasi residual signal to obtain a third frequency domain signal;
performing second band-pass filtering on the third frequency domain signal to obtain a fourth frequency domain signal;
carrying out inverse Fourier transform on the fourth frequency domain signal to obtain a time domain signal to be decomposed for next decomposition;
the first band-pass filter has a band-pass filtering frequency range of
p is the number of modal components, fP,end(ω) is the maximum frequency of the first time domain signal or the time domain signal to be decomposed when the p-th modal component is decomposed;
the second band-pass filter has a band-pass filter frequency range of
fP,0(ω) is the minimum frequency of the first time domain signal or the time domain signal to be decomposed when the p-th modal component is decomposed;
the termination condition is that the time domain signal to be decomposed is approximate to a monotone function or has at most one extreme point;
the time-frequency signal decomposition method is a local mean decomposition method, an empirical mode decomposition method, an intrinsic time scale decomposition method or a local characteristic scale decomposition method.
2. A multi-modal signal analysis/separation apparatus using the multi-modal signal analysis/separation method according to claim 1, comprising:
the modal signal decomposition module is used for decomposing the multi-modal frequency measurement signal by using a time-frequency signal decomposition method to obtain a first modal component signal, wherein the multi-modal frequency measurement signal is a first time domain signal;
the modal aliasing detection module is used for judging whether the first modal component signal is a single modal component signal or not, judging that the first modal component signal is correctly decomposed if the first modal component signal is a single modal component signal, and generating a time domain signal to be decomposed for next decomposition based on the first time domain signal and the first modal component signal;
the aliasing mode separation module is used for performing band-pass filtering processing on the first mode component signal with mode aliasing to obtain an updated single mode component signal and generating a time domain signal to be decomposed for next decomposition;
the decomposition judging module is used for judging whether the time domain signal to be decomposed meets the termination condition of modal component decomposition until all the modal component signals are decomposed;
and the modal parameter identification module is used for carrying out modal parameter identification on each decomposed single-modal component signal.
3. A multi-modal signal analysis separation apparatus comprising a processor for executing a program, wherein the program executes to perform the multi-modal signal analysis separation method of claim 1.
4. A storage medium comprising a stored program, wherein the program when executed controls a device on which the storage medium is located to perform the multi-modal signal analysis and separation method of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810297110.7A CN108596215B (en) | 2018-04-04 | 2018-04-04 | Multi-modal signal analysis and separation method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810297110.7A CN108596215B (en) | 2018-04-04 | 2018-04-04 | Multi-modal signal analysis and separation method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108596215A CN108596215A (en) | 2018-09-28 |
| CN108596215B true CN108596215B (en) | 2020-11-17 |
Family
ID=63625393
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810297110.7A Active CN108596215B (en) | 2018-04-04 | 2018-04-04 | Multi-modal signal analysis and separation method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108596215B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110109058B (en) * | 2019-05-05 | 2021-04-06 | 中国航发湖南动力机械研究所 | Planar array deconvolution sound source identification method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104007315A (en) * | 2014-04-29 | 2014-08-27 | 昆明理工大学 | Improved empirical mode decomposition processing method |
| CN107607065A (en) * | 2017-09-22 | 2018-01-19 | 河海大学 | A kind of impact echo signal analysis method based on variation mode decomposition |
| CN107748734A (en) * | 2017-10-31 | 2018-03-02 | 电子科技大学 | One kind parsing empirical mode decomposition method |
-
2018
- 2018-04-04 CN CN201810297110.7A patent/CN108596215B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104007315A (en) * | 2014-04-29 | 2014-08-27 | 昆明理工大学 | Improved empirical mode decomposition processing method |
| CN107607065A (en) * | 2017-09-22 | 2018-01-19 | 河海大学 | A kind of impact echo signal analysis method based on variation mode decomposition |
| CN107748734A (en) * | 2017-10-31 | 2018-03-02 | 电子科技大学 | One kind parsing empirical mode decomposition method |
Non-Patent Citations (1)
| Title |
|---|
| 基于遗传算法参数优化的变分模态分解结合1.5维普的轴承故障诊断;边杰;《推进技术》;20170731;第38卷(第7期);第1618-1624页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108596215A (en) | 2018-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jacobsen et al. | Using enhanced frequency domain decomposition as a robust technique to harmonic excitation in operational modal analysis | |
| Gillich et al. | Early structural damage assessment by using an improved frequency evaluation algorithm | |
| Pang et al. | An optimized variational mode extraction method for rolling bearing fault diagnosis | |
| CN109340586B (en) | Method and system for detecting leakage of water supply pipeline | |
| Hartono et al. | Gear fault diagnosis using the general linear chirplet transform with vibration and acoustic measurements | |
| Xiao et al. | Research on fault feature extraction method of rolling bearing based on NMD and wavelet threshold denoising | |
| Shi et al. | The VMD-scale space based hoyergram and its application in rolling bearing fault diagnosis | |
| Hemmati et al. | Rolling element bearing fault diagnostics using acoustic emission technique and advanced signal processing | |
| Ma et al. | A hybrid fault diagnosis method based on singular value difference spectrum denoising and local mean decomposition for rolling bearing | |
| CN108596215B (en) | Multi-modal signal analysis and separation method, device, equipment and storage medium | |
| Yu et al. | A fast filtering method based on adaptive impulsive wavelet for the gear fault diagnosis | |
| Ma et al. | Blade crack detection using variational model decomposition and time-delayed feedback nonlinear tri-stable stochastic resonance | |
| Behzad et al. | Inaccessible rolling bearing diagnosis using a novel criterion for Morlet wavelet optimization | |
| Wang et al. | Early weak fault diagnosis of rolling element bearing based on resonance sparse decomposition and multi-objective information frequency band selection method | |
| Wang et al. | Time synchronous averaging based on cross-power spectrum | |
| US8838445B1 (en) | Method of removing contamination in acoustic noise measurements | |
| US8995230B2 (en) | Method of extracting zero crossing data from full spectrum signals | |
| Zhao et al. | High-order iterative rearrangement transform for time–frequency characterization of bearing fault impact | |
| CN108020741B (en) | Method and device for identifying damping characteristics of double-frequency harmonic attenuation signal | |
| Xia et al. | Modal parameter identification based on hilbert vibration decomposition in vibration stability of bridge structures | |
| Song et al. | Fault feature recognition of centrifugal compressor with cracked blade based on SNR estimation and adaptive stochastic resonance | |
| Shi et al. | A fractal-dimension-based envelope demodulation for rolling element bearing fault feature extraction from vibration signals | |
| Liang et al. | Impulse feature extraction of bearing faults based on convolutive nonnegative matrix factorization | |
| Anttonen | Fourier transform techniques for noise reduction | |
| Brown et al. | Comparison of modelled pursuits with ESPRIT and the matrix pencil method in the modelling of medical percussion signals |
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 |