A method and system for detecting mechanical wear status of a Roots-type compressor

Abstract

The present invention provides a method for detecting the mechanical wear status of a Roots-type compressor, comprising obtaining an original vibration signal transmitted from the Roots-type compressor; segmenting the original vibration signal into multiple short signals based on a preset adaptive window segmentation method; decomposing the multiple short signals based on a preset symplectic geometric mode decomposition method to obtain initial symplectic geometric components of each short signal; discarding noise components from the initial symplectic geometric components of each short signal based on a preset component selection criterion of cyclic kurtosis entropy, and further reconstructing the signal to obtain a de-noised signal; plotting the Hilbert envelope spectrum of the de-noised signal, and performing characteristic frequency analysis on the Hilbert envelope spectrum to detect the mechanical wear status of the Roots-type compressor. The present invention can not only extract multiple fault features, but also weak fault features, thereby improving the accuracy, reliability, and stability of mechanical wear status detection.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical equipment wear state detection, and in particular to a method and system for detecting the mechanical wear state of a Roots-type compressor. Background Art

[0002] Roots compressors are rotary machines that rely on the relative rotation of two impellers to continuously compress and transport gas. Due to their stable operating characteristics, high volumetric efficiency, and strong adaptability, they are widely used in evaporation processes in industries such as electronics, petroleum, steel, building materials, chemicals, food, medical treatment, and papermaking.

[0003] As core equipment in the evaporation process, Roots compressors operate under complex conditions such as high speed and high temperature for extended periods, making them highly susceptible to damage and failure. Common Roots compressor failure modes include damage to key, vulnerable parts such as the impeller, bearings, gears, and drive shaft; loose anchor bolts; motor winding shorts, winding grounding or interphase shorts; broken rotor bars; and abnormal leakage.

[0004] Under harsh operating conditions, bearings, gears, impellers, and other components on Roots compressors often wear, crack, or even break. If these faults are not discovered and addressed promptly, they can hinder the operation of the compressor, causing significant economic losses at best and serious safety accidents at worst. Therefore, real-time monitoring and testing of the operating status and performance of bearings, gears, and impellers in Roots compressors is crucial.

[0005] The structure of Roots compressor is relatively complex, and the detection of its mechanical wear status can be divided into three processes: fault signal data acquisition, fault signal feature extraction and fault status identification.

[0006] Currently, the main methods for detecting the mechanical wear of rotating components include time-domain analysis, frequency-domain analysis, wavelet packet transform, integrated empirical mode decomposition, and minimum entropy deconvolution. However, because the early signs of Roots compressor failure are not obvious and are easily masked by other strong background noise, the fault signal often exhibits non-stationary characteristics, making conventional detection methods ineffective in processing non-stationary signals.

[0007] Therefore, it is necessary to use a new method in the detection of mechanical wear status of Roots compressors, which can not only extract multiple fault features, but also extract weak fault features, thereby improving the accuracy, reliability and stability of mechanical wear status detection and reducing the possibility of accidents caused by failure of rotating parts. Summary of the Invention

[0008] The technical problem to be solved by the embodiments of the present invention is to provide a method and system for detecting the mechanical wear status of a Roots-type compressor, which can not only extract multiple fault characteristics, but also extract weak fault characteristics, thereby improving the accuracy, reliability and stability of mechanical wear status detection and reducing the possibility of accidents caused by failure of rotating parts.

[0009] In order to solve the above technical problems, an embodiment of the present invention provides a method for detecting the mechanical wear state of a Roots-type compressor, the method comprising the following steps:

[0010] Obtain the original vibration signal transmitted from the Roots compressor;

[0011] Based on a preset adaptive window segmentation method, the original vibration signal is segmented into multiple short signals;

[0012] Decomposing the multiple short signals based on a preset symplectic geometric pattern decomposition method to obtain an initial symplectic geometric component of each short signal, discarding a noise component in the initial symplectic geometric component of each short signal according to a preset component selection criterion of cyclic kurtosis entropy, and further reconstructing the signal to obtain a noise-reduced signal;

[0013] The Hilbert envelope spectrum of the noise reduction signal is plotted, and characteristic frequency analysis is performed on the Hilbert envelope spectrum to detect the mechanical wear state of the Roots-type compressor.

[0014] The specific steps of segmenting the original vibration signal into multiple short signals based on the preset adaptive window segmentation method include:

[0015] The first step is to set the initial window length W and the moving step length V; wherein, the initial window length is set to W = f s / f shift , f s is the sampling frequency, f shift is the rotation frequency; the moving step length V is half of the initial window length W;

[0016] Step 2: Determine an initial window on the original vibration signal, and adaptively expand the window to the right with the window length W as the basic unit, and merge adjacent windows in sequence for expansion;

[0017] Step 3: Determine the CKE of the original vibration signal in each expanded window; if the CKE of the original vibration signal in the currently expanded window becomes smaller, retain the currently expanded window as the initial window and further add the next window for expansion; otherwise, intercept the window before the currently expanded window as an initial window, re-identify the currently expanded window as a new initial window, and continue to merge its corresponding adjacent windows for expansion;

[0018] Step 4: Repeat the above step 3 until all the initial windows slide across the entire original vibration signal, so that the original vibration signal is divided into multiple short signals.

[0019] The specific steps of decomposing the multiple short signals based on the preset symplectic geometric pattern decomposition method to obtain the initial symplectic geometric components of each short signal include:

[0020] Step 1: Assume that the current short signal is a={a1,a2,···,a n}; where n represents the length of signal a;

[0021] Then, define the trajectory matrix A of signal a as shown in the following formula (1):

[0022]

[0023] Where k represents the embedding dimension; the embedding dimension k is obtained by calculating the power spectral density of signal a and estimating the frequency of the maximum peak; where k = n / 3 if the normalized frequency is less than a given threshold, otherwise k = 1.2×(f s / f max );f s Expressed as sampling frequency, f max It is represented by the frequency of the maximum peak of the power spectrum density of signal a; τ represents the delay time; m = n-(k-1)τ;

[0024] The second step is to construct the Hamiltonian matrix B according to formula (2):

[0025]

[0026] Where C = A T A;

[0027] Step 3: Let D = B 2 , according to the definition of Hamiltonian matrix, D and B are both Hamiltonian matrices, and then a symplectic orthogonal matrix E is constructed, as shown in the following formula (3):

[0028]

[0029] Wherein, the matrix E is represented as an orthogonal symplectic matrix with symplectic matrix properties; F is represented as an upper triangular matrix, and the eigenvalue of the upper triangular matrix F is represented as λ i (i=1,2,···,k);f ij =0(i>j+1);

[0030] Step 4. If the matrix C is a real symmetric matrix, then the eigenvalue of C is equal to the eigenvalue of the upper triangular matrix F. According to the properties of the Hamiltonian matrix, the eigenvalue of C is The eigenvalues of matrix C can be expressed as β1 > β2 > ··· > β k ;

[0031] Let P i (i = 1, 2, ···, k) be the eigenvectors of matrix C, let P i (i = 1, 2, ···, k) be the eigenvectors of matrix C, and let S i = P i T A T and Z i = P i S i , then the initial reconstruction trajectory matrix Z is defined as follows:

[0032] Z = Z1 + Z2 + … + Z k (4);

[0033] Step 5: Since the initial reconstruction trajectory matrix Z is an m×k matrix, it is necessary to convert it into k initial symplectic geometric components Y i (i = 1, 2, ···, k) of length n through diagonal averaging:

[0034] Any element of Z is defined as z ij , where 1 ≤ i ≤ k, 1 ≤ j ≤ m, k * = min(m, k), m * = max(m, k), n = m + (k - 1)τ;

[0035] If m < k, let z ij * = z ij ; otherwise, let z ij * = z ji ; Therefore, the process of diagonal averaging is shown in the following formula (5):

[0036]

[0037] By processing the initial reconstruction trajectory matrix Z, k initial symplectic geometric components Y i (y1, y2, ···, y n ) are obtained, as shown in the following formula (6):

[0038] Y = Y1 + Y2 + … + Y k (6).

[0039] Among them, the specific steps of further reconstructing the denoised signal after discarding the noise components in the initial symplectic geometric components of each short signal according to the component selection criteria of the preset cyclic kurtosis entropy include:

[0040] Step 1: Calculate the initial symplectic geometry component Y i The cyclic kurtosis entropy is obtained by CKE1, CKE2, ···, CKE k , and construct the weight function to calculate each initial symplectic geometric component Y by the following formula (7) i The corresponding weight value L i ;

[0041]

[0042] Among them, CKE mean Expressed as the average value of CKE; i = 1, 2, ···, k;

[0043] In the second step, the final noise reduction signal component M can be obtained by the following formula (8):

[0044]

[0045] Step 3: reconstruct the noise reduction signal component M to obtain a noise reduction signal;

[0046] Among them, the initial symplectic geometric component Y i The calculation process of the cyclic kurtosis entropy is as follows:

[0047] Given a signal b={b1,b2,···,b n}, then the cycle kurtosis (CK) can be expressed as shown in the following formula (9):

[0048]

[0049] Wherein, RSE represents the autocorrelation function of the square envelope signal SE(b), and SE(b) = |b| 2 =|b+jHilber(b)| 2 , Hilbert represents Hilbert transform; Ra(0) is the autocorrelation function value of signal b when the delay coefficient is 0, and R b (τ)=E[b(ni)b(ni+τ)]; hT (h=1, 2, ···, n) represents the delay coefficient;

[0050] The CK value of any delay coefficient is calculated by the following formula (10), and the cyclic kurtosis entropy can be expressed as follows by the following formula (11):

[0051]

[0052]

[0053] Where, h = {1, 2, ···, n}.

[0054] The mechanical wear state of the Roots-type compressor includes bearing inner ring failure, bearing outer ring failure, rolling element failure, impeller meshing and gear failure.

[0055] An embodiment of the present invention further provides a system for detecting the mechanical wear state of a Roots-type compressor, comprising:

[0056] A signal acquisition unit is used to acquire the original vibration signal transmitted from the Roots compressor;

[0057] A signal segmentation unit, configured to segment the original vibration signal into a plurality of short signals based on a preset adaptive window segmentation method;

[0058] a signal feature extraction unit, configured to decompose the plurality of short signals based on a preset symplectic geometric pattern decomposition method to obtain an initial symplectic geometric component of each short signal, and discard a noise component in the initial symplectic geometric component of each short signal according to a preset component selection criterion of cyclic kurtosis entropy, and then further reconstruct the signal to obtain a noise-reduced signal;

[0059] The mechanical wear state detection unit is used to draw the Hilbert envelope spectrum of the noise reduction signal and perform characteristic frequency analysis on the Hilbert envelope spectrum to detect the mechanical wear state of the Roots-type compressor.

[0060] The mechanical wear state of the Roots-type compressor includes bearing inner ring failure, bearing outer ring failure, rolling element failure, impeller meshing and gear failure.

[0061] The implementation of the embodiments of the present invention has the following beneficial effects:

[0062] On the one hand, the present invention utilizes an adaptive window segmentation method to effectively discover weak fault features, and on the other hand, selects effective components for reconstruction based on cyclic kurtosis entropy to effectively remove noise interference, thereby accurately, reliably and stably detecting the mechanical wear state of a Roots-type compressor. This not only extracts a variety of fault features, but also weak fault features, thereby improving the accuracy, reliability and stability of mechanical wear state detection and reducing the possibility of accidents caused by failures of rotating parts. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, without paying any creative work, other drawings obtained based on these drawings still fall within the scope of the present invention.

[0064] Figure 1A flow chart of a method for detecting the mechanical wear state of a Roots-type compressor provided in an embodiment of the present invention;

[0065] Figure 2 A result diagram of a vibration signal collected when an inner ring failure occurs in a bearing in an application scenario of a method for detecting the mechanical wear state of a Roots-type compressor provided by an embodiment of the present invention;

[0066] Figure 3 A diagram showing the results of analyzing a fault signal of a Roots-type compressor bearing inner ring using an envelope analysis method in an application scenario of a method for detecting a mechanical wear state of a Roots-type compressor provided by an embodiment of the present invention;

[0067] Figure 4 A diagram showing the results of analyzing a fault signal of an inner ring bearing of a Roots-type compressor using the method of the embodiment of the present invention in an application scenario of a method for detecting a mechanical wear state of a Roots-type compressor provided by an embodiment of the present invention;

[0068] Figure 5 A schematic structural diagram of a mechanical wear status detection system for a Roots-type compressor provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be described in further detail below with reference to the accompanying drawings.

[0070] like Figure 1 As shown in the figure, a method for detecting the mechanical wear state of a Roots-type compressor is provided in an embodiment of the present invention, and the method includes the following steps:

[0071] Step S1: obtaining an original vibration signal transmitted from a Roots compressor;

[0072] The specific process is to fix the acceleration sensor on the Roots compressor and connect it to the multi-channel data acquisition and analyzer, so that the multi-channel data acquisition and analyzer can obtain the original vibration signal transmitted by the Roots compressor in real time.

[0073] Step S2: Segmenting the original vibration signal into multiple short signals based on a preset adaptive window segmentation method;

[0074] The specific process is as follows: the first step is to set the initial window length W and the moving step length V; wherein, the initial window length is set to W = f s / f shift , f s is the sampling frequency, f shift is the rotation frequency; the moving step length V is half of the initial window length W;

[0075] Step 2: Determine an initial window on the original vibration signal, and adaptively expand the window to the right with the window length W as the basic unit, and merge adjacent windows in sequence for expansion;

[0076] Step 3: Determine the CKE of the original vibration signal in each expanded window; if the CKE of the original vibration signal in the currently expanded window becomes smaller, retain the currently expanded window as the initial window and further add the next window for expansion; otherwise, intercept the window before the currently expanded window as an initial window, re-identify the currently expanded window as a new initial window, and continue to merge its corresponding adjacent windows for expansion;

[0077] Step 4: Repeat the above step 3 until all the initial windows slide across the entire original vibration signal, so that the original vibration signal is divided into multiple short signals.

[0078] It should be noted that the adaptive window segmentation method established based on CKE can adaptively segment the signal into multiple relatively optimal short signals, which can be used to mine weak fault features in the signal.

[0079] Step S3: decomposing the multiple short signals based on a preset symplectic geometric pattern decomposition method to obtain an initial symplectic geometric component of each short signal, discarding the noise component in the initial symplectic geometric component of each short signal according to a preset component selection criterion of cyclic kurtosis entropy, and further reconstructing to obtain a noise-reduced signal;

[0080] The specific process is as follows: first, based on a preset symplectic geometric mode decomposition method, the multiple short signals are decomposed to obtain the initial symplectic geometric components of each short signal. The specific steps are as follows:

[0081] Step 1: Assume that the current short signal is a={a1,a2,···,a n}; where n represents the length of signal a;

[0082] Then, define the trajectory matrix A of signal a as shown in the following formula (1):

[0083]

[0084] Where k represents the embedding dimension; the embedding dimension k is obtained by calculating the power spectral density of signal a and estimating the frequency of the maximum peak; where k = n / 3 if the normalized frequency is less than a given threshold, otherwise k = 1.2×(f s / f max );f s Expressed as sampling frequency, f max It is represented by the frequency of the maximum peak of the power spectrum density of signal a; τ represents the delay time; m = n-(k-1)τ;

[0085] The second step is to construct the Hamiltonian matrix B according to formula (2):

[0086]

[0087] Where C = A T A;

[0088] Step 3: Let D = B 2 , according to the definition of Hamiltonian matrix, D and B are both Hamiltonian matrices, and then a symplectic orthogonal matrix E is constructed, as shown in the following formula (3):

[0089]

[0090] Wherein, the matrix E is represented as an orthogonal symplectic matrix with symplectic matrix properties; F is represented as an upper triangular matrix, and the eigenvalue of the upper triangular matrix F is represented as λ i (i=1,2,···,k);f ij =0(i>j+1);

[0091] Step 4. If the matrix C is a real symmetric matrix, then the eigenvalue of C is equal to the eigenvalue of the upper triangular matrix F. According to the properties of the Hamiltonian matrix, the eigenvalue of C is The eigenvalues of the matrix C can be expressed as β1>β2>···>β k ;

[0092] Let P i (i=1,2,···,k) is the eigenvector of matrix C, let P i (i=1,2,···,k) are the eigenvectors of matrix C, and let S i =P i T A T And Z i =P i S i , then the initial reconstruction trajectory matrix Z is defined as follows:

[0093] Z=Z1+Z2+…+Z k (4);

[0094] Step 5: Since the initial reconstructed trajectory matrix Z is an m×k matrix, it needs to be converted into k initial symplectic geometric components Y of length n by diagonal averaging. i (i=1,2,···,k):

[0095] Any element of Z is defined as z ij , where 1≤i≤k, 1≤j≤m, k * =min(m,k),m *= max(m, k), n = m + (k - 1)τ;

[0096] If m < k, let z ij * = z ij ; Otherwise, let z ij * = z ji ; Therefore, the process of diagonal averaging is shown in Equation (5) below:

[0097]

[0098] By processing the initial reconstruction trajectory matrix Z, k initial symplectic geometric components Y i (y1, y2, ···, y n ) are obtained, as shown in Equation (6) below:

[0099] Y = Y1 + Y2 + … + Y k (6).

[0100] Finally, according to the preset component selection criteria of cyclic kurtosis entropy, after discarding the noise components in the initial symplectic geometric components of each short signal, further reconstruction is performed to obtain the denoised signal. The specific steps are as follows:

[0101] Step 1: Calculate the cyclic kurtosis entropy of the initial symplectic geometric component Y i to obtain CKE1, CKE2, ···, CKE k , and construct a weight function to calculate the weight value L i corresponding to each initial symplectic geometric component Y i ;

[0102]

[0103] where CKE mean represents the average value of CKE; i = 1, 2, ···, k;

[0104] Step 2: The final denoised signal component M can be obtained through Equation (8) below:

[0105]

[0106] Step 3: Reconstruct the denoised signal component M to obtain the denoised signal;

[0107] where the calculation process of the cyclic kurtosis entropy of the initial symplectic geometric component Y i is as follows:

[0108] Given a signal b = {b1, b2, ···, b n}, then the cycle kurtosis (CK) can be expressed as shown in the following formula (9):

[0109]

[0110] Wherein, RSE represents the autocorrelation function of the square envelope signal SE(b), and SE(b) = |b| 2 =|b+jHilber(b)| 2 , Hilbert represents Hilbert transform; Ra(0) is the autocorrelation function value of signal b when the delay coefficient is 0, and R b (τ)=E[b(ni)b(ni+τ)]; hT (h=1, 2, ···, n) represents the delay coefficient;

[0111] The CK value of any delay coefficient is calculated by the following formula (10), and the cyclic kurtosis entropy can be expressed as follows by the following formula (11):

[0112]

[0113]

[0114] Where, h = {1, 2, ···, n}.

[0115] Step S4: plotting the Hilbert envelope spectrum of the noise reduction signal, and performing characteristic frequency analysis on the Hilbert envelope spectrum to detect the mechanical wear state of the Roots-type compressor.

[0116] The specific process involves performing Hilbert envelope demodulation on the noise-reduced signal in a multi-channel data acquisition and analysis instrument to detect the mechanical wear status of the Roots compressor. Mechanical wear status of the Roots compressor includes, but is not limited to, bearing inner race faults, bearing outer race faults, rolling element faults, impeller meshing, and gear faults.

[0117] It should be noted that the mechanical wear state of the Roots compressor is detected by analyzing whether the noise reduction signal after Hilbert envelope demodulation appears at a frequency that matches the theoretical fault characteristic frequency; if so, a fault conclusion is drawn; otherwise, a normal conclusion is drawn.

[0118] Among them, the fault characteristic frequencies of the Roots compressor are: bearing inner ring fault frequency, outer ring fault frequency, rolling element fault frequency, impeller meshing frequency and gear fault frequency.

[0119] The outer race fault characteristic frequency BPFO is The inner race fault characteristic frequency BPFI is The rolling element fault characteristic frequency BSF is The impeller meshing frequency IEF is IEF=NIEF f shaft ; Gear fault characteristic frequency GF is GF=f shaft .

[0120] Where N r Indicates the number of rolling elements, f shaft Indicates the shaft speed, R d Indicates the roller diameter, P d represents the pitch diameter, α represents the contact angle, N IEF Indicates the number of impeller teeth.

[0121] like Figures 2 to 4 As shown, the application scenario of the mechanical wear state detection method of a Roots-type compressor provided in an embodiment of the present invention is further described as follows:

[0122] Taking a Roots compressor bearing with an inner ring fault as an example, the bearing inner ring fault frequency is 147.8 Hz. The diagnostic results of the method of the present invention are compared with those of the envelope analysis method, which demonstrates the effectiveness of the method of the present invention. Figure 2 This is the vibration signal collected when the inner ring of the bearing of a Roots compressor fails. Figure 3 The result diagram of the application of envelope analysis method to process the fault signal of the inner ring of the Roots compressor bearing is given. It can be seen from the figure that the characteristic frequency of the inner ring fault is not displayed. Figure 4 The figure shows the result obtained by processing the fault signal of the inner ring of the bearing of the Roots compressor by the method of the present invention. Figure 4 As shown in the figure, the fault characteristic frequency of the bearing inner race is 146.9Hz, and its double and triple frequencies are clearly displayed. Therefore, compared with the theoretical calculated values, it can be determined that the fault type of this Roots compressor is a bearing inner race fault.

[0123] like Figure 5 FIG. 1 is a diagram showing a mechanical wear state detection system for a Roots-type compressor according to an embodiment of the present invention, comprising:

[0124] The signal acquisition unit 110 is used to acquire the original vibration signal transmitted from the Roots compressor;

[0125] The signal segmentation unit 120 is configured to segment the original vibration signal into a plurality of short signals based on a preset adaptive window segmentation method;

[0126] a signal feature extraction unit 130 configured to decompose the multiple short signals based on a preset symplectic geometric pattern decomposition method to obtain an initial symplectic geometric component of each short signal, and discard the noise component in the initial symplectic geometric component of each short signal according to a preset component selection criterion of cyclic kurtosis entropy, and then further reconstruct the signal to obtain a noise-reduced signal;

[0127] The mechanical wear state detection unit 140 is configured to plot the Hilbert envelope spectrum of the noise reduction signal and perform characteristic frequency analysis on the Hilbert envelope spectrum to detect the mechanical wear state of the Roots-type compressor.

[0128] The mechanical wear state of the Roots-type compressor includes bearing inner ring failure, bearing outer ring failure, rolling element failure, impeller meshing and gear failure.

[0129] The implementation of the embodiments of the present invention has the following beneficial effects:

[0130] On the one hand, the present invention utilizes an adaptive window segmentation method to effectively discover weak fault features, and on the other hand, selects effective components for reconstruction based on cyclic kurtosis entropy to effectively remove noise interference, thereby accurately, reliably and stably detecting the mechanical wear state of a Roots-type compressor. This not only extracts a variety of fault features, but also weak fault features, thereby improving the accuracy, reliability and stability of mechanical wear state detection and reducing the possibility of accidents caused by failures of rotating parts.

[0131] It is worth noting that in the above system embodiment, the various units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0132] Those skilled in the art will understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc.

[0133] The above disclosure is merely a preferred embodiment of the present invention and certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for detecting the mechanical wear state of a Roots-type compressor, characterized in that: The method comprises the following steps: Obtain the original vibration signal transmitted from the Roots compressor; Based on a preset adaptive window segmentation method, the original vibration signal is segmented into multiple short signals; Decomposing the multiple short signals based on a preset symplectic geometric pattern decomposition method to obtain an initial symplectic geometric component of each short signal, discarding a noise component in the initial symplectic geometric component of each short signal according to a preset component selection criterion of cyclic kurtosis entropy, and further reconstructing the signal to obtain a noise-reduced signal; plotting a Hilbert envelope spectrum of the noise reduction signal and performing characteristic frequency analysis on the Hilbert envelope spectrum to detect a mechanical wear state of the Roots-type compressor; The specific steps of discarding the noise component in the initial symplectic geometric component of each short signal according to the preset component selection standard of cyclic kurtosis entropy and then further reconstructing to obtain the noise-reduced signal include: Step 1: Calculate the initial symplectic geometry components Y i The cyclic kurtosis entropy of CKE 1 ,CKE 2 ,···,CKE k , and construct the weight function to calculate each initial symplectic geometric component through the following formula (7): Y i Corresponding weight value L i ; （7）； in, CKE mean Expressed as CKE Average value ; i=1,2,···,k ; Step 2: Final noise reduction signal component M It can be obtained by the following formula (8): （8）； The third step is to convert the noise reduction signal component M Reconstruct and obtain the noise-reduced signal; Among them, the initial symplectic geometric components Y i The calculation process of the cyclic kurtosis entropy is as follows: Given a length of n signal b ={ b 1 , b 2 ,···, b n }, then the cycle kurtosis (CK) can be expressed as shown in the following formula (9): （9）； in, RSE Represented as a square envelope signal SE(b) The autocorrelation function of , Hilbert Expressed as Hilbert transform; Rb(0) The signal when the delay coefficient is 0 b The autocorrelation function value of ; hT ( h =1,2,···, n ) is expressed as the delay coefficient; The CK value of any delay coefficient is calculated by the following formula (10), and the cyclic kurtosis entropy can be expressed as follows by the following formula (11): （10）； （11）； in, h={1,2,···,n} .

2. The method for detecting the mechanical wear state of a Roots-type compressor according to claim 1, wherein: The specific steps of segmenting the original vibration signal into multiple short signals based on the preset adaptive window segmentation method include: Step 1: Set the initial window length W and moving step length V ; Among them, the initial window length is set to W = f s / f shift , f s is the sampling frequency, f shift is the rotation frequency; the moving step V is the initial window length W half of The second step is to determine the initial window on the original vibration signal and use the window length W The adaptive window is expanded to the right for the basic unit, and adjacent windows are merged in sequence for expansion; The third step is to determine the original vibration signal in the window after each expansion. CKE Size; if the original vibration signal in the window after the current expansion CKE If the window becomes smaller, the window after the current expansion is retained as the initial window, and the next window is further added for expansion; otherwise, the window before the current expansion is intercepted as an initial window, and the currently expanded window is re-identified as a new initial window and its corresponding adjacent windows are continuously merged for expansion; Step 4: Repeat the above step 3 until all the initial windows slide across the entire original vibration signal, so that the original vibration signal is divided into multiple short signals.

3. The method for detecting the mechanical wear state of a Roots-type compressor according to claim 2, wherein: The specific steps of decomposing the multiple short signals based on the preset symplectic geometric pattern decomposition method to obtain the initial symplectic geometric components of each short signal include: Step 1: Assume that the current short signal is a ={ a 1, a 2,···, a n };in, n Represented as a signal a length; Then, define the signal a The trajectory matrix A , as shown in the following formula (1): （1）； in, k Represented as embedding dimension; embedding dimension k By calculating the signal a The power spectral density of the power spectrum is obtained by estimating the frequency of the maximum peak; if the normalized frequency is less than a given threshold, then k=n / 3 ,otherwise, k=1.2×(f s / f max ) ; f s Expressed as sampling frequency, f max Represented as a signal a The frequency of the maximum peak of the power spectral density; τ Expressed as delay time; m = n -( k -1) τ ; The second step is to construct the Hamiltonian matrix according to formula (2). B : （2）； in, C=A T A ; Step 3: Order D=B 2 , according to the definition of Hamiltonian matrix, D and B Are all Hamiltonian matrices, and then construct a symplectic orthogonal matrix E , as shown in the following formula (3): （3）； Among them, the matrix E It is represented as an orthogonal symplectic matrix with symplectic matrix properties; F is represented as an upper triangular matrix, and the upper triangular matrix F The eigenvalues of λ i ( i =1,2,···, k ); f ij =0( i > j +1); Step 4: If the matrix C is a real symmetric matrix, then C The eigenvalues of the upper triangular matrix F The eigenvalue of ; According to the properties of the Hamiltonian matrix, C The characteristic value of β i = ( i =1,2,···, k ),matrix C The eigenvalues of can be expressed as β 1> β 2>···> β k ; make P i ( i =1,2,···, k ) is a matrix C The characteristic vector of P i ( i =1,2,···, k ) is a matrix C The eigenvector of , and let S i = P i T A T and Z i = P i S i , then the initial reconstruction trajectory matrix Z The definition is as follows: （4）； Step 5: Due to the initial reconstruction trajectory matrix Z for m×k Matrix, so it needs to be converted into a matrix of length by diagonal averaging. n of k Initial symplectic geometry components Y i ( i =1,2,···, k ): Any Z The elements are defined as z ij , where 1≤ i ≤ k , 1≤ j ≤ m , k * =min( m , k ), m * =max( m , k ), n = m +( k- 1) τ ; if m < k ,make z ij * = z ij Otherwise, let z ij * = z ji ; Therefore, the diagonal averaging process is shown in the following formula (5): （5）； By processing the initial reconstruction trajectory matrix Z ,get k Initial symplectic geometry components Y i ( y 1, y 2 , ···, y n ), as shown in the following formula (6): （6）。 4. The method for detecting the mechanical wear state of a Roots-type compressor according to claim 1, wherein: The mechanical wear conditions of the Roots-type compressor include bearing inner race failure, bearing outer race failure, rolling element failure, impeller meshing, and gear failure.

5. A mechanical wear state detection system for a Roots-type compressor, characterized in that: include: A signal acquisition unit is used to acquire the original vibration signal transmitted from the Roots compressor; A signal segmentation unit, configured to segment the original vibration signal into a plurality of short signals based on a preset adaptive window segmentation method; a signal feature extraction unit, configured to decompose the plurality of short signals based on a preset symplectic geometric pattern decomposition method to obtain an initial symplectic geometric component of each short signal, and discard a noise component in the initial symplectic geometric component of each short signal according to a preset component selection criterion of cyclic kurtosis entropy, and then further reconstruct the signal to obtain a noise-reduced signal; a mechanical wear state detection unit, configured to plot a Hilbert envelope spectrum of the noise reduction signal and perform characteristic frequency analysis on the Hilbert envelope spectrum to detect the mechanical wear state of the Roots-type compressor; The specific steps of discarding the noise component in the initial symplectic geometric component of each short signal according to the preset component selection standard of cyclic kurtosis entropy and then further reconstructing to obtain the noise-reduced signal include: Step 1: Calculate the initial symplectic geometry components Y i The cyclic kurtosis entropy of CKE 1 ,CKE 2 ,···,CKE k , and construct the weight function to calculate each initial symplectic geometric component through the following formula (7): Y i Corresponding weight value L i ; （7）； in, CKE mean Expressed as CKE Average value ; i=1,2,···,k ; Step 2: Final noise reduction signal component M It can be obtained by the following formula (8): （8）； The third step is to convert the noise reduction signal component M Reconstruct and obtain the noise-reduced signal; Among them, the initial symplectic geometric components Y i The calculation process of the cyclic kurtosis entropy is as follows: Given a length of n signal b ={ b 1 , b 2 ,···, b n }, then the cycle kurtosis (CK) can be expressed as shown in the following formula (9): （9）； in, RSE Represented as a square envelope signal SE(b) The autocorrelation function of , Hilbert Expressed as Hilbert transform; Rb(0) The signal when the delay coefficient is 0 b The autocorrelation function value of ; hT ( h =1,2,···, n ) is expressed as the delay coefficient; The CK value of any delay coefficient is calculated by the following formula (10), and the cyclic kurtosis entropy can be expressed as follows by the following formula (11): （10）； （11）； in, h={1,2,···,n} .

6. The mechanical wear state detection system for a Roots-type compressor according to claim 5, characterized in that: The mechanical wear conditions of the Roots-type compressor include bearing inner race failure, bearing outer race failure, rolling element failure, impeller meshing, and gear failure.

Images