Bearing fault diagnosis method and system based on particle swarm optimization generalized variational mode decomposition

Abstract

The invention discloses a bearing fault diagnosis method and system based on particle swarm optimization generalized variational mode decomposition, and relates to the field of fault diagnosis. The problems between excessive filtering and mode aliasing in an existing variational mode decomposition method are solved. The method comprises the following steps: collecting a vibration time domain signal in the operation process of the circulating water pump, and inputting the vibration time domain signal into a generalized variational mode decomposition model to obtain a series of decomposition modes; constructing a decomposition mode function, taking the maximum value of a 1.5-dimensional spectrum weighted harmonic-to-noise ratio as a target, obtaining an optimal setting parameter of generalized variational mode decomposition through a particle swarm algorithm, and re-decomposing the vibration time domain signal by a method for determining the optimal setting parameter based on the generalized variational mode decomposition to obtain an optimized decomposition mode; a 1.5-dimensional spectrum weighted harmonic-to-noise ratio formula is constructed; determining an optimal decomposition mode based on the maximum value of a 1.5-dimensional spectrum weighted harmonic-to-noise ratio formula; and determining a fault diagnosis result according to the optimal decomposition mode weighted harmonic-to-noise ratio result. The method is also suitable for verifying the fault type accuracy of the bearing.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a bearing fault diagnosis method and system based on particle swarm optimization generalized variational mode decomposition. Background Art

[0002] The circulating water pump is one of the important auxiliary equipment in nuclear power plants. Its main function is to transport cooling water to the condenser and other cooling equipment to ensure the safe and stable operation of nuclear power units. The circulating water pump usually operates under high load and continuous operation. The reliability of operation is crucial to the safety and economy of nuclear power plants. Rolling bearings are one of the key components of circulating water pumps in nuclear power plants. Their operating status directly affects the performance and life of circulating water pumps. Once a fault occurs, it may cause bearing overheating, abnormal pump vibration, and even cause serious equipment failure, threatening the safety and stability of nuclear power units. Therefore, the research and application of rolling bearing fault diagnosis technology can further improve the reliability of circulating water pumps in nuclear power plants and provide a solid guarantee for the safe and stable operation of nuclear power units. Traditional signal processing methods have the problem of simultaneously weakening fault feature information while separating and extracting fault components, so that it is difficult to effectively extract feature information, resulting in inaccurate fault diagnosis results.

[0003] In terms of signal processing, the commonly used methods are mainly divided into three categories: time domain, frequency domain and time-frequency domain. The time domain decomposition method is a signal processing technology that decomposes the signal into different time domain components. These methods mainly include empirical mode decomposition EMD, local mean decomposition LMD, intrinsic time scale decomposition ITD, selective filtering decomposition IFD, adaptive local iterative filtering method ALIF, etc. The frequency domain filtering method is to decompose the signal into different frequency components and then process them separately. The methods based on this idea mainly include discrete wavelet transform DWT, wavelet packet transform WPT, empirical wavelet transform EWT, spectral kurtosis SK, fast spectral FK, variational mode decomposition VMD, adaptive chirp mode decomposition ACMD frequency domain filtering methods, etc. The time-frequency analysis method converts one-dimensional information into a two-dimensional joint distribution of time and frequency, which can more effectively describe the time-varying frequency characteristics of non-stationary signals. The commonly used time-frequency analysis methods currently include short-time Fourier transform STFT, smoothed pseudo-Wigner-Ville distribution SPWVD, redistribution transform RS, continuous wavelet transform CWT, synchronous compression transform SST, etc.

[0004] VMD is an emerging signal processing method based on the variational principle. It realizes feature extraction by decomposing complex signals into a set of finite intrinsic mode functions (IMFs). Compared with traditional signal processing methods, VMD has many advantages in rolling bearing fault diagnosis. VMD can adaptively decompose complex signals by optimizing the decomposition process and gradually extracting the frequency information of the signal, thus avoiding the problem of mode aliasing. VMD can accurately separate different modes through frequency domain analysis and improve the decomposition accuracy. VMD directly decomposes in the frequency domain without assuming the stationarity of the signal. It can adapt to the non-stationary characteristics of rolling bearing fault signals and effectively extract key frequency components in impact signals and modulation signals. VMD can accurately control the number and frequency range of decomposed modes by setting the number of modes and frequency bandwidth limits, thus avoiding information redundancy or omission of important features.

[0005] However, due to the complex and changeable working environment of the circulating water pump in the nuclear power system, the collected vibration signals have the characteristics of diverse components, uncertain bandwidth of each component, and strong background noise, which leads to the submersion of fault information. The excessive filtering caused by the excessive penalty factor when performing modal separation in the VMD method will lead to the simultaneous weakening of the fault feature information while filtering out the noise. The excessively small penalty factor will lead to modal aliasing. Therefore, it is urgent to improve VMD to solve the contradiction between excessive filtering and modal aliasing. Summary of the invention

[0006] The present invention aims to solve the problems of over-filtering and modal aliasing in the variational modal decomposition method in the prior art. To solve the above technical problems, the present invention is implemented by the following technical solutions:

[0007] Solution 1: The present invention proposes a bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition, the method comprising the following steps:

[0008] Step 1, collecting the vibration time domain signal during the operation of the circulating water pump;

[0009] Step 2: input the vibration time domain signal collected in step 1 into the generalized variational mode decomposition model to obtain a series of decomposed modes;

[0010] Step 3: Based on the series of decomposed modes obtained in step 2, a decomposition modal function is constructed, and the maximum value of the 1.5-dimensional spectral weighted harmonic noise ratio is taken as the target. The optimal setting parameters of the generalized variational modal decomposition model are obtained by particle swarm optimization, where the optimal setting parameters include decomposition order, mode number and penalty factor;

[0011] Step 4: re-decompose the vibration time domain signal based on the method for optimally setting parameters of the generalized variational modal decomposition model described in step 3 to obtain a series of optimized decomposition modes;

[0012] Step 5: construct a 1.5-dimensional spectrum weighted harmonic noise ratio formula based on the vibration time domain signal re-decomposed in step 4;

[0013] Step 6, determining the optimal decomposition mode based on the maximum value of the 1.5-dimensional spectrum weighted harmonic noise ratio formula described in step 5;

[0014] Step 7: Determine the fault diagnosis result according to the optimal decomposition modal weighted harmonic noise ratio result determined in step 6.

[0015] Further, a preferred implementation is provided, in step 3, taking the maximum value of the 1.5-dimensional spectral weighted harmonic noise ratio as the target, the method for obtaining the optimal setting parameters of the generalized variational mode decomposition by the particle swarm algorithm is:

[0016]

[0017] Where B(ω) is the diagonal slice after Fourier transformation, R 3x (τ1,τ2)=E{x(t)x(t+τ1)x(t+τ2)}, E is the mathematical expectation, τ1,τ2 are different time delays, R 3x (τ1, τ2) is the main diagonal slice; taking τ1 = τ2 = τ, the main diagonal slice of the third-order cumulant is: R 3x (τ,τ)=E{x(t)x(t+τ)x(t+τ)}.

[0018] Further, a preferred implementation is provided, in step 5, based on the re-decomposed vibration time domain signal in step 4, a 1.5-dimensional spectrum weighted harmonic noise ratio formula is constructed as follows:

[0019]

[0020] Among them, WHNR f is the weighted harmonic noise ratio, x is the 1.5-dimensional spectrum demodulation result, f a is the intermediate variable, r x To calculate the autocorrelation of the input signal, τ p For r x The position offset of the autocorrelation function, r x (τ p ) is the amplitude corresponding to the offset, r x (0) is the total energy of the autocorrelation function, i is the weighted range, f is the theoretical fault frequency, and f w is the HNR weight under the current fault frequency, and n is a positive integer.

[0021] Further, a preferred implementation is provided, in which the method for obtaining the optimal setting parameters of generalized variational mode decomposition by particle swarm algorithm in S3 is:

[0022] S3.1, initialize particle parameters, including the initial velocity and position of particles, the number of iterations of particles, population size, inertia factor and acceleration factor;

[0023] S3.2, update the particle position and velocity, set the parameter value range that needs to be optimized for generalized variational mode decomposition, and use it as the value range for updating the particle velocity and position;

[0024] S3.3, using 1.5-dimensional spectral weighted harmonic noise ratio as fitness function;

[0025] S3.4, calculate the fitness of each particle and find the individual extreme value and group extreme value;

[0026] S3.5. Update the position and velocity of each particle and iterate continuously until the optimal solution is found.

[0027] Furthermore, a preferred implementation is provided, in which the fault diagnosis result is determined according to the optimal decomposition modal weighted harmonic-to-noise ratio result determined in step 6 by determining the weighted harmonic-to-noise ratio thresholds corresponding to different faults.

[0028] Furthermore, a preferred embodiment is provided to determine the weighted harmonic-to-noise ratio thresholds corresponding to different faults, including the weighted harmonic-to-noise ratio threshold corresponding to the inner ring fault, the weighted harmonic-to-noise ratio threshold corresponding to the outer ring fault, the weighted harmonic-to-noise ratio threshold corresponding to the rolling element fault, and the weighted harmonic-to-noise ratio threshold corresponding to the retaining cage fault.

[0029] Further, a preferred embodiment is provided, wherein the weighted harmonic-to-noise ratio threshold corresponding to the inner race fault, the weighted harmonic-to-noise ratio threshold corresponding to the outer race fault, the weighted harmonic-to-noise ratio threshold corresponding to the rolling element fault, and the weighted harmonic-to-noise ratio threshold corresponding to the cage fault specifically include:

[0030] Determine the weighted harmonic noise ratio WHNR corresponding to the inner race fault fi Is it less than the weighted harmonic noise ratio threshold corresponding to the inner ring fault? If the weighted harmonic noise ratio WHNR corresponding to the inner ring fault fi If the weighted harmonic noise ratio WHNR corresponding to the inner ring fault is smaller than the weighted harmonic noise ratio threshold corresponding to the inner ring fault, it means that it is normal; if the weighted harmonic noise ratio WHNR corresponding to the inner ring fault is fi If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the inner ring fault, it means that there is an inner ring fault;

[0031] Determine the weighted harmonic noise ratio WHNR corresponding to the outer ring fault fo Is it less than the weighted harmonic noise ratio threshold corresponding to the outer ring fault? If the weighted harmonic noise ratio WHNR corresponding to the outer ring fault fo If the weighted harmonic noise ratio WHNR corresponding to the outer ring fault is smaller than the weighted harmonic noise ratio threshold corresponding to the outer ring fault, it is normal; if the weighted harmonic noise ratio WHNR corresponding to the outer ring fault is fo If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the outer race fault, it means that there is an outer race fault;

[0032] Determine the weighted harmonic noise ratio (WHNR) corresponding to the rolling element fault fr Is it less than the weighted harmonic noise ratio threshold corresponding to the rolling element fault? If the weighted harmonic noise ratio WHNR corresponding to the rolling element fault fr If the weighted harmonic noise ratio WHNR corresponding to the rolling element fault is less than the weighted harmonic noise ratio threshold corresponding to the rolling element fault, it is normal; if the weighted harmonic noise ratio WHNR corresponding to the rolling element fault is fr If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the rolling element fault, it means that there is a rolling element fault;

[0033] Determine the weighted harmonic noise ratio WHNR corresponding to the cage failure fc Is it less than the weighted harmonic noise ratio threshold corresponding to the cage fault? If the weighted harmonic noise ratio WHNR corresponding to the cage fault fc If the weighted harmonic noise ratio WHNR corresponding to the cage fault is less than the weighted harmonic noise ratio threshold corresponding to the cage fault, it is normal; if the weighted harmonic noise ratio WHNR corresponding to the cage fault is fc If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the cage fault, it means that there is a cage fault.

[0034] Solution 2: A bearing fault diagnosis system based on particle swarm optimization generalized variational mode decomposition, the system comprising:

[0035] A data acquisition module is used to collect vibration time domain signals during the operation of the circulating water pump;

[0036] A decomposition mode module is used to input the vibration time domain signal collected by the data acquisition module into the generalized variational mode decomposition model to obtain a series of decomposition modes;

[0037] The particle swarm optimization module is used to construct the decomposition mode function based on a series of decomposition modes obtained by the decomposition mode module. The optimal setting parameters of the generalized variational mode decomposition model are obtained by the particle swarm algorithm with the maximum value of the 1.5-dimensional spectral weighted harmonic noise ratio as the goal, where the optimal setting parameters include the decomposition order, the number of modes and the penalty factor;

[0038] The optimized decomposition mode module and the method for determining the optimal setting parameters of the generalized variational mode decomposition model described in the particle swarm optimization module re-decompose the vibration time domain signal to obtain a series of optimized decomposition modes;

[0039] A weighted harmonic noise ratio formula determination module is used to construct a 1.5-dimensional spectrum weighted harmonic noise ratio formula based on the vibration time domain signal re-decomposed by the optimized decomposition mode module;

[0040] An optimal decomposition mode module, used for determining the optimal decomposition mode based on the maximum value of the 1.5-dimensional spectrum weighted harmonic noise ratio formula described in the weighted harmonic noise ratio formula determination module;

[0041] The fault diagnosis module is used to determine the fault diagnosis result according to the optimal decomposition mode weighted harmonic-to-noise ratio result determined by the optimal decomposition mode module.

[0042] Solution three: A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes any one of the methods described in Solution one.

[0043] Solution 4: A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of Solution 1.

[0044] The present invention is beneficial in that:

[0045] The bearing fault diagnosis method based on generalized variational modal decomposition (GVDM) optimized by particle swarm optimization described in the present invention solves the contradiction between over-filtering and modal aliasing in the variational modal decomposition method in the prior art. The vibration signal is decomposed by establishing a generalized variational modal decomposition model. The decomposed mode has more fault information than the variational modal decomposition, which improves the accuracy of bearing fault diagnosis. The particle swarm algorithm is used to automatically optimize the parameters to achieve rapid and accurate diagnosis of the bearing fault type.

[0046] The present invention is also applicable to the field of verifying the accuracy of bearing fault types. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flow chart of a bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition described in implementation mode one. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the implementation methods of the present application clearer, the technical solutions in the implementation methods of the present application will be clearly and completely described below in conjunction with the drawings in the implementation methods of the present application. Obviously, the described implementation methods are only part of the implementation methods of the present application, not all of the implementation methods.

[0049] Embodiment 1: This embodiment provides a bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition, the method comprising the following steps:

[0050] Step 1, collecting the vibration time domain signal during the operation of the circulating water pump;

[0051] Step 2: input the vibration time domain signal collected in step 1 into the generalized variational mode decomposition model to obtain a series of decomposed modes;

[0052] Step 3: Based on the series of decomposed modes obtained in step 2, a decomposition modal function is constructed, and the maximum value of the 1.5-dimensional spectral weighted harmonic noise ratio is taken as the target. The optimal setting parameters of the generalized variational modal decomposition model are obtained by particle swarm optimization, where the optimal setting parameters include decomposition order, mode number and penalty factor;

[0053] Step 4: re-decompose the vibration time domain signal based on the method for optimally setting parameters of the generalized variational modal decomposition model described in step 3 to obtain a series of optimized decomposition modes;

[0054] Step 5: construct a 1.5-dimensional spectrum weighted harmonic noise ratio formula based on the vibration time domain signal re-decomposed in step 4;

[0055] Step 6, determining the optimal decomposition mode based on the maximum value of the 1.5-dimensional spectrum weighted harmonic noise ratio formula described in step 5;

[0056] Step 7: Determine the fault diagnosis result according to the optimal decomposition modal weighted harmonic noise ratio result determined in step 6.

[0057] Implementation method 2: This implementation method is a further limitation of the bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition described in implementation method 1. In step 3, the maximum value of the 1.5-dimensional spectral weighted harmonic noise ratio is taken as the target, and the method for obtaining the optimal setting parameters of generalized variational mode decomposition through the particle swarm algorithm is:

[0058]

[0059] Where B(ω) is the diagonal slice after Fourier transformation, R 3x (τ1,τ2)=E{x(t)x(t+τ1)x(t+τ2)}, E is the mathematical expectation, τ1,τ2 are different time delays, R 3x (τ1, τ2) is the main diagonal slice; taking τ1 = τ2 = τ, the main diagonal slice of the third-order cumulant is: R 3x (τ,τ)=E{x(t)x(t+τ)x(t+τ)}.

[0060] Implementation method three: This implementation method is a further limitation of the bearing fault diagnosis method based on particle swarm optimization generalized variational modal decomposition described in implementation method one. In step 5, the 1.5-dimensional spectrum weighted harmonic noise ratio formula is constructed based on the re-decomposed vibration time domain signal in step 4:

[0061]

[0062] Among them, WHNR f is the weighted harmonic noise ratio, x is the 1.5-dimensional spectrum demodulation result, f a is the intermediate variable, r x To calculate the autocorrelation of the input signal, τp For r x The position offset of the autocorrelation function, r x (τ p ) is the amplitude corresponding to the offset, r x (0) is the total energy of the autocorrelation function, i is the weighted range, f is the theoretical fault frequency, and f w is the HNR weight under the current fault frequency, and n is a positive integer.

[0063] Implementation method 4: This implementation method is a further limitation of the bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition described in implementation method 1. The method for obtaining the optimal setting parameters of generalized variational mode decomposition by particle swarm algorithm in S3 is:

[0064] S3.1, initialize particle parameters, including the initial velocity and position of particles, the number of iterations of particles, population size, inertia factor and acceleration factor;

[0065] S3.2, update the particle position and velocity, set the parameter value range that needs to be optimized for generalized variational mode decomposition, and use it as the value range for updating the particle velocity and position;

[0066] S3.3, using 1.5-dimensional spectral weighted harmonic noise ratio as fitness function;

[0067] S3.4, calculate the fitness of each particle and find the individual extreme value and group extreme value;

[0068] S3.5. Update the position and velocity of each particle and iterate continuously until the optimal solution is found.

[0069] Implementation method five: This implementation method is a further limitation of the bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition described in implementation method one. The fault diagnosis result is determined according to the optimal decomposition mode weighted harmonic-to-noise ratio result determined in step 6 by determining the weighted harmonic-to-noise ratio threshold corresponding to different faults.

[0070] Implementation method six. This implementation method is a further limitation of the bearing fault diagnosis method based on particle swarm optimization generalized variational modal decomposition described in implementation method one, and determines the weighted harmonic-to-noise ratio thresholds corresponding to different faults, including the weighted harmonic-to-noise ratio threshold corresponding to the inner ring fault, the weighted harmonic-to-noise ratio threshold corresponding to the outer ring fault, the weighted harmonic-to-noise ratio threshold corresponding to the rolling element fault, and the weighted harmonic-to-noise ratio threshold corresponding to the retaining cage fault.

[0071] Embodiment 7. This embodiment further limits the bearing fault diagnosis method based on generalized variational mode decomposition of particle swarm optimization described in embodiment 6. The weighted harmonic noise ratio threshold corresponding to the inner ring fault, the weighted harmonic noise ratio threshold corresponding to the outer ring fault, the weighted harmonic noise ratio threshold corresponding to the rolling element fault, and the weighted harmonic noise ratio threshold corresponding to the cage fault specifically include:

[0072] Determine the weighted harmonic noise ratio WHNR corresponding to the inner race fault fi Is it less than the weighted harmonic noise ratio threshold corresponding to the inner ring fault? If the weighted harmonic noise ratio WHNR corresponding to the inner ring fault fi If the weighted harmonic noise ratio WHNR corresponding to the inner ring fault is smaller than the weighted harmonic noise ratio threshold corresponding to the inner ring fault, it means that it is normal; if the weighted harmonic noise ratio WHNR corresponding to the inner ring fault is fi If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the inner ring fault, it means that there is an inner ring fault;

[0073] Determine the weighted harmonic noise ratio WHNR corresponding to the outer ring fault fo Is it less than the weighted harmonic noise ratio threshold corresponding to the outer ring fault? If the weighted harmonic noise ratio WHNR corresponding to the outer ring fault fo If the weighted harmonic noise ratio WHNR corresponding to the outer ring fault is smaller than the weighted harmonic noise ratio threshold corresponding to the outer ring fault, it is normal; if the weighted harmonic noise ratio WHNR corresponding to the outer ring fault is fo If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the outer race fault, it means that there is an outer race fault;

[0074] Determine the weighted harmonic noise ratio (WHNR) corresponding to the rolling element fault fr Is it less than the weighted harmonic noise ratio threshold corresponding to the rolling element fault? If the weighted harmonic noise ratio WHNR corresponding to the rolling element fault fr If the weighted harmonic noise ratio WHNR corresponding to the rolling element fault is less than the weighted harmonic noise ratio threshold corresponding to the rolling element fault, it is normal; if the weighted harmonic noise ratio WHNR corresponding to the rolling element fault is fr If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the rolling element fault, it means that there is a rolling element fault;

[0075] Determine the weighted harmonic noise ratio WHNR corresponding to the cage failure fc Is it less than the weighted harmonic noise ratio threshold corresponding to the cage fault? If the weighted harmonic noise ratio WHNR corresponding to the cage fault fc If the weighted harmonic noise ratio WHNR corresponding to the cage fault is less than the weighted harmonic noise ratio threshold corresponding to the cage fault, it is normal; if the weighted harmonic noise ratio WHNR corresponding to the cage fault is fc If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the cage fault, it means that there is a cage fault.

[0076] Embodiment 8: This embodiment proposes a bearing fault diagnosis system based on particle swarm optimization generalized variational mode decomposition, the system comprising:

[0077] A data acquisition module is used to collect vibration time domain signals during the operation of the circulating water pump;

[0078] A decomposition mode module is used to input the vibration time domain signal collected by the data acquisition module into the generalized variational mode decomposition model to obtain a series of decomposition modes;

[0079] The particle swarm optimization module is used to construct the decomposition mode function based on a series of decomposition modes obtained by the decomposition mode module. The optimal setting parameters of the generalized variational mode decomposition model are obtained by the particle swarm algorithm with the maximum value of the 1.5-dimensional spectral weighted harmonic noise ratio as the goal, where the optimal setting parameters include the decomposition order, the number of modes and the penalty factor;

[0080] The optimized decomposition mode module and the method for determining the optimal setting parameters of the generalized variational mode decomposition model described in the particle swarm optimization module re-decompose the vibration time domain signal to obtain a series of optimized decomposition modes;

[0081] A weighted harmonic noise ratio formula determination module is used to construct a 1.5-dimensional spectrum weighted harmonic noise ratio formula based on the vibration time domain signal re-decomposed by the optimized decomposition mode module;

[0082] An optimal decomposition mode module, used for determining the optimal decomposition mode based on the maximum value of the 1.5-dimensional spectrum weighted harmonic noise ratio formula described in the weighted harmonic noise ratio formula determination module;

[0083] The fault diagnosis module is used to determine the fault diagnosis result according to the optimal decomposition mode weighted harmonic-to-noise ratio result determined by the optimal decomposition mode module.

[0084] Embodiment 9. This embodiment proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the method described in any one of embodiments 1 to 7.

[0085] Embodiment 10: This embodiment proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the method described in any one of embodiments 1 to 7 are implemented.

[0086] Implementation eleven: This implementation provides an example, which is used to explain the above implementations one to eight. The specific example is as follows:

[0087] See also Figure 1 To illustrate this embodiment, the method described in this embodiment includes the following steps:

[0088] The purpose of this embodiment is to provide a bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition. The particle swarm optimization algorithm is introduced because the generalized variational mode decomposition algorithm has one more hyperparameter compared to the variational mode decomposition algorithm. In order to avoid complex parameter adjustment, the particle swarm algorithm is used to automatically optimize the parameters to achieve rapid and accurate diagnosis of the bearing fault type.

[0089] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0090] Explanation of technical terms: Variational mode decomposition (VMD); Generalized Variational mode decomposition (GVMD); Weighted Harmonic noise ratio (WHNR); Intrinsic Mode Function (IMF); Empirical Mode Decomposition (EMD).

[0091] Example:

[0092] like Figure 1 As shown, this embodiment discloses a bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition, the method comprising:

[0093] Step S1: collecting vibration time domain signals during the operation of the circulating water pump;

[0094] Step S2: input the vibration time domain signal into the generalized variational mode decomposition model to obtain a series of decomposed modes;

[0095] Step S3: Taking the maximum value of the 1.5-dimensional spectrum weighted harmonic noise ratio of the decomposed modal function under a specific fault as the target, the optimal setting parameters of the generalized variational modal decomposition are obtained by particle swarm algorithm, wherein the optimal parameters include the decomposition order, the number of modes and the penalty factor;

[0096] Step S4: re-decompose the vibration time domain signal using the optimized generalized variational mode decomposition method to obtain a series of optimized decomposed modes;

[0097] Step S5: Calculate the 1.5-dimensional spectrum weighted harmonic noise ratio of the decomposed modal function under a specific fault;

[0098] Step S6: determining the optimal decomposition mode based on the maximum value of the weighted harmonic noise ratio corresponding to the specific fault;

[0099] Step S7: Determine the fault diagnosis result according to the optimal decomposition modal weighted harmonic noise ratio result.

[0100] The following is a detailed discussion of each step:

[0101] Step S1: Collect the vibration time domain signal during the operation of the circulating water pump. In this embodiment, the prefabricated fault parts are respectively installed on the pump body of the circulating water pump, and vibration measuring points are arranged in two orthogonal directions (X axis and Y axis) in the middle of the bearing housing and in the vertical direction (Z axis) on the base. Vibration sensors are installed on the vibration measuring points, and the vibration time domain signal during the operation of the circulating water pump is monitored by the vibration sensor. Specifically, the sampling duration is t, the sampling frequency is fz, and the above vibration time domain signal is a high-frequency vibration time domain signal.

[0102] Step S2: Input the vibration time domain signal into the generalized variational mode decomposition model for decomposition to obtain a series of intrinsic mode functions.

[0103] The overall framework of the generalized variational modal decomposition model is a variational problem. The purpose of decomposition is to minimize the sum of the estimated bandwidths of each mode. It is assumed that each "mode" is a finite bandwidth with a different center frequency. Each mode and its center frequency are continuously updated, and each mode is gradually demodulated to the corresponding baseband. Finally, each mode, that is, the corresponding center frequency, is extracted together.

[0104] The generalized variational mode decomposition model is proposed on the basis of the variational mode decomposition model. The main idea is to control the narrowband standard by controlling the derivative order. The rest is the same as the variational mode decomposition model principle. The original signal is input to obtain a series of intrinsic mode functions.

[0105] Step S3: Taking the total weighted harmonic noise ratio under all fault modes as the objective function, initialize the particle swarm optimization algorithm parameters. In this embodiment, the number of particles is set to 100, the number of iterations is set to 10, the inertia weight is set to 0.5, the local learning factor is set to 0.2, the global learning factor is set to 0.5, the decomposition order range is set to (0.5, 10), the decomposition number range is set to (0, 10), and the penalty factor range is set to (10, 10^10).

[0106] Substitute the characteristic frequency corresponding to each fault into the weighted harmonic-to-noise ratio formula to obtain the weighted harmonic-to-noise ratio corresponding to different faults;

[0107] The total weighted harmonic-to-noise ratio under all fault modes is calculated based on the weighted harmonic-to-noise ratio corresponding to different faults. The specific formula is:

[0108] WHNR=WHNR fi +WHNR fo +WHNR fr +WHNR fc ;

[0109] Where WHNR is the total weighted harmonic noise ratio under all fault modes, WHNR fi is the weighted harmonic noise ratio corresponding to the inner race fault, WHNR fo is the weighted harmonic noise ratio corresponding to the outer race fault, WHNR fr is the weighted harmonic noise ratio corresponding to the rolling element fault, WHNR fc is the weighted harmonic noise ratio corresponding to the cage fault.

[0110] Step S4: re-decompose the vibration time domain signal using the optimal parameters of the generalized variational modal decomposition obtained in step S3 to obtain a series of optimized decomposition modes;

[0111] Step S5: Determine the fault diagnosis result according to the optimal decomposition mode weighted harmonic noise ratio calculation, specifically including:

[0112] Determine weighted harmonic-to-noise ratio thresholds corresponding to different faults, specifically including: a weighted harmonic-to-noise ratio threshold corresponding to an inner race fault, a weighted harmonic-to-noise ratio threshold corresponding to an outer race fault, a weighted harmonic-to-noise ratio threshold corresponding to a rolling element fault, and a weighted harmonic-to-noise ratio threshold corresponding to a cage fault;

[0113] In this implementation, the method of weighted harmonic noise ratio threshold is given according to actual needs;

[0114] Determine whether the weighted harmonic-to-noise ratio of a specific fault in the optimal mode is less than the corresponding weighted harmonic-to-noise ratio threshold. If the weighted harmonic-to-noise ratio corresponding to the fault is less than the corresponding weighted harmonic-to-noise ratio threshold, it means that it is normal; if the weighted harmonic-to-noise ratio corresponding to the fault is greater than or equal to the weighted harmonic-to-noise ratio threshold corresponding to the fault, it means that the corresponding fault exists.

[0115] Figure 1Any process or method description in the flowchart described in or otherwise described herein can be understood as representing a module, fragment or portion of a code including one or more executable instructions for implementing the steps of a custom logic function or process, and the scope of the preferred embodiment of the present invention includes other implementations, in which the functions may be performed in a substantially simultaneous manner or in a reverse order according to the functions involved, which should be understood by a person skilled in the art of the present invention. The logic and / or steps represented in the flowchart or otherwise described herein illustrate the possible implementation architecture, functions and operations of the apparatus and methods according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It is also noted that each block in the block diagram or flowchart, and combinations of blocks in the block diagram or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions. For example, what may be considered as an ordered list of executable instructions for implementing a logical function may be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other system that can fetch instructions from and execute instructions in an instruction execution system, apparatus, or device).

[0116] Those skilled in the art will appreciate that the above are only preferred embodiments of the present invention, and the various embodiments of the present disclosure and / or the features described in the claims may be combined or combined in various ways, even if such combinations or combinations are not explicitly described in the present disclosure. It is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art may still modify the technical solutions described in the aforementioned embodiments, or perform equivalent substitutions on some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

[0117] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition, characterized in that: The method comprises the following steps: Step 1, collecting the vibration time domain signal during the operation of the circulating water pump; Step 2: input the vibration time domain signal collected in step 1 into the generalized variational mode decomposition model to obtain a series of decomposed modes; Step 3: Based on the series of decomposed modes obtained in step 2, a decomposition modal function is constructed, and the maximum value of the 1.5-dimensional spectral weighted harmonic noise ratio is taken as the target. The optimal setting parameters of the generalized variational modal decomposition model are obtained by particle swarm optimization, where the optimal setting parameters include decomposition order, mode number and penalty factor; Step 4: re-decompose the vibration time domain signal based on the method for optimally setting parameters of the generalized variational modal decomposition model described in step 3 to obtain a series of optimized decomposition modes; Step 5: construct a 1.5-dimensional spectrum weighted harmonic noise ratio formula based on the vibration time domain signal re-decomposed in step 4; Step 6, determining the optimal decomposition mode based on the maximum value of the 1.5-dimensional spectrum weighted harmonic noise ratio formula described in step 5; Step 7: Determine the fault diagnosis result according to the optimal decomposition modal weighted harmonic noise ratio result determined in step 6.

2. The bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition according to claim 1 is characterized in that: In step 3, the maximum value of the 1.5-dimensional spectral weighted harmonic noise ratio is taken as the target, and the method of obtaining the optimal setting parameters of the generalized variational mode decomposition through the particle swarm algorithm is as follows: Where B(ω) is the diagonal slice after Fourier transformation, R 3x (τ1,τ2)=E{x(t)x(t+τ1)x(t+τ2)}, E is the mathematical expectation, τ1,τ2 are different time delays, R 3x (τ1, τ2) is the main diagonal slice; taking τ1 = τ2 = τ, the main diagonal slice of the third-order cumulant is: R 3x (τ,τ)=E{x(t)x(t+τ)x(t+τ)}.

3. The bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition according to claim 1 is characterized in that: In step 5, the formula for constructing the 1.5-dimensional spectrum weighted harmonic noise ratio based on the re-decomposed vibration time domain signal in step 4 is: Among them, WHNR f is the weighted harmonic noise ratio, x is the 1.5-dimensional spectrum demodulation result, f a is the intermediate variable, r x To calculate the autocorrelation of the input signal, τ p For r x The position offset of the autocorrelation function, r x (τ p ) is the amplitude corresponding to the offset, r x (0) is the total energy of the autocorrelation function, i is the weighted range, f is the theoretical fault frequency, and f w is the HNR weight under the current fault frequency, and n is a positive integer.

4. The bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition according to claim 1 is characterized in that: The method for obtaining the optimal setting parameters of generalized variational mode decomposition through particle swarm algorithm in S3 is: S3.1, initialize particle parameters, including the initial velocity and position of particles, the number of iterations of particles, population size, inertia factor and acceleration factor; S3.2, update the particle position and velocity, set the parameter value range that needs to be optimized for generalized variational mode decomposition, and use it as the value range for updating the particle velocity and position; S3.3, using 1.5-dimensional spectral weighted harmonic noise ratio as fitness function; S3.4, calculate the fitness of each particle and find the individual extreme value and group extreme value; S3.

5. Update the position and velocity of each particle and iterate continuously until the optimal solution is found.

5. The bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition according to claim 1 is characterized in that: The fault diagnosis result is determined according to the optimal decomposition modal weighted harmonic-to-noise ratio result determined in step 6 by determining the weighted harmonic-to-noise ratio thresholds corresponding to different faults.

6. The bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition according to claim 1 is characterized in that: Determining the weighted harmonic-to-noise ratio thresholds corresponding to different faults includes the weighted harmonic-to-noise ratio threshold corresponding to the inner race fault, the weighted harmonic-to-noise ratio threshold corresponding to the outer race fault, the weighted harmonic-to-noise ratio threshold corresponding to the rolling element fault and the weighted harmonic-to-noise ratio threshold corresponding to the cage fault.

7. The bearing fault diagnosis method based on particle swarm optimization generalized variational mode decomposition according to claim 6 is characterized in that: The weighted harmonic-to-noise ratio threshold corresponding to the inner race fault, the weighted harmonic-to-noise ratio threshold corresponding to the outer race fault, the weighted harmonic-to-noise ratio threshold corresponding to the rolling element fault, and the weighted harmonic-to-noise ratio threshold corresponding to the cage fault specifically include: Determine the weighted harmonic noise ratio WHNR corresponding to the inner race fault fi Is it less than the weighted harmonic noise ratio threshold corresponding to the inner ring fault? If the weighted harmonic noise ratio WHNR corresponding to the inner ring fault fi If the weighted harmonic noise ratio WHNR corresponding to the inner ring fault is smaller than the weighted harmonic noise ratio threshold corresponding to the inner ring fault, it means that it is normal; if the weighted harmonic noise ratio WHNR corresponding to the inner ring fault is fi If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the inner ring fault, it means that there is an inner ring fault; Determine the weighted harmonic noise ratio WHNR corresponding to the outer ring fault fo Is it less than the weighted harmonic noise ratio threshold corresponding to the outer ring fault? If the weighted harmonic noise ratio WHNR corresponding to the outer ring fault fo If the weighted harmonic noise ratio WHNR corresponding to the outer ring fault is smaller than the weighted harmonic noise ratio threshold corresponding to the outer ring fault, it means that it is normal; if the weighted harmonic noise ratio WHNR corresponding to the outer ring fault is fo If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the outer race fault, it means that there is an outer race fault; Determine the weighted harmonic noise ratio (WHNR) corresponding to the rolling element fault fr Is it less than the weighted harmonic noise ratio threshold corresponding to the rolling element fault? If the weighted harmonic noise ratio WHNR corresponding to the rolling element fault fr If the weighted harmonic noise ratio WHNR corresponding to the rolling element fault is less than the weighted harmonic noise ratio threshold corresponding to the rolling element fault, it is normal; if the weighted harmonic noise ratio WHNR corresponding to the rolling element fault is fr If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the rolling element fault, it means that there is a rolling element fault; Determine the weighted harmonic noise ratio WHNR corresponding to the cage failure fc Is it less than the weighted harmonic noise ratio threshold corresponding to the cage fault? If the weighted harmonic noise ratio WHNR corresponding to the cage fault fc If the weighted harmonic noise ratio WHNR corresponding to the cage fault is less than the weighted harmonic noise ratio threshold corresponding to the cage fault, it is normal; if the weighted harmonic noise ratio WHNR corresponding to the cage fault is fc If it is greater than or equal to the weighted harmonic noise ratio threshold corresponding to the cage fault, it means that there is a cage fault.

8. A bearing fault diagnosis system based on particle swarm optimization generalized variational mode decomposition, characterized in that: The system comprises: A data acquisition module is used to collect vibration time domain signals during the operation of the circulating water pump; A decomposition mode module is used to input the vibration time domain signal collected by the data acquisition module into the generalized variational mode decomposition model to obtain a series of decomposition modes; The particle swarm optimization module is used to construct the decomposition mode function based on a series of decomposition modes obtained by the decomposition mode module. The optimal setting parameters of the generalized variational mode decomposition model are obtained by the particle swarm algorithm with the maximum value of the 1.5-dimensional spectral weighted harmonic noise ratio as the goal, where the optimal setting parameters include the decomposition order, the number of modes and the penalty factor; The optimized decomposition mode module and the method for determining the optimal setting parameters of the generalized variational mode decomposition model described in the particle swarm optimization module re-decompose the vibration time domain signal to obtain a series of optimized decomposition modes; A weighted harmonic noise ratio formula determination module is used to construct a 1.5-dimensional spectrum weighted harmonic noise ratio formula based on the vibration time domain signal re-decomposed by the optimized decomposition mode module; An optimal decomposition mode module, used for determining the optimal decomposition mode based on the maximum value of the 1.5-dimensional spectrum weighted harmonic noise ratio formula described in the weighted harmonic noise ratio formula determination module; The fault diagnosis module is used to determine the fault diagnosis result according to the optimal decomposition mode weighted harmonic-to-noise ratio result determined by the optimal decomposition mode module.

9. A computer device comprising a memory and a processor, characterized in that A computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images