Plunger pump motor bearing fault diagnosis method and system based on parameter optimization characteristic modal decomposition

Abstract

The invention discloses a plunger pump motor bearing fault diagnosis method based on parameter optimization characteristic mode decomposition. The method comprises the following steps: designing an optimization objective function of a grey wolf optimization algorithm; initializing the population number and the maximum number of iterations of the grey wolf optimization algorithm; setting a range of characteristic mode decomposition parameters; setting a search boundary of a grey wolf optimization algorithm according to the parameter range, and randomly initializing individual positions of a grey wolf population; randomly updating gray wolf individual positions in the search boundary; taking the current position as a characteristic mode decomposition parameter combination, and decomposing the preprocessed vibration signal by adopting characteristic mode decomposition to obtain a decomposed mode component under the current parameter combination; if the termination condition is met, outputting the current value of the parameter as an optimal parameter combination; and calculating an envelope spectrum peak factor of the modal component under the optimal parameter combination, screening the component with the maximum index as the optimal modal component, and calculating an envelope spectrum of the optimal modal component. According to the method, the problem that important parameters of characteristic modal decomposition are set depending on human experience is solved, and meanwhile, the defect that the characteristic modal decomposition does not have self-adaptability is overcome.

Description

Technical Field

[0001] The invention relates to a plunger pump motor bearing fault diagnosis method based on parameter optimization characteristic mode decomposition, and belongs to the field of intelligent fault diagnosis and health management of industrial equipment. Background Art

[0002] As a key device in the hydraulic system, the piston pump's bearing operates under high load and complex working conditions and is one of the most prone to failure. If the early failure of the bearing is not discovered in time, it may lead to equipment performance degradation, downtime losses and even serious safety hazards. Therefore, the development of efficient and reliable early bearing fault diagnosis technology is of great engineering significance.

[0003] Piston pump bearings usually work in complex working conditions with high pressure, high speed and severe vibration. Their vibration signals are often interfered by impact noise and background noise. Traditional signal processing methods are not accurate enough in extracting fault feature information, which makes it difficult to effectively diagnose early faults. Traditional bearing fault diagnosis methods rely on spectrum analysis or empirical formulas, which are usually difficult to accurately extract bearing fault feature signals from impact noise and complex background noise. Especially in equipment with complex vibration signals such as piston pumps, these methods have poor diagnostic effects and are difficult to meet actual industrial needs.

[0004] Modal decomposition techniques such as empirical mode decomposition (EMD) and variational mode decomposition (VMD) have been widely used in the analysis of non-stationary signals, but their decomposition performance depends on the selection of algorithm parameters. The selection of its parameters depends on experience. Improper settings may lead to mode aliasing or feature loss, reduce the diagnostic accuracy, and lack of global optimization mechanism, which easily leads to inaccurate feature extraction or information loss, making it difficult to meet the needs of efficient fault diagnosis under complex working conditions. Eigenmode decomposition (FMD) is an emerging signal processing method that can decompose complex signals in the time domain and frequency domain. Its core idea is to decompose the original signal into several physically meaningful intrinsic mode components (IMFs) by constructing an optimization objective function. These components reflect the main characteristics of the signal in a specific frequency band. FMD has significant advantages in processing non-stationary and nonlinear signals, especially for vibration signals containing impact noise and background noise under complex working conditions. It has strong anti-interference ability. Compared with traditional methods, FMD can extract key feature information from signals more accurately; although FMD itself has superior decomposition ability, its performance depends to a large extent on parameter settings (such as the number of decomposed modes, penalty factor, etc.).

[0005] How to use FMD to achieve more efficient and accurate extraction of plunger pump motor bearing fault features and provide a reliable basis for early fault diagnosis has become a problem that needs to be solved. Summary of the invention

[0006] The present invention provides a plunger pump motor bearing fault diagnosis method based on parameter optimized characteristic mode decomposition, which solves the problem that FMD needs to rely on human experience to set parameters and does not have parameter adaptability, and separates signal components containing rich fault characteristic information from bearing fault signals under reciprocating impact conditions, thereby achieving the best diagnostic effect and being applied in early fault detection of plunger pump bearings.

[0007] The technical solution of the present invention is:

[0008] According to a first aspect of the present invention, a method for diagnosing a piston pump motor bearing fault based on parameter-optimized eigenmode decomposition is provided, comprising the following steps:

[0009] Step 1: preprocessing the vibration signal collected by the vibration sensor to obtain a preprocessed vibration signal;

[0010] Step 2: Design the optimization objective function of the gray wolf optimization algorithm; then initialize the population number and the maximum number of iterations of the gray wolf optimization algorithm;

[0011] Step 3: Set the range of characteristic mode decomposition parameters; set the search boundary of the gray wolf optimization algorithm according to the parameter range, and then randomly initialize the individual positions of the gray wolf population; wherein the characteristic mode decomposition parameters include the number of decomposition modes M and the filter length L;

[0012] Step 4: Randomly update the individual position of the gray wolf within the search boundary; based on the current position as the characteristic mode decomposition parameter combination, the preprocessed vibration signal is decomposed by characteristic mode decomposition to obtain the decomposed modal components under the current parameter combination;

[0013] Step 5: Calculate the objective function based on the output signal obtained in step 4 to determine whether the termination condition is met: if so, the current values ​​of the output parameters M and L are taken as the optimal parameter combination;

[0014] Step 6: Introduce the envelope spectrum peak factor as the screening index of the optimal modal component;

[0015] Step 7: Calculate the envelope spectrum peak factor of the modal component under the optimal parameter combination, select the component with the largest index as the optimal modal component, calculate the envelope spectrum of the optimal modal component, and obtain the characteristic frequency in the envelope spectrum.

[0016] Furthermore, the optimization objective function is specifically:

[0017] For a given signal x(t), firstly perform envelope extraction to obtain the envelope signal e(t);

[0018] For the envelope signal e(t), the envelope entropy is defined as: Among them, p iis the probability distribution of the envelope signal, and N is the number of discrete points of the signal;

[0019] Based on the envelope entropy, the specific optimization objective function is defined as: Minimize H(e).

[0020] Furthermore, the value range of the decomposition mode number M is set to 2 to 6, and the value range of the filter length L is set to 20 to 60.

[0021] According to a second aspect of the present invention, a plunger pump motor bearing fault diagnosis system based on parameter-optimized eigenmode decomposition is provided, comprising a module of any one of the above-mentioned plunger pump motor bearing fault diagnosis methods based on parameter-optimized eigenmode decomposition.

[0022] According to a third aspect of the present invention, a processor is provided, the processor being used to run a program, wherein when the program is run, the method for diagnosing a bearing fault of a piston pump motor based on parameter-optimized characteristic mode decomposition as described above is executed.

[0023] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium comprising a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned methods for diagnosing piston pump motor bearings based on parameter-optimized characteristic mode decomposition.

[0024] The beneficial effects of the present invention are:

[0025] In view of the complex working environment of the plunger pump, the weak characteristic signal of the early fault of the bearing, and the difficulty in feature extraction, the present invention proposes a rolling bearing fault diagnosis method based on parameter optimization characteristic mode decomposition. This method solves the problem that the characteristic mode decomposition relies on human experience to set its important parameters. At the same time, the method designs the objective function based on the minimum envelope entropy and adaptively finds the optimal parameter combination through the gray wolf optimization algorithm, which makes up for its lack of adaptability. In addition, the simulation and measured signal analysis results show that the method of the present invention can still effectively extract weak fault characteristic information under the interference of reciprocating impact and realize accurate fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flow chart of the method of the present invention;

[0027] Figure 2 is an iterative curve of the objective function provided according to an embodiment of the present invention;

[0028] Figure 3 This is a diagram showing the diagnosis result of the early fault of the motor bearing of a simulated plunger pump according to the present invention;

[0029] Figure 4 This is a diagram showing the diagnostic results of the measured early failure of the motor bearing of the plunger pump according to the present invention;

[0030] Figure 5 It is a comparison chart of indicators of various decomposed components under the measured signal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other arbitrarily without conflict.

[0032] Example 1: Figure 1-Figure 5 As shown, a plunger pump motor bearing fault diagnosis method based on parameter optimization characteristic mode decomposition, the method comprises the following steps:

[0033] Step 1: Install the vibration sensor at a preset position of the plunger pump bearing to ensure that the signal collection point can reflect the operating status of the plunger pump bearing; pre-process the vibration signal collected by the vibration sensor to obtain a pre-processed vibration signal; the pre-processing is to remove the DC component and high-frequency noise in turn and retain the signal characteristics of the key frequency band; in the specific operation, the DC component is removed by removing the mean value, and the high-frequency noise is removed by a band-pass filter;

[0034] Step 2: Design the optimization objective function of the gray wolf optimization algorithm; then initialize the population number and the first maximum number of iterations of the gray wolf optimization algorithm;

[0035] Furthermore, the optimization objective function is specifically:

[0036] For a given signal x(t), the envelope is first extracted by Hilbert transform and other methods to obtain the envelope signal e(t). The envelope signal e(t) describes the amplitude change of the original preprocessed vibration signal.

[0037] For the envelope signal e(t), the envelope entropy is defined as: Among them, p i is the probability distribution of the envelope signal, and N is the number of discrete points of the signal;

[0038] Based on the envelope entropy, the specific optimization objective function is defined as: Minimize H(e).

[0039] It should be noted that the envelope function is the outer envelope curve of the signal, which can extract the main trend components of the signal; entropy is an indicator to measure the uncertainty of the signal. The larger the entropy of the signal, the more complex its information content and the higher its uncertainty; conversely, the smaller the entropy, the simpler the structure of the signal and the stronger its regularity. In view of the fact that the fault characteristics of the plunger pump bearing have the characteristics of strong regularity, the present invention takes minimizing the envelope entropy as the objective function.

[0040] The gray wolf optimization algorithm simulates the process of gray wolves encircling and hunting prey, which is actually the process of finding the optimal position of the alpha wolf. Its mathematical model can be expressed as follows: X(t+1) = X p (t)-A|BX p (t)-X(t)|, where: t represents the current iteration number, X(t) is the current position vector of the wolf pack; X(t+1) is the position vector of the wolf pack after iterative update; Xp(t) is the position vector of the current target prey; A and B are the convergence and swing factors respectively.

[0041] Step 3: Set the range of characteristic mode decomposition (FMD) parameters, where M is the number of decomposition modes and L is the filter length; set the search boundary of the gray wolf optimization algorithm according to the parameter range, and then randomly initialize the individual positions of the gray wolf population; illustratively, set M to 2 to 6 and set L to 20 to 60;

[0042] Step 4: Randomly update the individual position of the gray wolf within the search boundary, that is, randomly determine the characteristic mode decomposition parameter value; based on the current position as the characteristic mode decomposition (FMD) parameter combination, use FMD to decompose the preprocessed vibration signal to obtain the decomposed modal components under the current parameter combination;

[0043] Go down to the ground, step 4, specifically:

[0044] Step 4.1, using the Hanning window to initialize the FIR filter group, the original pre-processed vibration signal frequency band is divided into K segments, and the second iteration number i=1 is set; in the embodiment of the present invention, K is initially set to 7, which satisfies the maximum value greater than M;

[0045] Step 4.2: Use the pre-processed vibration signal Perform deconvolution operation, and the modal component signal obtained by deconvolution is the decomposed mode; among them, represents the signal obtained by deconvolution of the kth segment at the i-th iteration; represents the kth filter coefficient in the i-th iteration, k = 1, 2, ..., K, * represents convolution;

[0046] Step 4.3: Deconvolve the K-segment signal to obtain K modal components, calculate the correlation coefficient between every two modal components, construct a K×K correlation matrix, select the two modal components with the largest correlation coefficient, and estimate the fault period T according to the autocorrelation theory. k i , calculate their related kurtosis, retain the larger modal component of the two related kurtosis, and set K = K-1;

[0047] Step 4.4, determine whether the current modal number K reaches the preset modal number M: if reached, the decomposition stops and the final decomposition result is output; if not, the filter coefficients are updated using the original signal, the modal components obtained by decomposition, and the estimated signal failure period; the second iteration number is set to i=i+1, and return to step 4.2 to continue the above operation.

[0048] Step 5: Calculate the objective function according to the output signal obtained in step 4 to determine whether the termination condition is met: if so, the current values ​​of the output parameters M and L are used as the optimal parameter combination; the termination condition is to meet the first maximum iteration number or accuracy requirement;

[0049] Step 6: Introduce the envelope spectrum peak factor as the screening index of the optimal modal component;

[0050] The envelope spectrum crest factor is an indicator used to measure the shape of the signal envelope spectrum. When the impact period of the vibration signal becomes more significant, the signal-to-noise ratio will also increase. At this time, the peak value of the spectrum line at the impact frequency of the envelope spectrum will increase accordingly. r ,f s / 2] is called the envelope spectrum peak factor, which is expressed as: Where: f r is the frequency conversion; f s is the sampling frequency; n is a constant, generally taken as 2; E m is the energy of the envelope spectrum. The higher the value of this indicator, the higher the signal-to-noise ratio and energy of the signal, and the more obvious the periodic impact component.

[0051] Step 7: Calculate the envelope spectrum peak factor of the modal component under the optimal parameter combination, select the component with the largest index as the optimal modal component, calculate the envelope spectrum of the optimal component, and obtain the characteristic frequency in the envelope spectrum.

[0052] It can be seen from the application of the above technical scheme that fault feature extraction is the core link of mechanical system fault diagnosis, and its goal is to extract key features that can characterize the fault type and severity from complex and noisy signals. This process directly determines the accuracy and reliability of the diagnostic results. The existing fault feature extraction method evaluates the signal through time domain indicators (such as mean, root mean square value, kurtosis, etc.), which is easily interfered by noise and difficult to capture the instantaneous information of non-stationary signals, and has limited effect on early fault identification. On the one hand, the present invention introduces the minimum envelope entropy as the objective function, and then uses the gray wolf optimization algorithm to optimize the parameter combination for impact noise and background noise in a complex working environment to improve the accuracy of characteristic mode decomposition; on the other hand, the present invention uses the maximum envelope spectrum peak factor as the evaluation index, selects the optimal mode component from the modal component, and extracts the characteristic frequency of the optimal mode component through envelope demodulation to enhance the recognition of the bearing fault characteristic signal.

[0053] Furthermore, the characteristic frequency is compared with its theoretical value to achieve fault diagnosis.

[0054] To further illustrate the effectiveness of the present invention, the following is given:

[0055] 1. The following is an analysis of the impact simulation signal of the inner ring fault of the rolling bearing under reciprocating impact conditions. The specific analysis is as follows:

[0056] (1) The mathematical model expression used to simulate the impact signal of the inner ring fault of the rolling bearing is as follows:

[0057]

[0058] Where: x1(t) represents the simulation signal of the inner ring fault of the rolling bearing, s(t) is the periodic pulse, n(t) is the noise, A i is 1 / f r is the modulation amplitude of the period, A0 is the inner circle displacement constant, that is, the initial amplitude; h(t) is an exponential decay pulse, T is the period, τ i is the time offset of the ith pulse.

[0059] Set the system resonant frequency to f n =500Hz, model attenuation coefficient α = 200, center offset C A =0.1, initial phase φ A =π / 2,φ ω = 0. Other simulation parameters are shown in the table. Gaussian white noise with a signal-to-noise ratio of -5db is added, and the signal length is 1x8192.

[0060] The mathematical model used to simulate the reciprocating impact signal of the plunger pump is as follows:

[0061]

[0062] Table 1 Simulation parameters

[0063]

[0064] Based on the method of the present invention, the simulation signal is processed and analyzed. First, the parameters of the gray wolf optimization algorithm are initialized, and the parameter range is set. The decomposition mode number K is [2, 6], and the filter length L is [20, 60]. Iteration is performed with the minimum envelope entropy as the fitness function. When the first iteration number reaches the fifth time, the function value reaches a minimum of 2.9986. At this time, the optimal parameters [K, L] = [5, 24] are obtained. The parameters are brought into FMD to decompose the simulation signal. The optimal decomposition component is selected according to the maximum envelope spectrum peak factor index, and envelope demodulation analysis is performed. The optimal modal component and envelope spectrum obtained are shown in the figure. Figure 3 As shown, through Figure 3 It can be seen from the optimal modal component in the upper part that the periodic pulses of the early fault of the plunger pump bearing are significant, and the spectral line amplitudes at 2 times and 4 times the rotation frequency in its envelope spectrum are large, and the spectral line at the characteristic frequency of 130Hz of the inner ring fault of the plunger pump bearing is also relatively significant, which matches the theoretical frequency of the inner ring fault, indicating that the characteristic frequency obtained by the method of the present invention can be effectively used for fault diagnosis of simulation signals.

[0065] 2. Measured signal analysis: The specific analysis is as follows:

[0066] The motor end bearing of a plunger pump device in an oil production plant was used as the research object for experimental verification. The plunger pump device consists of a drive motor, a pulley, 5 plunger pumps and a base. The device is driven by a motor when working, and the crankshaft is driven to rotate through the pulley, and the crankshaft drives the plunger rod to reciprocate, so that the plunger reciprocates inside the pump body. This continuous change in load will have a certain impact on the motor.

[0067] The model parameters of the experimental bearing are shown in Table 2. The test bearing is a deep groove ball bearing of model 6322. During the experiment, the motor shaft speed is 1043r / min, the frequency is 17.383Hz, the signal sampling frequency is set to 12800Hz, the number of sampling points is 8192, the sampling interval is 3min, and a total of 14322 vibration signal files are collected. Calculate the theoretical characteristic frequency f of the inner ring fault of the rolling bearing i It is 128.14Hz.

[0068] Table 2 Bearing parameters

[0069] model Pitch circle diameter / mm Rolling element diameter / mm Number of rolling elements Contact Angle 6322 175 40 12 0

[0070] The method proposed in the present invention is used to analyze the early fault signal of the plunger pump bearing, and the Gray Wolf optimization algorithm is used to optimize the two parameters of the decomposition mode number K and the filter length L of the characteristic mode decomposition. The minimum envelope entropy is used as the fitness function for iteration. When the first iteration number reaches the fourth time, the fitness function reaches a minimum of 3.51. At this time, the optimal decomposition mode number and filter length (K=3, L=22) are obtained. The fitness values ​​under different iteration numbers are as follows: Figure 2 Substituting the optimal parameters into FMD to decompose the signal, the optimal modal components and envelope spectra are shown in Figure 4 As shown, through Figure 4 The optimal modal components in the upper part show that the early failure periodic pulse of the piston pump bearing is significant. Figure 4 It can be seen from the envelope spectrum in the lower part that the amplitude of the spectrum line at 2 times the rotation frequency is larger, and the spectrum line at the characteristic frequency of 130Hz of the inner ring fault of the plunger pump bearing is also more significant, which matches the theoretical frequency of the inner ring fault, indicating that the characteristic frequency obtained by the method of the present invention can be effectively used for fault diagnosis of the measured signal.

[0071] From the first and second above, it can be seen that the effectiveness of the method is verified by analyzing the simulated signal and the measured signal. In addition, compared with the traditional method based on variational mode decomposition and fast spectral kurtosis, the method of the present invention has higher performance and stronger early fault detection capability in plunger pump bearing fault diagnosis.

[0072] 3. In order to further illustrate the effect of the envelope spectrum peak factor of the present invention, a comparative experiment is given as follows:

[0073] Based on the measured signal data at the second point, the envelope spectrum peak factor and kurtosis index are compared. The comparison results are as follows: Figure 5 As shown, through Figure 5 It can be seen that the kurtosis index results under the three modal components are very close and cannot be effectively distinguished; while the difference in the envelope spectrum peak factor under different modal components is more significant, which shows that the envelope spectrum peak factor selected by the present invention has higher performance than the traditional kurtosis index. As can be seen from the figure, the optimal component is IMF1 according to the maximum envelope spectrum peak factor, and envelope demodulation analysis is performed. Figure 4 The spectral lines at the first-order and second-order characteristic frequencies of the bearing inner ring fault in its envelope spectrum are prominent, which indicates that there is damage to the bearing inner ring.

[0074] Embodiment 2: A plunger pump motor bearing fault diagnosis system based on parameter-optimized characteristic mode decomposition, comprising a module of a plunger pump motor bearing fault diagnosis method based on parameter-optimized characteristic mode decomposition as described in any one of Embodiment 1; specifically comprising: a first module for executing step 1: preprocessing the vibration signal collected by the vibration sensor to obtain the preprocessed vibration signal; a second module for executing step 2: designing the optimization objective function of the gray wolf optimization algorithm; then initializing the population number and the maximum number of iterations of the gray wolf optimization algorithm; a third module for executing step 3: setting the range of characteristic mode decomposition parameters; setting the search boundary of the gray wolf optimization algorithm according to the parameter range, and then randomly initializing the individual positions of the gray wolf population; wherein the characteristic mode decomposition parameters include the decomposition mode number M and the filter length L; a fourth ... a fourth module for executing step 3: setting the range of characteristic mode decomposition parameters; setting the search boundary of the gray wolf optimization algorithm according to the parameter range, and then randomly initial A module for executing step 4: randomly updating the individual position of the gray wolf within the search boundary; using the current position as the characteristic mode decomposition parameter combination, using the characteristic mode decomposition to decompose the preprocessed vibration signal, and obtaining the modal component after decomposition under the current parameter combination; a fifth module for executing step 5: calculating the objective function according to the output signal obtained in step 4, and judging whether the termination condition is met: if so, the current values ​​of the output parameters M and L are taken as the optimal parameter combination; a sixth module for executing step 6: introducing the envelope spectrum peak factor as the screening index of the optimal modal component; a seventh module for executing step 7: calculating the envelope spectrum peak factor of the modal component under the optimal parameter combination, screening the component with the largest index as the optimal modal component, calculating the envelope spectrum of the optimal modal component, and obtaining the characteristic frequency in the envelope spectrum. For the parts not described in detail in the above modules, please refer to the relevant descriptions of other embodiments.

[0075] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0076] Embodiment 3: A processor is used to run a program, wherein when the program is running, the plunger pump motor bearing fault diagnosis method based on parameter-optimized characteristic mode decomposition as described in any one of Embodiments 1 is executed.

[0077] Embodiment 4: A computer-readable storage medium, the computer-readable storage medium comprising a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the plunger pump motor bearing fault diagnosis method based on parameter-optimized characteristic mode decomposition as described in any one of Embodiments 1.

[0078] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0079] The specific implementation modes of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above implementation modes, and various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present invention.

Claims

1. A plunger pump motor bearing fault diagnosis method based on parameter optimization characteristic mode decomposition, characterized in that: The following steps are involved: Step 1: preprocessing the vibration signal collected by the vibration sensor to obtain a preprocessed vibration signal; Step 2: Design the optimization objective function of the gray wolf optimization algorithm; then initialize the population number and the maximum number of iterations of the gray wolf optimization algorithm; Step 3: Set the range of characteristic mode decomposition parameters; set the search boundary of the gray wolf optimization algorithm according to the parameter range, and then randomly initialize the individual positions of the gray wolf population; wherein the characteristic mode decomposition parameters include the number of decomposition modes M and the filter length L; Step 4: Randomly update the individual position of the gray wolf within the search boundary; based on the current position as the characteristic mode decomposition parameter combination, the preprocessed vibration signal is decomposed by characteristic mode decomposition to obtain the decomposed modal components under the current parameter combination; Step 5: Calculate the objective function based on the output signal obtained in step 4 to determine whether the termination condition is met: if so, the current values ​​of the output parameters M and L are taken as the optimal parameter combination; Step 6: Introduce the envelope spectrum peak factor as the screening index of the optimal modal component; Step 7: Calculate the envelope spectrum peak factor of the modal component under the optimal parameter combination, select the component with the largest index as the optimal modal component, calculate the envelope spectrum of the optimal modal component, and obtain the characteristic frequency in the envelope spectrum.

2. The plunger pump motor bearing fault diagnosis method based on parameter optimization characteristic mode decomposition according to claim 1 is characterized in that: The optimization objective function is specifically: For a given signal x(t), firstly perform envelope extraction to obtain the envelope signal e(t); For the envelope signal e(t), the envelope entropy is defined as: Among them, p i is the probability distribution of the envelope signal, and N is the number of discrete points of the signal; Based on the envelope entropy, the specific optimization objective function is defined as: Minimize H(e).

3. The plunger pump motor bearing fault diagnosis method based on parameter optimization characteristic mode decomposition according to claim 1 is characterized in that: The value range of the decomposition mode number M is set to 2 to 6, and the value range of the filter length L is set to 20 to 60.

4. A plunger pump motor bearing fault diagnosis system based on parameter optimization characteristic mode decomposition, characterized in that: A module comprising a plunger pump motor bearing fault diagnosis method based on parameter-optimized characteristic mode decomposition as described in any one of claims 1-3.

5. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the plunger pump motor bearing fault diagnosis method based on parameter-optimized characteristic mode decomposition as described in any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the plunger pump motor bearing fault diagnosis method based on parameter-optimized characteristic mode decomposition as described in any one of claims 1-3.

Images