Rotating machine fault diagnosis method and device and storage medium

Abstract

The invention discloses a rotating machine fault diagnosis method and device and a storage medium in the technical field of mechanical fault diagnosis. The method comprises the following steps: acquiring a discretized demodulation signal; solving the discretized demodulation signal through an augmented Lagrangian multiplier method and an alternating direction multiplier method, and obtaining an iteratively optimized demodulation signal; calculating an instantaneous frequency increment based on the demodulated signal after iterative optimization, applying smoothness constraint and low-rank constraint to the instantaneous frequency increment, and solving through an iterative solution algorithm to obtain an estimated instantaneous frequency; and based on a pre-acquired theoretical fault characteristic frequency and the estimated instantaneous frequency, determining the fault type of the rotating machine according to a matching result. According to the method, the technical problems that the instantaneous frequency presents a non-stationary feature and is easily influenced by background noise, so that the accurate instantaneous frequency is difficult to extract, and the potential fault of the rotating machine cannot be accurately diagnosed can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical fault diagnosis, and particularly to a method, device and storage medium for diagnosing faults of rotating machinery. Background Art

[0002] Gear and bearing components in rotating machinery play an important role in modern industrial applications. In order to avoid catastrophic accidents, it is of great significance to develop advanced rotating machinery fault diagnosis technologies to ensure the reliability of the system.

[0003] When gears and bearings operate under time-varying rotational speeds, these vibration signals are nonlinearly frequency modulated and exhibit complex characteristics. These nonlinear signals with complex time-varying amplitudes and frequencies are usually referred to as nonlinearly frequency modulated signals. The time-varying instantaneous frequency in the nonlinearly frequency modulated signal can reveal the vibration law of the signal, and the instantaneous frequency is a key index for identifying faults of gears and bearings under time-varying rotational speed conditions. However, the instantaneous frequency in the nonlinearly frequency modulated signal presents non-stationary characteristics and is easily affected by background noise, resulting in difficulty in extracting accurate instantaneous frequency, and thus it is impossible to accurately diagnose potential faults of rotating machinery.

[0004] Therefore, there is an urgent need for a method, device and storage medium for diagnosing faults of rotating machinery to solve the above technical problems. Summary of the Invention

[0005] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method, device and storage medium for diagnosing faults of rotating machinery, which can solve the technical problems that the instantaneous frequency in the nonlinearly frequency modulated signal presents non-stationary characteristics and is easily affected by background noise, resulting in difficulty in extracting accurate instantaneous frequency, and thus it is impossible to accurately diagnose potential faults of rotating machinery.

[0006] To achieve the above purpose, the present invention is implemented by the following technical solutions:

[0007] In a first aspect, the present invention provides a method for diagnosing faults of rotating machinery, including:

[0008] Obtain vibration signals;

[0009] Perform a demodulation operation on the vibration signals to obtain demodulated signals corresponding to the vibration signals, and apply a smoothness constraint and a sparsity constraint to the demodulated signals to obtain the constrained demodulated signals;

[0010] Perform discretization processing on the constrained demodulated signals to obtain the discretized demodulated signals;

[0011] Solve the discretized demodulated signals by the augmented Lagrangian multiplier method and the alternating direction multiplier method to obtain the iteratively optimized demodulated signals;

[0012] Calculate the instantaneous frequency increment based on the iteratively optimized demodulation signal, impose smoothness constraints and low-rank constraints on the instantaneous frequency increment, and solve for the estimated instantaneous frequency through an iterative solution algorithm;

[0013] Based on the pre-acquired theoretical fault characteristic frequency and the estimated instantaneous frequency, determine the type of rotating machinery fault.

[0014] Further, imposing smoothness constraints and sparsity constraints on the demodulation signal to obtain the constrained demodulation signal includes:

[0015] The expression of the demodulation signal is:

[0016] ,

[0017] ,

[0018] Smoothness constraint:

[0019] Q1: ,

[0020] Sparsity constraint:

[0021] Q2: ,

[0022] Q3: ,

[0023] Where, represents the th mode, M represents the total number of modes, represents the time, respectively represent the first demodulation signal and the second demodulation signal of the th mode passing through the smoothness constraint at time , g(t) represents the constraint function, respectively represent 's second derivative, represents the instantaneous amplitude of the th mode at time , and respectively represent the modulation frequency and demodulation frequency of the th mode at time , the parameter s represents a variable in the interval [0, t] of the variable limit integral, and respectively represent the modulation frequency and demodulation frequency of the th mode at time s, represents the initial phase, and are the first demodulation operator and the second demodulation operator respectively, Indicates The mode in The first demodulated signal and the second demodulated signal which are subjected to the sparsity constraint at each moment, are the first penalty coefficient and the second penalty coefficient respectively, is the Fourier transform matrix.

[0024] Furthermore, the constrained demodulated signal is discretized to obtain the demodulated signal after the discretization process, which includes:

[0025] Q4: ,

[0026] Q5: ,

[0027] Q6: ,

[0028] ,

[0029] ,

[0030] ,

[0031] ,

[0032] ,

[0033] ,

[0034] ,

[0035] in, represents the sampling duration from time 0 to time N-1, g represents the vibration signal, which is the representation form of the vibration signal after discretization processing; and Respectively represent The first demodulated signal and the second demodulated signal after the mode discretization processing and smoothness constraint, and Respectively represent A first demodulated signal and a second demodulated signal after a mode discretization process and a sparsity constraint; and express The general form after discretization; Expressed as The modulation frequency of each mode, are the first and second penalty coefficients, respectively, is the second-order difference matrix; represents a diagonal matrix, is the transpose; represents the kernel matrix, is expressed as at the demodulation operator at the moment, and the specific representation form is , is expressed as at the demodulation operator at the moment, and the specific representation form is , represents the phase in the trigonometric function at time t.

[0036] Furthermore, the demodulation signal after the discretization process is solved by the augmented Lagrangian multiplier method and the alternating direction multiplier method, and obtaining the iteratively optimized demodulation signal includes:

[0037] The Q4 is solved by the augmented Lagrangian multiplier and the alternating direction multiplier method to obtain:

[0038] ,

[0039] ,

[0040] ,

[0041] wherein, represents the vibration signal, respectively correspond to the first demodulation signal value and the second demodulation signal value of the q-th mode at the th iteration, respectively represent the first kernel matrix value and the second kernel matrix value corresponding to the q-th mode at the th iteration;

[0042] Introduce the first auxiliary function Construct the Q5 as a constrained optimization problem and solve it by the augmented Lagrangian multiplier and the alternating direction multiplier method to obtain:

[0043] ,

[0044] ,

[0045] ,

[0046] Introduce the second auxiliary function h to construct the Q6 as a constrained optimization problem and solve it by the augmented Lagrangian multiplier and the alternating direction multiplier method to obtain:

[0047] ,

[0048] ,

[0049] ,

[0050] Among them, represents the first iteration number, represents the third penalty coefficient, represents the first Lagrange multiplier, represents the second iteration number, represents the fourth penalty coefficient, represents the second Lagrange multiplier, represents the soft threshold function, represents the third iteration number, and h is the second auxiliary function, is the fifth penalty coefficient, is the third Lagrange multiplier.

[0051] Furthermore, based on the demodulation signal after iterative optimization, calculate the instantaneous frequency increment, impose smoothness constraints and low-rank constraints on the instantaneous frequency increment, and solve to obtain the estimated instantaneous frequency through an iterative solution algorithm, including:

[0052] Based on the demodulation signal after iterative optimization, obtaining the instantaneous frequency increment through the arctangent demodulation method includes:

[0053] ,

[0054] Discretize the instantaneous frequency increment to obtain the discretized instantaneous frequency increment, add smoothness constraints to the discretized instantaneous frequency increment, solve to obtain the desired instantaneous frequency increment, and obtain the initially optimized estimated instantaneous frequency according to the desired instantaneous frequency increment.

[0055] ,

[0056] ,

[0057] ,

[0058] ,

[0059] Among them, is the sixth penalty coefficient, , is the desired instantaneous frequency increment, represents the identity matrix, is expressed as the initially optimized estimated instantaneous frequency for the th iteration, is expressed as the initially optimized estimated instantaneous frequency for the th iteration;

[0060] Construct a noise observation matrix according to the initially optimized estimated instantaneous frequency, and add a low-rank constraint to the noise observation matrix. Obtaining the estimated instantaneous frequency includes:

[0061] Q7: ,

[0062] where, is the estimated instantaneous frequency, is a low-rank matrix, is the noise observation matrix composed of the initial optimization , is the seventh penalty coefficient, represents the matrix of the th singular value, represents the Frobenius norm of the matrix ;

[0063] Shrink the singular values of the noise observation matrix to solve for the Q7 to obtain:

[0064] ,

[0065] where, are all orthogonal matrices obtained by applying singular value decomposition to the matrix , is the largest singular value of the , Each row of represents the estimated instantaneous frequency.

[0066] Furthermore, based on the pre-obtained theoretical fault characteristic frequency and the estimated instantaneous frequency, determining the type of rotating machinery fault includes:

[0067] Calculate the theoretical fault characteristic frequency of the rotating machinery according to the structural parameters and rotational frequency of the rotating machinery, compare the estimated instantaneous frequency with the theoretical fault characteristic frequency, and determine the type of rotating machinery fault according to the matching result.

[0068] In a second aspect, the present invention provides a rotating machinery fault diagnosis device, including:

[0069] A signal acquisition module for acquiring vibration signals;

[0070] A demodulated signal optimization module for performing a demodulation operation on the vibration signal to obtain a demodulated signal corresponding to the vibration signal, and applying a smoothness constraint and a sparsity constraint to the demodulated signal to obtain a constrained demodulated signal;

[0071] A discretization module for discretizing the constrained demodulated signal to obtain a discretized demodulated signal;

[0072] A solution module, configured to solve the discretized demodulation signal by using the augmented Lagrangian multiplier method and the alternating direction multiplier method, and obtain the iteratively optimized demodulation signal;

[0073] An estimated instantaneous frequency acquisition module, configured to calculate an instantaneous frequency increment based on the iteratively optimized demodulation signal, impose smoothness constraints and low-rank constraints on the instantaneous frequency increment, and solve through an iterative solution algorithm to obtain an estimated instantaneous frequency;

[0074] A fault confirmation module, configured to determine the type of rotating machinery fault based on the pre-acquired theoretical fault characteristic frequency and the estimated instantaneous frequency.

[0075] In a third aspect, the present invention provides an electronic terminal, including a processor and a memory connected to the processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the method described in any one of the above are executed.

[0076] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0077] Compared with the prior art, the beneficial effects achieved by the present invention:

[0078] The present invention first proposes a rotating machinery fault diagnosis method. By imposing sparse constraints on the demodulation signal, the estimation accuracy of the demodulation signal is improved and the subsequent instantaneous frequency estimation error is reduced; by imposing low-rank constraints on the instantaneous frequency, the inherent proportional relationship of the instantaneous frequency is revealed and the inaccurate initial instantaneous frequency is optimized to reach the true instantaneous frequency. The hybrid constraints enable the present invention to effectively improve the instantaneous frequency estimation accuracy and noise robustness, and effectively process non-linear frequency modulation signals with closely spaced instantaneous frequencies. Therefore, the present invention can still accurately extract the instantaneous frequency under various complex interference conditions, realize the accurate diagnosis of rotating machinery faults, and ensure the safe operation of the entire mechanical system. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 is a flowchart of a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0080] Figure 2 is a schematic diagram of a planetary gearbox test bench in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0081] Figure 3 is a schematic diagram of the structure of a planetary gearbox in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0082] Figure 4It is the time-domain waveform diagram of the vibration signal in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0083] Figure 5 It is the schematic diagram of the rotation frequency of the motor in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0084] Figure 6 It is the time-frequency diagram of the short-time Fourier transform of the vibration signal in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0085] Figure 7 It is the initial instantaneous frequency diagram obtained by ridge line detection from the short-time Fourier transform time-frequency diagram in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0086] Figure 8 It is the schematic diagram of the estimated instantaneous frequency in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0087] Figure 9 It is the time-frequency schematic diagram of the estimated instantaneous frequency in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0088] Figure 10 It is the schematic diagram of the estimated instantaneous frequency by the traditional variational non-linear frequency modulation mode decomposition in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0089] Figure 11 It is the time-frequency schematic diagram of the estimated instantaneous frequency by the traditional variational non-linear frequency modulation mode decomposition in a rotating machinery fault diagnosis method provided by an embodiment of the present invention;

[0090] Among them, 1. Driving motor; 2. Input encoder; 3. Planetary gearbox; 4. Output encoder; 5. Electromagnetic brake. Specific implementation mode

[0091] The technical solution of the present invention will be described in detail below through the drawings and specific embodiments. It should be understood that the specific features in the embodiments of the present application and the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations on the technical solution of the present application. Without conflict, the technical features in the embodiments of the present application and the embodiments can be combined with each other.

[0092] The term "and / or" in the present invention is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist, and B exists alone. In addition, the character " / " in the present invention generally represents an "or" relationship between the associated objects before and after.

[0093] Embodiment 1:

[0094] The rotating machinery in the present invention is introduced by taking a planetary gearbox as an example. Figure 2 Shown is a planetary gearbox test bench, which includes an electromagnetic brake 5, an output encoder 4, a planetary gearbox 3, an accelerometer (not shown in the figure) installed on the planetary gearbox, an input encoder 2, and a drive motor 1. The electromagnetic brake is used to load the output shaft, and the accelerometer is used to collect vibration signals. The signals are sampled at 5000 Hz, and the sampling time is 2.5 s. In addition, during the experiment, by manually adjusting the motor controller, the rotational speed undergoes an acceleration and deceleration process. Figure 3 The structural schematic diagram of the planetary gearbox is given. This planetary gearbox includes a sun gear Z with 20 teeth S , a ring gear Z with 82 teeth R and three planetary gears Z with 31 teeth evenly distributed p . When the sun gear and the planetary carrier are respectively connected to the input shaft and the output shaft, the ring gear is in a stationary state. According to the structural parameters of the planetary gearbox, the theoretical fault characteristic frequencies of the planetary gear, the sun gear, and the ring gear are calculated as , where is the rotational frequency.

[0095] Figure 1 is the flowchart of the rotating machinery fault diagnosis method in Embodiment 1 of the present invention. This flowchart only shows the logical sequence of the method described in this embodiment. On the premise of non - conflict, in other possible embodiments of the present invention, the steps shown or described can be completed in a different order from Figure 1 that shown.

[0096] The rotating machinery fault diagnosis method provided in this embodiment can be applied to a terminal and can be executed by a mechanical equipment fault identification device. This device can be implemented in software and / or hardware, and this device can be integrated in the terminal, for example: any smart phone, tablet computer or computer device with communication functions. The method of this embodiment specifically includes the following steps:

[0097] Step 1: Obtain vibration signals. Install an accelerometer on the device under test to collect vibration signals, then intercept 2.5 s of vibration signals and remove the mean value. Denote the vibration signals as . The time - domain waveform diagram of the vibration signals is as shown in Figure 4 , and the rotational frequency diagram of the motor is as shown in Figure 5 .

[0098] Step 2: Input the vibration signals into a pre - constructed fault diagnosis model, perform a demodulation operation on the vibration signals to obtain demodulated signals corresponding to the vibration signals, and apply smoothness constraints and sparsity constraints to the demodulated signals to obtain the constrained demodulated signals. The expression of the demodulated signals is:

[0099] ,

[0100] ,

[0101] Among them, when , has the narrowest bandwidth, that is, the demodulated signal. Therefore, the difference between these two demodulated signals is based on the trigonometric transformation formula and the difference angle formula in the demodulation technology. Sine and cosine demodulation respectively correspond to the quadrature component and the in-phase component of the signal, and the demodulation results are essentially different. The demodulated signals processed subsequently are the same.

[0102] Smoothness constraint:

[0103] Q1: ,

[0104] Sparsity constraint:

[0105] Q2: ,

[0106] Q3: ,

[0107] Among them, represents the th mode, M represents the total number of modes, represents the time, respectively represent the first demodulated signal and the second demodulated signal of the th mode after the smoothness constraint at time. g(t) represents the constraint function, respectively represent 's second derivative, represents the instantaneous amplitude of the th mode at time, and respectively represent the modulation frequency and the demodulation frequency of the th mode at time. The parameter s represents a variable in the definite integral in the interval [0, t]. and respectively represent the modulation frequency and the demodulation frequency of the th mode at s time, represents the initial phase, and are the first demodulation operator and the second demodulation operator respectively, represents the first demodulated signal and the second demodulated signal of the th mode after the sparsity constraint at time, They are the first penalty coefficient and the second penalty coefficient, respectively. is the Fourier transform matrix;

[0108] It should be noted here that the demodulated signal after constraint is the demodulated signal after smoothness constraint and sparsity constraint.

[0109] Step 3: Discretize the demodulated signal after constraint to obtain the demodulated signal after discretization:

[0110] Q4: ,

[0111] Q5: ,

[0112] Q6: ,

[0113] ,

[0114] ,

[0115] ,

[0116] ,

[0117] ,

[0118] ,

[0119] ,

[0120] Among them, represents the sampling duration from time 0 to time N - 1, g represents the vibration signal, which is the representation form after discretization of the vibration signal; and respectively represent the first demodulated signal and the second demodulated signal after smoothness constraint after discretization of the th mode, and respectively represent the first demodulated signal and the second demodulated signal after sparsity constraint after discretization of the th mode; and represent the general form after discretization; is expressed as the modulation frequency of the th mode, are the first and second penalty coefficients respectively, is the second-order difference matrix; represents the diagonal matrix, is the transpose; denotes the kernel matrix, is expressed as the demodulation operator at moment, and the specific expression form is , is expressed as the demodulation operator at moment, and the specific expression form is , represents the phase in the trigonometric function at time t.

[0121] Step 4: Solve the demodulation signal after the discretization process by the augmented Lagrangian multiplier method and the alternating direction multiplier method to obtain the iteratively optimized demodulation signal:

[0122] Solve the Q4 through the augmented Lagrangian multiplier and the alternating direction multiplier method to obtain:

[0123] ,

[0124] ,

[0125] ,

[0126] where, represents the vibration signal, respectively correspond to the first demodulation signal value and the second demodulation signal value of the q-th mode at the -th iteration, respectively represent the first kernel matrix value and the second kernel matrix value corresponding to the q-th mode at the -th iteration (note that for the simplicity of symbols, the iteration counter (i.e., the superscript n) of each variable is omitted on the right side of the equation);

[0127] Introduce the first auxiliary function Construct the Q5 into a constrained optimization problem and solve it through the augmented Lagrangian multiplier and the alternating direction multiplier method to obtain:

[0128] ,

[0129] ,

[0130] ,

[0131] Introduce the second auxiliary function h to construct the Q6 into a constrained optimization problem and solve it through the augmented Lagrangian multiplier and the alternating direction multiplier method to obtain:

[0132] ,

[0133] ,

[0134] ,

[0135] Among them, represents the first iteration number, represents the third penalty coefficient, represents the first Lagrange multiplier, represents the second iteration number, represents the fourth penalty coefficient, represents the second Lagrange multiplier, represents the soft threshold function, represents the third iteration number, and h is the second auxiliary function, is the fifth penalty coefficient, is the third Lagrange multiplier.

[0136] Step Five: Calculate the instantaneous frequency increment based on the iteratively optimized demodulation signal, impose smoothness constraints and low-rankness constraints on the instantaneous frequency increment, and solve for the estimated instantaneous frequency through an iterative solution algorithm, including:

[0137] Based on the iteratively optimized demodulation signal, obtaining the instantaneous frequency increment through the arctangent demodulation method includes:

[0138] ,

[0139] Perform discretization processing on the instantaneous frequency increment to obtain the discretized instantaneous frequency increment, add smoothness constraints to the discretized instantaneous frequency increment, solve for the desired instantaneous frequency increment, and obtain the initially optimized estimated instantaneous frequency based on the desired instantaneous frequency increment,

[0140] ,

[0141] ,

[0142] ,

[0143] ,

[0144] Among them, is the sixth penalty coefficient, , is the desired instantaneous frequency increment, represents the identity matrix, is represented as the initial optimization of the estimated instantaneous frequency for iteration times, is represented as the initial optimization of the estimated instantaneous frequency for iteration times;

[0145] Construct a noise observation matrix based on the estimated instantaneous frequency optimized initially, and add a low-rank constraint to the noise observation matrix. Obtaining the estimated instantaneous frequency includes:

[0146] Q7: ,

[0147] where, is the estimated instantaneous frequency, is a low-rank matrix, is the noise observation matrix composed of the initially optimized , is the seventh penalty coefficient, represents the th singular value of the matrix , represents the Frobenius norm of the matrix ;

[0148] Shrink the singular values of the noise observation matrix to solve for the above Q7 to obtain:

[0149] ,

[0150] where, are all orthogonal matrices obtained by applying singular value decomposition to the matrix , is the largest singular value of the , Each row of represents the estimated instantaneous frequency.

[0151] Step Six: Based on the pre-obtained theoretical fault characteristic frequencies and the estimated instantaneous frequencies, determine the rotating machinery fault type according to the matching result:

[0152] Calculate the theoretical fault characteristic frequencies of the rotating machinery according to the structural parameters and rotational frequency of the rotating machinery, compare the estimated instantaneous frequencies with the theoretical fault characteristic frequencies, and determine the rotating machinery fault type according to the matching result;

[0153] It should be noted that calculating the theoretical fault characteristic frequencies here is just substituting the parameters into the publicly available rotating machinery fault characteristic frequency formulas (in this embodiment, specifically: for gears: sun gear, planet gear, and ring gear; for bearings: fault characteristic frequency formulas for inner ring, outer ring, and roller). This belongs to the prior art and will not be elaborated here;

[0154] As shown in the above steps, the instantaneous frequencies estimated by the method proposed in this patent are only compared with the theoretical fault characteristic frequencies here to determine the rotating machinery fault location. This also belongs to the prior art and will not be elaborated here.

[0155] Figure 6 and Figure 7The time-frequency diagrams of the short-time Fourier transform of the vibration signals and the initial instantaneous frequency diagrams obtained by ridge line detection from the time-frequency diagrams of the short-time Fourier transform are respectively given. It can be seen from the initial instantaneous frequency diagrams that the initial instantaneous frequencies are mutually aliased and seriously deviate from the true instantaneous frequencies. Using these initial instantaneous frequencies, the instantaneous frequencies estimated by the method of the present invention and their time-frequency diagrams are respectively as Figure 8 and Figure 9 shown. From Figure 8 it can be seen that the method of the present invention accurately estimates the rotational frequency , the fault characteristic frequency of the ring gear and its harmonics , indicating that the ring gear fault of the planetary gearbox is accurately diagnosed; at the same time Figure 9 gives a clear time-frequency diagram of the estimated instantaneous frequency, in which the influence of noise is eliminated. These information verify that the method of the present invention can effectively improve the accuracy of instantaneous frequency estimation even in the case of severe noise and inaccurate initial instantaneous frequencies, and realize the fault diagnosis of the gearbox. The instantaneous frequencies estimated by the traditional variational non-linear frequency modulation mode decomposition and their time-frequency diagrams are as Figure 10 and Figure 11 shown. From Figure 10 no fault-related frequencies can be identified, and the distribution of its estimated instantaneous frequencies is consistent with the initial instantaneous frequencies, which indicates that the traditional variational non-linear frequency modulation mode decomposition has poor accuracy in estimating instantaneous frequencies due to its own limitations and cannot realize the fault diagnosis of the planetary gearbox.

[0156] It should be noted that Figure 10 in the Initial is translated as initial value, and Estimated is translated as estimated value.

[0157] The method of the present invention can be designed as a rotating machinery fault diagnosis software, installed in the upper computer, the upper computer is connected to the signal acquisition device, and the upper computer processes the vibration signal in real time according to the above steps to timely diagnose the rotating machinery fault and determine the fault type, so as to ensure the operation safety of the entire mechanical system.

[0158] Example Two:

[0159] The second embodiment of the present invention provides a rotating machinery fault diagnosis device, including:

[0160] A signal acquisition module for acquiring vibration signals;

[0161] A demodulated signal optimization module for inputting the vibration signal into a pre-constructed fault diagnosis model, extracting a demodulated signal from the decomposed non-linear frequency modulation mode, and applying a smoothness constraint and a sparsity constraint to the demodulated signal to obtain a constrained demodulated signal;

[0162] A discretization module, configured to perform discretization processing on the demodulated signal after constraint to obtain the demodulated signal after discretization processing;

[0163] A solution module, configured to solve the demodulated signal after discretization processing by using the augmented Lagrangian multiplier method and the alternating direction multiplier method to obtain the demodulated signal after iterative optimization;

[0164] An estimated instantaneous frequency acquisition module, configured to calculate an instantaneous frequency increment based on the demodulated signal after iterative optimization, impose smoothness constraint and low-rank constraint on the instantaneous frequency increment, and solve to obtain an estimated instantaneous frequency through an iterative solution algorithm;

[0165] A fault confirmation module, configured to determine the type of rotating machinery fault based on the pre-acquired theoretical fault characteristic frequency and the estimated instantaneous frequency according to the matching result.

[0166] The rotating machinery fault diagnosis device provided in the second embodiment of the present invention can execute the rotating machinery fault diagnosis method provided in the first embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0167] Embodiment Three:

[0168] The third embodiment of the present invention further provides an electronic terminal, including a processor and a memory connected to the processor. A computer program is stored in the memory, and the processor is configured to operate according to the instruction to execute the steps of the method in the first embodiment.

[0169] The electronic terminal provided in the third embodiment of the present invention can execute the rotating machinery fault diagnosis method provided in the first embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0170] Embodiment Four:

[0171] The fourth embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the method in the first embodiment, and has corresponding functional modules and beneficial effects for executing the method.

[0172] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a device, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0173] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce a means for implementing the functions specified in one or more flows Figure 1 one or more flows and / or blocks Figure 1 or a means for implementing the functions specified in one or more blocks.

[0174] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction means that implements the functions specified in one or more flows Figure 1 one or more flows and / or blocks Figure 1 or a means for implementing the functions specified in one or more blocks.

[0175] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more flows Figure 1 one or more flows and / or blocks Figure 1 or a means for implementing the functions specified in one or more blocks.

[0176] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principles of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A method for diagnosing faults in rotating machinery, characterized in that, Including: Obtain vibration signals; Perform a demodulation operation on the vibration signal to obtain a demodulation signal corresponding to the vibration signal, and apply a smoothness constraint and a sparsity constraint to the demodulation signal to obtain a constrained demodulation signal; Perform discretization processing on the constrained demodulation signal to obtain a discretized demodulation signal; Solve the discretized demodulation signal by the augmented Lagrangian multiplier method and the alternating direction multiplier method to obtain an iteratively optimized demodulation signal; Calculate the instantaneous frequency increment based on the iteratively optimized demodulation signal, apply a smoothness constraint and a low-rank constraint to the instantaneous frequency increment, and solve by an iterative solution algorithm to obtain an estimated instantaneous frequency; Determine the type of rotating machinery fault based on the pre-obtained theoretical fault characteristic frequency and the estimated instantaneous frequency.

2. The rotating machinery fault diagnosis method according to claim 1, characterized in that Applying a smoothness constraint and a sparsity constraint to the demodulation signal to obtain a constrained demodulation signal includes: The expression of the demodulation signal is: ， ， Smoothness constraint: Q1： ， Sparsity constraint: Q2： ， Q3： ， Among them, represents the th mode, M represents the total number of modes, represents the time, respectively represent the th mode's first demodulation signal and second demodulation signal after smoothness constraint at time, g(t) represents the constraint function, respectively represent 's second-order derivatives, represents the th mode's time's instantaneous amplitude, and respectively represent the th mode's time's modulation frequency and demodulation frequency, the parameter s represents a variable in the definite integral in the interval [0, t], and respectively represent the th mode's modulation frequency and demodulation frequency at s time, represents the initial phase, and are the first demodulation operator and the second demodulation operator respectively, represents the th mode's first demodulation signal and second demodulation signal after sparsity constraint at time, are the first penalty coefficient and the second penalty coefficient respectively, is the Fourier transform matrix.

3. The rotating machinery fault diagnosis method according to claim 2, wherein Performing discretization processing on the constrained demodulation signal to obtain a discretized demodulation signal includes: Q4： ， Q5： ， Q6： ， ， ， ， ， ， ， ， Among them, represents the sampling duration from time 0 to time N - 1, g represents the vibration signal, which is the representation form after the discretization process of the vibration signal; and respectively represent the first demodulation signal and the second demodulation signal after the discretization process of the th mode and passing through the smoothness constraint, and respectively represent the first demodulation signal and the second demodulation signal after the discretization process of the th mode and passing through the sparsity constraint; and represent the general form after the discretization process; is expressed as the modulation frequency of the th mode, are the first and second penalty coefficients respectively, is the second-order difference matrix; represents the diagonal matrix, is the transpose; represents the kernel matrix, is expressed as the demodulation operator at time, and the specific representation form is , is expressed as the demodulation operator at time, and the specific representation form is , represents the phase in the trigonometric function at time t.

4. The rotating machinery fault diagnosis method according to claim 3, wherein Solving the discretized demodulation signal by the augmented Lagrangian multiplier method and the alternating direction multiplier method to obtain an iteratively optimized demodulation signal includes: Solving the Q4 by the augmented Lagrangian multiplier and the alternating direction multiplier method to obtain: ， ， ， Among them, represents the vibration signal, respectively corresponding to the first demodulation signal value and the second demodulation signal value of the q-th mode at the k-th iteration, respectively representing the first kernel matrix value and the second kernel matrix value corresponding to the q-th mode at the k-th iteration; Introduce the first auxiliary function Construct the Q5 as a constrained optimization problem and solve it by the augmented Lagrangian multiplier and the alternating direction multiplier method to obtain: ， ， ， Introduce a second auxiliary function h to construct the Q6 into a constrained optimization problem, and solve by the augmented Lagrangian multiplier and the alternating direction multiplier method to obtain: ， ， ， Among them, represents the first iteration number, represents the third penalty coefficient, represents the first Lagrange multiplier, represents the second iteration number, represents the fourth penalty coefficient, represents the second Lagrange multiplier, represents the soft threshold function, represents the third iteration number, h is the second auxiliary function, is the fifth penalty coefficient, is the third Lagrange multiplier.

5. The rotating machinery fault diagnosis method according to claim 4, characterized in that Calculating the instantaneous frequency increment based on the iteratively optimized demodulation signal, applying a smoothness constraint and a low-rank constraint to the instantaneous frequency increment, and solving by an iterative solution algorithm to obtain an estimated instantaneous frequency includes: Based on the iteratively optimized demodulation signal, obtaining the instantaneous frequency increment by the arctangent demodulation method includes: ， Perform discretization processing on the instantaneous frequency increment to obtain a discretized instantaneous frequency increment, apply a smoothness constraint to the discretized instantaneous frequency increment, solve to obtain an expected instantaneous frequency increment, and obtain an initially optimized estimated instantaneous frequency according to the expected instantaneous frequency increment; ， ， ， ， Among them, is the sixth penalty coefficient, , is the desired instantaneous frequency increment, represents the identity matrix, is expressed as the initial optimized estimated instantaneous frequency after iterations, is expressed as the initial optimized estimated instantaneous frequency after iterations; Construct a noise observation matrix according to the initially optimized estimated instantaneous frequency, apply a low-rank constraint to the noise observation matrix, and obtain an estimated instantaneous frequency includes: Q7： ， Among them, to estimate the instantaneous frequency, is a low-rank matrix, is the noise observation matrix composed of the initial optimization ; is the seventh penalty coefficient, represents the matrix the th singular value of, represents the Frobenius norm of the matrix ; Shrink the singular value of the noise observation matrix to solve the Q7 to obtain: , Among them, are all orthogonal matrices obtained by applying singular value decomposition to the matrix , is the largest singular value of Each row of represents the estimated instantaneous frequency.

6. The rotating machinery fault diagnosis method according to claim 5, wherein Determining the type of rotating machinery fault based on the pre-obtained theoretical fault characteristic frequency and the estimated instantaneous frequency includes: Calculate the theoretical fault characteristic frequency of the rotating machinery according to the structural parameters and rotation frequency of the rotating machinery, compare the estimated instantaneous frequency with the theoretical fault characteristic frequency, and determine the type of rotating machinery fault according to the matching result.

7. A rotating machinery fault diagnosis device, characterized in that, Including: A signal acquisition module for acquiring vibration signals; A demodulation signal optimization module for performing a demodulation operation on the vibration signal to obtain a demodulation signal corresponding to the vibration signal, and applying a smoothness constraint and a sparsity constraint to the demodulation signal to obtain a constrained demodulation signal; A discretization module for performing discretization processing on the constrained demodulation signal to obtain a discretized demodulation signal; A solution module, configured to solve the discretized demodulation signal by using the augmented Lagrangian multiplier method and the alternating direction multiplier method, and obtain the iteratively optimized demodulation signal; An estimated instantaneous frequency acquisition module, configured to calculate an instantaneous frequency increment based on the iteratively optimized demodulation signal, impose smoothness constraints and low-rank constraints on the instantaneous frequency increment, and solve to obtain an estimated instantaneous frequency through an iterative solution algorithm; A fault confirmation module, configured to determine the type of rotating machinery fault based on a pre-acquired theoretical fault characteristic frequency and the estimated instantaneous frequency.

8. An electronic terminal, characterized in that, It includes a processor and a memory connected to the processor. A computer program is stored in the memory. When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are executed.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images