A method, apparatus and storage medium for diagnosing mechanical equipment faults

Abstract

This invention discloses a method, apparatus, and storage medium for diagnosing mechanical equipment faults in the field of mechanical fault diagnosis technology. The method includes: acquiring a preprocessed encoder signal; inputting the preprocessed encoder signal into a weighted dual-domain sparse decomposition model, and obtaining distinguishable periodic fault impact features through the objective function of the weighted dual-domain sparse decomposition model; calculating the fault characteristic frequency when a fault occurs at different locations of the monitored equipment according to the structure and operating parameters of the monitored equipment, and calculating the corresponding fault characteristic period based on the fault characteristic frequency; and determining the fault location of the monitored equipment by comparing the interval of the fault characteristic period with that of the periodic fault impact features. This invention can solve the problem that existing technologies cannot simultaneously consider the time-domain and frequency-domain sparsity of encoder signals, thereby enabling effective decoupling of different components of the encoder signal and improving the ability to extract and identify weak fault signals.

Description

Technical Field

[0001] This invention relates to the field of mechanical fault diagnosis technology, and in particular to a method, apparatus and storage medium for diagnosing mechanical equipment faults. Background Technology

[0002] Mechanical equipment, as an important component of modern industry, is widely used in fields such as wind power, metallurgy, mining, rail transportation, and aerospace. Its operating status directly affects the system's performance, safety, and service life. Once mechanical equipment malfunctions, it may reduce transmission efficiency and accelerate mechanical wear, or even lead to equipment shutdown or catastrophic accidents.

[0003] In recent years, with the continuous improvement of control precision requirements, a large number of rotating machinery devices have integrated high-precision rotary encoders to acquire kinematic information such as angles and rotational speeds. Research shows that encoder signals contain rich dynamic response characteristics. When abnormalities occur in mechanical components during operation (such as tooth surface spalling or tooth root cracks), the encoder signal will change accordingly. Therefore, if abnormal information can be effectively extracted from the encoder signal, the health status of mechanical components can be identified.

[0004] To facilitate the identification of mechanical faults, the raw encoder signals recording the angular displacement of the rotating shaft are typically converted into more diagnostically valuable kinematic variables, such as instantaneous angular velocity (IAS) and instantaneous angular acceleration (IAA). The presence of mechanical faults will alter the dynamic behavior of the equipment, causing periodic fluctuations in IAS and IAA.

[0005] However, encoder signals are inevitably affected by various interference components in practical applications, leading to interference problems in the derived IAS, IAA, and other variable signals, thus affecting the accuracy of fault diagnosis. Furthermore, the insufficient utilization of the inherent characteristics of the signal structure limits its feature extraction performance under strong interference conditions. Specifically, fault impulse signals typically exhibit a periodic sparse structure in the time domain, while harmonic interference components exhibit typical frequency-domain sparsity. Existing technologies cannot simultaneously consider the decomposition models of time-domain and frequency-domain sparsity to achieve effective decoupling of different components, further limiting the ability to extract and identify weak fault signals.

[0006] Therefore, there is an urgent need for a method, device, and storage medium for diagnosing mechanical equipment faults to solve the above-mentioned technical problems. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method, device and storage medium for diagnosing mechanical equipment faults, which can solve the technical problem that the prior art cannot simultaneously take into account the time domain and frequency domain sparsity of encoder signals.

[0008] To achieve the above objectives, the present invention is implemented using the following technical solution:

[0009] In a first aspect, the present invention provides a method for diagnosing mechanical equipment faults, comprising:

[0010] Obtain the preprocessed encoder signal;

[0011] The preprocessed encoder signal is input into a pre-constructed weighted dual-domain sparse decomposition model, and the distinguishable periodic fault impact characteristics are obtained through the objective function of the weighted dual-domain sparse decomposition model.

[0012] Based on the structure and operating parameters of the monitored equipment, calculate the fault characteristic frequency when a fault occurs at different locations of the monitored equipment, and calculate the corresponding fault characteristic period based on the fault characteristic frequency.

[0013] Based on the fault characteristic period, the fault location of the monitored equipment is determined by comparing it with the interval of the periodic fault impact characteristic.

[0014] Further, the original encoder signal is acquired;

[0015] The original encoder signal is converted into an instantaneous angular velocity signal using a first-order differential algorithm. The instantaneous angular velocity signal is then subjected to time-domain synchronous averaging to obtain the preprocessed encoder signal.

[0016] Furthermore, the objective function F1 expression of the weighted dual-domain sparse decomposition model includes:

[0017] ,

[0018] in, This represents the pre-processed encoder signal. Indicates harmonic components, This represents the inverse Fourier transform operator. Represents the frequency domain sparse coefficients. Indicates the impact characteristics of periodic faults. Denotes the second norm, and Both represent regularization parameters, which control the sparsity constraint strength in the frequency domain and time domain, respectively. Denotes the non-convex penalty function. As the first weighting coefficient, This is the second weighting coefficient. Index representing a periodic group, This indicates the number of elements in a periodic group. This represents the element-wise dot product operation. and Both represent penalty parameters that control the degree of non-convexity of the penalty function. This represents a periodic binary vector used to constrain the periodic group sparsity characteristics of impact signals in the time domain. Indicates the first The first weighting coefficient of each period group Indicates the first The second weighting coefficient of each period group Indicates the first The frequency domain sparse coefficients of a periodic group Indicates the first The periodic fault impact characteristics of each periodic group include: Each element.

[0019] Furthermore, the expression for the periodic binary vector includes:

[0020] ,

[0021] , and The following conditions must be met:

[0022] ,

[0023] ,

[0024] in, This represents the number of non-zero points in a single fault characteristic cycle. The number of zeros in a single fault characteristic cycle. It is the sampling frequency. Indicates the fault characteristic frequency, A periodic binary vector The number of fault characteristic cycles included.

[0025] Furthermore, the first The expression for the first weighting coefficient of each period group includes:

[0026] ,

[0027] The first The expressions for the second weighting coefficient of each period group include:

[0028] ,

[0029] in, It is a pre-set positive parameter used to avoid the denominator being zero.

[0030] Furthermore, the solution method for the weighted bi-domain sparse decomposition model includes:

[0031] Introducing auxiliary variables The objective function F1 of the unconstrained optimization problem is transformed into a constrained optimization problem F2, the expression of which is:

[0032] ,

[0033] The constrained optimization problem F2 is decomposed into the following sub-constrained optimization problems using the alternating direction multiplier method:

[0034] ,

[0035] ,

[0036] ,

[0037] ,

[0038] in, For penalty parameters, For intermediate scaling variables;

[0039] Solve the sub-constraint optimization problem to obtain the distinct periodic fault impact characteristics.

[0040] In a second aspect, the present invention provides a mechanical equipment fault diagnosis device, comprising:

[0041] The signal acquisition module is used to acquire the preprocessed encoder signal;

[0042] The feature extraction module is used to input the preprocessed encoder signal into a pre-constructed weighted dual-domain sparse decomposition model, and obtain the distinguishable periodic fault impact features through the objective function of the weighted dual-domain sparse decomposition model.

[0043] The fault characteristic cycle calculation module is used to calculate the fault characteristic frequency when a fault occurs at different locations of the monitored equipment based on the structure and operating parameters of the monitored equipment, and to calculate the corresponding fault characteristic cycle based on the fault characteristic frequency.

[0044] The fault location confirmation module is used to determine the fault location of the monitored equipment by comparing the fault characteristic period with the interval of the periodic fault impact characteristic.

[0045] Thirdly, 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 of the preceding claims are performed.

[0046] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0047] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0048] This invention proposes a fault diagnosis method for mechanical equipment. In a pre-constructed weighted dual-domain sparse decomposition model, it fully utilizes the differences in morphological characteristics between fault impact features and interference components in the time and frequency domains to distinguish fault impact features from harmonic components, thereby obtaining the differentiated periodic fault impact features. Then, based on the structure and operating parameters of the monitored equipment, the fault characteristic frequencies at different locations of the monitored equipment are calculated. Based on the fault characteristic frequencies, the fault characteristic period is calculated. Finally, based on the fault characteristic period, the interval of the periodic fault impact features is compared to determine the fault location of the monitored equipment. This method takes into account the sparsity of the encoder signal in both the time and frequency domains, improves the effective decoupling of different components of the encoder signal, and enhances the ability to extract and identify weak fault signals.

[0049] By introducing weighting coefficients, periodic binary vectors, and non-convex penalty functions, non-convex regular terms are constructed to constrain the periodic group sparsity characteristics of fault impact features in the time domain and the frequency domain sparsity characteristics of harmonic interference components in the frequency domain. Attached Figure Description

[0050] Figure 1 This application provides a flowchart of a fault diagnosis method for mechanical equipment.

[0051] Figure 2 This is a schematic diagram of the original encoder signal in a mechanical equipment fault diagnosis method provided in an embodiment of this application;

[0052] Figure 3 This application provides a structural diagram of a planetary gearbox test bench in a mechanical equipment fault diagnosis method.

[0053] Figure 4 This is a schematic diagram of the instantaneous angular velocity signal obtained by first-order difference in a mechanical equipment fault diagnosis method provided in an embodiment of this application;

[0054] Figure 5 This is a schematic diagram of the time-domain synchronous averaging result in a mechanical equipment fault diagnosis method provided in an embodiment of this application;

[0055] Figure 6 A schematic diagram of the harmonic components of a weighted dual-domain sparse decomposition model in a mechanical equipment fault diagnosis method provided in this application embodiment;

[0056] Figure 7 This is a schematic diagram of the periodic fault impact characteristics of a weighted dual-domain sparse decomposition model in a mechanical equipment fault diagnosis method provided in this application embodiment;

[0057] Figure 8 This is a schematic diagram of harmonic components obtained by the Q-switching factor wavelet transform decomposition method provided in an embodiment of this application;

[0058] Figure 9 A schematic diagram of periodic fault impact characteristics obtained by the Q-switching factor wavelet transform decomposition method according to an embodiment of this application;

[0059] Figure 10 This is a schematic diagram of the analysis results obtained by the maximum correlation kurtosis deconvolution method provided in an embodiment of this application.

[0060] The components include: 1. Servo motor; 2. Input encoder; 3. Planetary gearbox; 4. Output encoder; and 5. Brake. Detailed Implementation

[0061] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0062] In this invention, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B together, or B alone. Additionally, in this invention, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0063] Example 1:

[0064] Figure 1 This is a flowchart of the mechanical equipment fault diagnosis method in Embodiment 1 of the present invention. This flowchart only illustrates the logical sequence of the method described in this embodiment. Provided there are no conflicts, different methods may be used in other possible embodiments of the present invention. Figure 1 Complete the steps shown or described in the order indicated.

[0065] The mechanical equipment 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 can be integrated into the terminal, such as any smartphone, tablet, or computer device with communication capabilities. The method in this embodiment specifically includes the following steps:

[0066] Step 1: Obtain the preprocessed encoder signal, specifically including:

[0067] like Figure 2 As shown, the original encoder signal is obtained from the monitored device, and its structure, operating parameters, sampling frequency, etc. are recorded to provide data support for subsequent processing.

[0068] like Figure 3 The diagram shows the overall structure of the planetary gearbox experimental platform, used to acquire raw encoder signals. It includes a servo motor 1, a planetary gearbox 3, an input encoder 2, an output encoder 4, a coupling (not shown), and a brake 5. This planetary gearbox experimental platform allows for adjustable speed and load to simulate the operating conditions of a planetary gearbox under different working conditions in actual engineering. Encoders are mounted on the input and output shafts respectively to collect rotational position information. The planetary gearbox includes a sun gear (20 teeth), three equally spaced planetary gears (31 teeth each), and a ring gear (82 teeth). The sun gear is rigidly connected to the input shaft, driving the planetary gears to rotate. The planetary gears are connected to the output shaft via a bracket to transmit power, while the ring gear remains stationary. The experimental platform can simulate various typical fault modes, such as tooth wear, spalling, and broken teeth. Data acquisition is performed using a Heidenhain IK220 high-performance counter card at a sampling frequency of 5000Hz. In this experiment, the servo motor 1 speed was set to 20Hz.

[0069] A planetary gear with single-tooth wear failure was tested on the experimental platform. Figure 3 The raw signal from the output shaft encoder is shown, lasting 3 seconds. Since the encoder output is a cumulative angular displacement, its waveform exhibits an approximately linear upward trend, which cannot directly reveal potential fault information.

[0070] The original encoder signal was converted into an instantaneous angular velocity signal using a first-order differential algorithm, and the result is as follows. Figure 4 As shown, however, due to the presence of numerous harmonic components and background noise in the signal, the periodic fault impact characteristics are still severely obscured and difficult to identify. This indicates that the instantaneous angular velocity obtained solely through differential operation fails to effectively reveal diagnostic information related to gear faults. The main reason for this is that while differential operation can enhance fault characteristics, it also amplifies the interference effects of noise and harmonics.

[0071] Subsequently, the instantaneous angular velocity signal is subjected to time-domain synchronous averaging to enhance the periodic fault impact characteristics that may exist in the encoder signal, and obtain the pre-processed encoder signal.

[0072] Figure 5 The result is a time-domain synchronous average, in which the periodic impact characteristics are still significantly masked by harmonics and their overtone interference, and therefore cannot be directly used for fault location and identification.

[0073] Step 2: Input the preprocessed encoder signal into the pre-constructed weighted dual-domain sparse decomposition model, and obtain the distinguishable periodic fault impact characteristics through the objective function of the weighted dual-domain sparse decomposition model.

[0074] The objective function F1 expression of the weighted dual-domain sparse decomposition model includes:

[0075] ,

[0076] in, This represents the pre-processed encoder signal. Indicates harmonic components, This represents the inverse Fourier transform operator, used to transform sparse coefficients in the frequency domain. Mapping back to the time domain signal, Represents the frequency domain sparse coefficients. Indicates the impact characteristics of periodic faults. Denotes the second norm, and Both represent regularization parameters, which control the sparsity constraint strength in the frequency domain and time domain, respectively. The non-convex penalty function is represented in this embodiment. Let it be atan, so as to avoid underestimating large values ​​to the greatest extent possible, while maximizing the sparsity of the coefficients. As the first weighting coefficient, This is the second weighting coefficient. Index representing a periodic group, This indicates the number of elements in a periodic group. This represents the element-wise dot product operation. and Both represent penalty parameters that control the degree of non-convexity of the penalty function. This represents a periodic binary vector used to constrain the periodic group sparsity characteristics of impact signals in the time domain. Indicates the first The first weighting coefficient of each period group Indicates the first The second weighting coefficient of each period group Indicates the first The frequency domain sparse coefficients of a periodic group Indicates the first The periodic fault impact characteristics of each periodic group include: Each element.

[0077] Specifically, frequency domain sparse regularization term The periodic group sparsity regularization term is used to constrain the frequency domain sparsity of harmonic components in the frequency domain. This method is used to constrain the periodic group sparsity of fault impact features in the time domain. It combines group sparsity structure with non-convex penalty function, which helps to enhance the recognition ability of impact features with obvious periodic structure.

[0078] Specifically, The expression is:

[0079] ,

[0080] Where R is the set of real numbers.

[0081] Specifically, the expression for the periodic binary vector includes:

[0082] ,

[0083] , and The following conditions must be met:

[0084] ,

[0085] ,

[0086] in, This represents the number of non-zero points in a single fault characteristic cycle. The number of zeros in a single fault characteristic cycle. It is the sampling frequency. Indicates the fault characteristic frequency, A periodic binary vector The number of fault characteristic cycles included.

[0087] Specifically, the first The expression for the first weighting coefficient of each period group includes:

[0088] ,

[0089] The first The expressions for the second weighting coefficient of each period group include:

[0090] ,

[0091] in, These are pre-set positive parameters used to avoid zero denominators; they are weighting coefficients introduced in the weighted two-domain sparse decomposition model. and This further enhances the model's ability to utilize sparse priors. The weighting coefficients can strengthen the sparse prior structure of harmonic components and impulse features in the frequency and time domains, as well as the sparse prior structure of periodic groups, highlighting significant features. This allows the weighted dual-domain sparse decomposition model to extract features more effectively and improve the decoupling and separation accuracy of the two types of features.

[0092] The specific methods for obtaining the distinguishable periodic fault impact characteristics are derived through a solution method, which includes:

[0093] Introducing auxiliary variables The objective function F1 of the unconstrained optimization problem is transformed into a constrained optimization problem F2, the expression of which is:

[0094] ,

[0095] The solution of the constrained optimization problem F2 is transformed into an iterative solution of the sub-constrained optimization problem F3 by using the alternating direction multiplier method:

[0096] ,

[0097] The sub-constrained optimization problem F3 is a standard least squares problem, and its explicit solution is:

[0098] ;

[0099] Where T is the transpose operator.

[0100] The solution of the constrained optimization problem F2 is transformed into an iterative solution of the sub-constrained optimization problem F4 using the alternating direction multiplier method.

[0101] ,

[0102] The optimal solution to the sub-constrained optimization problem F4 is derived using the controlled minimization method, expressed as follows:

[0103] ,

[0104] In the formula, Indicates the number of iterations. For iteration The periodic failure impact characteristics, For the second diagonal matrix:

[0105] ,

[0106] in, For the first One element, This represents the nth row and nth column of the matrix.

[0107] The solution of the constrained optimization problem F2 is transformed into an iterative solution of the sub-constrained optimization problem F5 using the alternating direction multiplier method.

[0108] ,

[0109] ,

[0110] in, For penalty parameters, For intermediate scaling variables;

[0111] The optimal solution to the sub-constrained optimization problem F5 is derived using the controlled minimization method, and is expressed as:

[0112] ,

[0113] in, Given the first diagonal matrix, the solution is:

[0114] ,

[0115] in, Indicates and Different second nonconvex penalty functions.

[0116] In summary, by solving the sub-constraint optimization problem, we can obtain the distinguishable periodic fault impact characteristics, i.e. , Figure 6 and Figure 7 The decomposition results of the weighted dual-domain sparse decomposition model are shown, and it can be seen that the harmonic components and periodic fault impact characteristics are accurately separated.

[0117] Step 3: Based on the structure and operating parameters of the monitored equipment, calculate the fault characteristic frequency when a fault occurs at different locations of the monitored equipment, and calculate the corresponding fault characteristic period based on the fault characteristic frequency;

[0118] It should be noted that calculating the fault characteristic frequency when a fault occurs at different locations of the monitored equipment based on structural parameters and calculating the corresponding fault characteristic period based on the fault characteristic frequency are both conventional existing technologies and will not be elaborated here.

[0119] Step 4: Based on the fault characteristic period, compare it with the interval of the periodic fault impact characteristic to determine the fault location of the monitored equipment.

[0120] Under the condition that the servo motor speed is 20Hz, the fault characteristic frequencies of the sun gear, planet gears, and ring gear are calculated based on the structural parameters of the planetary gearbox, as shown in Table 1, which serves as the basis for subsequent fault location. Specifically, and These are the servo motor speed and the output shaft speed, respectively. , and These are the fault characteristic frequencies of the sun gear, planet gears, and ring gear, respectively.

[0121] Table 1 Fault Characteristic Frequencies of Planetary Gearbox

[0122]

[0123] In addition, it can be seen that Figure 6 and Figure 7 The impact feature signal with a clear periodic structure was successfully extracted, and the time interval of the extracted impact signal was [not specified]. The correlation with the planetary gear failure cycle indicates that the planetary gear failure was accurately identified, verifying the advantages of this method in terms of fault identification accuracy and signal-to-noise ratio enhancement.

[0124] To further illustrate the superiority of the proposed method, the Q-factor wavelet transform decomposition method and the maximum correlation kurtosis deconvolution method are introduced as comparative methods. Figure 8 and Figure 9 The results shown are from the Q-factor wavelet transform decomposition method. Although it achieves signal separation, the impulse characteristics are still significantly affected by random noise, with only some periodic features remaining. This indicates that the Q-factor wavelet transform method has certain limitations in extracting weak periodic impulses, mainly because it fails to explicitly constrain the sparse structure of the periodic group of the impulse signal, resulting in insufficient suppression of the time-domain aliasing between harmonic components and transient impulse components.

[0125] also, Figure 10 The analysis results of the maximum correlation kurtosis deconvolution method are presented. In addition to the periodic impact features caused by the fault, other interfering impact features can be observed. Moreover, the background noise is complex and has amplitude distortion, which reduces the identifiability of the fault impact features and limits its applicability in weak fault identification.

[0126] Example 2:

[0127] Embodiment 2 of the present invention provides a mechanical equipment fault diagnosis device, comprising:

[0128] The signal acquisition module is used to acquire the preprocessed encoder signal;

[0129] The feature extraction module is used to input the preprocessed encoder signal into a pre-constructed weighted dual-domain sparse decomposition model, and obtain the distinguishable periodic fault impact features through the objective function of the weighted dual-domain sparse decomposition model.

[0130] The fault characteristic cycle calculation module is used to calculate the fault characteristic frequency when a fault occurs at different locations of the monitored equipment based on the structure and operating parameters of the monitored equipment, and to calculate the corresponding fault characteristic cycle based on the fault characteristic frequency.

[0131] The fault location confirmation module is used to determine the fault location of the monitored equipment by comparing the fault characteristic period with the interval of the periodic fault impact characteristic.

[0132] The mechanical equipment fault diagnosis device provided in Embodiment 2 of the present invention can execute the mechanical equipment fault diagnosis method provided in Embodiment 1 of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0133] Example 3:

[0134] Embodiment 3 of the present invention also provides an electronic terminal, including a processor and a memory connected to the processor, wherein a computer program is stored in the memory, and the processor is used to perform operations according to the instructions to execute the steps of the method described in Embodiment 1.

[0135] The electronic terminal provided in Embodiment 3 of the present invention can execute the mechanical equipment fault diagnosis method provided in Embodiment 1 of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0136] Example 4:

[0137] Embodiment 4 of the present invention also provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the steps of the method described in Embodiment 1, and has the corresponding functional modules and beneficial effects of the method.

[0138] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied 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.

[0139] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0142] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for diagnosing mechanical equipment faults, characterized in that, include: Obtain the preprocessed encoder signal; The preprocessed encoder signal is input into a pre-constructed weighted dual-domain sparse decomposition model, and the distinguishable periodic fault impact characteristics are obtained through the objective function of the weighted dual-domain sparse decomposition model. Based on the structure and operating parameters of the monitored equipment, calculate the fault characteristic frequency when a fault occurs at different locations of the monitored equipment, and calculate the corresponding fault characteristic period based on the fault characteristic frequency. Based on the fault characteristic period, the fault location of the monitored equipment is determined by comparing it with the interval of the periodic fault impact characteristic. The objective function F1 expression of the weighted bi-domain sparse decomposition model includes: ， in, This represents the pre-processed encoder signal. Indicates harmonic components, This represents the inverse Fourier transform operator. Represents the frequency domain sparse coefficients. Indicates the impact characteristics of periodic faults. Denotes the second norm, and Both represent regularization parameters, which control the sparsity constraint strength in the frequency domain and time domain, respectively. Denotes the non-convex penalty function. As the first weighting coefficient, This is the second weighting coefficient. Index representing a periodic group, This indicates the number of elements in a periodic group. This represents the element-wise dot product operation. and Both represent penalty parameters that control the degree of non-convexity of the penalty function. This represents a periodic binary vector used to constrain the periodic group sparsity characteristics of the impact signal in the time domain. Indicates the first The first weighting coefficient of each period group Indicates the first The second weighting coefficient of each period group Indicates the first The frequency domain sparse coefficients of a periodic group Indicates the first The periodic fault impact characteristics of each periodic group include: One element; The solution method for the weighted bi-domain sparse decomposition model includes: Introducing auxiliary variables The objective function F1 of the unconstrained optimization problem is transformed into a constrained optimization problem F2, the expression of which is: ， The constrained optimization problem F2 is decomposed into the following sub-constrained optimization problems using the alternating direction multiplier method: ， ， ， ， in, For penalty parameters, For intermediate scaling variables; Solve the sub-constraint optimization problem to obtain the distinct periodic fault impact characteristics.

2. The mechanical equipment fault diagnosis method according to claim 1, characterized in that, Obtain the raw encoder signal; The original encoder signal is converted into an instantaneous angular velocity signal using a first-order differential algorithm. The instantaneous angular velocity signal is then subjected to time-domain synchronous averaging to obtain the preprocessed encoder signal.

3. The mechanical equipment fault diagnosis method according to claim 1, characterized in that, The expression for the periodic binary vector includes: ， , and The following conditions must be met: ， ， in, This represents the number of non-zero points in a single fault characteristic cycle. The number of zeros in a single fault characteristic cycle. It is the sampling frequency. Indicates the fault characteristic frequency, A periodic binary vector The number of fault characteristic cycles included.

4. The mechanical equipment fault diagnosis method according to claim 1, characterized in that, The first The expression for the first weighting coefficient of each period group includes: ， The first The expressions for the second weighting coefficient of each period group include: ， in, It is a pre-set positive parameter used to avoid the denominator being zero.

5. A mechanical equipment fault diagnosis device, characterized in that, include: The signal acquisition module is used to acquire the preprocessed encoder signal; The feature extraction module is used to input the preprocessed encoder signal into a pre-constructed weighted dual-domain sparse decomposition model, and obtain the distinguishable periodic fault impact features through the objective function of the weighted dual-domain sparse decomposition model. The fault characteristic cycle calculation module is used to calculate the fault characteristic frequency when a fault occurs at different locations of the monitored equipment based on the structure and operating parameters of the monitored equipment, and to calculate the corresponding fault characteristic cycle based on the fault characteristic frequency. The fault location confirmation module is used to determine the fault location of the monitored equipment by comparing the interval of the fault characteristic period with the interval of the periodic fault impact characteristic. The objective function F1 expression of the weighted bi-domain sparse decomposition model includes: ， in, This represents the pre-processed encoder signal. Indicates harmonic components, This represents the inverse Fourier transform operator. Represents the frequency domain sparse coefficients. Indicates the impact characteristics of periodic faults. Denotes the second norm, and Both represent regularization parameters, which control the sparsity constraint strength in the frequency domain and time domain, respectively. Denotes the non-convex penalty function. As the first weighting coefficient, This is the second weighting coefficient. Index representing a periodic group, This indicates the number of elements in a periodic group. This represents the element-wise dot product operation. and Both represent penalty parameters that control the degree of non-convexity of the penalty function. This represents a periodic binary vector used to constrain the periodic group sparsity characteristics of impact signals in the time domain. Indicates the first The first weighting coefficient of each period group Indicates the first The second weighting coefficient of each period group Indicates the first The frequency domain sparse coefficients of a periodic group Indicates the first The periodic fault impact characteristics of each periodic group include: One element; The solution method for the weighted bi-domain sparse decomposition model includes: Introducing auxiliary variables The objective function F1 of the unconstrained optimization problem is transformed into a constrained optimization problem F2, the expression of which is: ， The constrained optimization problem F2 is decomposed into the following sub-constrained optimization problems using the alternating direction multiplier method: ， ， ， ， in, For penalty parameters, For intermediate scaling variables; Solve the sub-constraint optimization problem to obtain the distinct periodic fault impact characteristics.

6. An electronic terminal, characterized in that, It includes 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, it performs the steps of the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 4.

Images