A fault detection method, device, apparatus and computer storage medium

Abstract

The application provides a fault detection method, device and equipment and a computer storage medium. The method comprises the following steps: filtering an initial vibration signal through an initial filter, determining an initial iteration gradient and an initial iteration step length according to an initial filtering variance and the initial vibration signal. Based on the initial iteration gradient and the initial iteration step length, the filter is iterated multiple times with the target of minimizing the filtering variance. The filtered filter can filter the initial vibration signal to determine a target filtering signal. The target mechanical equipment is detected for faults according to the target filtering signal. Iterating the filter with the target of minimizing the variance makes the filter accurately filter out the fault signal in the vibration signal, effectively improving the filtering accuracy of the noise signal and the detection accuracy and efficiency of the fault signal. Based on the filtered signal, the fault problem in the mechanical equipment can be more accurately and efficiently detected, providing strong safety guarantee for the normal operation of the mechanical equipment.

Description

Technical Field

[0001] This application belongs to the field of signal analysis, and in particular relates to a fault detection method, apparatus, device, and computer storage medium. Background Technology

[0002] Currently, most fault detection methods for mechanical equipment rely on fault analysis of data such as vibration signals during equipment operation. For example, deconvolution algorithms (DA) are used to accurately filter out fault signals from vibration signals during equipment operation by continuously optimizing and iterating filters.

[0003] However, most current deconvolution algorithms for filtering fault signals use kurtosis and signal periodicity as objective functions, which rely heavily on high signal-to-noise ratio (SNR) vibration signals. In low SNR vibration signals, real fault signals and noise are complexly intertwined, and traditional algorithms cannot accurately distinguish the various signal types, severely impacting detection accuracy and efficiency. Furthermore, noise signals in low SNR vibration signals may also exhibit periodicity; based on the aforementioned algorithms, periodic noise is easily misidentified as fault signals, resulting in low detection accuracy and poor robustness. Summary of the Invention

[0004] This application provides a fault detection method, apparatus, device, and computer storage medium, which can accurately and efficiently detect faults in mechanical equipment.

[0005] In a first aspect, embodiments of this application provide a fault detection method, including:

[0006] Acquire the initial vibration signal corresponding to the target mechanical equipment;

[0007] The initial vibration signal is filtered by a preset initial filter to obtain the initial filtered signal.

[0008] Based on the initial filtering variance and the initial vibration signal corresponding to the initial filtering signal, determine the initial iteration gradient and the initial iteration step size corresponding to the initial filter;

[0009] Based on the initial iteration gradient and initial iteration step size, with the goal of minimizing the filter variance, the initial filter is iterated multiple times to determine the filter after iteration. The filter variance is the filter variance corresponding to the filtered signal obtained by filtering the initial vibration signal.

[0010] The initial vibration signal is filtered by the iterative filter to obtain the target filtered signal;

[0011] Based on the target filtered signal, fault detection is performed on the target mechanical equipment.

[0012] In some embodiments, determining the initial iteration gradient and initial iteration step size corresponding to the initial filter based on the initial filter variance and the initial vibration signal corresponding to the initial filter signal includes:

[0013] Based on the initial filter variance and the initial vibration signal, the partial derivative of the initial filter variance with respect to the filter parameters corresponding to the initial filter is determined and used as the initial iteration gradient.

[0014] The initial iteration step size is determined based on the initial iteration gradient, the initial filtered signal, and the initial vibration signal.

[0015] In some embodiments, determining the initial iteration step size based on the initial iteration gradient, the initial filtered signal, and the initial vibration signal includes:

[0016] Determine the inner product of the initial iterative gradient and the initial vibration signal, and determine the difference between the initial filtered signal and the mean value corresponding to the initial filtered signal;

[0017] The initial iteration step size is determined based on the inner product and the difference.

[0018] In some embodiments, based on the initial iteration gradient and the initial iteration step size, and with the objective of minimizing the filter variance, the initial filter is iterated multiple times to determine the iterated filter, including:

[0019] Based on the initial iteration gradient and the initial iteration step size, with the goal of minimizing the filtering variance of the filtered signal obtained by filtering the initial vibration signal after the first iteration, the initial parameters of the initial filter are iterated for the first time to obtain the filter after the first iteration.

[0020] Based on the initial vibration signal, the filter variance corresponding to the filtered signal obtained by filtering the initial vibration signal with the filter after the iteration is taken as the objective. The filter after the first iteration is iterated multiple times to determine the filter after the iteration.

[0021] In some embodiments, based on the initial vibration signal, and with the objective of minimizing the filtering variance corresponding to the filtered signal obtained by filtering the initial vibration signal using the iteratively filtered filter, multiple iterations are performed on the filter after the initial iteration, including:

[0022] For each iteration, the initial vibration signal is filtered using the filter from the previous iteration to obtain the filtered signal from the previous iteration.

[0023] Based on the filtering variance of the filtered signal corresponding to the filter after the previous iteration and the initial vibration signal, determine the iteration gradient and iteration step size corresponding to the current iteration.

[0024] Based on the iteration gradient and iteration step size corresponding to this iteration, the parameters of the filter after the previous iteration are iterated with the goal of minimizing the filtering variance of the filtered signal obtained by the filter after this iteration on the initial vibration signal.

[0025] In some embodiments, the filter is iterated multiple times after the initial iteration to determine the iterated filter, including:

[0026] For each iteration, determine whether the iteration gradient corresponding to that iteration is greater than the preset iteration gradient threshold;

[0027] If so, continue iterating;

[0028] If not, based on the iteration gradient and iteration step size corresponding to this iteration, with the goal of minimizing the filtering variance of the filtered signal obtained by filtering the initial vibration signal after this iteration, the parameters of the filter after the previous iteration are iterated, and the filter after this iteration is used as the iterated filter.

[0029] In some embodiments, fault detection of target mechanical equipment is performed based on the target filtered signal, including:

[0030] Determine the envelope spectrum data corresponding to the target filtered signal;

[0031] Based on the envelope spectrum data, the fault signals of the target mechanical equipment during operation are determined.

[0032] In some embodiments, acquiring the initial vibration signal corresponding to the target mechanical equipment includes:

[0033] Vibration data generated by the target mechanical equipment at a preset acquisition interval is collected by a preset vibration sensor and used as the initial vibration signal.

[0034] Secondly, embodiments of this application provide a fault detection device, including:

[0035] The signal acquisition unit is used to acquire the initial vibration signal corresponding to the target mechanical equipment.

[0036] The initial filtering unit is used to filter the initial vibration signal through a preset initial filter to obtain the initial filtered signal.

[0037] The iterative information determination unit is used to determine the initial iteration gradient and initial iteration step size corresponding to the initial filter based on the initial filter variance and the initial vibration signal corresponding to the initial filter signal.

[0038] The iterative unit is used to perform multiple parameter iterations on the initial filter based on the initial iterative gradient and the initial iterative step size, with the goal of minimizing the filter variance, to determine the filter after iteration. The filter variance is the filter variance corresponding to the filtered signal obtained by filtering the initial vibration signal.

[0039] The filtering unit is used to filter the initial vibration signal through the iterative filter to obtain the target filtered signal;

[0040] The fault detection unit is used to detect faults in the target mechanical equipment based on the target filtered signal.

[0041] Thirdly, embodiments of this application provide a fault detection device, the device comprising:

[0042] Processor and memory storing programs or instructions;

[0043] The processor implements the above methods when executing programs or instructions.

[0044] Fourthly, embodiments of this application provide a machine-readable storage medium storing a program or instructions that, when executed by a processor, implement the method described above.

[0045] Fifthly, embodiments of this application provide a computer program product in which instructions, when executed by a processor of an electronic device, cause the electronic device to perform the above-described method.

[0046] The technical solutions provided by the embodiments of this application bring at least the following beneficial effects:

[0047] This application provides a fault detection method, apparatus, device, and computer storage medium, which can filter the acquired initial vibration signal using an initial filter. Based on the variance of the initial filtered signal and the initial vibration signal, an initial iteration gradient and an initial iteration step size are determined. Based on the initial iteration gradient and initial iteration step size, the initial filter is iterated multiple times with the goal of minimizing the filtering variance. The iterated filter can then be used to filter the initial vibration signal to determine the target filtered signal.

[0048] Finally, the target mechanical equipment can be fault detected based on the filtered target signal. Iterating the filter in a way that minimizes variance allows the filter to accurately filter out fault signals in the vibration signal of the mechanical equipment, effectively improving the filtering accuracy of noise signals and the detection efficiency of fault signals. Based on the filtered signal, faults in the mechanical equipment can be detected more accurately and efficiently, thus providing strong safety assurance for the normal operation of the mechanical equipment. Attached Figure Description

[0049] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 A flowchart illustrating a fault detection method provided in one embodiment of this application;

[0051] Figure 2(a) is one of the schematic diagrams of an initial vibration signal acquisition process provided in an embodiment of this application;

[0052] Figure 2(b) is a second schematic diagram of an initial vibration signal acquisition process provided in one embodiment of this application;

[0053] Figure 2(c) is a third schematic diagram of an initial vibration signal acquisition process provided in one embodiment of this application;

[0054] Figure 3(a) is a waveform diagram of a target filtering signal provided in an embodiment of this application;

[0055] Figure 3(b) is a waveform diagram of envelope spectrum data provided in an embodiment of this application;

[0056] Figure 4 This is a schematic diagram of the overall framework of a fault detection method provided in one embodiment of this application;

[0057] Figure 5 A schematic diagram of the structure of a fault detection device provided in another embodiment of this application;

[0058] Figure 6 This is a schematic diagram of the hardware structure of an electronic device provided in yet another embodiment of this application. Detailed Implementation

[0059] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0060] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0061] At present, most fault detection of mechanical equipment is achieved through deconvolution algorithms (DA). However, the traditional deconvolution algorithm mainly uses kurtosis and signal periodicity as objective functions. The core principle is to separate fault features from vibration signals by optimizing filters.

[0062] However, this method has significant limitations. In low signal-to-noise ratio environments, real fault signals and noise are highly intertwined, and the noise itself may have periodic characteristics. This makes it difficult for traditional algorithms to accurately distinguish between fault signals and interference noise during the operation of mechanical equipment. They are very prone to misjudging periodic noise as fault signals, resulting in poor robustness and significantly reducing the detection accuracy and efficiency for mechanical equipment faults.

[0063] Furthermore, even if there are detection methods that can effectively identify low signal-to-noise ratio and periodic noise, most of them require prior periodic data of the noise signal, which has serious limitations and greatly restricts the application effect in complex working conditions and when the prior period of noise cannot be obtained.

[0064] Based on the technical problems mentioned above, this application provides a fault detection method, apparatus, device, and computer storage medium. Specifically, the acquired initial vibration signal can be filtered using an initial filter. The initial iteration gradient and initial iteration step size are determined based on the variance of the initial filtered signal and the initial vibration signal.

[0065] Then, based on the initial iteration gradient and initial iteration step size, the initial filter is iterated multiple times with the objective of minimizing the filter variance. The iterated filter can be used to filter the initial vibration signal to determine the target filtered signal. Finally, based on the target filtered signal obtained, fault detection can be performed on the target mechanical equipment.

[0066] The technical solution provided in this application can iterate the filter in a way that minimizes variance. This effectively enables the iterated filter to accurately filter out fault signals from the vibration signals of the target mechanical equipment, thereby improving the filtering accuracy of noise signals and the detection accuracy and efficiency of fault signals. Based on the filtered signal obtained by this application's technical solution, faults in the target mechanical equipment can be detected more accurately and efficiently, thus providing strong safety assurance for the normal operation of the target mechanical equipment.

[0067] The execution entity used in the embodiments of this application can be a terminal device, such as a desktop computer or laptop computer, or a remote device, such as a server. In addition, the execution entity used in the embodiments of this application can also be a software entity, such as a client or software program installed on a terminal device. No specific limitation is made here regarding the execution entity used in applying the fault detection method, apparatus, device, and computer storage medium provided in the embodiments of this application; the entity can be flexibly selected according to the application scenario and actual needs.

[0068] It should be noted that the embodiments provided in this application do not strictly limit the specific application scenarios of the fault detection methods, devices, equipment and computer storage media mentioned above, and can be flexibly applied according to actual needs.

[0069] For example, in a scenario involving fault detection of the epicyclic gearbox in wind turbines, the technical solution provided in this application can filter the vibration signal collected by the epicyclic gearbox during the operation of the wind turbine using an initial filter. Based on the filtering variance corresponding to the initial filtered signal and the vibration signal of the epicyclic gearbox, the initial iteration gradient and the initial iteration step size are determined.

[0070] Based on the initial iteration gradient and initial iteration step size, the filter undergoes multiple rounds of parameter iteration with the objective of minimizing the filter variance. The resulting filter can then process the vibration signal from the planetary gearbox, accurately obtaining the target filtered signal. Furthermore, based on the target filtered signal, fault signals within the planetary gearbox can be accurately detected.

[0071] The technical solution provided in this application can efficiently and accurately filter out potential fault signals in the planetary gearbox of a wind turbine, precisely eliminating interference from invalid signals such as noise. Based on the detected faults, the efficiency and effectiveness of subsequent quality control and risk assessment can be effectively improved, providing strong safety assurance for the normal operation of the wind turbine.

[0072] It should be noted that the application scenarios described in the above embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will understand that with the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems. The fault detection method provided by the embodiments of this application can be applied to various application scenarios that require the detection of operational faults in mechanical equipment, such as electric motors, generators, industrial gearboxes, transmissions, centrifuges, etc.

[0073] Figure 1 This is a flowchart illustrating a fault detection method provided in one embodiment of this application.

[0074] like Figure 1 As shown, the fault detection method provided in this application embodiment includes steps S101 to S105.

[0075] S101: Obtain the initial vibration signal corresponding to the target mechanical equipment.

[0076] In step S101, the technical solution provided in this application embodiment can obtain the initial vibration signal of the target mechanical equipment during operation through a preset signal acquisition method.

[0077] The initial vibration signal can be vibration data collected from the target mechanical equipment at a preset sampling frequency within a preset sampling time. Specifically, it can be represented as x0∈R N x0 represents the initial vibration signal, R indicates that the initial vibration signal x0 is a set of real numbers, and N represents the data dimension corresponding to the initial vibration signal x0, which can be determined by calculating the product of the sampling frequency and the sampling time.

[0078] This application does not strictly limit the specific content of the signal acquisition method for obtaining the initial vibration signal in the embodiments. In one embodiment provided in this application, a vibration sensor may be added to the target mechanical equipment so that the vibration sensor can accurately collect the vibration data corresponding to the target mechanical equipment at a preset sampling frequency within the sampling time, and use it as the initial vibration signal.

[0079] To facilitate understanding of the initial vibration signal acquisition process, we will take the planetary gearbox in the wind turbine in the above example as the target mechanical device for detailed explanation. For details, please refer to Figures 2(a), 2(b), and 2(c).

[0080] Figures 2(a), 2(b), and 2(c) are schematic diagrams of an initial vibration signal acquisition process provided in an embodiment of this application.

[0081] Figure 2(a) is a schematic diagram of the wind turbine generator. In the diagram, 201 is the drive motor of the wind turbine generator, 202 is a tachometer, 203 is the planetary gearbox of the wind turbine generator, and 204 is a vibration sensor attached to the surface of the planetary gearbox 203 housing. 205 is a parallel shaft gearbox, and 206 is a brake.

[0082] The wind turbine shown in Figure 2(a) operates as follows: a drive motor 201 provides power, causing the planetary gearbox 203 and the parallel shaft gearbox 205 to move. A tachometer 202 records and feeds back the equipment's rotational speed information to rationally control the motor speed, while a brake 206 is used to apply loads. Vibration data during the wind turbine's operation can be accurately collected by a vibration sensor 204 located outside the planetary gearbox 203, thus obtaining the initial vibration signal.

[0083] Figure 2(b) is a schematic diagram of the initial vibration signal of the planetary gearbox 20 under the preset sampling conditions of a motor speed of 50Hz for the drive motor 201, a sampling frequency of 48000Hz for the vibration sensor 204, and a sampling time interval of 0.8s. Figure 2(c) is a schematic diagram of the envelope spectrum of the initial vibration signal determined based on Figure 2(b).

[0084] The technical solution provided in this application can perform subsequent filter iterations and parameter optimizations based on an initial vibration signal similar to that shown in Figure 2(b). The iterated filter can accurately identify fault signals in the initial vibration signal, improving fault detection accuracy and efficiency.

[0085] S102: The initial vibration signal is filtered by a preset initial filter to obtain an initial filtered signal.

[0086] In step S102, the technical solution provided in this application embodiment can first filter the initial vibration signal obtained in the above steps using an initial filter that has not undergone parameter optimization, and determine the initial filtered signal corresponding to the initial vibration signal.

[0087] The filter parameters corresponding to the initial filter can be f = [0, 0…1…0, 0]. T ∈R k f represents the filter parameters corresponding to the initial filter, [0,0…1…0,0] represents the specific initial parameter vector corresponding to the initial filter, T represents the transpose sign, and R kThis indicates that the initial filter parameters are a real vector with k dimensions. It is clear that the unoptimized filter parameters contain only one non-zero element. In subsequent iterations, each element in the filter parameters can be iterated sequentially, allowing the final iterated filter to accurately filter out the fault signal from the initial vibration signal.

[0088] Regarding the specific process of determining the initial filtered signal, in one embodiment provided in this application, for the initial vibration signal x0 obtained through step S101, (k-1) / 2 zero-value elements can be padded to both ends of the initial vibration signal x0. Then, the padded initial vibration signal X∈R N*k By performing matrix transformation, the Hankel matrix corresponding to the initial vibration signal x0 can be obtained. The specific Hankel matrix corresponding to the initial vibration signal x0 can be found in formula (1):

[0089]

[0090] Here, X represents the Hankel matrix corresponding to the initial vibration signal x0. It can be seen that the elements on each sub-diagonal of this matrix are equal. Matrix transformation can effectively enhance the signal characteristics in the initial vibration signal, greatly improving the accuracy and efficiency of signal feature extraction from the initial vibration signal in subsequent filter iterations and in determining the target filter signal, thus improving the filtering effect and consequently enhancing fault detection accuracy and efficiency.

[0091] Furthermore, by inputting the Hankel matrix corresponding to the determined initial vibration signal x0 into the initial filter, the initial filtered signal can be determined through the initial filter. The determination process can be referred to in formula (2):

[0092]

[0093] Where y can be used to represent the initial filtered signal, and f is the filtering parameter corresponding to the initial filter. Through the calculation process of formula (2), the initial filtered signal y corresponding to the initial vibration signal x0 can be determined.

[0094] S103: Determine the initial iteration gradient and initial iteration step size corresponding to the initial filter based on the initial filter variance and the initial vibration signal corresponding to the initial filter signal.

[0095] In step S103, the technical solution provided in this application embodiment can determine the corresponding initial filter variance based on the initial filter signal determined in the above steps. Then, based on the initial filter variance and the initial vibration signal, the initial iteration gradient and initial iteration step size corresponding to the initial filter are calculated.

[0096] The initial iterative gradient has the same meaning as the gradient in current filter-related technologies, that is, it can be used to represent the rate of change of the objective function relative to each parameter under the current filter parameters. In the embodiments of this application, the initial iterative gradient can be used to represent the rate of change of the filter variance relative to each parameter corresponding to the initial filter.

[0097] Similarly, the initial iteration step size has the same meaning as the step size / learning rate in current filter-related technologies, that is, it represents the magnitude of parameter update during the iteration process. In the embodiments of this application, the initial iteration step size can be used to represent the magnitude of parameter update of the initial filter during the first iteration.

[0098] Regarding the specific process of determining the initial iteration gradient, in one embodiment provided in this application, the partial derivatives of the initial filter variance with respect to the filter parameters corresponding to the initial filter can be accurately calculated based on the initial filter variance corresponding to the initial filter signal and the initial vibration signal.

[0099] In order to enable the filter to accurately filter out the signal with the minimum variance, in this embodiment, the partial derivative of the initial filter variance with respect to the filter parameters corresponding to the initial filter can be used as the initial iteration gradient of the initial filter.

[0100] The specific process for determining the initial iterative gradient can be found in formula (3):

[0101]

[0102] Where var represents the initial filter variance corresponding to the initial filtered signal, and y represents the initial filtered signal. This is used to represent the mean value corresponding to the initial filtered signal y. Through the calculation and derivation process of the above formula (3), the initial iteration gradient corresponding to the initial filter can be accurately calculated as:

[0103]

[0104] X T y is the transpose of the Hankel matrix corresponding to the initial vibration signal x0 determined in the above embodiments, y is the initial filtered signal, and N is the sampling data dimension of the initial vibration signal x0.

[0105] It should be noted that the process of determining the initial iterative gradient described above is almost identical to the process of determining the iterative gradient for each iteration in subsequent iterations, except that the filtered signal in the formula changes accordingly.

[0106] The reason for choosing the filter variance corresponding to the filtered signal as the objective function in the filter iteration process is that the fault signals generated by mechanical equipment during operation are usually regular impact signals. For example, a faulty gear produces a tiny vibration every revolution. The periodicity makes the distribution of the fault signal more concentrated, and the signal tail decays faster, similar to a uniform distribution, forming a "thin-tailed" sub-Gaussian property.

[0107] The tail probability distribution of a sub-Gaussian signal is lower than that of a normal Gaussian signal. Therefore, fault signals are more similar to sub-Gaussian signals, while noise signals are mostly Gaussian signals. The unique probability distribution characteristics of sub-Gaussian signals cause data points in similar signals to tend to be near the signal mean, resulting in a smaller overall variance. Therefore, it can be concluded that fault signals resembling sub-Gaussian signals in mechanical equipment vibration signals also have lower variance values.

[0108] Based on the above reasons, in the embodiments provided in this application, the vibration signal during the operation of the target mechanical equipment can be filtered with the goal of minimizing the variance, so as to filter out the sub-Gaussian signal with a smaller variance, and the fault signal in the vibration signal can be accurately determined.

[0109] By determining the filter iteration gradient using variance as the objective function, Gaussian-distributed noise signals in vibration signals can be accurately filtered out without being affected by periodicity. This allows for the precise filtering of fault signals even in vibration signals with low signal-to-noise ratios. This effectively improves the efficiency and accuracy of fault detection in mechanical equipment, providing strong safety assurance for the normal operation of such equipment.

[0110] The above describes the process and principle for determining the initial iteration gradient and the corresponding iteration gradients for subsequent rounds. Regarding the specific process for determining the initial iteration step size corresponding to the initial filter, in one embodiment provided in this application, the initial iteration step size for the first iteration of the initial filter can be calculated based on the inner product between the determined initial iteration gradient and the previously obtained initial vibration signal, and the difference between the initial filtered signal and the signal mean corresponding to the initial filtered signal.

[0111] Similarly, in subsequent iterations, the calculation method for the iteration step size in each iteration can be exactly the same as the calculation method for the initial iteration step size.

[0112] Specifically, the process of determining the initial iteration step size will be described in detail below based on the formula derivation. In order to maximize the iteration rate of the filter, it should be ensured that the filter parameters can be optimized in the direction of the fastest decrease of the filter variance during each iteration. Therefore, it can be concluded that when the filter performs parameter iteration in each round, the optimal step size should make the rate of change of the objective function (i.e., the filter variance in this embodiment) in the direction of the step size zero, that is, the partial derivative of the filter variance with respect to the step size in this embodiment is zero, as shown in formula (4):

[0113]

[0114] It should be noted that, for the first iteration of the initial filter, var in formula (4) does not represent the initial filter variance corresponding to the initial filtered signal, but rather the filter variance corresponding to the filtered signal after the filter processes the initial vibration signal after the first iteration. α represents the initial iteration step size corresponding to the first iteration process.

[0115] The parameter optimization formula for each iteration of the filter in this embodiment can be found in formula (5):

[0116]

[0117] in, Here, f represents the filter parameters after each iteration, and f represents the filter parameters before the iteration. Let α be the iteration gradient for each iteration, and α be the iteration step size. The filter parameters can be iterated based on formula (5) for both the initial and subsequent iterations.

[0118] Assume that the filtered signal obtained by the filter after the first iteration of the initial vibration signal is: Based on the above formulas, the following formula (6) and its corresponding derivation process can be calculated:

[0119]

[0120] Based on formula (6) and the above formula (4), formula (7) can be derived:

[0121]

[0122] Specifically, regarding the initial iteration process, It can represent the inner product value corresponding to the initial vibration signal and the initial iterative gradient, while This can represent the signal mean corresponding to the initial filtered signal. Based on formula (7), the following formulas (8) and (9) are defined:

[0123]

[0124] Based on formulas (8) and (9), formula (7) can be further derived, calculated and simplified to obtain the following formula (10):

[0125]

[0126] Based on formula (10), the expression formula corresponding to the initial iteration step size can be determined, as shown in formula (11):

[0127]

[0128] In the initial iteration process of the initial filter, α is the initial iteration step size, W is the arithmetic matrix calculated from the inner product value corresponding to the initial vibration signal and the initial iteration gradient as shown in the above formula (8), and r is the difference matrix between the initial filtered signal and the signal mean corresponding to the initial filtered signal as shown in formula (9).

[0129] It can be seen that the initial iteration step size calculated by formula (11) is a negative value. Therefore, in the filter parameter optimization function shown in formula (5), the filter parameters after the first iteration are... The filter parameters f before iteration and the initial iteration gradient The sum of the product of the initial iteration step size α and the initial iteration step size α. Since the initial iteration step size is negative, the parameter iteration in formula (5) ensures that the filter parameters can iterate rapidly in the direction of minimizing variance, whether it is the first iteration or each subsequent iteration.

[0130] Furthermore, in this embodiment, the steepest descent step size of each iteration can be determined by the steepest descent method, so that the objective function, specifically the filter variance, can converge quickly, which greatly accelerates the iteration rate of the filter and, to a certain extent, significantly improves the efficiency and accuracy of subsequent fault detection for mechanical equipment.

[0131] Based on the above formula (11), the iteration step size corresponding to each iteration of the filter can be calculated. The above embodiment is only an example for easy understanding, taking the first iteration as an example.

[0132] S104: Based on the initial iteration gradient and initial iteration step size, with the goal of minimizing the filter variance, perform multiple parameter iterations on the initial filter to determine the filter after iteration.

[0133] In step 104, the technical solution provided in this application embodiment can, based on the above-described process of determining the initial iteration gradient and initial iteration step size, perform multiple rounds of filter parameter iteration on the initial filter with the goal of minimizing the filtering variance corresponding to the filtered signal after each round of iteration on the initial vibration signal. The filter after multiple rounds of iteration can be used for subsequent fault detection of the target mechanical equipment.

[0134] In one embodiment provided in this application, the initial iteration gradient and the initial iteration step size are determined according to the above steps, and the parameters of the filter after the first iteration can be accurately calculated by formula (5) in the above embodiment.

[0135] Then, the initial vibration signal is filtered using the filter obtained after the first iteration to obtain the filtered signal for the second iteration. The iteration gradient and iteration step size for the second iteration are then calculated using formulas (3) and (11) in the above embodiment, and the parameters are adjusted and optimized using formula (5). This process is repeated to obtain the filter after iteration.

[0136] Specifically, in one embodiment, for each iteration process, the technical solution provided in this application can determine the iteration gradient and iteration step size corresponding to the iteration process of that iteration based on the filtered signal after the filter with optimized parameters of the previous iteration is used to filter the initial vibration signal.

[0137] Then, based on the iteration gradient and iteration step size corresponding to the iteration process, with the goal of minimizing the filtering variance of the filter signal obtained by filtering the initial vibration signal after the iteration, the parameters of the filter corresponding to the previous iteration are optimized and adjusted. The specific parameter optimization process can be referred to the above formula (5).

[0138] After multiple parameter iterations, the filter can be formally used in subsequent fault detection of the target mechanical equipment. It can accurately filter out Gaussian-distributed noise signals with high variance in the initial vibration signal, thereby accurately retaining the fault signal in the initial vibration signal. By iterating the filter parameters using the steepest descent method, the iteration rate and signal recognition capability of the filter are effectively improved, greatly enhancing the detection efficiency and accuracy of the subsequent fault signal detection process.

[0139] The specific criteria for determining the completion of the iteration of the filter during the aforementioned multi-round iteration process are not strictly limited in the embodiments of this application. For example, in one embodiment provided in this application, a stopping criterion can be set for the iteration gradient corresponding to each round of iteration to obtain a filter with fully iterated parameters.

[0140] Specifically, for each round of parameter iteration, before optimizing the filter parameters, it can be determined whether the iteration gradient corresponding to the iteration round determined by the above formula (3) is greater than the preset iteration gradient threshold.

[0141] When the iteration gradient corresponding to the current iteration round is greater than the preset iteration gradient threshold, it indicates that the gradient of the current iteration round has not decreased to the ideal state, and the filtering effect of the filter on the low variance signal in the initial vibration signal still needs to be optimized. In this case, the filter after the previous iteration is iterated again according to the iteration gradient corresponding to the current iteration round and the iteration step size determined by formula (11). After this iteration, the iteration gradient and iteration step size corresponding to the next round are determined.

[0142] When the iteration gradient corresponding to the current iteration round is determined to be no greater than the preset iteration gradient threshold, it indicates that the gradient of the current iteration round has met the stopping criteria. After this iteration, the relationship between the filter variance and the filter parameters of the filtered signal after filtering the initial vibration signal is no longer significant, and the filter variance after this iteration can almost approach the vibration signal with the smallest variance in the initial vibration signal, thus the objective function of the filter reaches a minimum value.

[0143] In this case, the filter after the previous iteration is iterated again according to the iteration gradient and iteration step size corresponding to the iteration round. The filter after this iteration can be used as the filter after the overall multi-round parameter iteration process.

[0144] By using the above process of determining the completed filter based on the iterative gradient, the iteration efficiency of the filter can be effectively improved, avoiding too many useless iterations and greatly saving iteration time and resource consumption. Simultaneously, it can ensure accurate convergence of the iterated filter, accurately identifying fault signals with small variance from the initial vibration signal, thereby effectively improving the detection efficiency and accuracy of subsequent fault detection processes.

[0145] Besides determining the completion of iteration based on the iterative gradient and a preset threshold as described above, other methods can also be used to determine the completion of the filter iteration. For example, in some other embodiments provided in this application, the completion of the filter iteration can also be determined by setting a maximum number of iterations or by determining the convergence of the filter variance based on the filtered signal. The specific method can be flexibly adjusted according to the application scenario and actual needs.

[0146] S105: The initial vibration signal is filtered by the iterative filter to obtain the target filtered signal.

[0147] In step S105, the technical solution provided in this application embodiment can use the filter obtained by the iteration of steps S101 to S104 above to filter the initial vibration signal again, thereby accurately filtering out the target filtered signal that can represent the fault signal in the initial vibration signal.

[0148] The specific filtering process is almost identical to the process described above, where the corresponding filtering signal is determined in each iteration, as shown in formula (2) in the above embodiment. The difference is that it is not necessary to calculate the iteration gradient and iteration step size again. The filtering process can be directly referred to the corresponding content and formula above, so it will not be elaborated further here.

[0149] S106: Perform fault detection on the target mechanical equipment based on the target filtered signal.

[0150] In step S106, the technical solution provided in this application embodiment can identify fault problems of the target mechanical equipment based on the target filtered signal determined in step S105, thereby accurately determining the fault problems that may exist in the target mechanical equipment within the sampling interval.

[0151] In one embodiment provided in this application, the envelope spectrum data corresponding to the target filtered signal can be determined based on the determined target filtered signal.

[0152] Among them, envelope spectrum data can represent the characteristic frequency, signal energy distribution, and non-stationary signal distribution of the fault signal in the target filtered signal.

[0153] The specific method for determining the envelope spectrum data corresponding to the target filtered signal is not strictly limited in this application. For example, in one embodiment provided in this application, the envelope signal corresponding to the target filtered signal can be determined first. Then, the envelope signal is frequency domain transformed, which can be achieved through methods such as Fast Fourier Transform (FFT), to finally obtain the envelope spectrum data corresponding to the target filtered signal. Of course, other feasible methods for determining the envelope spectrum data are also possible, and can be flexibly selected according to the application scenario and signal type.

[0154] Furthermore, based on the envelope spectrum data corresponding to the target filtered signal, operational faults in the target mechanical equipment can be accurately detected. The specific detection method for the fault is not strictly limited in this application embodiment. In one embodiment, different types of fault detection can be performed based on a preset fault signal frequency threshold, or other feasible fault detection methods can be used, flexibly selected according to the application scenario and the actual signal type.

[0155] To facilitate understanding of the specific representation of the target filtered signal and its envelope spectrum data, a schematic diagram of a target filtered signal and its corresponding envelope spectrum data will be used for illustration, based on Figure 2(b) shown in the above embodiment. See Figures 3(a) and 3(b) for details.

[0156] Figure 3(a) is a waveform diagram of a target filtering signal provided in an embodiment of this application.

[0157] Figure 3(b) is a waveform diagram of envelope spectrum data provided in an embodiment of this application.

[0158] As shown in Figures 3(a) and 3(b), the parameter iteration through the above steps enables the filter to accurately filter the target filtered signal shown in Figure 3(a) from the initial vibration signal shown in Figure 2(b). Then, based on the target filtered signal shown in Figure 3(a), the envelope spectrum data shown in Figure 3(b) can be determined. In the subsequent fault detection process, the potential faults in the target mechanical equipment during operation can be accurately detected based on the envelope spectrum data.

[0159] The above fault detection process can accurately detect faults in target mechanical equipment within the sampling time based on the target filtered signal determined in the above steps. By using a filter obtained through iteration with the goal of minimizing the filter variance, the target filtered signal can accurately represent the faults in the initial vibration signal, effectively improving the detection accuracy and efficiency of the fault detection process, avoiding false or missed faults, and thus ensuring the normal operation and safety of the target mechanical equipment.

[0160] The above describes the specific operation details of each step in the fault detection method provided in this application embodiment. To facilitate understanding of the overall implementation flow of the fault detection method, a schematic diagram of the overall framework of the fault detection method is provided below for reference. Figure 4 As shown in the image.

[0161] Figure 4 This is a schematic diagram of the overall framework of a fault detection method provided in one embodiment of the present application, including steps S401 to S408.

[0162] S401: Obtain the initial vibration signal corresponding to the envelope spectrum data.

[0163] Step S401 allows for the precise acquisition of vibration data of the target mechanical equipment within the sampling time interval using methods such as vibration sensors.

[0164] S402: The initial vibration signal is filtered by an initial filter to determine the initial filtered signal.

[0165] S403: Determine the iterative gradient.

[0166] S404: Determine the iteration step size.

[0167] Through steps S402 to S404, the initial vibration signal can be filtered by a filter. Based on the determined filtering variance of the filtered signal and the initial vibration signal, the iterative gradient (e.g., the above formula (3)) and the gradient step size (e.g., the above formula (11)) can be determined respectively.

[0168] S405: Optimize filter parameters based on the iterative gradient and iteration step size.

[0169] Step S405 optimizes the filter parameters in each iteration process based on the iteration gradient and iteration step size, with the goal of minimizing the filter variance, effectively improving the filter's ability to identify fault signals.

[0170] S406: Determine whether the iteration is complete based on the iteration gradient.

[0171] The judgment process in step S406 accurately determines when to stop the iteration, effectively improving the iteration rate and the filter's ability to detect fault signals. When the iteration gradient does not meet the preset threshold, parameter iteration is repeated, calculating the iteration gradient and iteration step size to ensure that the filter parameters after iteration can achieve precise convergence by minimizing variance.

[0172] S407: The initial vibration signal is filtered by the iterative filter to determine the target filtered signal.

[0173] S408: Based on the envelope spectrum data corresponding to the target filtered signal, perform fault detection on the target mechanical equipment.

[0174] Through steps S407 to S408, the target filtered signal corresponding to the initial vibration signal can be accurately determined through iterative filtering. Based on the envelope spectrum data corresponding to the target filtered signal, fault detection is performed on the target mechanical equipment. This accurately identifies the fault signal in the initial vibration signal, improves fault detection efficiency and accuracy, and provides strong safety assurance for the normal operation of the target mechanical equipment.

[0175] The above describes the specific implementation of the fault detection method provided in this application. The technical solution provided in this application can iterate the filter in a way that minimizes the variance, which effectively enables the iterated filter to accurately filter out fault signals from the vibration signals of the target mechanical equipment, thereby improving the filtering accuracy of noise signals and the detection accuracy and efficiency of fault signals.

[0176] This application can effectively solve the problem that current technology cannot accurately identify faults in vibration signals with low signal-to-noise ratio. Regardless of whether the vibration signal has a high or low signal-to-noise ratio, the filter obtained by iteratively based on the technical solution of this application can accurately filter the target filtered signal that can accurately represent the fault of the target mechanical equipment, effectively improving the applicability of the overall detection process and its adaptability to different types of vibration signals, and enhancing the robustness of the detection process.

[0177] The filtered signal obtained based on the technical solution of this application can detect faults in mechanical equipment more accurately and efficiently, thereby providing a strong safety guarantee for the normal operation of mechanical equipment.

[0178] Based on the fault detection method provided in the above embodiments, this application also provides an embodiment of a fault detection device.

[0179] Figure 5 This is a schematic diagram of the structure of a fault detection device provided in another embodiment of this application. The fault detection device 500 includes:

[0180] The signal acquisition unit 501 is used to acquire the initial vibration signal corresponding to the target mechanical equipment;

[0181] The initial filtering unit 502 is used to filter the initial vibration signal through a preset initial filter to obtain an initial filtered signal;

[0182] The iterative information determination unit 503 is used to determine the initial iteration gradient and initial iteration step size corresponding to the initial filter based on the initial filtering variance and the initial vibration signal corresponding to the initial filtering signal.

[0183] The iteration unit 504 is used to perform multiple parameter iterations on the initial filter based on the initial iteration gradient and the initial iteration step size, with the goal of minimizing the filter variance, to determine the filter after iteration. The filter variance is the filter variance corresponding to the filter signal obtained by filtering the initial vibration signal.

[0184] The filtering unit 505 is used to filter the initial vibration signal through the iterated filter to obtain the target filtered signal;

[0185] The fault detection unit 506 is used to detect faults in the target mechanical equipment based on the target filtered signal.

[0186] In some embodiments, the iteration information determining unit 503 is specifically used for:

[0187] Based on the initial filter variance and the initial vibration signal, the partial derivative of the initial filter variance with respect to the filter parameters corresponding to the initial filter is determined and used as the initial iteration gradient.

[0188] The initial iteration step size is determined based on the initial iteration gradient, the initial filtered signal, and the initial vibration signal.

[0189] In some embodiments, the iteration information determining unit 503 is specifically used for:

[0190] Determine the inner product of the initial iterative gradient and the initial vibration signal, and determine the difference between the initial filtered signal and the mean value corresponding to the initial filtered signal;

[0191] The initial iteration step size is determined based on the inner product and the difference.

[0192] In some embodiments, the iteration unit 504 is specifically used for:

[0193] Based on the initial iteration gradient and the initial iteration step size, with the goal of minimizing the filtering variance of the filtered signal obtained by filtering the initial vibration signal after the first iteration, the initial parameters of the initial filter are iterated for the first time to obtain the filter after the first iteration.

[0194] Based on the initial vibration signal, the filter variance corresponding to the filtered signal obtained by filtering the initial vibration signal with the filter after the iteration is taken as the objective. The filter after the first iteration is iterated multiple times to determine the filter after the iteration.

[0195] In some embodiments, the iteration unit 504 is specifically used for:

[0196] For each iteration, the initial vibration signal is filtered using the filter from the previous iteration to obtain the filtered signal from the previous iteration.

[0197] Based on the filtering variance of the filtered signal corresponding to the filter after the previous iteration and the initial vibration signal, determine the iteration gradient and iteration step size corresponding to the current iteration.

[0198] Based on the iteration gradient and iteration step size corresponding to this iteration, the parameters of the filter after the previous iteration are iterated with the goal of minimizing the filtering variance of the filtered signal obtained by the filter after this iteration on the initial vibration signal.

[0199] In some embodiments, the iteration unit 504 is specifically used for:

[0200] For each iteration, determine whether the iteration gradient corresponding to that iteration is greater than the preset iteration gradient threshold;

[0201] If so, continue iterating;

[0202] If not, based on the iteration gradient and iteration step size corresponding to this iteration, with the goal of minimizing the filtering variance of the filtered signal obtained by filtering the initial vibration signal after this iteration, the parameters of the filter after the previous iteration are iterated, and the filter after this iteration is used as the iterated filter.

[0203] In some embodiments, the fault detection unit 506 is specifically used for:

[0204] Determine the envelope spectrum data corresponding to the target filtered signal;

[0205] Based on the envelope spectrum data, the fault signals of the target mechanical equipment during operation are determined.

[0206] In some embodiments, the signal acquisition unit 501 is specifically used for:

[0207] Vibration data generated by the target mechanical equipment at a preset acquisition interval is collected by a preset vibration sensor and used as the initial vibration signal.

[0208] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. They are devices corresponding to the above-mentioned fault detection method. All implementation methods in the above-mentioned method embodiments are applicable to the embodiments of this device. For details on its specific functions and the technical effects it brings, please refer to the method embodiment section. It will not be repeated here.

[0209] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0210] Figure 6 This is a schematic diagram of the hardware structure of an electronic device provided in yet another embodiment of this application.

[0211] The device may include a processor 601 and a memory 602 storing programs or instructions.

[0212] When the processor 601 executes the program, it implements the steps in any of the above method embodiments.

[0213] For example, the program can be divided into one or more modules / units, one or more of which are stored in memory 602 and executed by processor 601 to complete this application. The one or more modules / units can be a series of program instruction segments capable of performing a specific function, which describe the execution process of the program in the device.

[0214] Specifically, the processor 601 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0215] Memory 602 may include mass storage for data or instructions. For example, and not limitingly, memory 602 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 602 may include removable or non-removable (or fixed) media. Where appropriate, memory 602 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 602 is non-volatile solid-state memory.

[0216] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) machine-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the methods according to one aspect of this disclosure.

[0217] The processor 601 implements any of the methods described above by reading and executing programs or instructions stored in the memory 602.

[0218] In one example, the electronic device may also include a communication interface 603 and a bus 604. The processor 601, memory 602, and communication interface 603 are connected via the bus 604 and communicate with each other.

[0219] The communication interface 603 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0220] Bus 604 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 604 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0221] Furthermore, in conjunction with the methods in the above embodiments, this application embodiment can provide a machine-readable storage medium for implementation. This machine-readable storage medium stores a program or instructions; when executed by a processor, the program or instructions implement any of the methods in the above embodiments. This machine-readable storage medium can be read by a machine such as a computer.

[0222] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above method embodiments and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0223] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0224] This application provides a computer program product stored in a machine-readable storage medium. The program product is executed by at least one processor to implement the various processes of the above method embodiments and achieve the same technical effects. To avoid repetition, it will not be described again here.

[0225] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0226] The functional modules shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer grids such as the Internet, intranets, etc.

[0227] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0228] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by a computer program or instructions. These programs or instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0229] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A fault detection method, characterized in that, include: Acquire the initial vibration signal corresponding to the target mechanical equipment; The initial vibration signal is filtered by a preset initial filter to obtain an initial filtered signal; Based on the initial filtering variance corresponding to the initial filtering signal and the initial vibration signal, determine the initial iteration gradient and initial iteration step size corresponding to the initial filter; Based on the initial iteration gradient and the initial iteration step size, with the goal of minimizing the filter variance, the initial filter is iterated multiple times to determine the filter after iteration. The filter variance is the filter variance corresponding to the filtered signal obtained by filtering the initial vibration signal. The initial vibration signal is filtered by the iterative filter to obtain the target filtered signal. Based on the target filtered signal, fault detection is performed on the target mechanical equipment; The step of determining the initial iteration gradient and initial iteration step size corresponding to the initial filter based on the initial filter variance corresponding to the initial filter signal and the initial vibration signal includes: Based on the initial filter variance and the initial vibration signal, the partial derivative of the initial filter variance with respect to the filter parameters corresponding to the initial filter is determined and used as the initial iteration gradient; Determine the inner product of the initial iterative gradient and the initial vibration signal, and determine the difference between the initial filtered signal and the mean of the initial filtered signal; The initial iteration step size is determined based on the inner product value and the difference value.

2. The method according to claim 1, characterized in that, Based on the initial iteration gradient and the initial iteration step size, with the objective of minimizing the filter variance, the initial filter undergoes multiple parameter iterations to determine the iteratively refined filter, including: Based on the initial iteration gradient and the initial iteration step size, with the goal of minimizing the filtering variance corresponding to the filtered signal obtained by filtering the initial vibration signal after the first iteration, the initial parameters of the initial filter are iterated for the first time to obtain the filter after the first iteration. Based on the initial vibration signal, with the goal of minimizing the filtering variance corresponding to the filtered signal obtained by filtering the initial vibration signal with the iterative filter, the filter after the first iteration is iterated multiple times to determine the iterative filter.

3. The method according to claim 2, characterized in that, Based on the initial vibration signal, with the objective of minimizing the filtering variance corresponding to the filtered signal obtained by filtering the initial vibration signal using the iterative filter, multiple iterations are performed on the filter after the initial iteration, including: For each iteration, the initial vibration signal is filtered using the filter from the previous iteration to obtain the filtered signal corresponding to the filter from the previous iteration. Based on the filtering variance of the filter signal corresponding to the filter after the previous iteration and the initial vibration signal, determine the iteration gradient and iteration step size corresponding to the current iteration. Based on the iteration gradient and iteration step size corresponding to this iteration, the parameters of the filter after the previous iteration are iterated with the goal of minimizing the filtering variance of the filtered signal obtained by filtering the initial vibration signal after this iteration.

4. The method according to claim 2, characterized in that, The filter after the initial iteration is iterated multiple times to determine the iterated filter, including: For each iteration, determine whether the iteration gradient corresponding to that iteration is greater than the preset iteration gradient threshold; If so, continue iterating; If not, based on the iteration gradient and iteration step size corresponding to this iteration, with the goal of minimizing the filtering variance of the filtered signal obtained by filtering the initial vibration signal after this iteration, the parameters of the filter after the previous iteration are iterated, and the filter after this iteration is used as the iterated filter.

5. The method according to claim 1, characterized in that, Based on the target filtered signal, fault detection is performed on the target mechanical equipment, including: Determine the envelope spectrum data corresponding to the target filtered signal; Based on the envelope spectrum data, the fault signal of the target mechanical equipment during operation is determined.

6. The method according to claim 1, characterized in that, Acquire the initial vibration signal corresponding to the target mechanical equipment, including: The vibration data generated by the target mechanical equipment at a preset acquisition interval is collected by a preset vibration sensor and used as the initial vibration signal.

7. An electronic device, characterized in that, The device includes: a processor and a memory storing programs or instructions; When the processor executes the program or instructions, it implements the method as described in any one of claims 1-6.

8. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores a program or instructions that, when executed by a processor, implement the method as described in any one of claims 1-6.

9. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device causes the electronic device to perform the method as described in any one of claims 1-6.

Images