Adaptive Fault Diagnosis Method for Rotating Machinery

Abstract

The present invention relates to the technical field of signal detection, and specifically provides an adaptive fault diagnosis method for rotating machinery. The method comprises: obtaining a vibration signal of a rotating machinery, preprocessing the vibration signal, and obtaining a signal time-frequency feature; calculating an instantaneous time-frequency estimate based on an adaptive window function; calculating an instantaneous frequency based on the signal time-frequency feature and the instantaneous time-frequency estimate; smoothing and optimizing the instantaneous frequency to obtain a time-frequency estimate; reconstructing the vibration signal based on the time-frequency estimate to obtain a reconstructed signal, and determining a fault diagnosis result of the rotating machinery based on the reconstructed signal. This is to solve the problem of low accuracy in fault analysis of the vibration signal of a rotating machinery using a synchronous extraction transformation method in the related art.

Description

Technical Field

[0001] The invention relates to the technical field of signal detection, and in particular to an adaptive fault diagnosis method for rotating machinery. Background Art

[0002] Rotating machinery is widely used in many fields such as automobile manufacturing, aerospace, and household appliances. Especially in industrial production systems, for connection and transmission, it is essential to use rotating machinery equipment containing bearings, gears (gearboxes), and shaft parts. At the same time, these devices are in complex working conditions such as variable loads and speeds for a long time, and are prone to failure. Because the vibration signals of rotating machinery under complex working conditions show non-stationary or even nonlinear behavior, and the signals are cross-coupled, it becomes extremely difficult to analyze their faults.

[0003] The Synchro Extracting Transform (SET) method can retain the most relevant time-frequency coefficients of the signal's time-frequency characteristics, remove most of the scattered energy, and improve the readability of the signal's time-frequency diagram. However, due to the low time-frequency resolution of the original constant window width, the Synchro Extracting Transform method can only provide a constant time-frequency resolution in the entire time-frequency domain, which is insufficient to characterize non-stationary signals with large fluctuations in the instantaneous frequency characteristic range, and cannot accurately characterize the time-frequency characteristics of the actual vibration signal of the rotating machinery.

[0004] With regard to the problem of low accuracy in fault analysis of vibration signals of rotating machinery using synchronous extraction and transformation methods in related technologies, no effective solution has been proposed so far. Summary of the invention

[0005] An adaptive fault diagnosis method for a rotating machine provided by an embodiment of the present invention at least solves the problem of low accuracy in fault analysis of vibration signals of a rotating machine using a synchronous extraction transformation method in the related art.

[0006] An adaptive fault diagnosis method for a rotating machinery provided by an embodiment of the present invention comprises: acquiring a vibration signal of the rotating machinery, preprocessing the vibration signal to obtain a signal time-frequency characteristic; calculating an instantaneous time-frequency estimation value based on an adaptive window function; calculating an instantaneous frequency based on the signal time-frequency characteristic and the instantaneous time-frequency estimation value; smoothing and optimizing the instantaneous frequency to obtain a time-frequency estimation value; reconstructing the vibration signal based on the time-frequency estimation value to obtain a reconstructed signal, and determining a fault diagnosis result of the rotating machinery based on the reconstructed signal.

[0007] The adaptive fault diagnosis method of rotating machinery provided by the embodiment of the present invention obtains the vibration signal of the rotating machinery, pre-processes the vibration signal, and obtains the time-frequency characteristics of the signal, including: sampling the vibration signal based on a preset threshold to obtain a signal to be processed; performing a short-time Fourier transform on the signal to be processed based on a Gaussian function to obtain the time-frequency characteristics of the signal.

[0008] The adaptive fault diagnosis method for rotating machinery provided by the embodiment of the present invention calculates instantaneous time-frequency estimation values ​​based on an adaptive window function, including: taking the normalized angle function as the intermediate frequency slope of the original window function at the corresponding moment, and constructing an adaptive window function in combination with the time scale of the time-frequency distribution; multiplying the adaptive window function with the original window function to obtain a new window function; and calculating the instantaneous time-frequency estimation values ​​based on the new window function.

[0009] The adaptive fault diagnosis method for rotating machinery provided by the embodiment of the present invention calculates the instantaneous frequency based on the time-frequency characteristics of the signal and the instantaneous time-frequency estimation value, including: calculating the instantaneous frequency estimation value in the time dimension based on the new window function corresponding to each sampling point, the time corresponding to each sampling point, and the instantaneous time-frequency estimation value; calculating the instantaneous frequency estimation value in the frequency dimension based on the first-order derivative of the new window function corresponding to each sampling point and the instantaneous time-frequency estimation value; calculating the instantaneous frequency based on the instantaneous frequency estimation value in the time dimension and the instantaneous frequency estimation value in the frequency dimension.

[0010] The adaptive fault diagnosis method for rotating machinery provided by the embodiment of the present invention calculates the instantaneous frequency based on the instantaneous frequency estimation value in the time dimension and the instantaneous frequency estimation value in the frequency dimension, including: taking the instantaneous frequency estimation value that meets the preset threshold within the preset time as the element of the two-dimensional matrix; calculating the instantaneous frequency based on the two-dimensional matrix.

[0011] The adaptive fault diagnosis method for rotating machinery provided by the embodiment of the present invention performs smoothing optimization on the instantaneous frequency to obtain a time-frequency estimation value, including: performing Euclidean norm calculation on the product of the second-order difference operator and the instantaneous frequency mean to obtain a first norm value; performing Euclidean norm calculation on the difference between the instantaneous frequency mean and the instantaneous frequency to obtain a second norm value; and calculating the time-frequency estimation value based on the square of the first norm value, the square of the second norm value, and a smoothing parameter.

[0012] The adaptive fault diagnosis method for rotating machinery provided by the embodiment of the present invention reconstructs the vibration signal based on the time-frequency estimation value to obtain the reconstructed signal, including: constructing an expression of the association between the reconstructed signal and time based on the amplitude of the signal's time-frequency characteristics, the first-order derivative of the phase function of the signal's time-frequency characteristics, and the time-frequency estimation value.

[0013] The adaptive fault diagnosis method of rotating machinery provided by the embodiment of the present invention determines the fault diagnosis result of the rotating machinery based on the reconstructed signal, including: drawing a time-frequency waveform diagram of the reconstructed signal based on an expression; performing feature extraction on the time-frequency waveform diagram to obtain an envelope diagram of the reconstructed signal; and determining the fault location based on the corresponding frequency of the peak value in the envelope diagram.

[0014] An electronic device provided by an embodiment of the present invention includes: a processor and a memory storing a program, wherein the program includes instructions, and when the instructions are executed by the processor, the processor executes any of the above methods.

[0015] The present invention provides a computer program product according to an embodiment of the present invention, including a computer program / instruction. When the computer program / instruction is executed by a processor, any of the above methods is implemented.

[0016] The invention provides an adaptive fault diagnosis method for rotating machinery, which calculates the instantaneous time-frequency estimation value based on the adaptive window function, and can be used to characterize the non-stationary time-varying signal of the rotating machinery; the instantaneous frequency is smoothly optimized to enhance the noise resistance of the processing result; and the fault condition of the rotating machinery can be accurately detected based on the reconstructed signal. This solves the problem of low accuracy in fault analysis of the vibration signal of the rotating machinery using the synchronous extraction transformation method in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other embodiments can be obtained based on these drawings without creative work.

[0018] Figure 1 It is a flow chart of the steps of an adaptive fault diagnosis method for rotating machinery in an embodiment of the present invention.

[0019] Figure 2 It is a schematic diagram of a vibration signal used for testing in an embodiment of the invention.

[0020] Figure 3 It is based on the synchronous extraction transformation method Figure 2 The time-frequency diagram is obtained by processing the vibration signal shown.

[0021] Figure 4 The method provided by the embodiment of the present invention is based on Figure 2 The time-frequency diagram is obtained by processing the vibration signal shown.

[0022] Figure 5 is based on Figure 2 Waveform diagram of the reconstructed signal obtained from the vibration signal shown.

[0023] Figure 6 yes Figure 5 Envelope plot of the reconstructed signal shown.

[0024] Figure 7 It is a schematic diagram of the structure of an electronic device in an embodiment of the invention. DETAILED DESCRIPTION

[0025] The embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0026] The vibration signals of rotating machinery under complex working conditions exhibit non-stationary or even nonlinear behavior, and the signals are cross-coupled, making fault analysis extremely difficult.

[0027] The synchronous extraction transformation method can retain the most relevant time-frequency coefficients of the signal time-frequency characteristics, remove most of the scattered energy, and improve the readability of the signal time-frequency diagram. However, due to the low time-frequency resolution of the original constant window width, the synchronous extraction transformation method can only provide a constant time-frequency resolution in the entire time-frequency domain, which is not enough to characterize non-stationary signals with large fluctuations in the instantaneous frequency characteristic range, and cannot accurately characterize the time-frequency characteristics of the actual vibration signal of the rotating machinery.

[0028] For example, when analyzing vibration signals during the startup or speed change process of rotating machinery, since the frequency components of the signal will change significantly in a short period of time, a constant time-frequency resolution is difficult to accurately capture these rapidly changing frequency characteristics, resulting in low fault detection accuracy.

[0029] To this end, the present invention provides an adaptive fault diagnosis method for a rotating machine. Figure 1 As shown, the above method includes:

[0030] Step S101, obtaining a vibration signal of a rotating machine, preprocessing the vibration signal, and obtaining a signal time-frequency feature.

[0031] Step S102: Calculate the instantaneous time-frequency estimation value based on the adaptive window function.

[0032] Step S103, calculating the instantaneous frequency based on the signal time-frequency characteristics and the instantaneous time-frequency estimation value.

[0033] Step S104, smoothing and optimizing the instantaneous frequency to obtain a time-frequency estimation value.

[0034] Step S105 , reconstructing the vibration signal based on the time-frequency estimation value to obtain a reconstructed signal, and determining a fault diagnosis result of the rotating machinery based on the reconstructed signal.

[0035] It is understandable that it is necessary to install appropriate sensors to obtain corresponding vibration signals according to the type of rotating machinery and the components to be detected.

[0036] For example, the displacement sensor is installed at the shaft end of the gearbox, the speed sensor is installed near the output shaft or input shaft of the gearbox, or the acceleration sensor is installed on the bearing seat of the gearbox. Among them, the gearbox is a device that realizes the conversion and transmission of speed and torque through gear meshing. Its basic structure includes gear components, shafts and bearings, and a box body.

[0037] The signal time-frequency characteristics combine the time domain characteristics and frequency domain characteristics, including the frequency composition and energy distribution of the signal at different times.

[0038] The adaptive window function means that the window function used can adapt to the change of shaft speed, so as to capture the vibration signal characteristics of the rotating machinery at different speeds.

[0039] By combining the adaptive window function and the original window function, the non-stationary signal within a period of time can be converted into a stationary signal with a regular speed change within the entire time span, thereby realizing the characterization of the non-stationary time-varying signal, thereby solving the problem of fuzzy instantaneous frequency characteristics when processing sudden change signals using the synchronous extraction transformation method and other time-frequency analysis methods based on an invariant window.

[0040] The fault diagnosis results of the rotating machinery include but are not limited to determining the specific faulty component.

[0041] The above method provided by the embodiment of the invention calculates the instantaneous time-frequency estimation value based on the adaptive window function, which can be used to characterize the non-stationary time-varying signal of the rotating machinery; the instantaneous frequency is smoothly optimized to enhance the noise resistance of the processing result; and the fault condition of the rotating machinery can be accurately detected based on the reconstructed signal.

[0042] It can be understood that the above method provided by the embodiment of the invention is an optimization based on the synchronous extraction transformation method, and has strong versatility. Whether it is a small gearbox or a large shaft system equipment, it can automatically adjust parameters according to specific operating conditions and signal characteristics, realize effective fault detection and diagnosis, and provide technical support for the reliable operation of rotating machinery.

[0043] Preferably, in step S101, a vibration signal of a rotating machine is obtained, and the vibration signal is preprocessed to obtain a signal time-frequency feature, including:

[0044] Step S1011, sampling the vibration signal based on a preset threshold to obtain a signal to be processed.

[0045] It is understandable that the preset threshold is the signal length, which is used for preliminary screening or judgment of the signal, and can be set by those skilled in the art based on experience and the requirements of the equipment used. The vibration signal is sampled based on the preset threshold to facilitate subsequent short-time Fourier transform.

[0046] Optionally, in order to further improve the detection accuracy, operations such as noise reduction and trend item removal are performed on the sampled signal to obtain a signal to be processed.

[0047] Step S1012, performing short-time Fourier transform on the signal to be processed based on the Gaussian function to obtain the signal time-frequency characteristics.

[0048] In digital signal processing, the Gaussian function with symmetry and monotonicity can act as a filter and can effectively suppress the noise that follows the normal distribution.

[0049] The short-time Fourier transform divides the signal into many small segments (time windows) by windowing, and then performs a Fourier transform on each small segment to obtain the frequency components of the signal in different time segments. It can not only obtain the frequency information of the signal, but also retain the time information, overcoming the deficiency of the traditional Fourier transform that it can only obtain the overall frequency information of the signal but cannot reflect the change of frequency over time.

[0050] The signal time-frequency characteristics are obtained by the following formula :

[0051] ;

[0052] Wherein, s(t) represents the signal to be processed, g(t-τ) represents the original window function based on the Gaussian function, τ represents the time variable, ω represents the instantaneous frequency, and i represents the imaginary term.

[0053] It can be understood that the original window function g(t-τ) relates time resolution and frequency resolution.

[0054] A narrower original window function g(t-τ) can provide better time resolution because it can more accurately capture the changes in the signal over a short period of time; but it will lead to a decrease in frequency resolution because the frequency component estimation is not accurate enough when Fourier analysis is performed on the signal over a shorter time interval.

[0055] The wider original window function g(t-τ) can provide better frequency resolution, but poorer time resolution because it captures the signal for a longer average time and cannot well reflect the rapid changes of the signal in a short time.

[0056] In practical applications, those skilled in the art may select the width of the original window function g(t-τ) in combination with prior knowledge, analysis purpose, and detection method.

[0057] Preferably, in step S102, calculating the instantaneous time-frequency estimation value based on the adaptive window function includes:

[0058] Step S1021, taking the normalized angle function as the intermediate frequency slope of the original window function at the corresponding moment, and combining the time scale of the time-frequency distribution to construct an adaptive window function.

[0059] The meshing frequency of rotating machinery is proportional to the shaft speed, and there are clear time-frequency components in the signal that are directly related to the instantaneous shaft speed. In order to reduce the signal smearing problem in the instantaneous frequency estimation, the fundamental frequency estimate in the original window function needs to be proportional to the rotation frequency of the real signal component.

[0060] Considering that the rotation frequency is difficult to obtain directly under actual complex working conditions, the normalized angle function is used to replace the ratio between the fundamental frequency estimate and the rotation frequency. In addition, the normalized angle function has a great influence on the time-frequency representation of the analyzed signal. The closer the angle of the windowed signal is to the real signal, the more concentrated the energy is. Therefore, the normalized angle function α(μ) is used as the intermediate frequency slope of the original window function g(t-τ) at time t=μ. At this time, the selected angle is the same as the real angle of the truncated signal, and the energy concentration reaches the highest level.

[0061] The adaptive window function is obtained by :

[0062] ;

[0063] Where T represents the time scale of the time-frequency distribution.

[0064] Step S1022: Adaptive window function Multiply it with the original window function g(t-τ) to get the new window function P * V (τ).

[0065] Step S1023, based on the new window function P * V (τ) Calculate the instantaneous time-frequency estimate , specifically:

[0066] ;

[0067] It can be understood that based on the new window function P * V (τ) can transform a non-stationary signal within a period of time into a stationary signal with a rotation speed variation law within the entire time span, thereby realizing the characterization of non-stationary time-varying signals and improving detection accuracy.

[0068] Furthermore, the normalized angle function α(μ) corresponds to the moment with the most concentrated energy in each time slice, and the selected angle should correspond to the maximum kurtosis. Therefore, the optimal parameters can be determined by using a dynamic parameter selection scheme based on kurtosis, which is:

[0069] ;

[0070] ;

[0071] In the formula, Indicates frequency span, superscript 4 indicates fourth power, and superscript 2 indicates second power.

[0072] It can be understood that by calculating the kurtosis of the instantaneous frequency distribution at each time node, the optimal combination of the normalized angle and the corresponding window function at each time node can be obtained, thereby obtaining the optimal adaptive window function. , and then get the best instantaneous time-frequency estimate Among them, it is necessary to satisfy ω(u)|tanα(u)| <T。

[0073] Preferably, in step S103, calculating the instantaneous frequency based on the signal time-frequency characteristics and the instantaneous time-frequency estimation value includes:

[0074] Step S1031, based on the new window function corresponding to each sampling point, the instantaneous time-frequency estimation value corresponding to each sampling point , calculate the instantaneous frequency estimate in the time dimension , specifically:

[0075] ;

[0076] Since the actual vibration signal is usually discrete data, N is the number of sampling points, and the sample n=0, 1, ..., N-1, then The expression is:

[0077] ;

[0078] In the formula, represents Fourier transform, represents the signal to be processed s(t) after initialization, represents element-wise multiplication, Indicates the specific time corresponding to n, Represents the window function P centered on n * V (τ).

[0079] Based on the first derivative of the new window function corresponding to each sampling point , instantaneous time-frequency estimate, calculates the instantaneous frequency estimate in the frequency dimension , specifically:

[0080] ;

[0081] ;

[0082] Step S1032, calculating the instantaneous frequency based on the instantaneous frequency estimation value in the time dimension and the instantaneous frequency estimation value in the frequency dimension, includes:

[0083] The instantaneous frequency estimation values ​​that meet the preset threshold within the preset time are used as elements of the two-dimensional matrix, and the instantaneous frequency is calculated based on the two-dimensional matrix.

[0084] ;

[0085] In the formula, represents the real part; k = 0, 1, ..., N-1, k and n represent a two-dimensional matrix The kth row and nth column of the two-dimensional matrix It can be understood as a time-frequency coefficient; , , It represents a threshold parameter and can be set by those skilled in the art based on empirical values.

[0086] Preferably, in step S104, the instantaneous frequency is smoothly optimized to obtain a time-frequency estimation value, including:

[0087] The Euclidean norm of the product of the second-order difference operator and the instantaneous frequency mean is calculated to obtain the first norm value .

[0088] The Euclidean norm of the difference between the instantaneous frequency mean and the instantaneous frequency is calculated to obtain the second norm value .

[0089] Computes a time-frequency estimate based on the square of the first norm value, the square of the second norm value, and a smoothing parameter.

[0090] Specifically, the time-frequency estimation value is calculated by the following formula:

[0091] ;

[0092] In the formula, Indicates the calculation of the Euclidean norm (2-norm), Represents time-frequency coefficients The instantaneous frequency is calculated, Instantaneous frequency The mean of represents the second-order difference operator, Represents the smoothing parameter, which is used to control the smoothness of the instantaneous frequency.

[0093] It can be understood that by controlling the smoothing parameter The size of the time-frequency graph can adaptively adjust the smoothness of the instantaneous frequency according to the characteristics of the signal, while reducing noise interference and retaining the frequency change information related to the fault in the signal, thereby improving the readability of the time-frequency graph and making the fault characteristics more clearly displayed.

[0094] For example, the second-order difference operator Specifically:

[0095] ;

[0096] Preferably, in step S105, reconstructing the vibration signal based on the time-frequency estimation value to obtain a reconstructed signal includes:

[0097] Based on the amplitude A of the signal's time-frequency characteristics and the phase function of the signal's time-frequency characteristics The first derivative of , and the time-frequency estimates , construct the reconstructed signal Expressions associated with time, specifically:

[0098] ;

[0099] ;

[0100] It can be understood that the above-mentioned reconstructed signal is a multi-component signal, which includes multiple frequency components related to the operating status of various components of the rotating machinery.

[0101] Preferably, in step S105, determining the fault diagnosis result of the rotating machinery based on the reconstructed signal includes:

[0102] Plot the reconstructed signal based on the above expression Time-frequency waveform diagram.

[0103] The feature extraction is performed on the time-frequency waveform to obtain the envelope diagram of the reconstructed signal.

[0104] The fault location is determined based on the corresponding frequency of the peak in the envelope diagram.

[0105] It can be understood that the fault frequency can be determined based on the peak value appearing in the envelope diagram, and each rotating part of the rotating machinery has a corresponding operating frequency, and the rotating part corresponding to the operating frequency and the fault frequency has a fault. Therefore, the above method provided by the embodiment of the invention can accurately identify the fault type, help to achieve predictive maintenance, thereby reducing maintenance costs and improving equipment reliability and production efficiency.

[0106] Exemplarily, the present invention also provides test data of the above method to verify its technical effect, which is as follows:

[0107] Take the high-speed shaft of the gearbox in the rotating machinery and the gear pair meshing with the intermediate shaft as an example. The vibration signal is as follows: Figure 2 As shown, the gear rotation frequency of the high-speed shaft is 5 Hz and the gear meshing frequency of the intermediate shaft is 20 Hz.

[0108] The time-frequency diagram obtained by processing the above vibration signal based on the synchronous extraction transformation method is as follows Figure 3 As shown, the time-frequency diagram obtained by processing the above vibration signal according to the method provided in steps S101-S104 is as follows Figure 4 Compared to Figure 3 , Figure 4 The frequency characteristics in the image are clearer, the energy concentration of the instantaneous frequency ridge is improved, it has better readability, and can effectively reveal the changes of multiple signal components.

[0109] The waveform of the reconstructed signal obtained based on the above vibration signal is as follows: Figure 5 As shown, the black waveform is the vibration signal, and the red waveform is the reconstructed signal. It can be seen that the reconstructed signal removes most of the noise components, indicating that step S104 can indeed enhance the noise resistance of the processing result.

[0110] The envelope diagram of the above reconstructed signal is as follows Figure 6 As shown, the envelope spectrum peaks appear at 5Hz and 20Hz, corresponding to the gear rotation frequency of the high-speed shaft and the gear meshing frequency of the intermediate shaft, respectively, indicating that a fault has occurred in the meshing gear pair of the high-speed shaft and the intermediate shaft of the gearbox in the rotating machinery. Therefore, the above method provided in the embodiment of the present invention can accurately characterize the fault condition of the rotating machinery.

[0111] The present invention also provides a computer program product, including a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to execute the method of the present invention.

[0112] The present invention also provides a non-transitory machine-readable medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to execute the method of the present invention.

[0113] The present invention also provides an electronic device, comprising: at least one processor; and a memory connected to the at least one processor. The memory stores a computer program executable by the at least one processor, and the computer program is used to enable the electronic device to perform the method of the present invention when executed by the at least one processor.

[0114] refer to Figure 7 , a block diagram of an electronic device that can be used as a server or client of an embodiment of the invention will now be described, which is an example of hardware devices that can be applied to various aspects of the invention. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the invention described and / or required herein.

[0115] like Figure 7 As shown, the electronic device includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 to a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the electronic device can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0116] Multiple components in the electronic device are connected to the I / O interface 705, including: an input unit 706, an output unit 707, a storage unit 708, and a communication unit 709. The input unit 706 can be any type of device that can input information to the electronic device, and the input unit 706 can receive input digital or character information, and generate key signal input related to user settings and / or function control of the electronic device. The output unit 707 can be any type of device that can present information, and can include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 708 can include but is not limited to a disk, an optical disk. The communication unit 709 allows the electronic device to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and can include but is not limited to a modem, a network card, an infrared communication device, and / or a wireless communication transceiver, such as a Bluetooth device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0117] The computing unit 701 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a CPU, a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing units, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 701 performs the various methods and processes described above. For example, in some embodiments, the method embodiments created by the present invention may be implemented as a computer program, which is tangibly contained in a machine-readable medium, such as a storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on an electronic device via a ROM 702 and / or a communication unit 709. In some embodiments, the computing unit 701 may be configured to perform the above-described method in any other appropriate manner (e.g., by means of firmware).

[0118] The computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the computer program, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0119] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable signal medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, or infrared systems, devices, or equipment, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0120] It should be noted that the term "including" and its variations used in the embodiments of the present invention are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". The modifications of "one" and "multiple" mentioned in the embodiments of the present invention are illustrative and not restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, they should be understood as "one or more". The descriptions of the terms "first", "second", etc. are for descriptive purposes only and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated.

[0121] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the embodiments of the present invention are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0122] The various steps described in the method implementation methods provided by the embodiments of the present invention can be performed in different orders and / or in parallel. In addition, the method implementation methods may include additional steps and / or omit the steps shown. The scope of protection of the present invention is not limited in this respect.

[0123] The term "embodiment" in this specification refers to specific features, structures or characteristics described in conjunction with the embodiment that can be included in at least one embodiment of the invention. The appearance of this phrase in various places in the specification does not necessarily mean the same embodiment, nor does it mean that it is mutually exclusive with other embodiments and is independent or optional. The various embodiments in this specification are described in a related manner, and the same and similar parts between the various embodiments refer to each other. In particular, for the device, equipment, and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts refer to the partial description of the method embodiment.

[0124] The above-described embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of protection. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the attached claims.

Claims

1. An adaptive fault diagnosis method for rotating machinery, characterized in that: include: Acquire a vibration signal of a rotating machine, preprocess the vibration signal, and obtain a signal time-frequency characteristic; The normalized angle function is used as the intermediate frequency slope of the original window function at the corresponding moment, and the adaptive window function is constructed in combination with the time scale of the time-frequency distribution; Multiplying the adaptive window function by the original window function to obtain a new window function; Calculating an instantaneous time-frequency estimate based on the new window function; Calculating the instantaneous frequency based on the signal time-frequency characteristics and the instantaneous time-frequency estimation value; Smoothing and optimizing the instantaneous frequency to obtain a time-frequency estimation value; The vibration signal is reconstructed based on the time-frequency estimation value to obtain a reconstructed signal, and a fault diagnosis result of the rotating machinery is determined based on the reconstructed signal.

2. The method according to claim 1, characterized in that: Acquire a vibration signal of a rotating machine, preprocess the vibration signal, and obtain a signal time-frequency feature, including: Sampling the vibration signal based on a preset threshold to obtain a signal to be processed; Performing short-time Fourier transform on the signal to be processed based on a Gaussian function to obtain the time-frequency characteristics of the signal.

3. The method according to claim 1, characterized in that: Calculating the instantaneous frequency based on the signal time-frequency characteristics and the instantaneous time-frequency estimation value, comprising: Based on the new window function corresponding to each sampling point, the time corresponding to each sampling point, and the instantaneous time-frequency estimation value, calculate the instantaneous frequency estimation value in the time dimension; Calculate the instantaneous frequency estimation value in the frequency dimension based on the first-order derivative of the new window function corresponding to each sampling point and the instantaneous time-frequency estimation value; The instantaneous frequency is calculated based on the instantaneous frequency estimation value in the time dimension and the instantaneous frequency estimation value in the frequency dimension.

4. The method according to claim 3, characterized in that: Calculating the instantaneous frequency based on the instantaneous frequency estimation value in the time dimension and the instantaneous frequency estimation value in the frequency dimension includes: Taking the instantaneous frequency estimation values ​​that meet the preset threshold within the preset time as elements of the two-dimensional matrix; The instantaneous frequency is calculated based on the two-dimensional matrix.

5. The method according to claim 1, characterized in that Smoothing and optimizing the instantaneous frequency to obtain a time-frequency estimation value includes: The Euclidean norm of the product of the second-order difference operator and the instantaneous frequency mean is calculated to obtain the first norm value; Performing Euclidean norm calculation on the difference between the instantaneous frequency mean and the instantaneous frequency to obtain a second norm value; The time-frequency estimation value is calculated based on the square of the first norm value, the square of the second norm value, and a smoothing parameter.

6. The method according to claim 1, characterized in that Reconstructing the vibration signal based on the time-frequency estimation value to obtain a reconstructed signal includes: Based on the amplitude of the signal time-frequency feature, the first-order derivative of the phase function of the signal time-frequency feature, and the time-frequency estimation value, an expression of the reconstructed signal associated with time is constructed.

7. The method according to claim 6, characterized in that Determining a fault diagnosis result of the rotating machinery based on the reconstructed signal includes: Draw a time-frequency waveform diagram of the reconstructed signal based on the expression; Performing feature extraction on the time-frequency waveform diagram to obtain an envelope diagram of the reconstructed signal; The fault location is determined based on the corresponding frequency of the peak in the envelope diagram.

8. An electronic device comprising: A processor and a memory storing a program, wherein the program comprises instructions, which, when executed by the processor, cause the processor to perform the method according to any one of claims 1 to 7.

9. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images