Base frequency identification method, device, computer equipment and computer readable storage medium

Abstract

The embodiments of the present application disclose a fundamental frequency recognition method, apparatus, computer equipment and computer-readable storage medium, which obtain a target vibration signal of a machine; determine a kernel function corresponding to the fundamental frequency recognition model according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, wherein the hyperparameters in the kernel function include periodic parameters corresponding to each signal component in the target vibration signal; obtain a likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal; optimize the likelihood function of the hyperparameter to obtain a recognition result of the hyperparameter; determine the fundamental frequency of the target vibration signal according to the recognition result, and extract the fundamental frequency from the target vibration signal.

Description

Technical Field

[0001] The present application relates to the field of signal processing, and in particular to a baseband identification method, device, computer equipment and computer-readable storage medium. Background Art

[0002] Machinery is very important to all walks of life, including mechanical rotating machinery and other types. Common rotating machinery includes generators, steam turbines, aircraft engines, water pumps, ventilators, etc. In order to ensure the safe operation of machinery and avoid serious safety problems caused by mechanical degradation, breakage and other failures, it is necessary to detect the operating status of the equipment. The fundamental frequency of mechanical vibration is a parameter unique to machinery, and it can itself be used as an indicator of the structural operating status. For example, the fundamental frequency and its harmonics can be used to determine whether the mechanical shaft is unbalanced, misaligned, loose, etc., so it is very important to accurately identify the fundamental frequency of the machinery. Summary of the invention

[0003] The embodiments of the present application provide a fundamental frequency identification method, apparatus, computer device, and computer-readable storage medium, which can extract the fundamental frequency from a target vibration signal.

[0004] In a first aspect, an embodiment of the present application provides a baseband identification method, comprising:

[0005] Obtain target vibration signal of machinery;

[0006] According to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, determining a kernel function corresponding to the fundamental frequency recognition model, wherein the hyperparameters in the kernel function include period parameters corresponding to each signal component in the target vibration signal;

[0007] Acquire a likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal;

[0008] Optimizing the likelihood function of the hyperparameter to obtain an identification result of the hyperparameter;

[0009] The fundamental frequency of the target vibration signal is determined according to the recognition result.

[0010] In one embodiment, the fundamental frequency recognition model includes at least one fundamental frequency signal model and an interference signal model, and determining the kernel function corresponding to the fundamental frequency recognition model according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to includes:

[0011] Determine a corresponding first kernel function for each baseband signal model, and determine a corresponding second kernel function for the interference signal model;

[0012] The kernel function is obtained by fusing the first kernel function corresponding to each baseband signal model and the second kernel function corresponding to the interference signal model.

[0013] In one embodiment, the target vibration signal includes at least one fundamental frequency signal component, and the recognition result includes a parameter value of a period parameter corresponding to each fundamental frequency signal component and a parameter value of a corresponding amplitude parameter;

[0014] Determining the fundamental frequency of the target vibration signal according to the recognition result includes:

[0015] Determining a target baseband signal component from the baseband signal components according to a parameter value corresponding to the amplitude parameter of each baseband signal component;

[0016] The fundamental frequency of the target vibration signal is determined according to the parameter value of the corresponding period parameter of the target fundamental frequency signal component.

[0017] In one embodiment, determining the fundamental frequency of the target vibration signal according to the parameter value of the corresponding period parameter of the target fundamental frequency signal component includes:

[0018] Determining a sampling frequency of the target vibration signal;

[0019] The fundamental frequency of the target vibration signal is determined according to the sampling frequency and the parameter value of the periodic parameter corresponding to the target fundamental frequency signal component.

[0020] In one embodiment, the optimizing process of the likelihood function of the hyperparameter to obtain the recognition result of the hyperparameter includes:

[0021] Calculating the partial derivative of the likelihood function with respect to the hyperparameter;

[0022] The partial derivative is subjected to gradient descent processing to update the hyperparameter so as to maximize the likelihood function and obtain an identification result of the hyperparameter.

[0023] In one embodiment, the step of obtaining a target vibration signal of a machine includes:

[0024] Acquiring an original vibration signal of the machine;

[0025] Performing intensity normalization processing on the original vibration signal to obtain a processed vibration signal;

[0026] The processed vibration signal is filtered to obtain the target vibration signal.

[0027] In one embodiment, the fundamental frequency identification model is a Gaussian process model.

[0028] In a second aspect, an embodiment of the present application further provides a baseband identification device, comprising:

[0029] A first acquisition unit, used to acquire a target vibration signal of a machine;

[0030] A determination unit, configured to determine, according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, a kernel function corresponding to the fundamental frequency recognition model, wherein the hyperparameters in the kernel function include period parameters corresponding to each signal component in the target vibration signal;

[0031] A second acquisition unit, configured to acquire a likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal;

[0032] An optimization unit, used for optimizing the likelihood function of the hyperparameter to obtain an identification result of the hyperparameter;

[0033] The recognition unit is used to determine the fundamental frequency of the target vibration signal according to the recognition result.

[0034] In a third aspect, an embodiment of the present application also provides a computer device, including a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute any baseband identification method provided in the embodiment of the present application.

[0035] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, and the computer program is loaded by a processor to execute any baseband identification method provided in the embodiment of the present application.

[0036] The embodiment of the present application obtains a target vibration signal of a machine; determines a kernel function corresponding to the fundamental frequency recognition model according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, wherein the hyperparameters in the kernel function include periodic parameters corresponding to each signal component in the target vibration signal; obtains a likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal; optimizes the likelihood function of the hyperparameter to obtain a recognition result of the hyperparameter; determines the fundamental frequency of the target vibration signal according to the recognition result, and realizes extraction of the fundamental frequency from the target vibration signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0038] Figure 1is a flow chart of a baseband identification method provided in an embodiment of the present application;

[0039] Figure 2 is a schematic diagram of a vibration signal provided in an embodiment of the present application;

[0040] Figure 3 is a spectrum diagram of a vibration signal provided in an embodiment of the present application;

[0041] Figure 4 This is a schematic diagram of the fundamental frequency recognition model training provided by an embodiment of the present application;

[0042] Figure 5 is a schematic diagram of a signal model provided in an embodiment of the present application;

[0043] Figure 6 is a spectrum diagram of the baseband signal component provided in an embodiment of the present application;

[0044] Figure 7 is a schematic diagram of a baseband identification device provided in an embodiment of the present application;

[0045] Figure 8 It is a schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0047] The embodiments of the present application provide a baseband identification method, device, computer equipment and computer readable storage medium. The baseband identification device can be integrated in a computer equipment, which can be a server or a terminal.

[0048] The terminal may include a personal computer (PC), a mobile phone, a tablet computer, a notebook computer, and a vehicle-mounted computer.

[0049] Among them, the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, as well as big data and artificial intelligence platforms.

[0050] It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.

[0051] This embodiment will be described from the perspective of a baseband identification device. The baseband identification device may be integrated into a computer device, which may be a server or a terminal.

[0052] The present application provides a method for identifying a baseband frequency, such as Figure 1 As shown, the specific process of the baseband identification method can be as follows:

[0053] 101. Obtain a target vibration signal of the machine.

[0054] The machine may include a device that can perform an execution function, and the machine may be, for example, a rotating machine. Specifically, the machine may be a generator, a steam turbine, an aircraft engine, a water pump, a ventilator, and the like.

[0055] The target vibration signal may be a signal collected by a sensor during the operation of the machine, or may be a signal obtained by preprocessing the signal collected by the sensor. That is, in one embodiment, the step of “obtaining the target vibration signal of the machine” includes:

[0056] Get the original vibration signal of the machine;

[0057] Performing intensity normalization processing on the original vibration signal to obtain a processed vibration signal;

[0058] The processed vibration signal is filtered to obtain the target vibration signal.

[0059] Among them, the original vibration signal can be a signal collected by the sensor during the operation of the machine. The original vibration signal is subjected to intensity normalization processing to adjust the intensity of the original vibration signal to a numerical range, so as to facilitate subsequent analysis and processing of the signal; and the vibration signal is filtered to remove interference signals. Specifically, the frequency range of the mechanical fundamental frequency can be determined according to the model of the machine, and the vibration signal is filtered based on the probability range to obtain the target vibration signal.

[0060] The target vibration signal can be expressed as , where the symbol represents transposition processing, m is the signal length, is the nth data point in the target vibration signal.

[0061] 102. According to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, a kernel function corresponding to the fundamental frequency recognition model is determined, and the hyperparameters in the kernel function include periodic parameters corresponding to each signal component in the target vibration signal.

[0062] The fundamental frequency recognition model may be a data distribution model, such as a Gaussian process model, etc. The fundamental frequency recognition model is a Gaussian process model.

[0063] Assume that the nth periodic signal component in the vibration signal The mean is 0 and the covariance is The Gaussian process is , the interference signal component z in the vibration signal meets the mean of 0, the covariance The Gaussian process is , assuming that the covariance function of the periodic signal component is , the covariance function of the interference signal component is Since the vibration signal contains periodic model components and interference signal components, the vibration signal conforms to the distribution , where the kernel function R is , where k is the number of periodic signal components and the hyperparameter is ,in, l Represents the length scale, which is used to control the smoothness of the signal within the period. Indicates the length of the cycle, that is, the data length of the signal in a complete cycle. represents the variance of the signal, that is, the amplitude of the signal. The number of hyperparameters is related to the number of signal components in the vibration signal. For example, if there are n periodic signal components in the vibration signal, then there are n corresponding σ and n .

[0064] According to the above process, the ability of Gaussian process to model data components can be used to model the periodic signal components and interference signal components in the target vibration signal based on the periodic characteristics of the fundamental frequency, and the kernel function of the target vibration signal is constructed based on the kernel function of the periodic signal component in the target vibration signal and the kernel function of the interference signal component. The kernel function of the target vibration signal is a pseudo-periodic function, which is composed of a square exponential kernel function and at least one periodic kernel function added together. By identifying the hyperparameters, the fundamental frequency of the target vibration signal can be obtained, and the fundamental frequency of the machine can be accurately estimated in complex environments (such as multi-frequency interference and high noise levels).

[0065] Optionally, the periodic signal component in the target vibration signal may correspond to a fundamental frequency signal model, the interference signal component may correspond to an interference signal model, the fundamental frequency recognition model may only include the fundamental frequency signal model, or include the fundamental frequency signal model and the interference signal model, and the number of interference signal models may be flexibly set according to application requirements.

[0066] In one embodiment, the number of fundamental frequency signal models in the fundamental frequency recognition model can be set according to the characteristics of the target vibration signal. For example, two fundamental frequency signal models, three fundamental frequency signal models, four fundamental frequency signal models, etc. can be set. The kernel function of each fundamental frequency signal model and the kernel function of the interference signal are mixed to obtain the kernel function of the target vibration signal. That is, in one embodiment, the fundamental frequency recognition model includes at least one fundamental frequency signal model and at least one interference signal model. The step of "determining the kernel function corresponding to the fundamental frequency recognition model according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to" may include:

[0067] Determine a corresponding first kernel function for each baseband signal model, and determine a corresponding second kernel function for each interference signal model;

[0068] The kernel function is obtained by fusing the first kernel function corresponding to each baseband signal model and the second kernel function corresponding to each interference signal model.

[0069] For example, the first kernel function corresponding to each fundamental frequency signal model and the second kernel function corresponding to each interference signal model can be determined according to the target vibration signal, and the second kernel function and at least one first kernel function are mixed to obtain a mixed kernel function, which is the kernel function of the fundamental frequency recognition model.

[0070] In one embodiment, there may be more than one baseband signal model and there may be only one interference signal model.

[0071] The second kernel function is mixed with at least one first kernel function, specifically, at least one second kernel function is added to at least one first kernel function, or the first kernel function and the second kernel function are weighted summed, or the first kernel function and the second kernel function are multiplied, etc.

[0072] 103. Obtain the likelihood function of hyperparameters based on the fundamental frequency recognition model and the target vibration signal.

[0073] In the fundamental frequency identification model, the kernel function is used to construct the covariance matrix, which can describe the correlation between observations. The fundamental frequency identification model can predict the output based on the covariance matrix and input data. The kernel function contains hyperparameters, which can control the shape and characteristics of the kernel function. Optimizing the hyperparameters can improve the prediction performance of the fundamental frequency identification model.

[0074] The likelihood function can be used to evaluate the fit of the model and to optimize hyperparameters. The likelihood function of the hyperparameters is obtained based on the fundamental frequency recognition model and the target vibration signal.

[0075] 104. Optimize the likelihood function of the hyperparameter to obtain the recognition result of the hyperparameter.

[0076] The purpose of optimizing the likelihood function is to find a set of parameter values ​​that maximizes the likelihood of the fundamental frequency recognition model on the target vibration signal. This set of parameter values ​​is the recognition result of the hyperparameters.

[0077] In one embodiment, the likelihood function may be optimized by Newton's method, quasi-Newton's method (such as BFGS algorithm), or expectation maximization (EM) algorithm, and the hyperparameters are obtained as a result.

[0078] Optionally, the likelihood function may be optimized by a gradient descent method. That is, in one embodiment, the step of "optimizing the likelihood function of the hyperparameter to obtain the recognition result of the hyperparameter" includes:

[0079] Calculate the partial derivatives of the likelihood function with respect to the hyperparameters;

[0080] The partial derivatives are subjected to gradient descent to update the hyperparameters to maximize the likelihood function and obtain the identification results of the hyperparameters.

[0081] Specifically, the partial derivatives of the likelihood function with respect to the hyperparameters can be calculated, and gradient descent processing can be performed based on the partial derivatives. The gradients can be calculated by batch gradient descent (Batch Gradient Descent), stochastic gradient descent (Stochastic Gradient Descent, SGD), or mini-batch gradient descent (Mini-Batch Gradient Descent), and the hyperparameters can be updated according to the calculated gradients to maximize the likelihood function. The iteration is continued until the error rate of the fundamental frequency recognition model is less than a preset threshold, and the recognition result of the hyperparameters is obtained.

[0082] In one example, the natural logarithm of the likelihood function may be calculated to obtain the log-likelihood function, and a set of hyperparameter parameter values ​​that maximize the log-likelihood function may be determined based on the Maximum Likelihood Estimation (MLE) to obtain the hyperparameter identification result.

[0083] Exemplarily, the log-likelihood function may be as follows:

[0084] ,

[0085] Calculate the log-likelihood function for the hyperparameters The partial derivative of :

[0086]

[0087] in is a hyperparameter θThe hyperparameters in are solved iteratively using gradient descent according to the partial derivative calculation formula.

[0088] 105. Determine the fundamental frequency of the target vibration signal according to the recognition result.

[0089] The recognition result may include the parameter value of the periodic parameter corresponding to each fundamental frequency signal component in the target vibration signal. The fundamental frequency of the target vibration signal can be determined according to the parameter value of the periodic parameter. For example, if the fundamental frequency recognition model only includes one fundamental frequency signal model, the fundamental frequency of the target vibration signal can be calculated based on the parameter value corresponding to the periodic parameter in the recognition result.

[0090] In one embodiment, the fundamental frequency recognition model has multiple fundamental frequency signal models, and the target fundamental frequency signal can be determined from the models according to the recognition result, and then the fundamental frequency can be determined, that is, the target vibration signal includes at least one fundamental frequency signal component, and the recognition result includes the parameter value of the period parameter corresponding to each fundamental frequency signal component, and the parameter value of the corresponding amplitude parameter; the step of "determining the fundamental frequency of the target vibration signal according to the recognition result" may include:

[0091] Determine a target baseband signal component from the baseband signal components according to a parameter value corresponding to the amplitude parameter of each baseband signal component;

[0092] The fundamental frequency of the target vibration signal is determined according to the parameter value of the corresponding period parameter of the target fundamental frequency signal component.

[0093] For example, the largest parameter value among the parameter values ​​corresponding to the amplitude parameters of each baseband signal component can be The corresponding fundamental frequency signal component is determined as the target fundamental frequency signal component, and the fundamental frequency of the target vibration signal is determined according to the parameter value of the corresponding period parameter of the target fundamental frequency signal component.

[0094] Optionally, the embodiments of the present application may also detect the fundamental frequencies of multiple devices simultaneously. Specifically, each fundamental frequency signal model in the fundamental frequency recognition model may correspond to the fundamental frequency signal of a device, and the fundamental frequencies of multiple devices may be determined according to the recognition results of the hyperparameters.

[0095] Specifically, the fundamental frequency can be calculated according to the sampling frequency and the parameter value of the corresponding period parameter of the target fundamental frequency signal component. That is, in one embodiment, the step of "determining the fundamental frequency of the target vibration signal according to the parameter value of the corresponding period parameter of the target fundamental frequency signal component" may include:

[0096] Determine the sampling frequency of the target vibration signal;

[0097] The fundamental frequency of the target vibration signal is determined according to the sampling frequency and the parameter value of the periodic parameter corresponding to the target fundamental frequency signal component.

[0098] The sampling frequency may be the frequency at which the sensor samples the vibration signal during mechanical operation, and the fundamental frequency of the target vibration signal is obtained by calculating the ratio of the sampling frequency to the parameter value of the periodic parameter corresponding to the target fundamental frequency signal component.

[0099] The embodiment of the present application constructs a fundamental frequency recognition model. The processing process of the fundamental frequency recognition model is all performed in the time domain. There is no need for complex spectrum conversion. The processing is simple and noise interference is taken into account. The recognition accuracy of the fundamental frequency is high.

[0100] In order to more clearly illustrate the baseband identification method provided in the embodiment of the present application, a specific example is given below for illustration. Figure 2 The vibration data shown may be a vibration signal collected by a sensing system when a motor is working. The sensing system collects the amplitude data of the motor at a fixed sampling frequency, and selects data with a length of L=81920 for fundamental frequency identification. According to the model of the motor, it can be inferred that the fundamental frequency range of the motor is 0 to 200 Hz.

[0101] The frequency spectrum of the motor vibration signal from 0 to 200 Hz is shown in the figure below: Figure 3 As shown, from Figure 3 It can be seen that the vibration signal contains multiple main frequency components. Due to the interference of noise, it is difficult to distinguish the fundamental frequency of the motor.

[0102] The vibration signal is assumed to satisfy the Gaussian process, and two fundamental frequency signal models are established based on the Gaussian process ( and ) and an interference signal model (z), and mix the kernel functions corresponding to the two fundamental frequency signal models with the kernel function of the interference signal model to obtain the mixed kernel function R of the vibration signal, and calculate the log-likelihood function.

[0103] The log-likelihood function is optimized by gradient descent to solve the hyperparameters. Figure 4 is the relationship between the number of iterations and the training error of the fundamental frequency recognition model, given by Figure 4 It can be seen that the error decreases with the increase of the number of iterations, and finally converges to 3%, and the recognition result of the hyperparameter is obtained. In the recognition result, ,because Greater than ,therefore The corresponding signal component is the fundamental frequency.

[0104] right Figure 2 The two fundamental frequency signal models and interference signal models identified by the vibration signal can be shown as Figure 5 As shown, Figure 5 The first signal component in the baseband signal model , Figure 5 The second signal component in the baseband signal model , Figure 5 The third signal component in the model is the interference signal z .

[0105] Baseband signal model The spectrum diagram is as follows Figure 6 As shown, it can be seen that the fundamental frequency identification method provided in the embodiment of the present application can decompose the interference signal component and other similar periodic signal components from the vibration signal, and can accurately extract the fundamental frequency from the vibration signal.

[0106] As can be seen from the above, the embodiment of the present application obtains the target vibration signal of the machinery; determines the kernel function corresponding to the fundamental frequency recognition model according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, and the hyperparameters in the kernel function include the periodic parameters corresponding to each signal component in the target vibration signal; obtains the likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal; optimizes the likelihood function of the hyperparameter to obtain the recognition result of the hyperparameter; determines the fundamental frequency of the target vibration signal according to the recognition result, and realizes the extraction of the fundamental frequency from the target vibration signal.

[0107] In order to better implement the baseband identification method provided in the embodiment of the present application, a baseband identification device is also provided in one embodiment. The meanings of the terms are the same as those in the above baseband identification method, and the specific implementation details can refer to the description in the method embodiment.

[0108] The baseband identification device can be integrated into a computer device, such as Figure 7 As shown, the fundamental frequency identification device may include: a first acquisition unit 301, a determination unit 302, a second acquisition unit 303, an optimization unit 304 and an identification unit 305, which are specifically as follows:

[0109] (1) A first acquisition unit 301 is used to acquire a target vibration signal of a machine.

[0110] In one embodiment, the first obtaining unit 301 may also be used for:

[0111] Obtaining the original vibration signal of the above-mentioned machinery;

[0112] Performing intensity normalization processing on the original vibration signal to obtain a processed vibration signal;

[0113] The processed vibration signal is filtered to obtain the target vibration signal.

[0114] (2) A determination unit 302 is used to determine a kernel function corresponding to the fundamental frequency recognition model according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, wherein the hyperparameters in the kernel function include period parameters corresponding to each signal component in the target vibration signal.

[0115] In one embodiment, the baseband identification model includes at least one baseband signal model and one interference signal model, and the determination unit 302 may also be used to:

[0116] Determine a corresponding first kernel function for each of the baseband signal models, and determine a corresponding second kernel function for the interference signal model;

[0117] The kernel function is obtained by fusing the first kernel function corresponding to each baseband signal model and the second kernel function corresponding to the interference signal model.

[0118] (3) A second acquisition unit 303, configured to acquire a likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal.

[0119] (4) An optimization unit 304 is used to optimize the likelihood function of the hyperparameters to obtain the recognition results of the hyperparameters.

[0120] In one embodiment, the optimization unit 304 may also be used to:

[0121] Calculate the partial derivatives of the above likelihood function with respect to the above hyperparameters;

[0122] The partial derivatives are subjected to gradient descent processing to update the hyperparameters so as to maximize the likelihood function and obtain the recognition results of the hyperparameters.

[0123] (5) An identification unit 305, configured to determine the fundamental frequency of the target vibration signal according to the identification result.

[0124] In one embodiment, the target vibration signal includes at least one fundamental frequency signal component, and the recognition result includes a parameter value of a period parameter corresponding to each fundamental frequency signal component and a parameter value of a corresponding amplitude parameter; the recognition unit 305 may also be used to:

[0125] Determine a target baseband signal component from the baseband signal components according to a parameter value corresponding to the amplitude parameter of each baseband signal component;

[0126] The fundamental frequency of the target vibration signal is determined according to the parameter value of the corresponding period parameter of the target fundamental frequency signal component.

[0127] In one embodiment, the identification unit 305 may also be used for:

[0128] Determining the sampling frequency of the target vibration signal;

[0129] The fundamental frequency of the target vibration signal is determined according to the sampling frequency and the parameter value of the periodic parameter corresponding to the target fundamental frequency signal component.

[0130] As can be seen from the above, the embodiment of the present application obtains the target vibration signal of the machine through the first acquisition unit 301; the determination unit 302 determines the kernel function corresponding to the fundamental frequency recognition model according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, and the hyperparameters in the kernel function include the periodic parameters corresponding to each signal component in the target vibration signal; the second acquisition unit 303 obtains the likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal; the optimization unit 304 optimizes the likelihood function of the hyperparameter to obtain the recognition result of the hyperparameter; the recognition unit 305 determines the fundamental frequency of the target vibration signal according to the recognition result, so as to extract the fundamental frequency from the target vibration signal.

[0131] The present application also provides a computer device, which may be a terminal or a server. Figure 8 As shown, it shows a schematic diagram of the structure of the computer device involved in the embodiment of the present application, specifically:

[0132] The computer device may include components such as a processor 1001 with one or more processing cores, a memory 1002 with one or more computer-readable storage media, a power supply 1003, and an input unit 1004. Those skilled in the art will appreciate that Figure 8 The computer device structure shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. Among them:

[0133] The processor 1001 is the control center of the computer device. It uses various interfaces and lines to connect various parts of the entire computer device. By running or executing software programs and / or modules stored in the memory 1002 and calling data stored in the memory 1002, it executes various functions of the computer device and processes data, thereby monitoring the computer device as a whole. Optionally, the processor 1001 may include one or more processing cores; preferably, the processor 1001 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and computer programs, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 1001.

[0134] The memory 1002 can be used to store software programs and modules. The processor 1001 executes various functional applications and data processing by running the software programs and modules stored in the memory 1002. The memory 1002 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, a computer program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 1002 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 1002 may also include a memory controller to provide the processor 1001 with access to the memory 1002.

[0135] The computer device also includes a power supply 1003 for supplying power to various components. Preferably, the power supply 1003 can be logically connected to the processor 1001 through a power management system, so as to manage charging, discharging, and power consumption through the power management system. The power supply 1003 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0136] The computer device may further include an input unit 1004, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0137] Although not shown, the computer device may further include a display unit, etc., which will not be described in detail herein. Specifically, in this embodiment, the processor 1001 in the computer device will load the executable files corresponding to the processes of one or more computer programs into the memory 1002 according to the following instructions, and the processor 1001 will run the computer programs stored in the memory 1002, thereby realizing various functions, as follows:

[0138] Obtain target vibration signal of machinery;

[0139] According to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, a kernel function corresponding to the fundamental frequency recognition model is determined, and the hyperparameters in the kernel function include period parameters corresponding to each signal component in the target vibration signal;

[0140] Obtaining the likelihood function of hyperparameters based on the fundamental frequency recognition model and the target vibration signal;

[0141] Optimize the likelihood function of the hyperparameter to obtain the recognition result of the hyperparameter;

[0142] The fundamental frequency of the target vibration signal is determined according to the recognition result, so as to extract the fundamental frequency from the target vibration signal.

[0143] In one embodiment, the step of “the fundamental frequency recognition model includes at least one fundamental frequency signal model and one interference signal model, and the kernel function corresponding to the fundamental frequency recognition model is determined according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to” may include:

[0144] Determine a corresponding first kernel function for each baseband signal model, and determine a corresponding second kernel function for the interference signal model;

[0145] The kernel function is obtained by fusing the first kernel function corresponding to each baseband signal model and the second kernel function corresponding to the interference signal model.

[0146] In one embodiment, the target vibration signal includes at least one fundamental frequency signal component, and the recognition result includes a parameter value of a period parameter corresponding to each fundamental frequency signal component and a parameter value of a corresponding amplitude parameter;

[0147] The step of “determining the fundamental frequency of the target vibration signal according to the recognition result” may include:

[0148] Determine a target baseband signal component from the baseband signal components according to a parameter value corresponding to the amplitude parameter of each baseband signal component;

[0149] The fundamental frequency of the target vibration signal is determined according to the parameter value of the corresponding period parameter of the target fundamental frequency signal component.

[0150] In one embodiment, the step of “determining the fundamental frequency of the target vibration signal according to the parameter value of the corresponding period parameter of the target fundamental frequency signal component” may include:

[0151] Determine the sampling frequency of the target vibration signal;

[0152] The fundamental frequency of the target vibration signal is determined according to the sampling frequency and the parameter value of the periodic parameter corresponding to the target fundamental frequency signal component.

[0153] In one embodiment, the step of “optimizing the likelihood function of the hyperparameter to obtain the recognition result of the hyperparameter” may include:

[0154] Calculate the partial derivatives of the likelihood function with respect to the hyperparameters;

[0155] The partial derivatives are subjected to gradient descent to update the hyperparameters to maximize the likelihood function and obtain the identification results of the hyperparameters.

[0156] In one embodiment, the step of “obtaining a target vibration signal of a machine” may include:

[0157] Get the original vibration signal of the machine;

[0158] Performing intensity normalization processing on the original vibration signal to obtain a processed vibration signal;

[0159] The processed vibration signal is filtered to obtain the target vibration signal.

[0160] In one embodiment, the fundamental frequency identification model is a Gaussian process model.

[0161] The specific implementation of the above operations can be found in the previous embodiments and will not be described in detail here.

[0162] As can be seen from the above, the computer device of the embodiment of the present application can obtain the target vibration signal of the machine; determine the kernel function corresponding to the fundamental frequency recognition model according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, and the hyperparameters in the kernel function include the periodic parameters corresponding to each signal component in the target vibration signal; obtain the likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal; optimize the likelihood function of the hyperparameter to obtain the recognition result of the hyperparameter; determine the fundamental frequency of the target vibration signal according to the recognition result, and realize the extraction of the fundamental frequency from the target vibration signal.

[0163] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in various optional implementations in the above-mentioned embodiments.

[0164] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by a computer program, or by controlling related hardware through a computer program. The computer program may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0165] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. The computer program can be loaded by a processor to execute any baseband identification method provided in the embodiment of the present application.

[0166] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0167] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0168] Since the computer program stored in the computer-readable storage medium can execute any of the baseband identification methods provided in the embodiments of the present application, the beneficial effects that can be achieved by any of the baseband identification methods provided in the embodiments of the present application can be achieved. For details, please refer to the previous embodiments and will not be repeated here.

[0169] The above is a detailed introduction to a baseband identification method, device, computer equipment and computer-readable storage medium provided in the embodiments of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for those skilled in the art, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as limiting the present application.

Claims

1. A baseband identification method, characterized in that: The method comprises: Obtain target vibration signal of machinery; According to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to, determining a kernel function corresponding to the fundamental frequency recognition model, wherein the hyperparameters in the kernel function include a standard deviation parameter corresponding to each signal component in the target vibration signal and a period parameter corresponding to each fundamental frequency signal component, and the parameter value of the standard deviation parameter is used to indicate the amplitude of the fundamental frequency signal component; Acquire a likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal; Optimizing the likelihood function of the hyperparameter to obtain an identification result of the hyperparameter, wherein the identification result includes a parameter value of each of the standard deviation parameters and a parameter value of each of the period parameters; Determine the baseband signal component corresponding to the largest parameter value among the parameter values ​​of the standard deviation parameters of the baseband signal components as the target baseband signal component; The fundamental frequency of the target vibration signal is determined according to the parameter value of the periodic parameter corresponding to the target fundamental frequency signal component.

2. The method according to claim 1, characterized in that The fundamental frequency recognition model includes at least one fundamental frequency signal model and at least one interference signal model, and determining a kernel function corresponding to the fundamental frequency recognition model according to the target vibration signal and the fundamental frequency recognition model that the target vibration signal conforms to includes: Determine a corresponding first kernel function for each of the baseband signal models, and determine a corresponding second kernel function for each of the interference signal models; The kernel function is obtained by fusing the first kernel function corresponding to each baseband signal model and the second kernel function corresponding to each interference signal model.

3. The method according to claim 1, characterized in that The determining the fundamental frequency of the target vibration signal according to the parameter value of the periodic parameter corresponding to the target fundamental frequency signal component comprises: Determining a sampling frequency of the target vibration signal; The fundamental frequency of the target vibration signal is determined according to the sampling frequency and the parameter value of the periodic parameter corresponding to the target fundamental frequency signal component.

4. The method according to claim 1, characterized in that The optimizing process of the likelihood function of the hyperparameter to obtain the recognition result of the hyperparameter includes: Calculating the partial derivative of the likelihood function with respect to the hyperparameter; The partial derivative is subjected to gradient descent processing to update the hyperparameter so as to maximize the likelihood function and obtain an identification result of the hyperparameter.

5. The method according to claim 1, characterized in that The step of obtaining a target vibration signal of a machine includes: Acquiring an original vibration signal of the machine; Performing intensity normalization processing on the original vibration signal to obtain a processed vibration signal; The processed vibration signal is filtered to obtain the target vibration signal.

6. The method according to any one of claims 1 to 5, characterized in that: The fundamental frequency identification model is a Gaussian process model.

7. A fundamental frequency identification device, characterized in that: include: A first acquisition unit, used to acquire a target vibration signal of a machine; a determination unit, configured to determine a kernel function corresponding to the fundamental frequency identification model according to the target vibration signal and the fundamental frequency identification model that the target vibration signal conforms to, wherein the hyperparameters in the kernel function include a standard deviation parameter corresponding to each signal component in the target vibration signal and a period parameter corresponding to each fundamental frequency signal component, and the parameter value of the standard deviation parameter is used to indicate the amplitude of the fundamental frequency signal component; A second acquisition unit, configured to acquire a likelihood function of the hyperparameter based on the fundamental frequency recognition model and the target vibration signal; An optimization unit, configured to optimize the likelihood function of the hyperparameter to obtain an identification result of the hyperparameter, wherein the identification result includes a parameter value of each of the standard deviation parameters and a parameter value of each of the period parameters; The identification unit is used to determine the fundamental frequency signal component corresponding to the largest parameter value among the parameter values ​​of the standard deviation parameters of each of the fundamental frequency signal components as the target fundamental frequency signal component; and determine the fundamental frequency of the target vibration signal according to the parameter value of the periodic parameter corresponding to the target fundamental frequency signal component.

8. A computer device, characterized in that: The invention comprises a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute the baseband identification method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and the computer program is loaded by a processor to execute the fundamental frequency identification method according to any one of claims 1 to 6.

Images