Wind turbine generator main shaft bearing vibration abnormity identification method and device
By denoising the vibration signal of the spindle bearing of the wind turbine unit, local mean decomposition and zero-cross feature calculation methods, the problem that a single sensor is difficult to identify abnormal vibration is solved, and more accurate vibration abnormality detection is achieved.
Patent Information
- Application Number
- CN202411611371.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-12
- Publication Date
- 2025-02-18
AI Technical Summary
In the prior art, a single sensor is difficult to accurately capture and analyze the abnormal vibration characteristics of the spindle bearing of the wind turbine.
By obtaining the original vibration signal, local mean decomposition is performed after denoising, the instantaneous amplitude and frequency are extracted, and the zero-crossing characteristics are calculated to identify the vibration abnormality of the spindle bearing.
It improves the accuracy of detection of vibration abnormalities of the spindle bearing of the wind turbine unit, which helps to take timely maintenance measures to ensure the stable operation and long-term reliability of the wind power system.
Smart Images
Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of vibration anomaly detection, and in particular to a method and device for identifying vibration anomalies of a main shaft bearing of a wind turbine generator set. Background Art
[0002] As an important form of renewable energy, wind power generation has been widely used and developed around the world. With the increase in the scale and number of wind turbines, effectively identifying and predicting the abnormal operation of core components of wind turbines has become a key challenge to improve the operating efficiency and reliability of wind turbines.
[0003] As one of the core components, the main shaft bearing of a wind turbine carries the rotor load of the wind turbine and converts wind energy into mechanical energy, and then realizes the energy conversion between mechanical energy and electrical energy through the generator. The main shaft bearing of a wind turbine mainly bears the heavy pressure from wind and mechanical loads, and is prone to problems such as bearing wear, bearing heating, reduced unit output power, and safety accidents. Traditional vibration monitoring technology usually relies on vibration sensors to monitor the vibration signals of the main shaft bearings in real time to detect and diagnose possible abnormal conditions. However, due to the complexity and challenges of the operating environment of wind turbines, data from a single sensor is often difficult to accurately capture and analyze the characteristics of abnormal vibrations. Summary of the invention
[0004] The embodiments of the present application provide a method and device for identifying abnormal vibration of a main shaft bearing of a wind turbine generator set, thereby solving the technical problem in the prior art that data from a single sensor is often difficult to accurately capture and analyze the characteristics of abnormal vibration.
[0005] In a first aspect, an embodiment of the present application provides a method for identifying abnormal vibration of a main shaft bearing of a wind turbine generator set, the method comprising: obtaining an original vibration signal; wherein the original vibration signal comprises the effective value of the vibration acceleration of the main shaft bearing in the vertical direction or the effective value of the vibration acceleration in the horizontal direction; denoising the original vibration signal to obtain a denoised signal; performing local mean decomposition on the denoised signal to obtain multiple local amplitude components; extracting instantaneous amplitude and instantaneous frequency from the multiple local amplitude components, and performing zero-crossing feature calculation on the instantaneous amplitude and the instantaneous frequency to obtain the frequency of the denoised signal crossing the zero point from positive to negative in the time domain; determining the out-of-limit state of the zero-crossing feature of the vibration signal of the main shaft bearing based on the relationship between the zero-crossing frequency and the frequency feature distribution range, and identifying the abnormal vibration phenomenon of the main shaft bearing.
[0006] In combination with the first aspect, in a possible implementation manner, denoising the original vibration signal to obtain the denoised signal includes: using fast wavelet transform to remove noise in the original vibration signal to obtain the denoised signal.
[0007] In combination with the first aspect, in a possible implementation, the use of fast wavelet transform to remove noise in the original vibration signal includes: performing multi-layer wavelet decomposition on the original vibration signal to obtain multiple wavelet packet nodes and multiple wavelet coefficients corresponding to the multiple wavelet packet nodes; adaptively selecting the multiple wavelet packet nodes; performing non-local mean filtering on the wavelet coefficients corresponding to the wavelet packet nodes retained after adaptive selection to form adjusted wavelet coefficients; and reconstructing the vibration signal according to the adjusted wavelet coefficients to form the denoised signal.
[0008] In combination with the first aspect, in a possible implementation, the adaptive selection of the multiple wavelet packet nodes includes: calculating the energy concentration of each of the wavelet packet nodes; removing the wavelet packet nodes whose energy concentration is less than a preset selection threshold, and retaining the wavelet packet nodes whose energy concentration reaches the preset selection threshold.
[0009] In combination with the first aspect, in a possible implementation, the performing local mean decomposition on the denoised signal to obtain multiple local amplitude components includes: iteratively separating the local amplitude components from the original vibration signal until the original vibration signal is represented as the sum of the multiple local amplitude components and a monotonic function; separating the local amplitude components from the original vibration signal includes: iteratively performing a first iteration step; wherein the first iteration step includes: using a time scale to control the window size of extracting the local mean, calculating the local mean and local envelope value at each time point; determining the amplitude at each time point and the local mean. residual, and applying the frequency analysis method to the residual to extract signal components in different frequency ranges; when the energy of the residual reaches a preset termination condition, terminating the first iteration step; processing the local mean and local envelope value with a sliding average function to obtain a local mean function and a local envelope function; iteratively executing the second iteration step until a pure frequency modulation function is obtained; wherein the second iteration step includes: demodulating the original vibration signal using the local mean function; multiplying the multiple envelope signals generated to obtain an envelope signal of the local amplitude component; multiplying the envelope signal with the pure frequency modulation function to obtain the local amplitude component.
[0010] In combination with the first aspect, in a possible implementation manner, the calculating of zero-crossing characteristics of the instantaneous amplitude and the instantaneous frequency includes: dividing each of the local amplitude components into frequency segments; and calculating the zero-crossing rate of each of the frequency segments.
[0011] In a second aspect, an embodiment of the present application provides a device for identifying abnormal vibration of a main shaft bearing of a wind turbine generator set, the device comprising: an acquisition module for acquiring an original vibration signal; wherein the original vibration signal comprises the effective value of the vibration acceleration of the main shaft bearing in the vertical direction or the effective value of the vibration acceleration in the horizontal direction; a denoising module for denoising the original vibration signal to obtain a denoised signal; a decomposition module for performing local mean decomposition on the denoised signal to obtain a plurality of local amplitude components; an extraction module for extracting an instantaneous amplitude and an instantaneous frequency from the plurality of local amplitude components; a zero crossing rate module for performing zero crossing feature calculation on the instantaneous amplitude and the instantaneous frequency to obtain the frequency of the denoised signal crossing the zero point from the positive direction to the negative direction in the time domain; and an identification module for determining the out-of-limit state of the zero crossing feature of the vibration signal of the main shaft bearing based on the relationship between the zero crossing frequency and the frequency feature distribution range, and identifying the abnormal vibration phenomenon of the main shaft bearing.
[0012] In a third aspect, an embodiment of the present application provides an electronic device, comprising: one or more processors; and a memory storing computer executable instructions, wherein the computer executable instructions, when executed by the one or more processors, implement the method for identifying abnormal vibration of a main shaft bearing of a wind turbine set as described in the first aspect or any possible implementation method of the first aspect.
[0013] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the method for identifying abnormal vibration of a main shaft bearing of a wind turbine set as described in the first aspect or any possible implementation method of the first aspect is implemented.
[0014] The technical solution provided in the embodiments of the present application has at least the following technical effects: The embodiment of the present application provides a method for identifying abnormal vibration of the main shaft bearing of a wind turbine. The method performs local mean decomposition after denoising the original vibration signal to obtain multiple local amplitude components, extracts the instantaneous amplitude and instantaneous frequency from the local amplitude components, calculates the zero-crossing characteristics of the instantaneous amplitude and instantaneous frequency, obtains the frequency of the denoised signal crossing the zero point from the positive direction to the negative direction in the time domain, determines the over-limit state of the zero-crossing characteristic of the vibration signal of the main shaft bearing based on the relationship between the zero-crossing frequency and the frequency characteristic distribution range, and identifies the abnormal vibration phenomenon of the main shaft bearing. The method does not rely on the detection value of a single sensor to directly identify the abnormality, but processes the vibration signal and determines whether the abnormal vibration phenomenon occurs through the zero-crossing rate, which makes the detection of the abnormal vibration of the main shaft bearing more accurate, helps to take maintenance measures at the right time, and guarantees the stable operation and long-term reliability of the wind power generation system to the greatest extent. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] 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 of the present application will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0016] Figure 1 A flow chart of a method for identifying abnormal vibration of a main shaft bearing of a wind turbine provided in an embodiment of the present application; Figure 2 A schematic diagram of a device for identifying abnormal vibration of a main shaft bearing of a wind turbine provided in an embodiment of the present application; Figure 3 A schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0017] 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 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 ordinary technicians in this field without creative work are within the scope of protection of this application.
[0018] The present application embodiment provides a method for identifying abnormal vibration of a main shaft bearing of a wind turbine generator set, such as Figure 1 As shown, the method includes steps S101 to S106.
[0019] S101, obtaining an original vibration signal, wherein the original vibration signal includes an effective value of the vibration acceleration of the main shaft bearing in the vertical direction or an effective value of the vibration acceleration in the horizontal direction.
[0020] For example, a data acquisition and monitoring control system (SCADA) is installed in the wind turbine, and the effective value of the vertical vibration acceleration and the effective value of the horizontal vibration acceleration of the main shaft bearing are obtained from the SCADA system.
[0021] S102: De-noise the original vibration signal to obtain a de-noised signal.
[0022] Due to external air flow, ground vibration and other reasons, the original vibration signal obtained may contain noise. After executing step S102, the noise can be removed to obtain a denoised signal, which can more truly reflect the vibration of the main shaft bearing. For example, when executing step S102, low-pass filter denoising, band-stop filter denoising, wavelet transform denoising and other methods can be used.
[0023] S103 , performing local mean decomposition on the denoised signal to obtain multiple local amplitude components.
[0024] In this method, local mean decomposition can be used to extract the fault characteristics of the spindle bearing. By decomposing multiple local amplitude components, the vibration of the spindle bearing at different frequencies can be analyzed to determine whether the spindle bearing has a fault.
[0025] S104. Extracting instantaneous amplitude and instantaneous frequency from the multiple local amplitude components.
[0026] Each local amplitude component obtained after local mean decomposition is obtained by using an envelope signal and a pure frequency modulation signal, and the instantaneous amplitude of the local amplitude component is the envelope signal.
[0027] In the process of local mean decomposition of the denoised signal, a pure FM signal will be obtained, and the instantaneous frequency can be solved by methods such as phase difference or phase unwrapping of the pure FM signal.
[0028] S105 , performing zero-crossing feature calculation on the instantaneous amplitude and the instantaneous frequency to obtain the frequency of the denoised signal crossing the zero point from the positive direction to the negative direction in the time domain.
[0029] S106. Determine the over-limit state of the zero-crossing characteristic of the vibration signal of the spindle bearing based on the relationship between the zero-crossing frequency and the frequency characteristic distribution range, and identify the abnormal vibration phenomenon of the spindle bearing.
[0030] Exemplarily, the frequency characteristic distribution range can be determined as (μ-3σ, μ+3σ) according to the 3σ criterion. If the zero-crossing frequency exceeds the above frequency characteristic distribution range, it is determined that the zero-crossing characteristic of the vibration signal of the spindle bearing is out of limit, and abnormal vibration occurs in the spindle bearing.
[0031] The method described in the embodiment of the present application does not rely on the detection value of a single sensor to directly identify abnormalities, but rather processes the vibration signal and determines whether abnormal vibration occurs through the zero crossing rate. This makes the vibration abnormality detection of the main shaft bearing more accurate, helps to take maintenance measures at a timely manner, and ensures the stable operation and long-term reliability of the wind power generation system to the greatest extent.
[0032] In the embodiment of the present application, step S102 specifically includes: using fast wavelet transform to remove noise in the original vibration signal to obtain a denoised signal.
[0033] Fast wavelet transform has the ability of multi-resolution analysis. It can observe the original vibration signal step by step from coarse to fine, which enables fast wavelet transform to very well characterize the non-stationary features of the original vibration signal, such as edges, peaks and breakpoints. When processing vibration signals, these non-stationary features are often key information, and fast wavelet transform can more effectively extract these features and remove noise at the same time. Compared with denoising methods such as low-pass filtering, fast wavelet transform can achieve better denoising effect.
[0034] Specifically, the process of removing noise from the original vibration signal by using fast wavelet transform includes steps S201 to S204.
[0035] S201, performing multi-layer wavelet decomposition on the original vibration signal to obtain a plurality of wavelet packet nodes and a plurality of wavelet coefficients corresponding to the plurality of wavelet packet nodes.
[0036] Through multi-layer wavelet decomposition, the original vibration signal can be decomposed into sub-signals of multiple frequency ranges. Among them, the wavelet packet node is the node generated in the wavelet decomposition process, and each wavelet packet node represents a sub-signal of a specific frequency range. The wavelet coefficients corresponding to the wavelet packet nodes represent the local characteristics of the original vibration signal at different scales. Specifically, multiple wavelet coefficients are obtained by the following formula: ;in, represents the wavelet coefficients, represents the sub-signal, represents the wavelet basis function, a represents the scale factor, and b represents the translation factor.
[0037] S202: Adaptively select multiple wavelet packet nodes.
[0038] After executing step S202, the wavelet node that best represents the characteristics of the original vibration signal can be effectively selected, thereby reducing the amount of data while maintaining the signal quality and improving the efficiency of signal processing.
[0039] S203 , performing non-local mean filtering on the wavelet coefficients corresponding to the wavelet packet nodes retained after the adaptive selection to form adjusted wavelet coefficients.
[0040] When executing step S203, first, the non-local mean is calculated by the following formula: ; Where M is the window size for calculating the non-local mean; N is the signal length of the wavelet packet node; t′ is the time point index in the window, from 1 to N; Represents a wavelet packet node.
[0041] Then, the similarity between one of the wavelet coefficients and the other wavelet coefficients is calculated, which can be measured by the square error, correlation coefficient, etc. of the non-local means corresponding to the two wavelet coefficients.
[0042] Finally, the similarity weight is determined by using a Gaussian function or other methods, and the adjusted wavelet coefficients are determined by weighted averaging of the wavelet coefficients.
[0043] Since step S203 utilizes information of similar structures in the original vibration signal, executing step S203 can remove noise while retaining details of the original vibration signal.
[0044] S204: Reconstruct the vibration signal according to the adjusted wavelet coefficients to form a denoised signal.
[0045] When reconstructing the vibration signal, first reconstruct the retained wavelet packet nodes according to the adjusted wavelet coefficients. The reconstruction formula of the wavelet packet nodes is as follows: ;in, represents the wavelet coefficient of the j-th layer, the k-th wavelet packet node at time t after adjustment; Represents the retained wavelet coefficient of the j-th layer, k-th wavelet packet node at time t.
[0046] Then, after reconstructing the retained wavelet packet nodes, the retained wavelet packet nodes are merged to form a denoised signal.
[0047] Furthermore, the following steps S301 to S302 are used to adaptively select multiple wavelet packet nodes.
[0048] S301, calculating the energy concentration of each wavelet packet node.
[0049] Specifically, the energy concentration of each wavelet packet node is calculated by the following formula: ; in, Represents the energy of the wavelet packet node at position (j, k); represents the wavelet packet node at (j, k); N is the number of time samples. In the above formula, the numerator represents the sum of the energy of the wavelet packets for all time samples, and the denominator represents the total energy of all wavelet packets for all positions (j′, k′) and all time samples.
[0050] S302 , removing wavelet packet nodes whose energy concentration is less than a preset selection threshold, and retaining wavelet packet nodes whose energy concentration reaches the preset selection threshold.
[0051] In the embodiment of the present application, when executing step S103, the local amplitude component is separated from the original vibration signal by iteration until the original vibration signal is represented as the sum of multiple local amplitude components and a monotonic function; the original vibration signal x(t) is represented as k amplitude components and a monotonic function The sum is: .
[0052] The local amplitude component is separated from the original vibration signal, including steps S501 to S506.
[0053] S501, iteratively execute the first iteration step; wherein the first iteration step includes: using the time scale to control the window size for extracting the local mean, calculating the local mean and the local envelope value at each time point; determining the residual between the amplitude at each time point and the local mean, and applying the frequency analysis method to the residual to extract signal components in different frequency ranges.
[0054] The local mean at each time point is calculated using the following formula: ;in, represents the original vibration signal, k represents the offset in the time window, It means rounding down, j is an odd number to ensure the symmetry of the time window.
[0055] S502: When the residual energy reaches a preset termination condition, terminate the first iteration step.
[0056] S503 , processing the local mean and the local envelope value with a sliding average function to obtain a local mean function and a local envelope function.
[0057] The local mean function is calculated by the following formula: ;in, represents the original vibration signal, k represents the offset in the time window, Indicates rounding down, j is an odd number to ensure the symmetry of the time window.
[0058] The local envelope function is given by the following formula: ;in, represents the original vibration signal, k represents the offset in the time window, It means rounding down, j is an odd number to ensure the symmetry of the time window.
[0059] S504, iteratively executing the second iteration step until a pure frequency modulation function is obtained; wherein the second iteration step includes: performing demodulation processing on the original vibration signal using a local mean function.
[0060] Specifically, the local mean function is used to perform demodulation using the following formula: .like is a pure frequency modulation function, then the second iteration step ends; if If it is not a pure frequency modulation function, the second iteration step is performed iteratively.
[0061] S505 : multiply the generated multiple envelope signals to obtain an envelope signal of a local amplitude component.
[0062] The envelope signal formula of the local amplitude component is described as: .
[0063] S506: Multiply the envelope signal by the pure frequency modulation function to obtain a local amplitude component.
[0064] The local amplitude component formula is described as: .
[0065] Separate the local amplitude component obtained in step S506 from the original vibration signal, and convert the obtained new signal , and the signal Repeat the above steps as the original vibration signal, as shown in the following formula: ; Separate all local amplitude components until they become a monotonic function.
[0066] In the embodiment of the present application, step S105 includes step S601 to step S602.
[0067] S601, dividing each local amplitude component into frequency segments. After obtaining multiple frequency segments in step S601, it is convenient to perform more detailed feature analysis and zero crossing rate calculation in each subsequent segment.
[0068] S602: Calculate the zero crossing rate of each frequency segment.
[0069] The embodiment of the present application also provides a wind turbine main shaft bearing vibration anomaly identification device 200, which includes an acquisition module 201, a denoising module 202, a decomposition module 203, an extraction module 204, a zero crossing rate module 205 and an identification module 206.
[0070] The acquisition module 201 is used to acquire the original vibration signal; wherein the original vibration signal includes the effective value of the vibration acceleration of the spindle bearing in the vertical direction or the effective value of the vibration acceleration in the horizontal direction. The denoising module 202 is used to denoise the original vibration signal to obtain a denoised signal. The decomposition module 203 is used to perform local mean decomposition on the denoised signal to obtain multiple local amplitude components. The extraction module 204 is used to extract the instantaneous amplitude and instantaneous frequency from the multiple local amplitude components. The zero crossing rate module 205 is used to calculate the zero crossing feature of the instantaneous amplitude and instantaneous frequency to obtain the frequency of the denoised signal crossing the zero point from the positive direction to the negative direction in the time domain. The identification module 206 is used to determine the out-of-limit state of the zero crossing feature of the vibration signal of the spindle bearing based on the relationship between the zero crossing frequency and the frequency feature distribution range, and identify the abnormal vibration phenomenon of the spindle bearing.
[0071] The denoising module 202 is specifically used to remove noise in the original vibration signal by using fast wavelet transform to obtain a denoised signal.
[0072] The denoising module 202 includes a wavelet decomposition module, an adaptive selection module, a reconstruction module and a non-local mean module. The wavelet decomposition module is used to perform multi-layer wavelet decomposition on the original vibration signal to obtain multiple wavelet packet nodes and multiple wavelet coefficients corresponding to the multiple wavelet packet nodes. The adaptive selection module is used to perform adaptive selection on multiple wavelet packet nodes. The non-local mean module is used to perform non-local mean filtering on the wavelet coefficients corresponding to the wavelet packet nodes retained after adaptive selection to form adjusted wavelet coefficients. The reconstruction module is used to reconstruct the vibration signal according to the wavelet coefficients to form a denoised signal.
[0073] Specifically, the adaptive selection module is used to: calculate the energy concentration of each wavelet packet node; remove the wavelet packets whose energy concentration is less than a preset selection threshold, and retain the wavelet packets whose energy concentration reaches the preset selection threshold.
[0074] The decomposition module 203 is specifically used to iteratively separate the local amplitude component from the original vibration signal until the original vibration signal is represented as the sum of multiple local amplitude components and a monotonic function. Specifically, separating the local amplitude component from the original vibration signal includes: iteratively executing the first iteration step; terminating the first iteration step when the energy of the residual reaches a preset termination condition; processing the local mean and the local envelope value with a sliding average function to obtain a local mean function and a local envelope function; iteratively executing the second iteration step until a pure frequency modulation function is obtained; multiplying the multiple envelope signals generated to obtain an envelope signal of the local amplitude component; multiplying the envelope signal with the pure frequency modulation function to obtain the local amplitude component. Among them, the first iteration step includes: using the time scale to control the window size for extracting the local mean, calculating the local mean and local envelope value at each time point; determining the residual between the amplitude at each time point and the local mean, and applying the frequency analysis method to the residual to extract signal components in different frequency ranges. Among them, the second iteration step includes: demodulating the original vibration signal using the local mean function; The zero-crossing rate module 205 is specifically used to: divide each local amplitude component into frequency segments; and calculate the zero-crossing rate of each frequency segment.
[0075] The devices or modules described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. For the convenience of description, the above devices are described in various modules according to their functions. When implementing the embodiments of the present application, the functions of each module can be implemented in the same or multiple software and / or hardware. Of course, the module that implements a certain function can also be implemented by combining multiple sub-modules or sub-units.
[0076] An embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for identifying abnormal vibration of a main shaft bearing of a wind turbine set as described above is implemented.
[0077] An embodiment of the present application provides a computer-readable storage medium, which stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the method for identifying abnormal vibration of a main shaft bearing of a wind turbine set as described above is implemented.
[0078] The above-mentioned storage media include but are not limited to random access memory (RAM), read-only memory (ROM), cache, hard disk drive (HDD) or memory card (Memory Card).
[0079] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other. Each embodiment focuses on the differences from other embodiments.
[0080] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit the present application. Although the present application has been described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some or all of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present application.
Claims
1. A method for identifying abnormal vibration of a main shaft bearing of a wind turbine generator set, characterized in that: include: Acquire an original vibration signal; wherein the original vibration signal includes an effective value of the vibration acceleration of the spindle bearing in the vertical direction or an effective value of the vibration acceleration in the horizontal direction; De-noising the original vibration signal to obtain a de-noised signal; Performing local mean decomposition on the denoised signal to obtain multiple local amplitude components; Extracting instantaneous amplitude and instantaneous frequency from the multiple local amplitude components, and performing zero-crossing feature calculation on the instantaneous amplitude and the instantaneous frequency to obtain the frequency of the denoised signal crossing the zero point from positive to negative in the time domain; Based on the relationship between the zero-crossing frequency and the frequency characteristic distribution range, the over-limit state of the zero-crossing characteristic of the vibration signal of the spindle bearing is determined, and the abnormal vibration phenomenon of the spindle bearing is identified.
2. The method according to claim 1, characterized in that The denoising the original vibration signal to obtain a denoised signal comprises: Fast wavelet transform is used to remove noise in the original vibration signal to obtain the denoised signal.
3. The method according to claim 2, characterized in that The method of removing noise from the original vibration signal by using fast wavelet transform comprises: Performing multi-layer wavelet decomposition on the original vibration signal to obtain a plurality of wavelet packet nodes and a plurality of wavelet coefficients corresponding to the plurality of wavelet packet nodes; Adaptively selecting the plurality of wavelet packet nodes; The wavelet coefficients corresponding to the wavelet packet nodes retained after adaptive selection are subjected to non-local mean filtering to form adjusted wavelet coefficients.
4. Reconstruct the vibration signal according to the adjusted wavelet coefficients to form the denoised signal.
5. The method according to claim 3, characterized in that: The adaptively selecting the plurality of wavelet packet nodes comprises: Calculating the energy concentration of each of the wavelet packet nodes; The wavelet packet nodes whose energy concentration is less than a preset selection threshold are removed, and the wavelet packet nodes whose energy concentration reaches the preset selection threshold are retained.
6. The method according to claim 1, characterized in that The performing local mean decomposition on the denoised signal to obtain a plurality of local amplitude components comprises: Iteratively separating the local amplitude component from the original vibration signal until the original vibration signal is represented as the sum of the multiple local amplitude components and a monotonic function; separating the local amplitude component from the original vibration signal comprises: Iteratively perform the first iteration step; wherein the first iteration step includes: using the time scale to control the window size for extracting the local mean, calculating the local mean and the local envelope value at each time point; determining the residual between the amplitude at each time point and the local mean, and applying a frequency analysis method to the residual to extract signal components in different frequency ranges; When the energy of the residual reaches a preset termination condition, terminating the first iteration step; Processing the local mean value and the local envelope value with a sliding average function to obtain a local mean function and a local envelope function; Iteratively executing the second iteration step until a pure frequency modulation function is obtained; wherein the second iteration step comprises: demodulating the original vibration signal using the local mean function; Multiplying the generated multiple envelope signals to obtain the envelope signal of the local amplitude component; The envelope signal is multiplied by the pure frequency modulation function to obtain the local amplitude component.
7. The method according to claim 1, characterized in that The calculating of zero-crossing characteristics of the instantaneous amplitude and the instantaneous frequency comprises: dividing each of the local amplitude components into frequency bins; The zero crossing rate of each of the frequency bins is calculated.
8. A device for identifying abnormal vibration of a main shaft bearing of a wind turbine generator set, characterized in that: include: An acquisition module, used for acquiring an original vibration signal; wherein the original vibration signal includes an effective value of the vibration acceleration of the spindle bearing in the vertical direction or an effective value of the vibration acceleration in the horizontal direction; A denoising module, used for denoising the original vibration signal to obtain a denoised signal; A decomposition module, used for performing local mean decomposition on the denoised signal to obtain multiple local amplitude components; An extraction module, used for extracting instantaneous amplitude and instantaneous frequency from the multiple local amplitude components; A zero crossing rate module, used to calculate the zero crossing characteristics of the instantaneous amplitude and the instantaneous frequency, and obtain the frequency of the denoised signal crossing the zero point from positive to negative in the time domain; The identification module is used to determine the over-limit state of the zero-crossing characteristic of the vibration signal of the spindle bearing based on the relationship between the zero-crossing frequency and the frequency characteristic distribution range, and identify the abnormal vibration phenomenon of the spindle bearing.
9. An electronic device, characterized in that: include: one or more processors; as well as A memory storing computer executable instructions, wherein the computer executable instructions, when executed by the one or more processors, implement the method for identifying abnormal vibration of a main shaft bearing of a wind turbine set according to any one of claims 1 to 6.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the method for identifying abnormal vibration of a main shaft bearing of a wind turbine set according to any one of claims 1 to 6 is implemented.