Water turbine set fault diagnosis method and device, electronic equipment and storage medium

Abstract

The invention belongs to the technical field of hydraulic turbine set fault diagnosis, and particularly discloses a hydraulic turbine set fault diagnosis method and device, electronic equipment and a storage medium. The method comprises the following steps: performing noise reduction preprocessing on first acoustic emission signal data when a hydraulic turbine set to be tested operates to obtain second acoustic emission signal data after noise reduction processing; performing de-noising processing by using the second acoustic emission signal data and the reference frequency domain feature information, and determining fault diagnosis time domain signal data; the reference frequency domain characteristic information is determined based on third acoustic emission signal data of the to-be-detected hydraulic turbine set under the fault-free working condition; and determining a fault diagnosis result of the hydraulic turbine set to be detected based on the time domain feature information of the fault diagnosis time domain signal data. By means of the fault diagnosis method and device, whether the faults such as cracks, cavitation and Karman vortexes of the water turbine flow passage component of the water turbine set to be detected occur or not can be effectively diagnosed, the diagnosis result is accurate and reliable, and the fault diagnosis effect of the water turbine set is improved.

Description

Technical Field

[0001] The present application belongs to the technical field of hydraulic turbine unit fault diagnosis, and more specifically, relates to a hydraulic turbine unit fault diagnosis method, device, electronic equipment and storage medium. Background Art

[0002] As the core equipment of a hydropower station, the operating status of a hydraulic turbine unit directly affects the safety and efficiency of the power station. However, during the operation of a hydraulic turbine unit, various faults may occur due to the influence of various factors, such as water flow impact, mechanical wear, material fatigue, etc. If these faults are not diagnosed and repaired in a timely and effective manner, they may cause equipment damage, shutdown, or even safety accidents. Therefore, fault diagnosis of hydraulic turbine units is of great significance.

[0003] At present, many hydropower stations have installed vibration-based fault monitoring and diagnosis systems. However, traditional vibration diagnosis methods have many limitations. For example, for cracks, cavitation, and Karman vortex faults in turbine flow components, the vibration signals generated by these faults may be very weak or complex. It is difficult to accurately identify the fault signals using conventional vibration testing methods, resulting in inaccurate diagnostic results and poor fault diagnosis effects.

[0004] Therefore, how to better realize fault diagnosis of hydraulic turbine units has become a technical problem that needs to be solved urgently in the industry. Summary of the invention

[0005] In view of the defects of the prior art, the purpose of this application is to better realize the fault diagnosis of the turbine unit, aiming to solve the problems of inaccurate diagnostic results and poor fault diagnosis effect of traditional turbine unit fault diagnosis methods.

[0006] To achieve the above objectives, in a first aspect, the present application provides a method for diagnosing faults of a hydraulic turbine unit, comprising: Performing noise reduction preprocessing on the first acoustic emission signal data when the turbine unit to be tested is in operation, to obtain the second acoustic emission signal data after the noise reduction processing; Denoising the second acoustic emission signal data and reference frequency domain characteristic information to determine fault diagnosis time domain signal data; the reference frequency domain characteristic information is determined based on the third acoustic emission signal data of the turbine unit to be tested under a fault-free working condition; Based on the time domain characteristic information of the fault diagnosis time domain signal data, a fault diagnosis result of the hydraulic turbine unit to be tested is determined.

[0007] Optionally, the performing denoising processing using the second acoustic emission signal data and reference frequency domain feature information to determine fault diagnosis time domain signal data includes: Performing a discrete Fourier transform on the second acoustic emission signal data to obtain frequency domain feature information corresponding to the second acoustic emission signal data; Performing spectral subtraction processing based on the frequency domain feature information and the reference frequency domain feature information to obtain frequency domain feature deviation information; The fault diagnosis time domain signal data is determined based on the amplitude of the frequency domain characteristic deviation information and the phase angle corresponding to the frequency domain characteristic information.

[0008] Optionally, determining the fault diagnosis result of the hydraulic turbine unit to be tested based on the time domain characteristic information of the fault diagnosis time domain signal data includes: Determine a comparison result between the time domain characteristic information of the fault diagnosis time domain signal data and a preset threshold range corresponding thereto; Based on the comparison result, determining the fault diagnosis result of the hydraulic turbine unit to be tested; The time domain feature information includes at least one of a mean, a standard deviation, a variance and a kurtosis.

[0009] Optionally, the first acoustic emission signal data when the turbine unit to be measured is running is subjected to noise reduction preprocessing to obtain the second acoustic emission signal data after noise reduction processing, including: The first acoustic emission signal data is subjected to noise reduction preprocessing by using a singular value decomposition algorithm to obtain the second acoustic emission signal data.

[0010] Optionally, the using a singular value decomposition algorithm to perform noise reduction preprocessing on the first acoustic emission signal data to obtain the second acoustic emission signal data includes: Based on the first acoustic emission signal data, determining a corresponding Toeplitz matrix; Performing singular value decomposition on the Toeplitz matrix to determine the transpose of a left singular matrix, a singular value matrix, and a right singular matrix of the Toeplitz matrix; According to the principle of retaining the maximum singular value, retaining a target number of singular values ​​from the singular value matrix, and determining a new singular value matrix based on the target number of singular values; Determine a denoised signal matrix based on the left singular matrix, the new singular value matrix and the transpose of the right singular matrix; The second acoustic emission signal data is obtained according to each element on the sub-diagonal of the denoised signal matrix.

[0011] Optionally, before performing denoising processing using the second acoustic emission signal data and the reference frequency domain feature information to determine the fault diagnosis time domain signal data, the method further includes: According to the sampling frequency of the first acoustic emission signal data, collecting the third acoustic emission signal data of the tested hydraulic turbine unit under the fault-free working condition; Performing noise reduction preprocessing on the third acoustic emission signal data to obtain fourth acoustic emission signal data after noise reduction processing; Performing discrete Fourier transform on the fourth acoustic emission signal data to obtain the reference frequency domain feature information.

[0012] In a second aspect, the present application provides a hydraulic turbine fault diagnosis device, comprising: A noise reduction preprocessing module is used to perform noise reduction preprocessing on the first acoustic emission signal data when the turbine unit to be tested is running, so as to obtain the second acoustic emission signal data after noise reduction preprocessing; A denoising processing module, configured to perform denoising processing using the second acoustic emission signal data and reference frequency domain characteristic information to determine fault diagnosis time domain signal data; the reference frequency domain characteristic information is determined based on the third acoustic emission signal data of the turbine unit to be tested under a fault-free working condition; A fault diagnosis module is used to determine the fault diagnosis result of the hydraulic turbine group to be tested based on the time domain characteristic information of the fault diagnosis time domain signal data.

[0013] In a third aspect, the present application provides an electronic device comprising: at least one memory for storing programs; and at least one processor for executing the programs stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method described in the first aspect or any possible implementation of the first aspect.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect.

[0015] In a fifth aspect, the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect.

[0016] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0017] In general, the above technical solutions conceived by this application have the following beneficial effects compared with the prior art: The present application provides a method, device, electronic device and storage medium for diagnosing faults of a hydraulic turbine set. By utilizing the characteristics of acoustic emission, the acoustic emission signal data of the hydraulic turbine set to be tested during operation is subjected to noise reduction processing, and the acoustic emission signal data under fault-free working conditions are combined for denoising processing, so that clean acoustic emission signal data can be effectively extracted from the original acoustic emission signal data of the hydraulic turbine set, thereby obtaining time domain signal data for fault diagnosis, and the time domain characteristics of the fault diagnosis time domain signal data are combined for analysis, so that it can be effectively diagnosed whether the hydraulic turbine set to be tested has cracks, cavitation, Karman vortex and other faults in the flow-through components of the hydraulic turbine set, and the diagnosis result is accurate and reliable, thereby improving the fault diagnosis effect of the hydraulic turbine set. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is one of the flow charts of the hydraulic turbine unit fault diagnosis method provided in the embodiment of the present application; Figure 2 This is the second flow chart of the hydraulic turbine fault diagnosis method provided in the embodiment of the present application; Figure 3 is a structural schematic diagram of a hydraulic turbine fault diagnosis device provided in an embodiment of the present application; Figure 4 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0020] The terms "first" and "second" in the specification and claims of this application are used to distinguish different objects rather than to describe a specific order of the objects. For example, the first acoustic emission signal data and the second acoustic emission signal data are used to distinguish different acoustic emission signal data rather than to describe a specific order of the acoustic emission signal data.

[0021] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0022] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two. For example, multiple key parts refer to two or more key parts of two or more turbine units.

[0023] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0024] Figure 1 is one of the flow charts of the hydraulic turbine fault diagnosis method provided in the embodiment of the present application, such as Figure 1 As shown, the method includes: Step S1, performing noise reduction preprocessing on first acoustic emission signal data when the turbine unit to be tested is running, to obtain second acoustic emission signal data after noise reduction processing; Step S2, performing denoising processing using the second acoustic emission signal data and reference frequency domain characteristic information to determine fault diagnosis time domain signal data; the reference frequency domain characteristic information is determined based on the third acoustic emission signal data of the turbine unit to be tested under a fault-free working condition; Step S3, determining the fault diagnosis result of the hydraulic turbine unit to be tested based on the time domain characteristic information of the fault diagnosis time domain signal data.

[0025] Specifically, the acoustic emission signal described in the embodiment of the present application is a signal corresponding to the physical phenomenon of transient elastic waves generated due to the rapid release of energy inside an object or material.

[0026] In the embodiments of the present application, the collection of acoustic emission signal data in the turbine unit is mainly targeted at the key components and areas prone to failure of the unit, including the runner and blades, water-guiding components, cavitation-prone areas and other key components. These areas are usually places of water flow impact, mechanical vibration and stress concentration, and they are more likely to generate acoustic emission signals.

[0027] The first acoustic emission signal data described in the embodiment of the present application refers to the original acoustic emission signal data generated when the turbine unit to be tested is running under any operating scenario, wherein there may be acoustic emission signal data generated under a unit failure condition.

[0028] The second acoustic emission signal data described in the embodiment of the present application refers to the acoustic emission signal data obtained after the first acoustic emission signal data is subjected to noise reduction preprocessing by using a noise reduction algorithm.

[0029] The third acoustic emission signal data described in the embodiment of the present application refers to the acoustic emission signal data collected under the fault-free working condition of the unit when determining that the tested turbine unit has no faults.

[0030] The reference frequency domain characteristic information described in the embodiment of the present application refers to the signal characteristic information in the frequency domain obtained by performing corresponding processing on the third acoustic emission signal data according to the same processing flow as the first acoustic emission signal data. It can be obtained by performing signal noise reduction and frequency domain transformation on the third acoustic emission signal data.

[0031] The fault diagnosis time domain signal data described in the embodiment of the present application refers to time domain signal data used to directly diagnose faults of the unit, which can be obtained by performing frequency domain transformation on the second acoustic emission signal data to obtain corresponding frequency domain feature information, and then denoising it and the reference frequency domain feature information.

[0032] The fault diagnosis result described in the embodiment of the present application is used to characterize whether there is a fault in the turbine unit to be tested, which may specifically include a diagnosis result that the unit is fault-free, and a diagnosis result that the unit is faulty.

[0033] In an embodiment of the present application, in step S1, an acoustic emission sensor may be pre-installed at a key position of the turbine unit to be tested. It should be noted that the installation position of the sensor should be as close to the acoustic emission source as possible to improve the signal acquisition quality. At the same time, according to the actual situation and monitoring requirements of the unit, sensors may be installed at multiple key positions to achieve comprehensive monitoring and diagnosis. Furthermore, the first acoustic emission signal data of the turbine unit to be tested during operation may be collected by each acoustic emission sensor.

[0034] Furthermore, in this embodiment, relevant noise reduction processing algorithms, such as singular value decomposition (SVD) and singular spectrum analysis (SSA), can be used to perform noise reduction preprocessing on the acoustic emission signal data collected when the turbine unit to be tested is running, so as to pre-complete a round of noise signal processing on the first acoustic emission signal data to obtain the acoustic emission signal data after noise reduction processing, that is, the second acoustic emission signal data can be obtained.

[0035] In an embodiment of the present application, in step S2, after obtaining the second acoustic emission signal data, a frequency domain transform may be performed on it, so that the frequency domain feature information of the second acoustic emission signal data may be extracted, and then, by comparing the frequency domain feature information of the second acoustic emission signal data with the reference frequency domain feature information, the second acoustic emission signal data may be further denoised, and fault diagnosis time domain signal data may be obtained.

[0036] Furthermore, in an embodiment of the present application, in step S3, time domain analysis is performed on the obtained fault diagnosis time domain signal data, and relevant time domain characteristic information of the fault diagnosis time domain signal data can be obtained. The time domain characteristic information of the signal can be used to effectively understand the time domain characteristics of the turbine unit to be tested, thereby determining whether the turbine unit to be tested has a fault and obtaining the final fault diagnosis result.

[0037] The hydraulic turbine fault diagnosis method of the embodiment of the present application utilizes the characteristics of acoustic emission to perform noise reduction processing on the acoustic emission signal data of the hydraulic turbine to be tested during operation, and performs noise reduction processing on the acoustic emission signal data under fault-free working conditions, so as to effectively extract clean acoustic emission signal data from the original acoustic emission signal data of the hydraulic turbine to obtain time domain signal data for fault diagnosis, and analyzes the time domain characteristics of the fault diagnosis time domain signal data, so as to effectively diagnose whether the hydraulic turbine to be tested has cracks, cavitation, Karman vortex and other faults in the flow-through components of the hydraulic turbine, and the diagnosis result is accurate and reliable, thereby improving the fault diagnosis effect of the hydraulic turbine.

[0038] Based on the content of the above embodiment, as an optional embodiment, step S1, performing noise reduction preprocessing on the first acoustic emission signal data when the turbine unit to be tested is running, to obtain the second acoustic emission signal data after noise reduction processing, includes: The SVD algorithm is used to perform noise reduction preprocessing on the first acoustic emission signal data to obtain the second acoustic emission signal data.

[0039] Specifically, in the embodiment of the present application, the acoustic emission signal data during the operation of the water turbine unit, namely, the first acoustic emission signal data, is collected by an acoustic emission sensor. The sampling frequency of the acoustic emission sensor is The following relationship must be satisfied: ; in, Indicates the frequency of acoustic emission generated when a turbine unit fails. Generally between 20kHz and 400kHZ; sampling frequency At least 1000kHZ.

[0040] Further, in an embodiment of the present application, the first acoustic emission signal data may be subjected to noise reduction preprocessing using an SVD algorithm to obtain the second acoustic emission signal data, wherein the SVD algorithm may include a conventional SVD algorithm and an improved SVD algorithm.

[0041] That is to say, in the embodiment of the present application, a conventional SVD algorithm or an improved SVD algorithm may be used to perform noise reduction preprocessing on the first acoustic emission signal data.

[0042] The method of the embodiment of the present application can significantly improve the signal-to-noise ratio of the signal, retain the main features of the signal, make the signal clearer, and improve the signal quality by using the SVD algorithm to perform noise reduction processing on the acoustic emission signal data.

[0043] Based on the content of the above embodiment, as an optional embodiment, the first acoustic emission signal data is subjected to noise reduction preprocessing by using a singular value decomposition algorithm to obtain the second acoustic emission signal data, including: Based on the first acoustic emission signal data, determining a corresponding Toeplitz matrix; Perform singular value decomposition on the Toeplitz matrix to determine the transpose of the left singular matrix, the singular value matrix and the right singular matrix of the Toeplitz matrix; According to the principle of retaining the maximum singular value, a target number of singular values ​​are retained from the singular value matrix, and a new singular value matrix is ​​determined based on the target number of singular values; Determine a denoised signal matrix based on the left singular matrix, the new singular value matrix and the transpose of the right singular matrix; The second acoustic emission signal data is obtained according to each element on the sub-diagonal line of the signal matrix after denoising.

[0044] Specifically, the target number described in the embodiment of the present application refers to the number of valid singular values ​​retained in the singular value matrix, which can be specifically determined according to the rank of the Toeplitz matrix.

[0045] In an embodiment of the present application, an improved SVD algorithm is used to perform SVD noise reduction preprocessing on the collected original acoustic emission signal data, namely, the first acoustic emission signal data, which may specifically include the following steps.

[0046] Step S101: construct a Toplittz matrix for each frame of acoustic emission signal in the first acoustic emission signal data. ,Right now: ; in, Represents acoustic emission signal with noise, 0≤ n ≤ ; N is the number of signals in one frame of acoustic emission signal; L For the prediction order, in order to prevent the original signal from being lost due to too many prediction signals, we choose .

[0047] Step S102: for the noisy matrix, that is, the Toeplitz matrix Perform SVD decomposition calculation, that is ,in, , for Y The orthogonal matrices are Y The transpose of the left singular matrix and the right singular matrix of ; for Y The singular value matrix of .

[0048] Step S103, calculate the matrix Y The effective rank r of .

[0049] Step S104: Select valid singular values ​​from the obtained singular value matrix :According to the principle of retaining the maximum singular value, retain the target number of singular values ​​from the singular value matrix, that is, retain The first r larger singular values ​​of the vector matrix, and the remaining element positions are all set to 0, so that a new singular value matrix can be obtained , which can be expressed as .

[0050] Step S105, calculate the best estimated denoising matrix , that is, construct the denoised signal matrix, where , where yes The effective singular values ​​of and They are The transpose of the left singular matrix and the right singular matrix of .

[0051] Step S106: according to each element on the sub-diagonal line of the denoised signal matrix, the denoised signal matrix : ; Extract the last signal of the first row in order of the sub-diagonal line To the last signal of the first column , that is, the second acoustic emission signal data after SVD noise reduction processing can be obtained .

[0052] The method of the embodiment of the present application constructs a Toeplitz matrix by considering setting an appropriate prediction order according to the number of signals in the acoustic emission signal frame, and retaining the same number of valid singular values ​​as the rank value when reconstructing a new singular value matrix. It can retain the original signal data to the greatest extent while performing noise reduction processing, thereby improving the noise reduction effect on the original acoustic emission signal data of the turbine group.

[0053] Based on the content of the above embodiment, as an optional embodiment, in step S2, before performing denoising processing using the second acoustic emission signal data and the reference frequency domain feature information to determine the fault diagnosis time domain signal data, the method further includes: According to the sampling frequency of the first acoustic emission signal data, the third acoustic emission signal data of the tested turbine unit under the fault-free working condition is collected; Performing noise reduction preprocessing on the third acoustic emission signal data to obtain fourth acoustic emission signal data after noise reduction processing; The fourth acoustic emission signal data is subjected to discrete Fourier transform to obtain reference frequency domain feature information.

[0054] Specifically, the fourth acoustic emission signal data described in the embodiment of the present application refers to the acoustic emission signal data obtained by performing noise reduction processing on the acoustic emission signal data of the measured turbine unit under a fault-free working condition.

[0055] In an embodiment of the present application, in step S2, before performing denoising processing using the second acoustic emission signal data and the reference frequency domain feature information to determine the fault diagnosis time domain signal data, it is also necessary to determine the frequency domain feature information corresponding to the acoustic emission signal of the turbine unit under a fault-free condition (i.e., a normal condition), obtain the reference frequency domain feature information, and use it as normal control information.

[0056] Further, in the embodiment of the present application, according to the sampling frequency of the first acoustic emission signal data , collect the acoustic emission signal data of the turbine unit to be tested under the condition of no fault, that is, the third acoustic emission signal data. Then, according to the specific implementation of the above SVD algorithm, the acquired third acoustic emission signal data can be subjected to noise reduction preprocessing to obtain the fourth acoustic emission signal data after noise reduction processing. .

[0057] Furthermore, in an embodiment of the present application, the fourth acoustic emission signal data obtained is Perform Fast Fourier Transform (FFT) to calculate the frequency domain characteristics of the acoustic emission signal data after noise reduction under normal working conditions, that is, obtain the reference frequency domain characteristic information , which can be expressed as:

[0058] in, Represents the input fourth acoustic emission signal data.

[0059] Finally, this reference frequency domain feature information is stored for a long time as a normal control signal M , and stored long term as a normal control signal.

[0060] The method of the embodiment of the present application can provide accurate and reliable reference frequency domain feature information by collecting acoustic emission signal data of the turbine unit to be tested under normal operating conditions at the same sampling frequency, and performing noise reduction processing according to the same noise reduction preprocessing method, thereby providing reliable source data for subsequent unit fault diagnosis, which is beneficial to improving the accuracy of turbine unit fault diagnosis.

[0061] Based on the content of the above embodiment, as an optional embodiment, step S2, using the second acoustic emission signal data and the reference frequency domain feature information to perform denoising processing to determine the fault diagnosis time domain signal data, includes: Performing a discrete Fourier transform on the second acoustic emission signal data to obtain frequency domain feature information corresponding to the second acoustic emission signal data; Perform spectral subtraction processing based on the frequency domain feature information and the reference frequency domain feature information to obtain frequency domain feature deviation information; The fault diagnosis time domain signal data is determined based on the amplitude of the frequency domain characteristic deviation information and the phase angle corresponding to the frequency domain characteristic information.

[0062] Specifically, the spectral subtraction method described in the embodiment of the present application is a method for processing acoustic emission signals based on the spectral subtraction idea of ​​traditional speech denoising. In the frequency domain, the spectrum of the noisy emission signal can be expressed as the superposition of the clean acoustic emission signal spectrum and the noise spectrum. By estimating the noise spectrum and subtracting the noise spectrum from the noisy spectrum, the spectrum of the clean acoustic emission signal can be obtained.

[0063] The frequency domain feature deviation information described in the embodiment of the present application refers to the frequency domain feature information obtained by performing spectral subtraction processing on the frequency domain feature information and the reference frequency domain feature information, which can be characterized as the frequency spectrum information of the clean acoustic emission signal.

[0064] More specifically, in an embodiment of the present application, a discrete FFT is performed on the second acoustic emission signal data after noise reduction to obtain frequency domain feature information corresponding to the second acoustic emission signal data. , the process can be expressed as: 1; in, For the sampling frequency The first i Frame No. The acoustic emission signal of a turbine unit.

[0065] Further, in an embodiment of the present application, spectral subtraction is performed based on the frequency domain feature information and the reference frequency domain feature information. Specifically, first Find the magnitude and phase angle of each component. The magnitude is , phase angle It can be expressed as: .

[0066] At the same time, the average energy of the normal control signal is calculated based on the reference frequency domain feature information, that is: ; in, Indicates the corresponding frame number, Indicates normal control signal.

[0067] Furthermore, the normal control signal Perform basic spectral subtraction with the real-time signal and process it according to the following spectral subtraction formula to obtain frequency domain feature deviation information ,Right now: ; in, and are two constants; represents the over-reduction factor, which can be taken =4; represents the gain compensation factor, which can be taken as =0.001.

[0068] Furthermore, based on the amplitude of the frequency domain characteristic deviation information and the phase angle corresponding to the frequency domain characteristic information, the fault diagnosis time domain signal data is determined. Taking the square root gives the signal amplitude , and then the amplitude Phase Angle After synthesis, the inverse Fourier transform (IFFT) is performed to obtain the final spectral subtraction signal, that is, the fault diagnosis time domain signal data. .

[0069] The method of the embodiment of the present application, by drawing on the technical idea of ​​spectral subtraction and combining the difference in the spectral characteristics of the acoustic emission signal and the noise for denoising, can effectively reduce noise interference, improve the quality of the acquired fault diagnosis signal, make the fault signal clearer, enhance the fault detection sensitivity, and help to further improve the efficiency and accuracy of turbine unit fault diagnosis.

[0070] Based on the content of the above embodiment, as an optional embodiment, step S3, based on the time domain characteristic information of the fault diagnosis time domain signal data, determines the fault diagnosis result of the turbine unit to be tested, including: Determine a comparison result between the time domain characteristic information of the fault diagnosis time domain signal data and the preset threshold range corresponding thereto; Based on the comparison results, the fault diagnosis results of the turbine unit to be tested are determined; The time domain feature information includes at least one of a mean, a standard deviation, a variance, and a kurtosis.

[0071] Specifically, in the embodiment of the present application, when obtaining the fault diagnosis time domain signal data After that, the fault diagnosis time domain signal data can be extracted The time domain characteristic information of the unit acoustic emission signal can be effectively obtained by using the time domain characteristic information of the unit acoustic emission signal. Among them, the time domain characteristic information includes the mean , Standard Deviation ,variance Kurtosis Table 1 shows the specific expressions of each time domain feature information.

[0072] Table 1

[0073] Furthermore, an upper limit and a lower limit empirical value are set for each time domain characteristic information, and a preset threshold range corresponding to each time domain characteristic information is determined. When at least one characteristic value exceeds its corresponding preset threshold range, it can be determined that a fault has occurred in the turbine unit to be tested. The process can be expressed as follows: ; ; ; ; in, , , , Represent the fault diagnosis time domain signal data The mean, variance, standard deviation and kurtosis of ]、[ ] 、[ ]and[ ] The preset threshold range corresponding to each feature value.

[0074] It should be noted that, since the characteristic values ​​of acoustic emission signals of different turbine units under normal operating conditions may vary, it is necessary to determine the preset threshold range corresponding to each characteristic value through empirical values ​​obtained through experiments.

[0075] Optionally, determine four time domain feature information of the fault diagnosis time domain signal data: , , , The comparison result with the preset threshold range corresponding to it is determined after comparison. , , , When both exceed their corresponding preset threshold ranges, it can be determined that a fault has occurred in the turbine unit to be tested, ensuring that the fault diagnosis result of the turbine unit is more reliable.

[0076] Compared with the traditional vibration detection method, the method of the embodiment of the present application can more easily detect the early stage cracks of turbine blades and other difficult-to-detect faults by using acoustic emission signals for detection. This is because the acoustic emission signal is directly derived from the object being tested itself and reflects the primary signal of the fault, which has a high degree of real-time and accuracy. Therefore, by using multiple time domain characteristic information of the acoustic emission signal to judge the turbine fault, the accuracy and reliability of the unit fault diagnosis can be significantly improved.

[0077] Figure 2 FIG. 2 is a flow chart of the method for diagnosing faults of a hydraulic turbine set provided in an embodiment of the present application. Figure 2 As shown, in the embodiment of the present application, on the one hand, according to the set sampling frequency , collect the normal control signal of the turbine unit to be tested, that is, collect the acoustic emission signal data under the fault-free working condition, and then, perform noise reduction preprocessing according to the specific implementation method of the aforementioned SVD algorithm, perform FFT processing on the processed acoustic emission signal data, and obtain reference frequency domain feature information as the normal control signal M For long term storage.

[0078] On the other hand, when the turbine unit to be tested is running again, the real-time collected acoustic emission signal data is also subjected to the aforementioned SVD noise reduction preprocessing, and then FFT processing is performed to obtain the corresponding frequency domain feature information. , and perform basic spectral subtraction operation with the normal control signal. Specifically, the average energy of the normal control signal is calculated based on the reference frequency domain feature information. At the same time, the frequency domain feature information is calculated The magnitude of the quadratic Then, the average energy The quadratic Perform spectral subtraction to obtain frequency domain feature deviation information , and take the square root of the spectrum subtraction result to get the signal amplitude , and then the amplitude and frequency domain feature information Phase Angle After synthesis, IFFT processing is performed to finally obtain the fault signal, that is, the fault diagnosis time domain signal data. .

[0079] Further, in an embodiment of the present application, the fault diagnosis time domain signal data is calculated The time domain characteristic information includes at least one of the mean, standard deviation, variance and kurtosis. Then, it is determined whether the various types of time domain characteristic information are within the corresponding preset threshold range. If so, it can be determined that the turbine unit to be tested is operating normally; otherwise, it can be determined that the turbine unit to be tested has failed.

[0080] The method of the embodiment of the present application can effectively diagnose faults occurring in a hydro-turbine unit in real time, and while reducing the complexity of the fault diagnosis model, improves the efficiency and accuracy of fault diagnosis.

[0081] The hydraulic turbine unit fault diagnosis device provided in the present application is described below. The hydraulic turbine unit fault diagnosis device described below and the hydraulic turbine unit fault diagnosis method described above can be referenced to each other.

[0082] Figure 3 is a schematic diagram of the structure of a hydraulic turbine fault diagnosis device provided in an embodiment of the present application, such as Figure 3 As shown, including: The noise reduction preprocessing module 10 is used to perform noise reduction preprocessing on the first acoustic emission signal data when the turbine unit to be tested is running, and obtain the second acoustic emission signal data after the noise reduction processing; A denoising processing module 20 is used to perform denoising processing using the second acoustic emission signal data and reference frequency domain characteristic information to determine fault diagnosis time domain signal data; the reference frequency domain characteristic information is determined based on the third acoustic emission signal data of the turbine unit to be tested under a fault-free working condition; The fault diagnosis module 30 is used to determine the fault diagnosis result of the hydraulic turbine unit to be tested based on the time domain characteristic information of the fault diagnosis time domain signal data.

[0083] It can be understood that the detailed functional implementation of each of the above-mentioned units / modules can be found in the introduction of the aforementioned method embodiment, and will not be repeated here.

[0084] It should be understood that the above-mentioned device is used to execute the method in the above-mentioned embodiment. The implementation principle and technical effect of the corresponding program module in the device are similar to those described in the above-mentioned method. The working process of the device can refer to the corresponding process in the above-mentioned method, which will not be repeated here.

[0085] The hydraulic turbine fault diagnosis device of the embodiment of the present application utilizes the characteristics of acoustic emission to perform noise reduction processing on the acoustic emission signal data of the hydraulic turbine to be tested when it is running, and performs noise reduction processing on the acoustic emission signal data under fault-free working conditions, so as to effectively extract clean acoustic emission signal data from the original acoustic emission signal data of the hydraulic turbine, thereby obtaining time domain signal data for fault diagnosis, and analyzes the time domain characteristics of the fault diagnosis time domain signal data, so as to effectively diagnose whether the hydraulic turbine to be tested has cracks, cavitation, Karman vortex and other faults in the flow-through components of the hydraulic turbine, and the diagnosis result is accurate and reliable, thereby improving the fault diagnosis effect of the hydraulic turbine.

[0086] Based on the method in the above embodiment, the embodiment of the present application provides an electronic device, such as Figure 4 As shown, the electronic device may include: a processor (Processor) 410, a communication interface (Communications Interface) 420, a memory (Memory) 430 and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the method in the above embodiment.

[0087] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0088] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method in the above embodiment.

[0089] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method in the above embodiment.

[0090] It is understandable that the processor in the embodiment of the present application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0091] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0092] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0093] It should be understood that the various numerical numbers involved in the embodiments of the present application are only used for the convenience of description and are not used to limit the scope of the embodiments of the present application.

[0094] It should be understood that expressions such as "including" and "may include" that may be used in the present application indicate the existence of the disclosed functions, operations, or constituent elements, and do not limit one or more additional functions, operations, and constituent elements. In the present application, terms such as "including" and / or "having" may be interpreted as indicating specific characteristics, numbers, operations, constituent elements, components, or combinations thereof, but may not be interpreted as excluding the existence or possibility of adding one or more other characteristics, numbers, operations, constituent elements, components, or combinations thereof.

[0095] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for diagnosing faults of a hydraulic turbine unit, characterized in that: include: Performing noise reduction preprocessing on the first acoustic emission signal data when the turbine unit to be tested is in operation, to obtain the second acoustic emission signal data after the noise reduction processing; Denoising the second acoustic emission signal data and reference frequency domain characteristic information to determine fault diagnosis time domain signal data; the reference frequency domain characteristic information is determined based on the third acoustic emission signal data of the turbine unit to be tested under a fault-free working condition; Based on the time domain characteristic information of the fault diagnosis time domain signal data, a fault diagnosis result of the hydraulic turbine unit to be tested is determined.

2. The method for diagnosing faults of a hydraulic turbine unit according to claim 1, characterized in that: The method of performing denoising processing using the second acoustic emission signal data and reference frequency domain feature information to determine fault diagnosis time domain signal data includes: Performing a discrete Fourier transform on the second acoustic emission signal data to obtain frequency domain feature information corresponding to the second acoustic emission signal data; Performing spectral subtraction processing based on the frequency domain feature information and the reference frequency domain feature information to obtain frequency domain feature deviation information; The fault diagnosis time domain signal data is determined based on the amplitude of the frequency domain characteristic deviation information and the phase angle corresponding to the frequency domain characteristic information.

3. The method for diagnosing faults of a hydraulic turbine unit according to claim 1, characterized in that: The determining of the fault diagnosis result of the hydraulic turbine unit to be tested based on the time domain characteristic information of the fault diagnosis time domain signal data comprises: Determine a comparison result between the time domain characteristic information of the fault diagnosis time domain signal data and a preset threshold range corresponding thereto; Based on the comparison result, determining the fault diagnosis result of the hydraulic turbine unit to be tested; The time domain feature information includes at least one of a mean, a standard deviation, a variance and a kurtosis.

4. The method for diagnosing faults of a hydraulic turbine unit according to claim 1, characterized in that: The first acoustic emission signal data when the turbine unit to be tested is running is subjected to noise reduction preprocessing to obtain the second acoustic emission signal data after noise reduction processing, including: The first acoustic emission signal data is subjected to noise reduction preprocessing by using a singular value decomposition algorithm to obtain the second acoustic emission signal data.

5. The method for diagnosing faults of a hydraulic turbine unit according to claim 4, characterized in that: The method of performing noise reduction preprocessing on the first acoustic emission signal data by using a singular value decomposition algorithm to obtain the second acoustic emission signal data includes: Based on the first acoustic emission signal data, determining a corresponding Toeplitz matrix; Performing singular value decomposition on the Toeplitz matrix to determine the transpose of a left singular matrix, a singular value matrix, and a right singular matrix of the Toeplitz matrix; According to the principle of retaining the maximum singular value, retaining a target number of singular values ​​from the singular value matrix, and determining a new singular value matrix based on the target number of singular values; Determine a denoised signal matrix based on the left singular matrix, the new singular value matrix and the transpose of the right singular matrix; The second acoustic emission signal data is obtained according to each element on the sub-diagonal of the denoised signal matrix.

6. The method for diagnosing faults of a hydraulic turbine unit according to any one of claims 1 to 5, characterized in that: Before performing denoising processing using the second acoustic emission signal data and the reference frequency domain feature information to determine the fault diagnosis time domain signal data, the method further includes: According to the sampling frequency of the first acoustic emission signal data, collecting the third acoustic emission signal data of the tested hydraulic turbine unit under the fault-free working condition; Performing noise reduction preprocessing on the third acoustic emission signal data to obtain fourth acoustic emission signal data after noise reduction processing; Performing discrete Fourier transform on the fourth acoustic emission signal data to obtain the reference frequency domain feature information.

7. A hydraulic turbine fault diagnosis device, characterized in that: include: A noise reduction preprocessing module is used to perform noise reduction preprocessing on the first acoustic emission signal data when the turbine unit to be tested is running, so as to obtain the second acoustic emission signal data after noise reduction preprocessing; A denoising processing module, configured to perform denoising processing using the second acoustic emission signal data and reference frequency domain characteristic information to determine fault diagnosis time domain signal data; the reference frequency domain characteristic information is determined based on the third acoustic emission signal data of the turbine unit to be tested under a fault-free working condition; A fault diagnosis module is used to determine the fault diagnosis result of the hydraulic turbine group to be tested based on the time domain characteristic information of the fault diagnosis time domain signal data.

8. An electronic device, characterized in that: include: at least one memory for storing a computer program; At least one processor is used to execute the program stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program runs on a processor, the processor is caused to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that When the computer program product runs on a processor, the processor is caused to execute the method according to any one of claims 1 to 6.

Images