Hydroelectric generating set fault detection method based on acoustics and vibration sensing principle and related equipment

Abstract

The embodiment of the invention provides a hydroelectric generating set fault detection method based on acoustics and vibration sensing principles and related equipment, and relates to the technical field of hydroelectric generating set fault detection, the method comprises the following steps: obtaining filter parameters based on historical working condition data, and setting band-pass filters under different working conditions based on the filter parameters; selecting a band-pass filter of a corresponding working condition based on the current working condition parameter; collecting current acoustic data and vibration data, and filtering noise based on the band-pass filter to obtain real acoustic data and vibration data; generating an acoustic sample library of faults under different working conditions based on historically collected fault sample data; and blade deformation and hydrodynamic anomaly detection is carried out based on the real acoustic data and vibration data and the acoustic sample library. According to the technical scheme, data acquisition can be more accurate; and meanwhile, the state of the hydroelectric generating set can be evaluated more comprehensively and accurately, so that the accuracy and efficiency of fault diagnosis are improved.

Description

Technical Field

[0001] The present application relates to the technical field of hydropower unit fault detection, and in particular to a hydropower unit fault detection method and related equipment based on acoustic and vibration sensing principles. Background Art

[0002] Acoustic and vibration data collection and analysis is a key part of the maintenance and condition monitoring of hydropower units. These data can be used to effectively monitor and predict the operating status and potential failures of the units. However, in practical applications, this process faces a variety of technical challenges, especially due to the complexity of the operating environment of hydropower units.

[0003] The background noise generated by hydropower units during operation is usually nonlinear, including turbulent noise of water flow, mechanical friction noise, etc. The nonlinear characteristics of these noise sources make it difficult for traditional acoustic data processing methods to accurately extract useful fault feature information. Although traditional bandpass filters can remove some irrelevant frequency signals according to a preset frequency range, their fixed frequency range limits their applicability and flexibility when facing complex and changing noise environments.

[0004] In order to overcome the shortcomings of traditional bandpass filters, adaptive filters have been proposed and applied to dynamically changing background noise. Adaptive filters can better handle nonlinear and non-stationary noise by adjusting their parameters to adapt to the statistical characteristics of the input signal. However, adaptive filters require complex calculation processes to continuously update their parameters, which may cause delays in real-time data processing. They are also very sensitive to the initial settings of the filter. Once the settings are improper, their performance will be greatly reduced.

[0005] In practical applications, using acoustic data or vibration data alone for fault diagnosis often has certain limitations. Therefore, it is necessary to fuse the signals of the two to more effectively conduct online testing, fault identification and early warning of the unit's operating status. Summary of the invention

[0006] The embodiments of the present application provide a hydropower unit fault detection method and related equipment based on acoustic and vibration sensing principles to solve the technical problems existing in the prior art.

[0007] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0008] According to a first aspect of an embodiment of the present application, a method for detecting faults of a hydropower unit based on acoustic and vibration sensing principles is provided, comprising:

[0009] Acquire filter parameters based on historical operating condition data, and set bandpass filters under different operating conditions based on the filter parameters;

[0010] Select a bandpass filter corresponding to the working condition based on the current working condition parameters;

[0011] Collecting current acoustic data and vibration data, and filtering out noise based on the bandpass filter to obtain real acoustic data and vibration data;

[0012] Generate an acoustic sample library of faults under different working conditions based on historically collected fault sample data;

[0013] Blade deformation and fluid dynamic anomaly detection are performed based on the real acoustic data and vibration data and the acoustic sample library.

[0014] In some embodiments of the present application, based on the above solution, the step of obtaining filter parameters based on historical operating condition data includes:

[0015] Obtain historical operating data;

[0016] Different historical operating condition data are input into the neural network model for training to obtain the filter parameters under different operating conditions.

[0017] In some embodiments of the present application, based on the above solution, the generation of an acoustic sample library of faults under different working conditions based on historically collected fault sample data includes:

[0018] According to the historically collected sound and vibration fault sample data under different loads and water heads, the autocorrelation spectrum method is used to accumulate the fault sample data under different working conditions and preprocess them to form an acoustic sample library of faults under different working conditions.

[0019] In some embodiments of the present application, based on the above-mentioned solution, blade deformation detection is performed based on the real acoustic data and vibration data and the acoustic sample library, including:

[0020] Based on the acoustic data and vibration data related to the blade deformation fault in the acoustic sample library, a blade deformation fault identification algorithm model is constructed;

[0021] Acoustic data and vibration data are detected based on the blade deformation fault identification algorithm model.

[0022] In some embodiments of the present application, based on the aforementioned solution, the blade deformation fault identification algorithm model is constructed based on the acoustic data and vibration data related to the blade deformation fault in the acoustic sample library, including:

[0023] The frequency domain of the acoustic data related to blade deformation failure in the acoustic sample library is integrated to calculate the energy and then the logarithm is taken to extract the overall trend of energy Ψ(S(t));

[0024] Perform exponential square inverse processing on the vibration data related to blade deformation failure in the acoustic sample library to obtain processed data Ω(V(t));

[0025] Using a deep neural network, the blade deformation fault identification algorithm model is constructed based on the overall trend Ψ(S(t)) and the processed data Ω(V(t));

[0026]

[0027] In formula (1), DNN represents a deep neural network, and α and β represent the weights of acoustic data and vibration data in the final feature vector, respectively.

[0028] In some embodiments of the present application, based on the aforementioned solution, fluid dynamic anomaly detection is performed based on the real acoustic data and vibration data and the acoustic sample library, including:

[0029] Based on the acoustic data and vibration data related to the blade deformation failure in the acoustic sample library, a fluid dynamics anomaly algorithm model is constructed;

[0030] The acoustic data and the vibration data are detected based on the fluid dynamics anomaly algorithm model.

[0031] In some embodiments of the present application, based on the aforementioned solution, the fluid dynamics anomaly algorithm model is constructed based on the acoustic data and vibration data related to the blade deformation fault in the acoustic sample library, including:

[0032] The energy of the acoustic data related to the blade deformation fault in the acoustic sample library is subjected to exponential decay of the weighted spectral mean, and the processing function F(S) of the abnormal acoustic data of fluid dynamics is obtained;

[0033] The energy of the vibration data related to blade deformation failure in the acoustic sample library is summed and processed by natural logarithm to obtain the processing function G(V) of fluid dynamic abnormal vibration data;

[0034] The processing function F(S) and the processing function G(V) are classified using random forest to obtain a fluid dynamics anomaly algorithm model;

[0035]

[0036] In formula (2), S(f) represents the expression of acoustic data in the frequency domain and time domain, V(t n ) is the expression of vibration data in the frequency domain and time domain, |S(f)|, |V(tn )| represents the amplitude of acoustic and vibration data, N represents the total number of vibration data sampling points, and RF represents the random forest model.

[0037] In some embodiments of the present application, based on the above solution, it also includes:

[0038] Set fault alarm threshold;

[0039] When the detection result exceeds the fault alarm threshold, a fault alarm is issued.

[0040] According to a second aspect of an embodiment of the present application, a hydropower unit fault detection device based on acoustic and vibration sensing principles is provided, comprising:

[0041] An acquisition unit, used for acquiring filter parameters based on historical operating condition data;

[0042] A setting unit, used for setting the bandpass filter under different working conditions based on the filter parameters;

[0043] A selection unit, used for selecting a bandpass filter corresponding to the working condition based on the current working condition parameters;

[0044] A collection unit, used for collecting current acoustic data and vibration data;

[0045] A noise processing unit, used for filtering out noise based on the bandpass filter to obtain real acoustic data and vibration data;

[0046] A generation unit, used for generating an acoustic sample library of faults under different working conditions based on historically collected fault sample data;

[0047] A detection unit is used to detect blade deformation and fluid dynamic anomaly based on the real acoustic data and vibration data and the acoustic sample library.

[0048] According to a third aspect of an embodiment of the present application, a computer-readable storage medium is provided, wherein the storage medium stores computer instructions, and when the computer instructions are executed on a computer, the computer executes the method according to the first aspect.

[0049] According to a fourth aspect of the embodiments of the present application, there is provided an electronic device, including: a memory and a processor;

[0050] The memory is used to store computer instructions;

[0051] The processor is used to call the computer instructions stored in the memory so that the electronic device executes the method as described in the first aspect.

[0052] The technical solution of the present application adopts a neural network algorithm to dynamically select and adjust the parameters of the bandpass filter according to the different operating conditions of the hydropower unit, so as to make data collection more accurate; at the same time, a method for combined analysis of acoustic and vibration data is proposed. By integrating various types of data, the status of the hydropower unit can be evaluated more comprehensively and accurately, thereby improving the accuracy and efficiency of fault diagnosis.

[0053] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0055] Figure 1 A schematic flow chart of a method for detecting faults of a hydropower unit based on acoustic and vibration sensing principles according to an embodiment of the present application is shown;

[0056] Figure 2 A block diagram of a hydropower unit fault detection device based on acoustic and vibration sensing principles according to an embodiment of the present application is shown;

[0057] Figure 3 A block diagram of an electronic device according to an embodiment of the present application is shown;

[0058] Figure 4 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0059] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0060] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0061] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0062] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0063] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] Some embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.

[0065] See also Figure 1 , shows a flow chart of a method for detecting faults of a hydropower unit based on acoustic and vibration sensing principles according to an embodiment of the present application.

[0066] like Figure 1 As shown, a method for detecting faults of hydropower units based on acoustic and vibration sensing principles is demonstrated, which specifically includes steps S100 to S500.

[0067] refer to Figure 1 , step S100, obtaining filter parameters based on historical operating condition data, and setting a bandpass filter under different operating conditions based on the filter parameters.

[0068] It should be noted that the operating data includes unit load, working water head, water pressure, flow data, etc.

[0069] It should be noted that the bandpass filter in this embodiment includes: an acoustic data bandpass filter and a vibration data bandpass filter.

[0070] In some feasible embodiments, based on the above solution, the step of obtaining filter parameters based on historical operating condition data includes:

[0071] Obtain historical operating data;

[0072] Different historical operating condition data are input into the neural network model for training to obtain the filter parameters under different operating conditions.

[0073] It should be noted that the filter parameters include an upper cutoff frequency and a lower cutoff frequency.

[0074] It should be noted that before the historical operating condition data is input into the neural network model, it is necessary to perform feature conversion on the historical operating condition data to obtain the feature conversion function T(x), as shown below:

[0075]

[0076] The characteristic conversion function T(x) is mainly used to enhance the nonlinear representation capability of the input data, including using logarithmic transformation on the load to increase its dynamic range, using square root transformation on the speed to smooth its changes, performing exponential decay on the water pressure to reduce its impact, and using natural logarithmic transformation on the water flow.

[0077] The historical operating condition data after feature conversion is loaded into the neural network model for training to obtain the corresponding filter parameters, which are expressed as:

[0078] z1=ReLU(W1·z0+b1); (2)

[0079] z2=ReLU(W2·z1+b2); (3)

[0080] y=O(z2)=Softmax(W3·z2+b3); (4)

[0081] f high ,f low =1000·y; (5)

[0082] In formula (1) to formula (5), x L 、x R Expressed as the load and speed data of the turbine, x P 、x QIt is represented as water pressure and water flow data, W1, W2, W3 are weights of the corresponding layers, b1, b2, b3 are biases of the corresponding layers, ReLU and Softmax are activation functions, and f high Expressed as the upper cutoff frequency, f low Expressed as the lower cutoff frequency.

[0083] Continue to refer Figure 1 , step S200, selecting a bandpass filter corresponding to the working condition based on the current working condition parameters.

[0084] It can be understood that the bandpass filter set in step S100 corresponds to different working conditions, so when the current working condition parameters are known, the bandpass filter corresponding to the "working condition" can be selected according to the "working condition" of the working condition parameters, including acoustic data bandpass filters and vibration data bandpass filters.

[0085] It should be noted that, before step S200, the current operating condition parameters need to be obtained.

[0086] Continue to refer Figure 1 , step S300, collecting current acoustic data and vibration data, and filtering out noise based on the bandpass filter to obtain real acoustic data and vibration data.

[0087] It should be noted that, in step S300, a collector may be used to collect current acoustic data and vibration data.

[0088] It should be noted that in step S200, the obtained bandpass filter includes an acoustic data bandpass filter and a vibration data bandpass filter. Therefore, in this step S300, the acoustic data bandpass filter is used to filter out the noise of the current acoustic data, and the vibration data bandpass filter is used to filter out the noise of the current vibration data, and finally the real acoustic data and the real vibration data are obtained respectively.

[0089] Continue to refer Figure 1 , step S400, generating an acoustic sample library of faults under different working conditions based on the historically collected fault sample data.

[0090] In some feasible embodiments, based on the above solution, the generation of an acoustic sample library of faults under different working conditions based on historically collected fault sample data includes:

[0091] According to the historically collected sound and vibration fault sample data under different loads and water heads, the autocorrelation spectrum method is used to accumulate the fault sample data under different working conditions and preprocess them to form an acoustic sample library of faults under different working conditions.

[0092] Continue to refer Figure 1, step S500, performing blade deformation and fluid dynamic anomaly detection based on the real acoustic data and vibration data and the acoustic sample library.

[0093] In some feasible embodiments, based on the above solution, blade deformation detection is performed based on the real acoustic data and vibration data and the acoustic sample library, including:

[0094] Based on the acoustic data and vibration data related to the blade deformation fault in the acoustic sample library, a blade deformation fault identification algorithm model is constructed;

[0095] Acoustic data and vibration data are detected based on the blade deformation fault identification algorithm model.

[0096] It should be noted that, in this embodiment, the detection of acoustic data and vibration data based on the blade deformation fault identification algorithm model is mainly manifested in inputting the acoustic data and vibration data into the blade deformation fault identification algorithm model for calculation, and finally obtaining a blade deformation score.

[0097] In some feasible embodiments, based on the above solution, the blade deformation fault identification algorithm model is constructed based on the acoustic data and vibration data related to the blade deformation fault in the acoustic sample library, including:

[0098] The frequency domain of the acoustic data related to blade deformation failure in the acoustic sample library is integrated to calculate the energy and then the logarithm is taken to extract the overall trend of energy Ψ(S(t)), which is expressed as:

[0099]

[0100] The vibration data related to blade deformation failure in the acoustic sample library is subjected to exponential square inverse processing to obtain the processed data Ω(V(t)), which is expressed as:

[0101]

[0102] Using a deep neural network, the blade deformation fault identification algorithm model is constructed based on the overall trend Ψ(S(t)) and the processed data Ω(V(t));

[0103]

[0104] In formula (6) to formula (8), S(f) represents the expression of acoustic data in the frequency domain and time domain, V(t n ) is the expression of vibration data in the frequency domain and time domain, |S(f)|, |V(t n)| represents the amplitude of acoustic and vibration data, N represents the total number of vibration data sampling points, DNN represents deep neural network, α and β represent the weights of acoustic data and vibration data in the final feature vector respectively.

[0105] In some feasible embodiments, based on the above scheme, fluid dynamic anomaly detection is performed based on the real acoustic data and vibration data and the acoustic sample library, including:

[0106] Based on the acoustic data and vibration data related to the blade deformation failure in the acoustic sample library, a fluid dynamics anomaly algorithm model is constructed;

[0107] The acoustic data and the vibration data are detected based on the fluid dynamics anomaly algorithm model.

[0108] It should be noted that, in this embodiment, the detection of acoustic data and vibration data based on the fluid dynamics anomaly algorithm model is mainly performed by inputting the acoustic data and vibration data into the fluid dynamics anomaly algorithm model for calculation, and finally obtaining a fluid dynamics anomaly score.

[0109] In some feasible embodiments, based on the above solution, the fluid dynamics anomaly algorithm model is constructed based on the acoustic data and vibration data related to the blade deformation failure in the acoustic sample library, including:

[0110] The energy of the acoustic data related to blade deformation faults in the acoustic sample library is subjected to exponential decay of the weighted spectral mean, and the processing function F(S) of the abnormal acoustic data of fluid dynamics is obtained, which is expressed as:

[0111]

[0112] The energy of the vibration data related to blade deformation failure in the acoustic sample library is summed and processed by natural logarithm to obtain the processing function G(V) of abnormal fluid dynamic vibration data, which is expressed as:

[0113]

[0114] The processing function F(S) and the processing function G(V) are classified using random forest to obtain a fluid dynamics anomaly algorithm model;

[0115]

[0116] In formula (9) to formula (11), S(f) represents the expression of acoustic data in the frequency domain and time domain, V(t n ) is the expression of vibration data in the frequency domain and time domain, |S(f)|, |V(t n)| represents the amplitude of acoustic and vibration data, N represents the total number of vibration data sampling points, and RF represents the random forest model.

[0117] In some feasible embodiments, based on the above solution, it also includes:

[0118] Step S600, setting a fault alarm threshold, and when the detection result exceeds the fault alarm threshold, a fault alarm is issued.

[0119] For example, the fault alarm threshold is set to 0.5, and when the blade deformation score or the fluid dynamics abnormality score exceeds the fault alarm threshold, a corresponding type of fault alarm is issued.

[0120] In summary, this application uses a neural network algorithm to screen out filter parameters under different working conditions, sets bandpass filters under different working conditions based on the filter parameters under the working conditions, and then selects the bandpass filter for the corresponding working condition according to the specific actual situation. It not only solves the problem that the traditional bandpass filter has a fixed frequency range that limits its applicability and flexibility when facing a complex and changing noise environment, but also the application determines the corresponding filter parameters in advance according to different working conditions before real-time detection, and solves the problem that the adaptive filter requires a complex calculation process to continuously update its parameters, resulting in delays in real-time data processing.

[0121] Moreover, unlike using a single type of data for fault diagnosis, this application uses both acoustic data and vibration data, processes and analyzes data through intelligent algorithms, and performs fault diagnosis more comprehensively. Acoustic data is more sensitive to certain types of faults, such as cavitation or fluid dynamic anomalies, while vibration data more clearly shows mechanical structural problems, such as bearing damage or partial structural detachment of the runner chamber. Combining these two data types can improve the coverage and accuracy of fault detection.

[0122] Below is a specific application process of this method:

[0123] Test and verification: The data comes from the test of a certain model model. Various simulated working conditions are set up on the model, such as normal operation, blade deformation, and fluid dynamics abnormality.

[0124] Working environment: Set up a variety of simulated working environments, such as normal operation, partial structure detachment of the runner chamber, and abnormal fluid dynamics. The acoustic test data of this project comes from the turbine model test bench. By simulating normal conditions and specific fault conditions, a test environment for normal operation of the model machine, blade deformation, and abnormal fluid dynamics is built to obtain corresponding acoustic data under different operating conditions. Its different working conditions are simulated by changing key parameters such as guide vane angle, head, and flow. The guide vane angle is manually adjusted by the guide vane opening angle, thereby changing the unit flow and unit output. Blade deformation is achieved by locally deforming the runner blades. The deformation part is the water outlet edge, and the deformation part area is 1% and 2% of the projection area of ​​other blades to simulate the size of the blade deformation area.

[0125] Data Collection: Collect acoustic and vibration data under each operating condition.

[0126] Filter settings: A neural network algorithm is used to determine the parameters of the bandpass filter for each operating condition, and a traditional static bandpass filter and an adaptive filter are set for comparison.

[0127] Test objectives:

[0128] Compare the fault detection rates of various methods under different working conditions.

[0129] Analyze the real-time data processing capabilities and latency of each method.

[0130] Evaluate the comprehensiveness and accuracy of fault detection.

[0131] The experimental data obtained are compared, as shown in Table 1:

[0132] Table 1 Experimental data comparison table

[0133]

[0134]

[0135] The fault detection rate of the method of the present invention under all working conditions is generally higher than that of the traditional static bandpass filter and adaptive filter. Under the "blade deformation" working condition, the fault detection rate of the method of the present invention is as high as 95%, while the traditional static bandpass filter and adaptive filter are 65% and 70% respectively. This shows that the present invention can identify faults more effectively, which is because the present invention can comprehensively utilize acoustic and vibration data, as well as filter parameters optimized by neural network algorithms, to improve the ability to capture fault signals.

[0136] The present invention also shows superiority in response time, wherein the response time of all working conditions is 0.1 second, which is much faster than 0.5 seconds of the adaptive filter and 0.2 seconds of the traditional static bandpass filter.

[0137] In terms of accuracy, the present invention is also superior to the traditional method. In the "fluid power abnormality" working condition, the accuracy of the present invention is 93%, while the accuracy of the traditional method is only 75% (static bandpass filter) and 79% (adaptive filter).

[0138] Unlike traditional methods that usually rely on a single type of data, the present invention can capture fault characteristics from multiple dimensions by using both acoustic and vibration data, increasing the comprehensiveness of fault diagnosis. The comprehensive analysis method of the present invention provides richer information and helps to accurately judge complex fault conditions, especially when a single data source may not be sufficient to provide a complete fault view.

[0139] The following describes an embodiment of the device of the present application, which can be used to implement a method for detecting hydropower unit faults based on acoustic and vibration sensing principles in the above embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the above method of the present application.

[0140] Reference Figure 2 As shown, according to an embodiment of the present application, a hydropower unit fault detection device 200 based on acoustic and vibration sensing principles includes:

[0141] An acquisition unit 201 is used to acquire filter parameters based on historical operating condition data;

[0142] A setting unit 202, configured to set the bandpass filter under different working conditions based on the filter parameters;

[0143] A selection unit 203 is used to select a bandpass filter corresponding to the working condition based on the current working condition parameters;

[0144] A collection unit 204, used to collect current acoustic data and vibration data;

[0145] A noise processing unit 205, configured to filter out noise based on the bandpass filter to obtain real acoustic data and vibration data;

[0146] A generating unit 206, configured to generate an acoustic sample library of faults under different working conditions based on the historically collected fault sample data;

[0147] The detection unit 207 is used to perform blade deformation and fluid dynamic anomaly detection based on the real acoustic data and vibration data and the acoustic sample library.

[0148] like Figure 3As shown, an embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of the above-mentioned method for detecting faults of hydropower units based on acoustic and vibration sensing principles are implemented.

[0149] Since the electronic device introduced in this embodiment is a device used to implement a hydropower unit fault detection device based on acoustic and vibration sensing principles in the embodiment of the present application, based on the method introduced in the embodiment of the present application, technical personnel in this field can understand the specific implementation mode of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is not introduced in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of the present application is within the scope of protection of this application.

[0150] During the specific implementation process, when the computer program 311 is executed by a processor, any implementation method in the embodiments corresponding to the first aspect can be implemented.

[0151] Figure 4 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown.

[0152] It should be noted that Figure 4 The computer system 400 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0153] like Figure 4 As shown, the computer system 400 includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 402 or the program loaded from the storage part 408 to the random access memory (RAM) 403, such as executing the method described in the above embodiment. In the RAM 403, various programs and data required for system operation are also stored. The CPU 401, ROM 402 and RAM 403 are connected to each other through a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0154] The following components are connected to the I / O interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed so that a computer program read therefrom is installed into the storage section 408 as needed.

[0155] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 409, and / or installed from a removable medium 411. When the computer program is executed by a central processing unit (CPU) 401, various functions defined in the system of the present application are executed.

[0156] It should be noted that the computer-readable medium shown in the embodiment of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, - but not limited to - an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by an instruction execution system, device or device or used in combination with it. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0157] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0158] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0159] As another aspect, the present application further provides a computer program product or a computer program, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes a hydropower unit fault detection method based on acoustic and vibration sensing principles described in the above embodiment.

[0160] As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements a hydropower unit fault detection method based on acoustic and vibration sensing principles described in the above embodiment.

[0161] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0162] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation methods of the present application.

[0163] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the embodiments disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary techniques in the art that are not disclosed in the present application. It should be understood that the present application is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for detecting faults of hydropower units based on acoustic and vibration sensing principles, characterized in that: include: Acquire filter parameters based on historical operating condition data, and set bandpass filters under different operating conditions based on the filter parameters; Select a bandpass filter corresponding to the working condition based on the current working condition parameters; Collecting current acoustic data and vibration data, and filtering out noise based on the bandpass filter to obtain real acoustic data and vibration data; Generate an acoustic sample library of faults under different working conditions based on historically collected fault sample data; Blade deformation and fluid dynamic anomaly detection are performed based on the real acoustic data and vibration data and the acoustic sample library.

2. The method according to claim 1, characterized in that The obtaining of filter parameters based on historical operating condition data includes: Obtain historical operating data; Different historical operating condition data are input into the neural network model for training to obtain the filter parameters under different operating conditions.

3. The method according to claim 1, characterized in that The method of generating an acoustic sample library of faults under different working conditions based on the historically collected fault sample data includes: According to the historically collected sound and vibration fault sample data under different loads and water heads, the autocorrelation spectrum method is used to accumulate the fault sample data under different working conditions and preprocess them to form an acoustic sample library of faults under different working conditions.

4. The method according to any one of claims 1 to 3, characterized in that: Performing blade deformation detection based on the real acoustic data and vibration data and the acoustic sample library includes: Based on the acoustic data and vibration data related to the blade deformation fault in the acoustic sample library, a blade deformation fault identification algorithm model is constructed; Acoustic data and vibration data are detected based on the blade deformation fault identification algorithm model.

5. The method according to claim 4, characterized in that The step of constructing a blade deformation fault identification algorithm model based on the acoustic data and vibration data related to the blade deformation fault in the acoustic sample library includes: The frequency domain of the acoustic data related to blade deformation failure in the acoustic sample library is integrated to calculate the energy and then the logarithm is taken to extract the overall trend of energy Ψ(S(t)); Perform exponential square inverse processing on the vibration data related to blade deformation failure in the acoustic sample library to obtain processed data Ω(V(t)); Using a deep neural network, the blade deformation fault identification algorithm model is constructed based on the overall trend Ψ(S(t)) and the processed data Ω(V(t)); In formula (1), DNN represents a deep neural network, and α and β represent the weights of acoustic data and vibration data in the final feature vector, respectively.

6. The method according to any one of claims 1 to 3, characterized in that: Performing fluid dynamic anomaly detection based on the real acoustic data and vibration data and the acoustic sample library includes: Based on the acoustic data and vibration data related to the blade deformation failure in the acoustic sample library, a fluid dynamics anomaly algorithm model is constructed; The acoustic data and the vibration data are detected based on the fluid dynamics anomaly algorithm model.

7. The method according to claim 6, characterized in that The method of constructing a fluid dynamics anomaly algorithm model based on the acoustic data and vibration data related to the blade deformation fault in the acoustic sample library includes: The energy of the acoustic data related to the blade deformation fault in the acoustic sample library is subjected to exponential decay of the weighted spectral mean, and the processing function F(S) of the abnormal acoustic data of fluid dynamics is obtained; The energy of the vibration data related to blade deformation failure in the acoustic sample library is summed and processed by natural logarithm to obtain the processing function G(V) of fluid dynamic abnormal vibration data; The processing function F(S) and the processing function G(V) are classified using random forest to obtain a fluid dynamics anomaly algorithm model; In formula (2), S(f) represents the expression of acoustic data in the frequency domain and time domain, V(t n ) is the expression of vibration data in the frequency domain and time domain, |S(f)|, |V(t n )| represents the amplitude of acoustic and vibration data, N represents the total number of vibration data sampling points, and RF represents the random forest model.

8. The method according to claim 1, characterized in that Also includes: Set fault alarm threshold; When the detection result exceeds the fault alarm threshold, a fault alarm is issued.

9. A hydropower unit fault detection device based on acoustic and vibration sensing principles, characterized in that: include: An acquisition unit, used for acquiring filter parameters based on historical operating condition data; A setting unit, used for setting the bandpass filter under different working conditions based on the filter parameters; A selection unit, used for selecting a bandpass filter corresponding to the working condition based on the current working condition parameters; A collection unit, used for collecting current acoustic data and vibration data; A noise processing unit, used for filtering out noise based on the bandpass filter to obtain real acoustic data and vibration data; A generation unit, used for generating an acoustic sample library of faults under different working conditions based on historically collected fault sample data; A detection unit is used to detect blade deformation and fluid dynamic anomaly based on the real acoustic data and vibration data and the acoustic sample library.

10. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer instructions; The processor is used to call the computer instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-8.

Images