Water turbine cavitation fault diagnosis method and system based on vibration monitoring

Abstract

The invention discloses a water turbine cavitation fault diagnosis method based on vibration monitoring, and the method comprises the steps: obtaining water turbine signals, carrying out the data combination of the water turbine signals through machine learning, obtaining to-be-matched data, carrying out the preprocessing of the to-be-matched data, and obtaining first data; constructing a machine learning model based on cavitation and cavitation erosion classification by using the first data; according to the built machine learning model based on cavitation and cavitation erosion classification, the operation state of the water turbine is detected, when the cavitation state of the water turbine is abnormal, an early warning signal is sent out, and the early warning signal is converted into a configuration output signal of a water turbine vibration detection protection module; the intelligent and automatic monitoring mode can reduce the frequency and intensity of manual inspection and improve the working efficiency.

Description

Technical Field

[0001] The invention relates to the technical field of fault diagnosis, and in particular to a method and system for diagnosing cavitation fault of a water turbine based on oscillation monitoring. Background Art

[0002] Cavitation can lead to a decrease in the mechanical properties of the fluid, causing vibration and noise, and leading to corrosion damage to the flow-through parts. Cavitation erosion in turbines is one of the main reasons for the decrease in turbine energy characteristics and unstable operation. Cavitation (also known as cavitation) damages the flow-through parts of turbines, deteriorates operating conditions, and significantly reduces efficiency and output. When a turbine operates under cavitation conditions, its flow-through parts experience low-frequency and large-amplitude pulsations, causing vibration and noise in the unit and plant. Cavitation has a great impact on turbine operation. The development trend of turbines is to increase the capacity of a single unit and the speed of the unit, reduce the size, and reduce costs, which will further aggravate the cavitation of turbines.

[0003] The current recommended standards for turbines face potential risks, mainly in two categories: 1. When the turbine quality does not reach the negotiated cavitation guarantee value, resulting in losses to turbine users; 2. When the turbine quality exceeds the negotiated cavitation guarantee value, resulting in additional costs to turbine manufacturers. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: the present invention solves the problem that the turbine quality does not reach the negotiated cavitation guarantee value, resulting in losses to the turbine user and the turbine quality exceeds the negotiated cavitation guarantee value, resulting in extra costs to the turbine manufacturer.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for diagnosing cavitation faults of a hydraulic turbine based on oscillation monitoring, comprising:

[0008] Acquire a turbine signal, perform data combination on the turbine signal through machine learning to obtain data to be matched, and pre-process the data to be matched to obtain first data;

[0009] Using the first data to construct a machine learning model based on cavitation classification;

[0010] According to the machine learning model based on cavitation erosion classification, the operating status of the turbine is detected, and a warning signal is issued when the cavitation state of the turbine is abnormal, and the warning signal is converted into a configuration output signal of the turbine swing detection and protection module.

[0011] As a preferred solution of the method for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring described in the present invention, obtaining the hydraulic turbine signal includes:

[0012] According to different working conditions of the turbine, a synchronous trigger mechanism with equal time intervals is set to synchronously collect turbine signals;

[0013] The turbine signals include high-frequency cavitation sound signals, low-frequency vibration signals, low-frequency displacement signals and low-frequency pressure pulsation signals.

[0014] As a preferred solution of the method for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring according to the present invention, the first data includes:

[0015] Perform data preprocessing on the data to be matched to obtain cleaned data, i.e., first data;

[0016] Extract features from the original signal of the turbine in the first data to obtain feature data;

[0017] A label is assigned to each data sample in the first data, wherein the label is whether it belongs to the cavitation erosion category.

[0018] As a preferred solution of the turbine cavitation fault diagnosis method based on oscillation monitoring described in the present invention, wherein: using the first data to construct a machine learning model based on cavitation erosion classification includes:

[0019] Divide the data set corresponding to the cleaned data into a training set, a validation set, and a test set;

[0020] Use the training set to train the machine learning model, and use the validation set to adjust the hyperparameters of the machine learning model;

[0021] Use the test set to evaluate the performance of the model and optimize the structure of the machine learning model based on the evaluation results.

[0022] As a preferred solution of the turbine cavitation fault diagnosis method based on oscillation monitoring described in the present invention, the machine learning model based on cavitation erosion classification includes:

[0023] Determine the data interface between the turbine swing monitoring and protection module based on the data requirements of the machine learning model;

[0024] Configure the corresponding data communication, select the computing environment for deployment to obtain the target environment, serialize the trained machine learning model, and load and deploy it in the target environment;

[0025] The machine learning model is integrated into the turbine swing monitoring and protection module, and the integrated turbine swing monitoring and protection module is tested using actual data to obtain accurate data.

[0026] As a preferred solution of the method for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring described in the present invention, the detection of the operating state of the hydraulic turbine includes:

[0027] The preprocessed data is fed into a machine learning model to perform real-time analysis of turbine cavitation status;

[0028] The machine learning model determines whether the cavitation state of the turbine is abnormal based on the input data;

[0029] If the machine learning model detects an abnormal cavitation state in the turbine, an early warning mechanism is triggered.

[0030] As a preferred solution of the turbine cavitation fault diagnosis method based on oscillation monitoring described in the present invention, the early warning mechanism includes:

[0031] If an abnormal cavitation state of the turbine is detected, an early warning signal is generated;

[0032] The warning signal can be a data code or a text message.

[0033] In a second aspect, the present invention provides a system for diagnosing cavitation faults of a hydraulic turbine based on oscillation monitoring, comprising:

[0034] A preprocessing module, used for acquiring a turbine signal, combining the turbine signal through machine learning to obtain data to be matched, and preprocessing the data to be matched to obtain first data;

[0035] A model building module, used to build a machine learning model based on cavitation classification using the first data;

[0036] The detection module is used to detect the operating status of the turbine according to the machine learning model based on cavitation and erosion classification, send out an early warning signal when the cavitation state of the turbine is abnormal, and convert the early warning signal into a configuration output signal of the turbine swing detection and protection module.

[0037] In a third aspect, the present invention provides a computing device, comprising:

[0038] Memory and processor;

[0039] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the turbine cavitation fault diagnosis method based on vibration monitoring are implemented.

[0040] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the turbine cavitation fault diagnosis method based on oscillation monitoring.

[0041] Beneficial effects of the present invention: The turbine cavitation fault diagnosis method based on swing monitoring proposed by the present invention sets a synchronous trigger mechanism with equal time intervals according to different working conditions of the turbine, ensuring the synchronous collection of high-frequency cavitation sound signals, low-frequency vibration signals, low-frequency displacement signals and low-frequency pressure pulsation signals. This synchronization helps to improve the accuracy of data analysis and model prediction. Not only traditional swing monitoring data is considered, but also multi-source information such as high-frequency cavitation sound signals, low-frequency vibration signals, low-frequency displacement signals and low-frequency pressure pulsation signals are integrated. This method of integrating multi-source information can more comprehensively reflect the operating status of the turbine and improve the accuracy of prediction. Through unsupervised learning or supervised learning methods, data matching and model construction are performed on multi-source signals, and the cavitation and erosion state of the turbine can be automatically identified and classified. The application of the machine learning model enables the module to have the ability of self-learning and self-adaptation, and can cope with complex and changeable working conditions. When the machine learning model G detects that the cavitation state of the turbine is abnormal, the module can send out an early warning signal in advance. This early warning function can timely discover potential problems, avoid fault expansion, and reduce maintenance costs and production losses. The early warning signal is converted into the configuration output signal of the turbine oscillation monitoring and protection module, which is convenient for interaction with other modules or operators. This integrated design improves the compatibility and scalability of the module, and facilitates subsequent functional upgrades and expansions. It realizes intelligent and automated monitoring of the turbine operation status. Through real-time analysis of the machine learning model, the module can automatically determine the cavitation status of the turbine and take corresponding measures. This intelligent and automated monitoring method can reduce the frequency and intensity of manual inspections and improve work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0043] Figure 1 The present invention provides an overall flow chart of a method for diagnosing turbine cavitation faults based on oscillation monitoring. DETAILED DESCRIPTION

[0044] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0045] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0046] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0047] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0048] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0049] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0050] Example 1

[0051] Reference Figure 1, which is an embodiment of the present invention, provides a method for diagnosing cavitation faults of a hydraulic turbine based on oscillation monitoring, comprising:

[0052] S100: Acquire a turbine signal, perform data combination on the turbine signal through machine learning to obtain data to be matched, and pre-process the data to be matched to obtain first data;

[0053] In the embodiment of the present application, a synchronous trigger mechanism with equal time intervals is set according to different working conditions of the turbine to synchronously collect turbine signals;

[0054] The turbine signals include high-frequency cavitation sound signals, low-frequency vibration signals, low-frequency displacement signals and low-frequency pressure pulsation signals.

[0055] Specifically, the equal time intervals are set according to specific circumstances, and a series of trigger points with equal time intervals are pre-set according to different working conditions of the turbine. These trigger points can be automatically triggered based on changes in operating parameters, or can be triggered by manual settings.

[0056] High-frequency cavitation sound signals: Use appropriate ultrasonic sensors to capture the high-frequency cavitation sound signals generated by the turbine under specific operating conditions. Such signals are usually related to the cavitation phenomenon and can provide information on the fluid dynamic characteristics inside the turbine.

[0057] Low-frequency vibration signals: Low-frequency vibration sensors are used to monitor the vibration of the turbine. These sensors can be installed in key parts of the unit, such as the frame, bearings, etc. to capture low-frequency vibration signals. The analysis of vibration signals can provide important information about the unit's operating status, structural health, and potential failures.

[0058] Low-frequency displacement signal: The displacement sensor is used to monitor the displacement changes of key components of the turbine. The displacement signal can reflect the dynamic performance and operation stability of the unit.

[0059] Low frequency pressure pulsation signal: Use pressure sensors to monitor pressure pulsation inside turbines or in tailwater tubes. Pressure pulsation signals can reveal fluid flow instability and possible fluid dynamic problems.

[0060] It should be noted that at each set trigger point, four signals, namely, high-frequency cavitation sound signal, low-frequency vibration signal, low-frequency displacement signal and low-frequency pressure pulsation signal, are collected synchronously, and the collected data are processed and analyzed in real time to evaluate the operating status and performance of the turbine.

[0061] In the embodiment of the present application, data preprocessing is performed on the data to be matched to obtain cleaned data, i.e., first data;

[0062] Extract features from the original signal of the turbine in the first data to obtain feature data;

[0063] Assign a label to each data sample in the first data, where the label indicates whether it belongs to the cavitation erosion category.

[0064] Specifically, if a supervised learning method is used, a label needs to be assigned to each data sample to indicate whether it belongs to the cavitation category;

[0065] If there is enough labeled data, a classification algorithm can be used. If there is no labeled data but you want to find clusters in the data, a clustering algorithm can be used. Unsupervised learning requires clustering and labels are manually assigned based on the clustering results.

[0066] It should be noted that by using unsupervised learning or supervised learning methods, data matching and model building of multi-source signals can automatically identify and classify the cavitation state of the turbine. The application of machine learning models enables the module to have the ability of self-learning and self-adaptation, and can cope with complex and changing working conditions.

[0067] S200: constructing a machine learning model based on cavitation classification using the first data;

[0068] In the embodiment of the present application, the data set corresponding to the cleaned data is divided into a training set, a validation set and a test set;

[0069] Use the training set to train the machine learning model, and use the validation set to adjust the hyperparameters of the machine learning model;

[0070] Use the test set to evaluate the performance of the model and optimize the structure of the machine learning model based on the evaluation results.

[0071] Specifically, use the validation set to adjust the model's hyperparameters and monitor the model's performance on the validation set to avoid overfitting or underfitting. Common evaluation indicators include accuracy, recall, F1 score, confusion matrix, etc. Based on the evaluation results, optimize the model's structure or parameters.

[0072] Deploy the trained model to a real environment for cavitation classification of hydraulic turbines, monitor the performance of the model, and update the model when necessary.

[0073] S300: According to the machine learning model based on cavitation classification, the operating status of the turbine is detected, and a warning signal is issued when the cavitation status of the turbine is abnormal, and the warning signal is converted into a configuration output signal of the turbine swing detection and protection module.

[0074] In the embodiment of the present application, according to the data requirements of the machine learning model, a data interface with the turbine oscillation monitoring and protection module is determined;

[0075] Configure the corresponding data communication, select the computing environment for deployment to obtain the target environment, serialize the trained machine learning model, and load and deploy it in the target environment;

[0076] The machine learning model is integrated into the turbine swing monitoring and protection module, and the integrated turbine swing monitoring and protection module is tested using actual data to obtain accurate data.

[0077] Specifically, the structure, function and data processing flow of the turbine swing monitoring and protection module are studied. The integration feasibility is evaluated, and the hardware, software and interface of the turbine swing monitoring and protection module are analyzed to see whether they meet the integration requirements of the machine learning model.

[0078] The data interface between the machine learning model and the existing system can be API, database connection, etc., and the corresponding data communication code can be written or configured to ensure that the machine learning model can obtain the required data in real time. According to the complexity and performance requirements of the machine learning model, the appropriate computing environment can be selected for deployment, which can be a server, cloud service, etc.

[0079] Write code to integrate the machine learning model into the turbine swing monitoring and protection module to achieve real-time data acquisition, processing, prediction, and result output. Use actual data or simulated data to test the integrated turbine swing monitoring and protection module to ensure that the machine learning model can work properly and output accurate results.

[0080] The performance of the turbine swing monitoring and protection module can be optimized according to the test results, which can be used to adjust the data processing process, optimize the model parameters, etc. The user interface or module interface can be adjusted as needed to facilitate operators to view and use the prediction results of the machine learning model.

[0081] In the embodiment of the present application, the preprocessed data is input into the machine learning model to perform real-time analysis of the cavitation state of the turbine;

[0082] The machine learning model determines whether the cavitation state of the turbine is abnormal based on the input data;

[0083] If the machine learning model detects an abnormal cavitation state in the turbine, an early warning mechanism is triggered.

[0084] Specifically, high-frequency cavitation sound signals, low-frequency vibration signals, low-frequency displacement signals, and low-frequency pressure pulsation signals are acquired in real time, and the real-time acquired data is preprocessed, including noise elimination, data filtering, standardization or normalization, to ensure that the data quality meets the input requirements of the machine learning model, and the preprocessed data is input into the machine learning model to perform real-time analysis of the cavitation state of the turbine. The machine learning model determines whether the cavitation state of the turbine is abnormal based on the input data.

[0085] If the machine learning model detects an abnormal cavitation state of the turbine, the early warning mechanism is triggered. According to the preset threshold, which can be 0.5, the early warning level and corresponding treatment measures are determined.

[0086] In the embodiment of the present application, if an abnormal cavitation state of the turbine is detected, a warning signal is generated;

[0087] The warning signal can be a data code or a text message.

[0088] Specifically, the warning signal may also be an identifier in other forms, and the warning signal contains sufficient information so that the turbine swing monitoring and protection module can clearly identify and understand the current abnormal cavitation state of the turbine.

[0089] It should be noted that the early warning signal is converted into a configuration output signal that can be understood by the turbine swing monitoring and protection module, which involves interacting with the communication interface, data format or specific control logic of the turbine swing monitoring and protection module. The configuration output signal is displayed on the monitoring interface of the turbine swing monitoring and protection module so that the operator can intuitively see the abnormal information of the turbine cavitation state. At the same time, the turbine swing monitoring and protection module can automatically execute corresponding protection measures according to the configuration output signal, such as reducing the turbine speed, adjusting the operating parameters, etc., to prevent further damage or failure.

[0090] The above is a schematic scheme of a method for diagnosing cavitation fault of a water turbine based on oscillation monitoring in this embodiment. It should be noted that the technical scheme of the device for diagnosing cavitation fault of a water turbine based on oscillation monitoring and the technical scheme of the method for diagnosing cavitation fault of a water turbine based on oscillation monitoring are of the same concept. The details of the technical scheme of the device for diagnosing cavitation fault of a water turbine based on oscillation monitoring in this embodiment that are not described in detail can all be referred to the description of the technical scheme of the method for diagnosing cavitation fault of a water turbine based on oscillation monitoring.

[0091] The device for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring in this embodiment includes:

[0092] A preprocessing module, used for acquiring a turbine signal, combining the turbine signal through machine learning to obtain data to be matched, and preprocessing the data to be matched to obtain first data;

[0093] A model building module, used to build a machine learning model based on cavitation classification using the first data;

[0094] The detection module is used to detect the operating status of the turbine according to the machine learning model based on cavitation and erosion classification, send out an early warning signal when the cavitation state of the turbine is abnormal, and convert the early warning signal into a configuration output signal of the turbine swing detection and protection module.

[0095] This embodiment also provides a computing device, which is suitable for diagnosing cavitation faults of a hydraulic turbine based on oscillation monitoring, and includes:

[0096] Memory and processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the turbine cavitation fault diagnosis method based on vibration monitoring as proposed in the above embodiment.

[0097] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for diagnosing cavitation faults of a hydraulic turbine based on oscillation monitoring as proposed in the above embodiment is implemented.

[0098] The storage medium proposed in this embodiment and the method for diagnosing turbine cavitation fault based on swing monitoring proposed in the above embodiment belong to the same inventive concept. The technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0099] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer's floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform the methods of various embodiments of the present invention.

[0100] Example 2

[0101] Referring to Table 1, a method for diagnosing turbine cavitation faults based on oscillation monitoring is provided to verify and illustrate the technical effects used in this method.

[0102] Obtain different monitoring signals, including high-frequency cavitation sound signal Kh, low-frequency vibration signal Kd, low-frequency displacement signal Kw, and low-frequency pressure pulsation signal Km;

[0103] The data under normal and abnormal conditions are compared, as well as the prediction accuracy of the machine learning model under different conditions, as shown in Table 1.

[0104] Table 1 Comparison of experimental data

[0105]

[0106] It can be seen from the above table:

[0107] Vibration signal Kd (μm): represents the low-frequency vibration intensity of the turbine measured at a specific time point.

[0108] Displacement signal Kw (mm): represents the actual displacement of a certain position of the turbine during operation.

[0109] Pressure pulsation signal Km (bar): indicates the pressure fluctuation inside or around the turbine.

[0110] Cavitation sound signal Kh (dB): Indicates the noise intensity caused by cavitation.

[0111] Actual status: This is the actual operating status (normal / abnormal) of the turbine confirmed by other means.

[0112] Predicted status: This is the turbine operating status (normal / abnormal) predicted by the machine learning model after analyzing the current data set.

[0113] Accuracy: This is the proportion of the model’s predictions that match the actual state.

[0114] By comparing the actual state and the predicted state, the accuracy and reliability of the machine learning model can be evaluated. If the model can accurately identify abnormal conditions, it can be used for early warning in practical applications, thereby helping maintenance personnel take timely measures to avoid more serious damage. By collecting and analyzing this data, the model can be further optimized to improve its adaptability and accuracy under different working conditions.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for cavitation fault diagnosis of a hydraulic turbine based on oscillation monitoring, characterized in that: include: Acquire a turbine signal, perform data combination on the turbine signal through machine learning to obtain data to be matched, and pre-process the data to be matched to obtain first data; Using the first data to construct a machine learning model based on cavitation classification; According to the machine learning model based on cavitation erosion classification, the operating status of the turbine is detected, and a warning signal is issued when the cavitation state of the turbine is abnormal, and the warning signal is converted into a configuration output signal of the turbine swing detection and protection module.

2. The method for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring according to claim 1, characterized in that: Acquiring turbine signals includes: According to different working conditions of the turbine, a synchronous trigger mechanism with equal time intervals is set to synchronously collect turbine signals; The turbine signals include high-frequency cavitation sound signals, low-frequency vibration signals, low-frequency displacement signals and low-frequency pressure pulsation signals.

3. The method for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring as claimed in claim 2, characterized in that: The first data includes: Perform data preprocessing on the data to be matched to obtain cleaned data, i.e., first data; Extract features from the original signal of the turbine in the first data to obtain feature data; A label is assigned to each data sample in the first data, wherein the label is whether it belongs to the cavitation erosion category.

4. The method for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring according to claim 2 or 3, characterized in that: Using the first data to construct a machine learning model based on cavitation classification includes: Divide the data set corresponding to the cleaned data into a training set, a validation set, and a test set; Use the training set to train the machine learning model, and use the validation set to adjust the hyperparameters of the machine learning model; Use the test set to evaluate the performance of the model and optimize the structure of the machine learning model based on the evaluation results.

5. The method for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring as claimed in claim 4, characterized in that: Machine learning models based on cavitation classification include: Determine the data interface between the turbine swing monitoring and protection module based on the data requirements of the machine learning model; Configure the corresponding data communication, select the computing environment for deployment to obtain the target environment, serialize the trained machine learning model, and load and deploy it in the target environment; The machine learning model is integrated into the turbine swing monitoring and protection module, and the integrated turbine swing monitoring and protection module is tested using actual data to obtain accurate data.

6. The method for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring as claimed in claim 5, characterized in that: The detection of the operating status of the turbine includes: The preprocessed data is fed into a machine learning model to perform real-time analysis of turbine cavitation status; The machine learning model determines whether the cavitation state of the turbine is abnormal based on the input data; If the machine learning model detects an abnormal cavitation state in the turbine, an early warning mechanism is triggered.

7. The method for diagnosing cavitation fault of a hydraulic turbine based on oscillation monitoring as claimed in claim 6, characterized in that: Early warning mechanisms include: If an abnormal cavitation state of the turbine is detected, an early warning signal is generated; The warning signal can be a data code or a text message.

8. A system for diagnosing cavitation faults in a hydraulic turbine based on oscillation monitoring, characterized in that: include: A preprocessing module, used for acquiring a turbine signal, combining the turbine signal through machine learning to obtain data to be matched, and preprocessing the data to be matched to obtain first data; A model building module, used to build a machine learning model based on cavitation classification using the first data; The detection module is used to detect the operating status of the turbine according to the machine learning model based on cavitation and erosion classification, send out an early warning signal when the cavitation state of the turbine is abnormal, and convert the early warning signal into a configuration output signal of the turbine swing detection and protection module.

9. An electronic device, characterized in that: The device comprises: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images