Nuclear power steam turbine rotor fault analysis, early warning methods, devices, equipment and media

Abstract

The present invention relates to the field of data processing technology, and more particularly to a method, apparatus, device, and medium for analyzing and warning faults of a nuclear power steam turbine rotor. The method comprises obtaining a first radial amplitude and a first axial amplitude of the steam turbine rotor based on a vibration amplitude, plotting a difference change diagram of the first radial amplitude and the first axial amplitude as the rotor speed changes, determining that the rotor has an unbalance fault, and using a trained classification algorithm model to identify the fault type to obtain a fault type identification result. The method comprises determining that the steam turbine rotor has an unbalance fault when a change threshold is met based on the change in the first radial amplitude and the first axial amplitude of the steam turbine rotor according to the rotor speed, and inputting a second radial amplitude and a second axial amplitude that meet the change threshold into the trained classification algorithm model to obtain a fault type identification result, thereby classifying the unbalance fault of the steam turbine rotor and accurately achieving fault identification of the nuclear power steam turbine rotor.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method, device, equipment and medium for analyzing and warning faults of a nuclear power steam turbine rotor. Background Art

[0002] As a key piece of equipment in nuclear power plants, the stable operation of nuclear power turbines plays a vital role in ensuring safe and efficient power generation. The operating status of the turbine rotor, a core component, directly impacts the performance and reliability of the entire turbine. In actual operation, the turbine rotor is prone to imbalance failures during long-term, high-speed rotation due to manufacturing errors, material inhomogeneities, wear, scaling, and other factors. Imbalance failures can lead to increased rotor vibration, which not only affects the normal operating efficiency of the turbine but can also cause damage to other components. In severe cases, it can even cause the entire nuclear power plant to shut down, resulting in significant economic losses and safety hazards.

[0003] Currently, most methods measure rotor vibration signals and perform spectrum analysis on them. The fault type when an imbalance fault occurs is determined based on the amplitude and phase characteristics of different frequency components in the spectrum. However, different faults may have similar characteristics on the spectrum, which can easily lead to misjudgment. In addition, spectrum analysis is mainly based on steady-state vibration signals, and its ability to analyze fault changes in dynamic processes is weak, and it cannot promptly reflect the development trend of the fault.

[0004] Therefore, how to classify the unbalance faults of steam turbine rotors and accurately identify the faults of nuclear power steam turbine rotors has become an urgent problem to be solved. Summary of the Invention

[0005] The embodiments of the present invention provide a nuclear power steam turbine rotor fault analysis and early warning method, device, equipment and medium to solve the problem of how to classify the unbalance fault of the steam turbine rotor, thereby accurately realizing the fault identification of the nuclear power steam turbine rotor.

[0006] In a first aspect, an embodiment of the present invention provides a method for analyzing a nuclear power steam turbine rotor fault, the method comprising:

[0007] Acquiring operating condition information and process information of the steam turbine rotor to be tested collected by a sensor during operation, extracting the vibration amplitude of the steam turbine rotor and the vibration frequency corresponding to the vibration amplitude from the operating condition information, and extracting the rotor speed corresponding to the vibration frequency from the process information;

[0008] obtaining a first radial amplitude and a first axial amplitude of the steam turbine rotor based on the vibration amplitude, plotting a difference change graph of the first radial amplitude and the first axial amplitude as the rotor speed changes, and determining whether a difference change between the first radial amplitude and the first axial amplitude in the difference change graph reaches a preset change threshold; if so, determining that the rotor has an unbalance fault;

[0009] Determining, according to the difference change graph, an amplitude to be analyzed that meets the preset change threshold and a second radial amplitude to be analyzed and a second axial amplitude to be analyzed corresponding to the amplitude to be analyzed;

[0010] Using a trained classification algorithm model, the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed are respectively input into the classification algorithm model to perform fault type identification, thereby obtaining a fault type identification result.

[0011] In a second aspect, an embodiment of the present invention provides a nuclear power steam turbine rotor fault early warning method, the nuclear power steam turbine rotor fault early warning method comprising:

[0012] After obtaining a fault type identification result based on the nuclear power steam turbine rotor fault analysis method described in the first aspect above, obtaining historical standard information, and determining a fixed threshold corresponding to the operating condition information when the rotor has an unbalance fault based on the historical standard information;

[0013] determining a safety factor affecting normal operation of the rotor according to the fixed threshold, and extracting an operating condition parameter value corresponding to the safety factor from the operating condition information according to the safety factor;

[0014] It is determined whether the operating condition parameter value meets the fixed threshold value, and if it does not meet the fixed threshold value, an early warning signal is generated.

[0015] In a third aspect, an embodiment of the present invention provides a nuclear power steam turbine rotor fault analysis device, the nuclear power steam turbine rotor fault analysis device comprising:

[0016] an information extraction module, configured to obtain operating condition information and process information of the steam turbine rotor to be tested collected by a sensor during operation, extract the vibration amplitude of the steam turbine rotor and the vibration frequency corresponding to the vibration amplitude from the operating condition information, and extract the rotor speed corresponding to the vibration frequency from the process information;

[0017] an imbalance determination module, configured to obtain a first radial amplitude and a first axial amplitude of the steam turbine rotor based on the vibration amplitude, plot a difference change diagram of the first radial amplitude and the first axial amplitude as the rotor speed changes, determine whether a difference change between the first radial amplitude and the first axial amplitude in the difference change diagram reaches a preset change threshold, and if so, determine that the rotor has an imbalance fault;

[0018] an amplitude determination module for analysis, configured to determine, based on the difference change graph, an amplitude for analysis that satisfies the preset change threshold, and a second radial amplitude for analysis and a second axial amplitude for analysis corresponding to the amplitude for analysis;

[0019] The fault type identification module is used to use a trained classification algorithm model to input the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed into the classification algorithm model respectively to perform fault type identification and obtain a fault type identification result.

[0020] In a fourth aspect, an embodiment of the present invention provides a nuclear power steam turbine rotor fault early warning device, the nuclear power steam turbine rotor fault early warning device comprising:

[0021] a fixed threshold determination module, configured to obtain historical standard information after obtaining a fault type identification result based on the nuclear power steam turbine rotor fault analysis method, and determine, based on the historical standard information, a fixed threshold corresponding to the operating condition information when an unbalance fault occurs in the rotor;

[0022] an operating condition parameter extraction module, configured to determine a safety factor affecting the normal operation of the rotor according to the fixed threshold, and extract, based on the safety factor, an operating condition parameter value corresponding to the safety factor from the operating condition information;

[0023] The warning generation module is used to determine whether the operating condition parameter value meets the fixed threshold value, and if it does not meet the fixed threshold value, generate a warning signal.

[0024] In a fifth aspect, an embodiment of the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned nuclear power turbine rotor fault analysis method or nuclear power turbine rotor fault early warning method when executing the computer program.

[0025] In a sixth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the nuclear power turbine rotor fault analysis method or the nuclear power turbine rotor fault warning method is implemented.

[0026] Compared with the prior art, the present invention has the following beneficial effects: by acquiring operating condition information and process information of a steam turbine rotor to be tested collected by a sensor during operation, extracting the vibration amplitude of the steam turbine rotor and the vibration frequency corresponding to the vibration amplitude from the operating condition information, extracting the rotor speed corresponding to the vibration frequency from the process information, obtaining a first radial amplitude and a first axial amplitude of the steam turbine rotor based on the vibration amplitude, plotting a difference change graph of the first radial amplitude and the first axial amplitude as the rotor speed changes, determining whether the difference change between the first radial amplitude and the first axial amplitude in the difference change graph reaches a preset change threshold, and if so, determining that the rotor has an unbalance fault, determining an amplitude to be analyzed that meets the preset change threshold and a second radial amplitude to be analyzed and a second axial amplitude to be analyzed that correspond to the amplitude to be analyzed based on the difference change graph, and using a trained classification algorithm model, inputting the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed into the classification algorithm model for fault type identification to obtain a fault type identification result. By determining that the steam turbine rotor is an unbalance fault when a change threshold is met based on the change in the first radial amplitude and the first axial amplitude of the steam turbine rotor according to the rotor speed, and inputting the second radial amplitude and the second axial amplitude that meet the change threshold into the trained classification algorithm model, a fault type recognition result is obtained, thereby classifying the unbalance fault of the steam turbine rotor, thereby accurately realizing the fault identification of the nuclear power steam turbine rotor. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. 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 paying any creative labor.

[0028] Figure 1 This is a schematic diagram of an application environment of a nuclear power steam turbine rotor fault analysis method provided by the first embodiment of the present invention;

[0029] Figure 2 This is a flow chart of a method for analyzing a nuclear power steam turbine rotor failure provided by the second embodiment of the present invention;

[0030] Figure 3 This is a flow chart of a method for analyzing a nuclear power steam turbine rotor failure provided by the third embodiment of the present invention;

[0031] Figure 4 This is a flow chart of a nuclear power steam turbine rotor fault early warning method provided by the fourth embodiment of the present invention;

[0032] Figure 5This is a flow chart of a nuclear power steam turbine rotor fault early warning method provided by the fifth embodiment of the present invention;

[0033] Figure 6 This is a flow chart of a nuclear power steam turbine rotor fault early warning method provided by the sixth embodiment of the present invention;

[0034] Figure 7 This is a structural schematic diagram of a nuclear power steam turbine rotor fault analysis device provided by the seventh embodiment of the present invention;

[0035] Figure 8 This is a structural diagram of a nuclear power steam turbine rotor fault early warning device provided by Embodiment 8 of the present invention;

[0036] Figure 9 This is a structural diagram of a computer device provided in Example 9 of the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0038] like Figure 1 The figure shows a schematic diagram of the application environment of a method for analyzing a nuclear power turbine rotor fault provided by the first embodiment of the present invention, wherein the client and the server are connected for communication, and the user can provide the server with conditions, requirements, and operation instructions for nuclear power turbine rotor fault analysis by operating the client, and the server is used to execute the method for analyzing a nuclear power turbine rotor fault of the present invention according to the relevant content sent by the client. The client includes but is not limited to various computer devices such as personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The computer device corresponding to the server can be implemented using an independent server or a server cluster composed of multiple servers.

[0039] like Figure 2 FIG. 1 is a flow chart of a method for analyzing a nuclear power steam turbine rotor fault according to a second embodiment of the present invention, wherein the method for analyzing a nuclear power steam turbine rotor fault is applied in Figure 1 The nuclear power steam turbine rotor fault analysis method may include the following steps:

[0040] Step S201, obtaining the operating condition information and process information of the turbine rotor to be tested collected by the sensor during operation, extracting the vibration amplitude of the turbine rotor and the vibration frequency corresponding to the vibration amplitude from the operating condition information, and extracting the rotor speed corresponding to the vibration frequency from the process information.

[0041] In nuclear power steam turbine systems, various types of sensors are installed to monitor the rotor's operating conditions in real time. Different types of sensors are responsible for collecting different types of data.

[0042] Operating condition information primarily reflects the current operating status of the turbine rotor, typically including data on vibration, temperature, and pressure. For example, vibration sensors can measure rotor vibration during operation, temperature sensors can monitor temperature changes in the rotor and related components, and pressure sensors can obtain system pressure values. This information can be used to determine whether the rotor is operating stably.

[0043] Process information focuses on data related to the turbine's operating process. Rotor speed is a key process parameter, affecting the turbine's power generation efficiency and power output. Rotor speed also varies under different operating conditions. By collecting process information, we can understand the turbine's operating mode and working status.

[0044] Vibration amplitude refers to the amplitude of rotor vibration, which reflects the intensity of rotor vibration. A larger vibration amplitude may mean that the rotor has faults such as imbalance and looseness. In the collected operating information, the key feature of vibration amplitude is extracted by processing the data output by the vibration sensor.

[0045] Vibration frequency represents the periodicity of rotor vibration—the number of vibrations per unit time. Different faults can cause rotor vibrations of varying frequencies. For example, rotor imbalance often causes vibrations at the same frequency as the rotor's rotational frequency, while bearing faults can produce vibrations at a specific frequency. By analyzing vibration frequency, we can not only extract the vibration amplitude but also determine the corresponding frequency, enabling further in-depth fault analysis.

[0046] For example, vibration sensors are installed at key locations on the turbine rotor, such as the bearing seat and coupling, to monitor rotor vibration during operation. A speed sensor is also installed on the rotor's main shaft to measure rotor speed. At a specific moment, the sensors begin collecting data. The vibration sensor converts the detected rotor vibration signal into an electrical signal and transmits it to the data acquisition system. The speed sensor then measures the rotor speed in real time and also transmits this data to the data acquisition system. At this point, the data acquisition system acquires both operating condition information (including vibration data) and process information (including rotor speed data).

[0047] Step S202: Obtain a first radial amplitude and a first axial amplitude of the steam turbine rotor based on the vibration amplitude, draw a graph showing the difference between the first radial amplitude and the first axial amplitude as the rotor speed changes, and determine whether the difference between the first radial amplitude and the first axial amplitude in the graph reaches a preset change threshold. If so, determine that the rotor is unbalanced.

[0048] The vibration of a steam turbine rotor during operation is a complex three-dimensional motion, but it can generally be decomposed into two primary vibration directions: radial (perpendicular to the rotor axis) and axial (along the rotor axis). Using specific sensors or signal processing methods, the first radial amplitude and the first axial amplitude are separated from the collected vibration amplitude data. For example, vibration sensors installed at different locations and orientations can be used to measure radial and axial vibrations separately. Data processing then yields the corresponding amplitudes. These two amplitudes reflect the intensity of the rotor's radial and axial vibrations, respectively.

[0049] At different rotor speeds, the rotor's radial and axial vibration amplitudes vary. The rotor speed data extracted in step S201 is correlated with the first radial amplitude and first axial amplitude data, and the radial and axial amplitudes corresponding to each speed are recorded. For each speed point, the difference between the first radial amplitude and the first axial amplitude is calculated. This difference reflects the degree of difference in the rotor's radial and axial vibrations at that speed. A difference change graph is plotted, with the rotor speed as the horizontal axis and the difference between the first radial amplitude and the first axial amplitude as the vertical axis. This graph allows for intuitive observation of the changing trend of the radial and axial vibration amplitude difference as the rotor speed changes.

[0050] The preset variation threshold is determined based on factors such as turbine design parameters, historical operating data, and relevant industry standards. It represents the maximum allowable variation in the difference between the first radial amplitude and the first axial amplitude under normal operation. Observe the difference variation graph and analyze the variation. If the variation in the difference exceeds the preset variation threshold within certain speed ranges, it indicates that the difference between the radial and axial vibrations of the rotor at that speed is outside the normal range.

[0051] When the difference between the first radial amplitude and the first axial amplitude reaches a preset threshold, it means that there is a large imbalance in the radial and axial forces of the rotor. This imbalance is likely caused by uneven rotor mass distribution, that is, a rotor imbalance fault.

[0052] For example, there may be local wear, scaling, or foreign matter attached to the rotor, which may cause additional centrifugal force to be generated during the rotation of the rotor, resulting in abnormal differences in radial and axial vibrations, and thus rotor imbalance failure.

[0053] Optionally, before obtaining the first radial amplitude and the first axial amplitude of the steam turbine rotor according to the vibration amplitude in step S202, the following steps may be further included:

[0054] Obtain the preset safe amplitude range and draw a graph showing the amplitude change of the vibration amplitude as the rotor speed changes.

[0055] It is determined whether the amplitude change in the amplitude change diagram meets the preset safety amplitude range. If the amplitude change does not meet the preset safety amplitude range, the first radial amplitude and the first axial amplitude of the turbine rotor are obtained according to the vibration amplitude.

[0056] The preset safety amplitude range is a vibration amplitude range determined based on multiple factors, including the normal operating characteristics of the turbine rotor and industry standards. This range defines the acceptable fluctuation range of the turbine rotor vibration amplitude at different rotor speeds. For example, when the rotor speed is between 1000 and 1500 rpm, the safety amplitude range might be set to 0.1 to 0.3 mm.

[0057] The amplitude variation graph plots rotor speed on the horizontal axis and vibration amplitude on the vertical axis. This graph allows us to visually visualize how the vibration amplitude changes with rotor speed. For example, we might observe that as rotor speed increases, the vibration amplitude first rises slowly, then suddenly increases around a certain speed. This visualization helps us quickly identify anomalies in the vibration amplitude, such as sudden changes or periodic fluctuations. These anomalies may be early signs of rotor failure.

[0058] The vibration amplitudes in the amplitude variation graph are compared one by one with the preset safe amplitude range. If the vibration amplitude remains within the safe amplitude range throughout the entire rotor speed range, the rotor's operating status is generally normal and no further detailed analysis is required. If the amplitude variation does not conform to the preset safe amplitude range, meaning that some vibration amplitudes exceed the safe range, this indicates that there may be a problem with the rotor's operating status and further in-depth analysis is required. At this point, the next step is to obtain the first radial amplitude and first axial amplitude of the turbine rotor based on the vibration amplitude, in order to gain a more detailed understanding of the rotor's vibration in different directions and provide more accurate data support for subsequent fault diagnosis.

[0059] Step S203 : determining the amplitude to be analyzed that meets a preset change threshold and the second radial amplitude to be analyzed and the second axial amplitude to be analyzed corresponding to the amplitude to be analyzed according to the difference change graph.

[0060] The difference change graph shows how the difference between the first radial amplitude and the first axial amplitude changes with rotor speed. The graph clearly shows the fluctuation trend of the difference. When the difference reaches the preset threshold, it indicates that the radial and axial vibration of the rotor at these corresponding speeds are abnormal, and these abnormal points are the focus of attention.

[0061] In the difference change graph, the difference corresponding to each speed point is checked one by one. When the difference at a speed point reaches or exceeds the preset change threshold, the vibration amplitude corresponding to that speed point is recorded. This vibration amplitude is the amplitude to be analyzed. It represents the overall vibration intensity when abnormal rotor vibration occurs.

[0062] After determining the amplitude to be analyzed, the second radial amplitude and second axial amplitude corresponding to the amplitude to be analyzed are found based on the recorded correspondence between speed and amplitude. These two amplitudes respectively reflect the specific vibration conditions of the rotor in the radial and axial directions when abnormal vibration occurs.

[0063] The second radial amplitude and the second axial amplitude can more accurately display the directional characteristics of the rotor vibration. Different fault types may cause the rotor to exhibit different radial and axial vibration characteristics. Therefore, these two amplitudes are used to accurately determine the fault type.

[0064] For example, the preset change threshold is 0.5mm. In the difference change diagram, when the rotor speed is 2500 rpm, the difference between the first radial amplitude and the first axial amplitude is 0.6mm, which exceeds the threshold. At this time, the corresponding vibration amplitude at this speed (for example, 1.2mm) is determined as the amplitude to be analyzed.

[0065] Step S204 , using the trained classification algorithm model, inputting the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed into the classification algorithm model respectively to perform fault type identification and obtain a fault type identification result.

[0066] In practical applications, various classification algorithms can be used to build models, such as decision trees, support vector machines, and neural networks. These algorithms each have their own unique characteristics. For example, decision trees are intuitive and easily interpretable, while neural networks excel at handling complex nonlinear relationships. The choice of algorithm depends on a comprehensive consideration of factors such as the characteristics of the data, the complexity of the problem, and the actual application scenario.

[0067] Before using the model to identify fault types, it needs to be trained. This training process typically requires a large amount of historical data, which contains characteristic information such as the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed corresponding to different fault types. By allowing the model to learn the patterns and patterns in this data, the model parameters are adjusted to enable it to accurately classify different fault types. For example, a neural network model is trained using labeled fault data. After multiple iterations of optimization, the model gradually learns to distinguish the characteristic patterns corresponding to different fault types.

[0068] The amplitude to be analyzed, the second radial amplitude, the second axial amplitude determined in step S203, and the rotor speed extracted in step S201 are used as input data. These data provide a specific quantitative description of the current operating state of the steam turbine rotor, including key information such as the intensity, direction, and speed of the rotor vibration.

[0069] To ensure that input data can be correctly processed by the classification algorithm model, some data format conversion or normalization is required. For example, data with different units and magnitudes can be normalized to keep them within the same numerical range. This prevents certain features from having excessively large or small values ​​that could negatively impact model training and prediction. The prepared input data is fed into the trained classification algorithm model, which analyzes and processes the data based on its learned patterns. The model performs a series of calculations and judgments on the input data, ultimately outputting a fault type identification result.

[0070] Fault type identification results are typically presented as a specific fault type name or number. For example, fault types include force imbalance, cantilever rotor imbalance, even imbalance, and dynamic imbalance. The classification algorithm model determines the most likely rotor fault type based on the input data characteristics and outputs this as the identification result.

[0071] In an embodiment of the present application, the operating condition information and process information of the turbine rotor to be tested collected by the sensor during operation are obtained, the vibration amplitude of the turbine rotor and the vibration frequency corresponding to the vibration amplitude are extracted from the operating condition information, and the rotor speed corresponding to the vibration frequency is extracted from the process information. According to the vibration amplitude, a first radial amplitude and a first axial amplitude of the turbine rotor are obtained, and a difference change diagram of the first radial amplitude and the first axial amplitude as the rotor speed changes is drawn. It is judged whether the difference change between the first radial amplitude and the first axial amplitude in the difference change diagram reaches a preset change threshold. If the change threshold is reached, it is determined that the rotor has an unbalance fault. According to the difference change diagram, the amplitude to be analyzed that meets the preset change threshold and the second radial amplitude to be analyzed and the second axial amplitude to be analyzed corresponding to the amplitude to be analyzed are determined. Using a trained classification algorithm model, the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed are respectively input into the classification algorithm model for fault type identification to obtain a fault type identification result. By determining that the steam turbine rotor is an unbalance fault when a change threshold is met based on the change in the first radial amplitude and the first axial amplitude of the steam turbine rotor according to the rotor speed, and inputting the second radial amplitude and the second axial amplitude that meet the change threshold into the trained classification algorithm model, a fault type recognition result is obtained, thereby classifying the unbalance fault of the steam turbine rotor, thereby accurately realizing the fault identification of the nuclear power steam turbine rotor.

[0072] like Figure 3 FIG. 2 is a flow chart of a method for analyzing a nuclear power steam turbine rotor fault according to a third embodiment of the present invention. The training process of the trained classification algorithm model in step S204 may include the following steps:

[0073] Step S301: Obtain a training set, which includes the labeling results of the corresponding fault type and the operating condition information and process information of at least one steam turbine rotor. The operating condition information includes the amplitude to be analyzed that meets the preset change threshold, the second radial amplitude corresponding to the amplitude to be analyzed, and the second axial amplitude corresponding to the amplitude to be analyzed. The process information includes the rotor speed corresponding to the vibration frequency.

[0074] In step S302, the amplitude to be analyzed, the second radial amplitude, the second axial amplitude and the rotor speed of each steam turbine rotor are respectively input into the encoder in the preset classification algorithm model for encoding to obtain a first feature corresponding to the amplitude to be analyzed, a second feature corresponding to the second radial amplitude, a third feature corresponding to the second axial amplitude and a fourth feature corresponding to the rotor speed.

[0075] Among them, the training set is the basis of the entire training process. It contains two important pieces of information. The first is the labeling result corresponding to the fault type, which is a clear "standard answer" that tells the model what type of fault should be output under a given input. The second is the operating condition information and process information of at least one steam turbine rotor. The operating condition information covers the amplitude to be analyzed that meets the preset change threshold, the second radial amplitude and the second axial amplitude corresponding to the amplitude to be analyzed. This information reflects the intensity and direction characteristics of the rotor vibration. The process information includes the rotor speed corresponding to the vibration frequency, which reflects the operating status of the rotor. These data are usually collected from the actual steam turbine operation monitoring system and have been accumulated and sorted for a long time to ensure the diversity and representativeness of the data, so that the model can learn the characteristic patterns under different operating conditions and fault types.

[0076] The preset classification algorithm model includes an encoder, which encodes the raw input data (the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed). The purpose of encoding is to convert this data of varying types and magnitudes into feature vectors that are easier for the model to process and understand. The encoder generates a first feature corresponding to the amplitude to be analyzed, a second feature corresponding to the second radial amplitude, a third feature corresponding to the second axial amplitude, and a fourth feature corresponding to the rotor speed. These features are abstract representations of the raw data, more effectively reflecting key information within the data and facilitating subsequent fusion and classification operations.

[0077] In step S303 , the first feature, the second feature, the third feature, and the fourth feature are input into a preset classification algorithm model for fusion to obtain a first fusion feature, and the fusion loss is calculated.

[0078] In step S304 , the first fusion feature is input into a classifier in a preset classification algorithm model to perform fault classification to obtain a first classification result. The classification task loss is calculated based on the first classification result and the labeling result.

[0079] Step S305 , updating the parameters in the preset classification algorithm model according to the fusion loss and the classification task loss to obtain an updated classification algorithm model.

[0080] The first, second, third, and fourth features are each input into a pre-set classification algorithm model for fusion. The goal of fusion is to integrate feature information from different aspects to form a more comprehensive and representative feature representation, namely the first fused feature. Fusion leverages the correlation and complementarity between features, improving the model's ability to identify fault types. During feature fusion, the model calculates a fusion loss, which measures the difference between the fused features and the ideal fusion state and reflects the quality of the fusion process. A lower fusion loss indicates that the fused features more accurately integrate information from the original features.

[0081] The first fused feature is used in the classifier within the pre-set classification algorithm model to classify the fault. Based on the characteristics of the fused feature, the classifier determines the turbine rotor fault type and obtains a first classification result. This first classification result is then compared with the labeled results in the training set to calculate the classification task loss. This loss measures the difference between the model's classification result and the actual fault type, reflecting the model's accuracy in the fault classification task. The smaller the classification task loss, the closer the model's classification result is to the actual situation.

[0082] The fusion loss and classification task loss update the parameters in the preset classification algorithm model. By adjusting the model parameters, the model's performance in feature fusion and fault classification becomes better and better. Optimization algorithms (such as stochastic gradient descent) are usually used to update the model parameters according to the gradient information of the loss function, and the parameter values ​​are adjusted in the direction of reducing the loss function.

[0083] Step S306, calculate the weighted sum of the classification task loss and the fusion loss, and use the weighted sum as the total loss, use the updated classification algorithm model as the preset classification algorithm model, and return to execute the step of inputting the amplitude to be analyzed, the second radial amplitude, the second axial amplitude and the rotor speed of each turbine rotor into the encoder in the preset classification algorithm model for encoding, until the updated classification algorithm model obtained when the total loss meets the preset conditions is the trained classification algorithm model.

[0084] The classification task loss and fusion loss are calculated to obtain a weighted sum, which is used as the total loss. This weighted sum comprehensively considers the performance of both feature fusion and fault classification, providing a more comprehensive assessment of the model's overall performance. The updated classification algorithm model is used as the new preset model, and the process returns to step S302 to continue the next round of encoding, fusion, classification, and loss calculation. This process is repeated until the total loss meets the preset conditions (e.g., the total loss is less than a set threshold). When the total loss meets the preset conditions, it indicates that the model has converged to a satisfactory state. The updated classification algorithm model is now a trained classification algorithm model and can be used for actual fault type identification tasks.

[0085] In this embodiment, the parameters of the model are gradually adjusted through multiple cycles, so that the model can better process input data and accurately identify the fault type of the steam turbine rotor.

[0086] like Figure 4 FIG. 1 is a flow chart of a nuclear power steam turbine rotor fault early warning method provided by a fourth embodiment of the present invention. After the nuclear power steam turbine rotor fault analysis method obtains a fault type identification result, the nuclear power steam turbine rotor fault early warning method may include the following steps:

[0087] Step S401: Acquire historical standard information, and determine a fixed threshold of operating condition information corresponding to a rotor imbalance fault based on the historical standard information.

[0088] Step S402 : determining safety factors that affect the normal operation of the rotor according to a fixed threshold value, and extracting operating condition parameter values ​​corresponding to the safety factors from the operating condition information according to the safety factors.

[0089] Step S403: determine whether the operating condition parameter value meets the fixed threshold value. If it does not meet the fixed threshold value, generate a warning signal.

[0090] The historical standard information is derived from a large amount of historical nuclear power steam turbine rotor operation data. This data records the rotor's operating status and whether any faults occurred under different operating conditions. By collecting and organizing this historical data, a comprehensive and representative dataset can be obtained, providing a basis for subsequent threshold determination.

[0091] Based on the historical standard information obtained, the system analyzes the operating conditions associated with rotor imbalance failures, such as the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed. Through statistical analysis and other methods, the critical values ​​or common ranges of these operating conditions when rotor imbalance failures occur are determined and used as fixed thresholds. For example, analysis has found that when rotor imbalance occurs, the amplitude to be analyzed typically exceeds 0.5 mm. Therefore, 0.5 mm can be set as the fixed threshold for the amplitude to be analyzed.

[0092] Based on the fixed threshold determined in step S401, the operating condition information is analyzed to determine which has a significant impact on the normal operation of the rotor. For example, if the fixed threshold indicates that the amplitude to be analyzed and the rotor speed change significantly during a rotor imbalance fault, then the amplitude to be analyzed and the rotor speed can be determined to be safety factors.

[0093] The operating condition parameter values ​​corresponding to the determined safety factors are extracted from the current operating condition information. For example, if the determined safety factors are the amplitude to be analyzed and the rotor speed, the specific values ​​of these two parameters are extracted from the current operating condition information.

[0094] The extracted operating condition parameter value is compared with the fixed threshold value determined in step S401. For example, the current amplitude to be analyzed is compared with the fixed threshold value of the amplitude to be analyzed, and the current rotor speed is compared with the fixed threshold value of the rotor speed.

[0095] If one or more operating parameter values ​​fail to meet fixed thresholds, exceeding the set safety range, indicating a potential risk in the rotor's operating state, a warning signal is generated. This warning signal can be issued in a variety of ways, such as audible and visual alarms, SMS notifications, and system pop-up notifications, allowing relevant personnel to promptly be notified of rotor anomalies and take appropriate measures.

[0096] In this embodiment, by using historical data to set reasonable thresholds, real-time monitoring and judgment of key operating parameters that affect the normal operation of the rotor are carried out, which can timely discover potential risks in rotor operation and issue early warnings, helping to take measures in advance to avoid the occurrence or expansion of faults and ensure the safe and stable operation of nuclear power turbines.

[0097] like Figure 5 FIG. 4 is a flow chart of a nuclear power steam turbine rotor fault early warning method provided by a fifth embodiment of the present invention. Based on the fourth embodiment above, after extracting the operating condition parameter value corresponding to the safety factor in the operating condition information in step S402, the nuclear power steam turbine rotor fault early warning method may further include the following steps:

[0098] Step S501 : extracting standard parameter values ​​of multiple steam turbine rotors during normal operation based on the rotor speed, and calculating an average value of the standard parameters to obtain an average result.

[0099] Step S502 , determining whether the operating parameter value of the steam turbine rotor to be tested conforms to the average result; if not, generating a warning signal.

[0100] Rotor speed is a key factor affecting the operating state of a steam turbine rotor. At different rotor speeds, various operating parameters have different normal value ranges. Therefore, based on the current rotor speed of the turbine under test, the corresponding operating parameter values ​​of multiple other turbine rotors operating normally at the same or similar speeds are extracted from historical data. For example, if the current rotor speed of the turbine under test is 3000 rpm, the historical data is used to identify the values ​​of the amplitude to be analyzed, the second radial amplitude, and other parameters of multiple turbine rotors operating normally at around 3000 rpm.

[0101] Multiple standard parameter values ​​are mathematically averaged to produce an average result. This average represents the typical values ​​of these parameters for a steam turbine rotor operating normally at that rotor speed. For example, ten amplitude values ​​for the rotor under normal operation at 3000 rpm were extracted: 0.1, 0.12, 0.09, 0.11, 0.13, 0.1, 0.11, 0.12, 0.1, and 0.11 mm. These values ​​were added and divided by 10 to obtain an average value of 0.11 mm.

[0102] The current operating parameter value of the turbine rotor under test is compared with the average result calculated in step S501. "Compliance" here generally means that the operating parameter value is within a reasonable fluctuation range around the average result. This range can be pre-set based on actual conditions, for example, ±10% of the average result. For example, if the average result of the amplitude to be analyzed is 0.11 mm and the fluctuation range is set to ±10%, then the reasonable range is 0.099 - 0.121 mm.

[0103] If the operating parameter values ​​of the turbine rotor under test fall outside this reasonable fluctuation range, that is, if they do not conform to the average results, it indicates that the rotor's operating status may have deviated from normal levels, posing a potential failure risk. At this point, the system generates an early warning signal, prompting personnel to pay attention and further inspect the rotor's operating status to promptly identify and address any potential problems.

[0104] This embodiment uses the average values ​​of multiple standard rotor parameters during normal operation as a reference, providing a more detailed and reasonable basis for determining the operating status of the rotor under test. This addition of an average value-based judgment dimension, in addition to existing fixed threshold judgments, makes fault warnings more comprehensive and accurate, helping to more effectively ensure the safe and stable operation of nuclear power turbine rotors.

[0105] like Figure 6 FIG. 4 is a flow chart of a method for early warning of a nuclear power steam turbine rotor fault provided by a sixth embodiment of the present invention. After extracting the operating condition parameter value corresponding to the safety factor from the operating condition information in step S402, the method for early warning of a nuclear power steam turbine rotor fault may further include the following steps:

[0106] Step S601: Analyze the current operating trend of the corresponding steam turbine rotor to be tested according to the operating parameter value.

[0107] Step S602: Obtain the historical standard operation trend of the steam turbine rotor based on the historical standard information.

[0108] Step S603: Analyze whether the current operating trend meets the historical standard operating trend. If it does not meet the historical standard operating trend, generate an early warning signal.

[0109] The operating parameters corresponding to the safety factors have been extracted previously, such as the amplitude to be analyzed, rotor speed, and second radial amplitude. By observing and analyzing the changes in these parameters over time, the current operating trend of the turbine rotor under test can be depicted.

[0110] Operating trends are manifested as increases, decreases, fluctuations, or stabilizations in parameter values. For example, if the amplitude to be analyzed continues to rise over a period of time, it indicates that the rotor's operating condition is gradually deteriorating; if the parameter value remains stable within a certain range, it indicates that the rotor is operating relatively smoothly.

[0111] The historical standard information contains a large amount of operating condition data of the steam turbine rotor during normal operation. By processing and analyzing this data, such as performing statistics and fitting operations on parameter values ​​in different time periods under the same operating conditions, the historical standard operating trends of the steam turbine rotor can be summarized.

[0112] The historical standard operating trend represents the typical operating mode of the turbine rotor under normal conditions and is an important reference for judging whether the current operating status is normal.

[0113] The current operating trend obtained in step S601 is compared with the historical standard operating trend obtained in step S602. "Conformity" here means that the two have a high degree of similarity in terms of change form, change rate, etc. For example, the historical standard operating trend shows that under a certain operating condition, the amplitude to be analyzed will slowly increase with the increase of the rotor speed, and the rate of increase is within a certain range; if the rate of increase of the amplitude to be analyzed in the current operating trend is too fast, or there is a fluctuation that is different from the historical trend, the current operating trend is considered to be inconsistent with the historical standard operating trend.

[0114] If the current operating trend does not conform to the historical standard operating trend, it indicates that the operating status of the turbine rotor under test may be abnormal and there is a potential failure risk. In this case, the system will generate an early warning signal to notify relevant personnel to promptly inspect and maintain the rotor to prevent the occurrence or escalation of the failure.

[0115] In this embodiment, by analyzing and comparing the operating trends, the method can capture the dynamic changes in the rotor operating status and discover in advance some potential fault hazards that have not yet been directly reflected by the operating parameter values, thereby further improving the effectiveness and reliability of nuclear power turbine rotor fault warning and ensuring the safe and stable operation of nuclear power production.

[0116] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0117] like Figure 7 FIG. 1 is a schematic diagram of a nuclear power steam turbine rotor fault analysis device according to a seventh embodiment of the present invention. The nuclear power steam turbine rotor fault analysis device corresponds one-to-one with the nuclear power steam turbine rotor fault analysis method according to the aforementioned embodiment. The nuclear power steam turbine rotor fault analysis device includes an information acquisition module 71, an imbalance determination module 72, an amplitude determination module 73, and a fault type identification module 74. Each functional module is described in detail below:

[0118] An information acquisition module 71 is configured to acquire operating condition information and process information of the steam turbine rotor under test collected by sensors during operation, extract the vibration amplitude of the steam turbine rotor and the vibration frequency corresponding to the vibration amplitude from the operating condition information, and extract the rotor speed corresponding to the vibration frequency from the process information;

[0119] an imbalance determination module 72 for obtaining a first radial amplitude and a first axial amplitude of the steam turbine rotor based on the vibration amplitude, plotting a difference change diagram of the first radial amplitude and the first axial amplitude as the rotor speed changes, and determining whether the difference change between the first radial amplitude and the first axial amplitude in the difference change diagram reaches a preset change threshold. If so, determining that the rotor is unbalanced;

[0120] The amplitude determination module 73 is configured to determine, based on the difference change graph, an amplitude to be analyzed that satisfies a preset change threshold and a second radial amplitude to be analyzed and a second axial amplitude to be analyzed corresponding to the amplitude to be analyzed;

[0121] The fault type identification module 74 is used to use the trained classification algorithm model to input the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed into the classification algorithm model to perform fault type identification and obtain a fault type identification result.

[0122] Optionally, the fault type identification module 74 further includes:

[0123] a training data determination unit, configured to obtain a training set, the training set including a labeling result corresponding to a fault type and operating condition information and process information of at least one steam turbine rotor, the operating condition information including an amplitude to be analyzed that meets a preset change threshold, a second radial amplitude corresponding to the amplitude to be analyzed, and a second axial amplitude corresponding to the amplitude to be analyzed, and the process information including a rotor speed corresponding to a vibration frequency;

[0124] a feature extraction unit, configured to input the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed of each steam turbine rotor into an encoder in a preset classification algorithm model for encoding, thereby obtaining a first feature corresponding to the amplitude to be analyzed, a second feature corresponding to the second radial amplitude, a third feature corresponding to the second axial amplitude, and a fourth feature corresponding to the rotor speed;

[0125] A fusion loss calculation unit, configured to input the first feature, the second feature, the third feature, and the fourth feature into a preset classification algorithm model for fusion, obtain a first fusion feature, and calculate a fusion loss;

[0126] A classification loss calculation unit is used to input the first fusion feature into a classifier in a preset classification algorithm model to perform fault classification, obtain a first classification result, and calculate the classification task loss based on the first classification result and the labeling result;

[0127] A model updating unit, configured to update parameters in a preset classification algorithm model according to the fusion loss and the classification task loss to obtain an updated classification algorithm model;

[0128] Return to the execution unit, which is used to calculate the weighted sum of the classification task loss and the fusion loss, and use the weighted sum as the total loss, use the updated classification algorithm model as the preset classification algorithm model, and return to execute the step of inputting the amplitude to be analyzed, the second radial amplitude, the second axial amplitude and the rotor speed of each turbine rotor into the encoder in the preset classification algorithm model for encoding, until the updated classification algorithm model obtained when the total loss meets the preset conditions is the trained classification algorithm model.

[0129] Optionally, the nuclear power steam turbine rotor fault analysis device further includes:

[0130] An amplitude variation diagram drawing module is used to obtain a preset safe amplitude range before obtaining a first radial amplitude and a first axial amplitude of the steam turbine rotor based on the vibration amplitude, and draw an amplitude variation diagram of the vibration amplitude as the rotor speed changes;

[0131] The range judgment module is used to judge whether the amplitude change in the amplitude change diagram meets the preset safety amplitude range. If the amplitude change does not meet the preset safety amplitude range, the first radial amplitude and the first axial amplitude of the turbine rotor are obtained according to the vibration amplitude.

[0132] like Figure 8 FIG. 1 is a schematic diagram of a nuclear power steam turbine rotor fault warning device according to an eighth embodiment of the present invention. This device corresponds exactly to the nuclear power steam turbine rotor fault warning method described in the aforementioned embodiment. The device includes a fixed threshold determination module 81, an operating parameter extraction module 82, and an early warning generation module 83. Each functional module is described in detail below:

[0133] The fixed threshold determination module 81 is used to obtain historical standard information after the fault type identification result is obtained by the nuclear power steam turbine rotor fault analysis method, and determine the fixed threshold of the corresponding operating condition information when the rotor imbalance fault occurs based on the historical standard information;

[0134] The operating condition parameter extraction module 82 is used to determine the safety factors that affect the normal operation of the rotor according to the fixed threshold value, and extract the operating condition parameter value corresponding to the safety factor from the operating condition information according to the safety factor;

[0135] The warning generation module 83 is used to determine whether the operating condition parameter value meets the fixed threshold value, and if it does not meet the fixed threshold value, generate a warning signal.

[0136] Optionally, the operating condition parameter extraction module 82 includes:

[0137] an average value calculation unit, for extracting standard parameter values ​​of multiple steam turbine rotors during normal operation based on the rotor speed, and calculating the average value of the standard parameters to obtain an average result;

[0138] The result judgment unit is used to judge whether the operating parameter value of the turbine rotor to be tested conforms to the average result, and generates an early warning signal if it does not conform to the average result.

[0139] Optionally, the operating condition parameter extraction module 82 includes:

[0140] The parameter analysis unit is used to analyze the current operating trend of the corresponding steam turbine rotor to be tested according to the operating parameter values;

[0141] An operation trend determination unit, for obtaining a historical standard operation trend of the steam turbine rotor based on historical standard information;

[0142] The trend judgment unit is used to analyze whether the current operating trend conforms to the historical standard operating trend. If it does not conform to the historical standard operating trend, an early warning signal is generated.

[0143] Regarding the specific limitations of the nuclear power steam turbine rotor fault analysis device and the nuclear power steam turbine rotor fault early warning device, please refer to the limitations of the nuclear power steam turbine rotor fault analysis method and the nuclear power steam turbine rotor fault early warning method above, and will not be repeated here. The various modules in the above-mentioned nuclear power steam turbine rotor fault analysis device and the nuclear power steam turbine rotor fault early warning device can be implemented in whole or in part through software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0144] like Figure 9 The figure shows a schematic diagram of the structure of a computer device provided in accordance with a ninth embodiment of the present invention. The computer device includes a processor, a memory, a network interface, and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for analyzing a nuclear power turbine rotor fault and a method for early warning of a nuclear power turbine rotor fault are implemented.

[0145] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for analyzing a nuclear power turbine rotor fault in the above embodiment is implemented. For example, Figures 2 to 3 When the processor executes the computer program, the nuclear power steam turbine rotor fault early warning method in the above embodiment is implemented, for example Figures 4 to 6 Alternatively, when the processor executes the computer program, the functions of each module / unit in the embodiment of the nuclear power steam turbine rotor fault analysis device are realized, for example Figure 7 The functions of the information acquisition module 71, the imbalance determination module 72, the amplitude determination module 73 to be analyzed, and the fault type identification module 74 shown in the figure, or the functions of each module / unit in the embodiment of the nuclear power steam turbine rotor fault early warning device are realized when the processor executes the computer program, for example Figure 8 The fixed threshold determination module 81, the operating condition parameter extraction module 82, and the warning generation module 83 shown are not described here in detail to avoid repetition.

[0146] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for analyzing a nuclear power turbine rotor fault in the above embodiment is implemented. Figures 2 to 3 When the computer program is executed by the processor, the nuclear power steam turbine rotor fault warning method in the above embodiment is realized, such as Figures 4 to 6 Alternatively, when the computer program is executed by the processor, the functions of each module / unit in the embodiment of the nuclear power steam turbine rotor fault analysis device are realized, for example Figure 7 The functions of the information acquisition module 71, the imbalance determination module 72, the amplitude determination module 73 to be analyzed, and the fault type identification module 74 are not described here in detail to avoid repetition. Alternatively, when the computer program is executed by the processor, the functions of the modules / units in the embodiment of the nuclear power steam turbine rotor fault early warning device are realized, for example Figure 8 The functions of the fixed threshold determination module 81, the operating parameter extraction module 82, and the warning generation module 83 are not described here in detail to avoid repetition. The computer-readable storage medium may be non-volatile or volatile.

[0147] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0148] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0149] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A method for analyzing a nuclear power steam turbine rotor fault, characterized in that: The nuclear power steam turbine rotor fault analysis method comprises: Acquiring operating condition information and process information of the steam turbine rotor to be tested collected by a sensor during operation, extracting the vibration amplitude of the steam turbine rotor and the vibration frequency corresponding to the vibration amplitude from the operating condition information, and extracting the rotor speed corresponding to the vibration frequency from the process information; obtaining a first radial amplitude and a first axial amplitude of the steam turbine rotor based on the vibration amplitude, plotting a difference change graph of the first radial amplitude and the first axial amplitude as the rotor speed changes, and determining whether a difference change between the first radial amplitude and the first axial amplitude in the difference change graph reaches a preset change threshold; if so, determining that the rotor has an unbalance fault; Determining, according to the difference change graph, an amplitude to be analyzed that meets the preset change threshold and a second radial amplitude to be analyzed and a second axial amplitude to be analyzed corresponding to the amplitude to be analyzed; Using a trained classification algorithm model, the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed are respectively input into the classification algorithm model to perform fault type identification, thereby obtaining a fault type identification result; The drawing of a graph showing a difference between the first radial amplitude and the first axial amplitude as the rotor speed changes includes: For each rotor speed, calculating a difference between a first radial amplitude and a first axial amplitude of the vibration amplitude corresponding to the rotor speed, and plotting a difference change graph with the rotor speed as the horizontal axis and the difference between the first radial amplitude and the first axial amplitude as the vertical axis; The determining, based on the difference change graph, the amplitude to be analyzed that meets the preset change threshold and the second radial amplitude to be analyzed and the second axial amplitude to be analyzed corresponding to the amplitude to be analyzed, includes: In the difference change diagram, the difference corresponding to each rotor speed is checked one by one. When the difference of a certain rotor speed reaches or exceeds the preset change threshold, the vibration amplitude corresponding to the rotor speed is used as the amplitude to be analyzed. After determining the amplitude to be analyzed, the second radial amplitude to be analyzed and the second axial amplitude to be analyzed of the amplitude to be analyzed are obtained based on the recorded correspondence between the speed and the amplitude.

2. The nuclear power steam turbine rotor fault analysis method according to claim 1, characterized in that: The training process of the trained classification algorithm model is as follows: Obtaining a training set, the training set including labeling results corresponding to the fault type and operating condition information and process information of at least one steam turbine rotor, the operating condition information including the amplitude to be analyzed that meets the preset change threshold, a second radial amplitude corresponding to the amplitude to be analyzed, and a second axial amplitude corresponding to the amplitude to be analyzed, and the process information including a rotor speed corresponding to the vibration frequency; Inputting the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed of each steam turbine rotor into an encoder in a preset classification algorithm model for encoding, respectively, to obtain a first feature corresponding to the amplitude to be analyzed, a second feature corresponding to the second radial amplitude, a third feature corresponding to the second axial amplitude, and a fourth feature corresponding to the rotor speed; Inputting the first feature, the second feature, the third feature, and the fourth feature into the preset classification algorithm model for fusion to obtain a first fusion feature, and calculating the fusion loss; Inputting the first fusion feature into the classifier in the preset classification algorithm model to perform fault classification to obtain a first classification result, and calculating the classification task loss based on the first classification result and the labeling result; updating the parameters of the preset classification algorithm model according to the fusion loss and the classification task loss to obtain an updated classification algorithm model; Calculate the weighted sum of the classification task loss and the fusion loss, and use the weighted sum as the total loss. Use the updated classification algorithm model as the preset classification algorithm model, and return to execute the step of inputting the amplitude to be analyzed, the second radial amplitude, the second axial amplitude and the rotor speed of each turbine rotor into the encoder in the preset classification algorithm model for encoding, until the updated classification algorithm model obtained when the total loss meets the preset conditions is a trained classification algorithm model.

3. The nuclear power steam turbine rotor fault analysis method according to claim 1, characterized in that: Before obtaining the first radial amplitude and the first axial amplitude of the steam turbine rotor according to the vibration amplitude, the method further includes: Obtaining a preset safe amplitude range, and drawing an amplitude variation diagram of the vibration amplitude as the rotor speed changes; Determine whether the amplitude change in the amplitude change diagram meets the preset safety amplitude range; if the amplitude change does not meet the preset safety amplitude range, obtain the first radial amplitude and the first axial amplitude of the turbine rotor based on the vibration amplitude.

4. A nuclear power steam turbine rotor fault early warning method, characterized in that: The nuclear power steam turbine rotor fault early warning method comprises: After obtaining a fault type identification result based on the nuclear power steam turbine rotor fault analysis method according to any one of claims 1 to 3, obtaining historical standard information, and determining a fixed threshold corresponding to the operating condition information when an unbalance fault occurs in the rotor based on the historical standard information; determining a safety factor affecting normal operation of the rotor according to the fixed threshold, and extracting an operating condition parameter value corresponding to the safety factor from the operating condition information according to the safety factor; It is determined whether the operating condition parameter value meets the fixed threshold value, and if it does not meet the fixed threshold value, an early warning signal is generated.

5. The nuclear power steam turbine rotor fault early warning method according to claim 4, characterized in that: After extracting the operating condition parameter value corresponding to the safety factor from the operating condition information, the method further includes: Based on the rotor speed, extracting standard parameter values ​​of multiple steam turbine rotors during normal operation, and calculating an average value of the standard parameters to obtain an average result; It is determined whether the operating parameter value of the steam turbine rotor to be tested conforms to the average result; if it does not conform to the average result, an early warning signal is generated.

6. The nuclear power steam turbine rotor fault early warning method according to claim 4, characterized in that: After extracting the operating condition parameter value corresponding to the safety factor from the operating condition information, the method further includes: Analyzing the current operating trend of the steam turbine rotor to be tested according to the operating condition parameter value; Obtaining a historical standard operating trend of the steam turbine rotor based on the historical standard information; Analyze whether the current operating trend meets the historical standard operating trend, and generate an early warning signal if it does not meet the historical standard operating trend.

7. A nuclear power steam turbine rotor fault analysis device, characterized in that: The nuclear power steam turbine rotor fault analysis device comprises: an information extraction module, configured to obtain operating condition information and process information of the steam turbine rotor to be tested collected by a sensor during operation, extract the vibration amplitude of the steam turbine rotor and the vibration frequency corresponding to the vibration amplitude from the operating condition information, and extract the rotor speed corresponding to the vibration frequency from the process information; an imbalance determination module, configured to obtain a first radial amplitude and a first axial amplitude of the steam turbine rotor based on the vibration amplitude, plot a difference change diagram of the first radial amplitude and the first axial amplitude as the rotor speed changes, determine whether a difference change between the first radial amplitude and the first axial amplitude in the difference change diagram reaches a preset change threshold, and if so, determine that the rotor has an imbalance fault; an amplitude determination module for analysis, configured to determine, based on the difference change graph, an amplitude for analysis that satisfies the preset change threshold, and a second radial amplitude for analysis and a second axial amplitude for analysis corresponding to the amplitude for analysis; a fault type identification module, configured to use a trained classification algorithm model to input the amplitude to be analyzed, the second radial amplitude, the second axial amplitude, and the rotor speed into the classification algorithm model to perform fault type identification and obtain a fault type identification result; The drawing of a graph showing a difference between the first radial amplitude and the first axial amplitude as the rotor speed changes includes: For each rotor speed, calculating a difference between a first radial amplitude and a first axial amplitude of the vibration amplitude corresponding to the rotor speed, and plotting a difference change graph with the rotor speed as the horizontal axis and the difference between the first radial amplitude and the first axial amplitude as the vertical axis; The determining, based on the difference change graph, the amplitude to be analyzed that meets the preset change threshold and the second radial amplitude to be analyzed and the second axial amplitude to be analyzed corresponding to the amplitude to be analyzed, includes: In the difference change diagram, the difference corresponding to each rotor speed is checked one by one. When the difference of a certain rotor speed reaches or exceeds the preset change threshold, the vibration amplitude corresponding to the rotor speed is used as the amplitude to be analyzed. After determining the amplitude to be analyzed, the second radial amplitude to be analyzed and the second axial amplitude to be analyzed of the amplitude to be analyzed are obtained based on the recorded correspondence between the speed and the amplitude.

8. A nuclear power steam turbine rotor fault early warning device, characterized in that: The nuclear power steam turbine rotor fault early warning device comprises: a fixed threshold determination module, configured to obtain historical standard information after obtaining a fault type identification result based on the nuclear power steam turbine rotor fault analysis method according to any one of claims 1 to 3, and determine, based on the historical standard information, a fixed threshold corresponding to the operating condition information when an unbalance fault occurs in the rotor; an operating condition parameter extraction module, configured to determine a safety factor affecting the normal operation of the rotor according to the fixed threshold, and extract, based on the safety factor, an operating condition parameter value corresponding to the safety factor from the operating condition information; The warning generation module is used to determine whether the operating condition parameter value meets the fixed threshold value, and if it does not meet the fixed threshold value, generate a warning signal.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the nuclear power steam turbine rotor fault analysis method described in any one of claims 1 to 3 or the nuclear power steam turbine rotor fault early warning method described in any one of claims 4 to 6 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the nuclear power steam turbine rotor fault analysis method according to any one of claims 1 to 3 or the nuclear power steam turbine rotor fault early warning method according to any one of claims 4 to 6 is implemented.

Images