Turbine generator fault maintenance method, system and related device
By combining wavelet threshold denoising and multimodal feature fusion with a deep learning model, the shortcomings of traditional steam turbine generator fault diagnosis and maintenance methods are addressed, achieving efficient and scientific fault prediction and maintenance, and improving the operational reliability and maintenance quality of steam turbine generators.
Patent Information
- Application Number
- CN202510895393.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-30
- Publication Date
- 2025-10-17
AI Technical Summary
Traditional methods for troubleshooting steam turbine generators often involve over- or under-troubleshooting, and the failure to address problems promptly can lead to their escalation. Furthermore, these methods lack scientific rigor and systematic approach, resulting in low maintenance efficiency and difficulty in ensuring quality.
Wavelet threshold denoising and multimodal feature fusion technology are used to process real-time operating parameters. Combined with the deep learning fault prediction model and fault database, the fault type is predicted and targeted maintenance plans are generated.
It improves the accuracy of fault prediction, generates more targeted maintenance plans, improves maintenance efficiency and quality, avoids the occurrence and development of equipment failures, and enhances the operational reliability of steam turbine generators.
Smart Images
Figure FT_1
Abstract
Description
TECHNICAL FIELD
[0001] The present application belongs to the field of steam turbine generators, and particularly relates to a steam turbine generator fault maintenance method, system and related device. BACKGROUND
[0002] In industrial production, steam turbine generators serve as key power equipment, and their stable operation is crucial for ensuring production efficiency. However, during long-term operation, steam turbine generators inevitably encounter various faults, which not only affect their normal operation but also pose a threat to production safety. Currently, steam turbine generators have diverse fault modes, including abnormal vibration, overspeed, steam seal leakage, bearing temperature being too high, vacuum level being too low, etc. These fault modes are often caused by multiple factors, such as rotor imbalance, bearing wear, uneven thermal expansion of the cylinder, and regulating system failure, etc.
[0003] In terms of maintenance methods, traditional methods mainly rely on planned maintenance and post-maintenance. Planned maintenance can periodically conduct comprehensive checks on steam turbine generators, but often leads to over-maintenance or insufficient maintenance, resulting in resource waste or failure to timely discover hidden dangers. Post-maintenance, on the other hand, involves emergency treatment after a fault occurs, which can solve immediate problems but cannot avoid the loss caused by the fault to production, and may lead to the expansion of the fault due to untimely treatment.
[0004] In addition, current steam turbine generator maintenance work also faces problems such as lack of technical personnel and reliance on subjective experience. Many thermal power plants lack professional maintenance personnel, and maintenance work is often based on experience, lacking scientificity and systematicness. This not only leads to low efficiency and quality of maintenance, but also may cause new faults due to misjudgment or omission. Therefore, exploring a more scientific and efficient steam turbine generator fault diagnosis and maintenance method is of great significance for improving the operation reliability of steam turbine generators and reducing maintenance costs. SUMMARY
[0005] The purpose of the present application is to provide a steam turbine generator fault maintenance method, system and related device, which solves the problem of over-maintenance, insufficient maintenance or possible expansion of faults due to untimely treatment in traditional steam turbine generator fault maintenance methods.
[0006] To achieve the above purpose, the technical solution adopted by the present application is as follows: In a first aspect, the present application provides a steam turbine generator fault maintenance method, comprising the following steps: The obtained real-time operation parameters of the steam turbine generator to be tested are sequentially subjected to denoising and multi-modal feature fusion processing to obtain a feature vector. The feature vector is taken as an input of a pre-constructed steam turbine generator fault prediction model, so as to obtain a type of fault of the to-be-tested steam turbine generator in a future time period; The predicted fault type is combined with a pre-constructed fault database, so as to obtain a current maintenance scheme corresponding to the to-be-tested steam turbine generator.
[0007] Preferably, the obtained real-time operation parameters of the to-be-tested steam turbine generator are sequentially subjected to denoising processing and multi-modal feature fusion processing, so as to obtain the feature vector, and the specific method is as follows: The real-time operation parameters of the to-be-tested steam turbine generator are subjected to denoising processing by using a wavelet threshold denoising method, so as to obtain denoised different modal operation parameters; The denoised different modal operation parameters are subjected to feature extraction, and the extracted different modal features are fused, so as to obtain a comprehensive feature vector.
[0008] Preferably, the pre-constructed steam turbine generator fault prediction model comprises an input layer, a CNN feature extraction module, an LSTM time sequence modeling module, a full connection layer and an output layer.
[0009] Preferably, the pre-constructed fault database is obtained by the following method: The historical operation data, fault records and maintenance logs of the steam turbine generator are obtained; The obtained historical operation data, fault records and maintenance logs are subjected to cleaning and standardization processing, so as to obtain processed operation data, fault records and maintenance logs; The processed operation data, fault records and maintenance logs are associated according to time or equipment ID, so as to form data information; The data information and expert knowledge are combined by using a fault tree analysis method, so as to establish a fault database, and the fault database comprises an equipment information table, an operation data table, a fault record table and a maintenance log table.
[0010] In a second aspect, the present application provides a steam turbine generator fault maintenance system, which comprises: A feature vector acquisition unit is configured to sequentially subject the obtained real-time operation parameters of the to-be-tested steam turbine generator to denoising processing and multi-modal feature fusion processing, so as to obtain a feature vector; A fault type prediction unit is configured to take the feature vector as an input of a pre-constructed steam turbine generator fault prediction model, so as to obtain a type of fault of the to-be-tested steam turbine generator in a future time period; A maintenance scheme acquisition unit is configured to combine the predicted fault type with a pre-constructed fault database, so as to obtain a current maintenance scheme corresponding to the to-be-tested steam turbine generator.
[0011] Preferably, the pre-constructed steam turbine generator fault prediction model comprises an input layer, a CNN feature extraction module, an LSTM time series modeling module, a full connection layer and an output layer.
[0012] Preferably, the pre-constructed fault database comprises a device information table, an operation data table, a fault record table and a maintenance log table.
[0013] In a third aspect, the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores computer instructions, and when the computer instructions are executed by the processor, the electronic device implements the method.
[0014] In a fourth aspect, the present application provides a computer program product, wherein the computer program product comprises computer executable instructions, and when the computer executable instructions are executed, the method is implemented.
[0015] In a fifth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores computer executable instructions, and when the computer executable instructions are executed by a processor, the method is implemented.
[0016] Compared with the prior art, the present application has the following beneficial effects: The method for fault maintenance of a steam turbine generator provided by the present application can obtain a current corresponding maintenance scheme of the steam turbine generator to be tested by combining the predicted fault type with the pre-constructed fault database. Due to the improved accuracy of fault prediction, the generated maintenance scheme is more targeted, and corresponding maintenance measures can be taken for specific fault types, thereby improving the maintenance efficiency and quality. The scheme can effectively avoid the occurrence and development of equipment faults and improve the operation reliability of the steam turbine generator by discovering early signs of equipment faults in time and taking corresponding maintenance measures. BRIEF DESCRIPTION OF DRAWINGS
[0017] Figure 1 is a flowchart of an embodiment of the present application. DETAILED DESCRIPTION
[0018] In the following description, specific details are set forth in order to provide a thorough understanding of the embodiments of the present application. However, persons skilled in the art will understand that the present application can be practiced without these specific details. In other instances, well-known systems, devices, circuits and methods have not been described in detail so as not to obscure the description of the present application.
[0019] It will be understood that the term “includes,” “comprises, ” “comprising,” “has,” “contains” and / or “including,” when used in the specification and / or the claims of this application, specifies the presence of stated features, integers, steps, operations, elements, and / or components but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.
[0020] It will also be understood that the term “and / or” as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items, and that the term “includes” or “comprises” means “includes but not limited to” or “comprises but not limited to”.
[0021] As used in this specification and claims, the terms “if’ and “when” can be interpreted to mean “upon” or “in response to a determination” or “in response to a detection” depending on the context. Similarly, the phrase “if it is determined” or “if [a described condition or event] is detected” can be interpreted to mean “upon determining” or “in response to a determination” or “upon detecting [the described condition or event]” or “in response to a detection [of the described condition or event]” depending on the context.
[0022] In addition, in the description of the specification and the appended claims, the terms “first,” “second,” “third,” and the like are used merely to distinguish different elements, and can not be interpreted as indicating or implying a relative importance.
[0023] Reference in the specification to “one embodiment” or “some embodiments” or “an embodiment” or “some embodiments” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrases “in one embodiment,” “in some embodiments,” “in other embodiments,” “in additional embodiments” or the like in various places in the specification are not necessarily all referring to the same embodiment, although they can. The terms “comprising,” “including,” “having” and the like are meant to be interpreted as “including but not limited to” unless otherwise indicated by the language of the specification.
[0024] Embodiment 1 The method for troubleshooting of a steam turbine generator provided by the embodiment aims to establish a troubleshooting logic decision tree between the failure mode and the troubleshooting mode of the steam turbine generator, combine failure feature analysis, risk assessment and cost-benefit analysis, and form a scientific and efficient steam turbine generator failure maintenance strategy. The method takes into account the disadvantages of traditional planned maintenance and post-maintenance, realizes the optimal allocation of maintenance resources through a fine and intelligent decision-making process, improves the reliability and economy of equipment operation, and includes the following steps in specific implementation: Step 1, the obtained real-time running parameters of the to-be-tested steam turbine generator are sequentially subjected to denoising processing and multi-modal feature fusion processing to obtain a feature vector; the real-time running parameters include vibration signals, temperature signals and pressure signals; The feature vector is taken as an input of a pre-constructed steam turbine generator fault prediction model to obtain a type of fault of the to-be-tested steam turbine generator in a future time period; Step 2, the predicted fault type is combined with a pre-constructed fault database to obtain a current maintenance scheme corresponding to the to-be-tested steam turbine generator.
[0025] The predicted fault type is combined with a pre-constructed fault database in this embodiment, and a current maintenance scheme corresponding to the to-be-tested steam turbine generator can be obtained. Due to the improved accuracy of fault prediction, the generated maintenance scheme is more targeted, and corresponding maintenance measures can be taken for specific fault types, thereby improving the maintenance efficiency and quality.
[0026] This scheme can effectively avoid the occurrence and development of equipment faults and improve the operation reliability of the steam turbine generator by discovering early signs of equipment faults in a timely manner and taking corresponding maintenance measures.
[0027] Embodiment 2 On the basis of embodiment 1, a steam turbine generator fault maintenance method provided in this embodiment identifies and classifies the fault modes of the steam turbine generator, and the specific method is as follows: Obtain historical running data, fault records and maintenance logs of the steam turbine generator; The obtained historical running data, fault records and maintenance logs are subjected to cleaning and standardization processing to obtain processed running data, fault records and maintenance logs; The processed running data, fault records and maintenance logs are associated according to time or equipment ID to form data information; The data information and expert knowledge are combined by using a fault tree analysis method to establish a fault database, and the fault database includes an equipment information table, a running data table, a fault record table and a maintenance log table.
[0028] Embodiment 3 On the basis of embodiment 1, a steam turbine generator fault maintenance method provided in this embodiment sequentially subjects real-time running parameters of a to-be-tested steam turbine generator to denoising processing and multi-modal feature fusion processing to obtain a feature vector, and the specific method is as follows: The real-time running parameters of the to-be-tested steam turbine generator are subjected to denoising processing by using a wavelet threshold denoising method to obtain denoised different modal running parameters; The denoised different modal running parameters are subjected to feature extraction, and the extracted different modal features are fused to construct a comprehensive feature vector.
[0029] The comprehensive feature vector obtained by wavelet threshold denoising and multi-modal feature extraction and fusion can provide high-quality input for the steam turbine generator fault prediction model, thereby improving the accuracy of fault prediction. Accurate fault prediction can provide a reliable basis for maintenance decision-making, enabling maintenance personnel to develop a reasonable maintenance plan in advance, avoid further expansion of equipment failure, and reduce downtime and maintenance costs.
[0030] Embodiment 4 Based on embodiment 3, the fault maintenance method for a steam turbine generator provided in this embodiment uses a wavelet threshold denoising method to denoise the real-time operating parameters of the steam turbine generator to be tested. The specific method is as follows: The real-time operating parameters are decomposed into wavelet spaces of different scales, and then the wavelet coefficients at each scale are threshold processed to remove the wavelet coefficients corresponding to the noise. Finally, the signal is reconstructed by inverse wavelet transform to obtain the denoised operating parameters.
[0031] Embodiment 5 Based on embodiment 1, the fault maintenance method for a steam turbine generator provided in this embodiment uses a pre-constructed steam turbine generator fault prediction model. The specific construction method is as follows: A hybrid model is constructed based on the combination of a long short-term memory network (LSTM) and a convolutional neural network (CNN) based on deep learning. Obtain historical operating data of the steam turbine generator, including normal operating data and operating data under different fault types. The historical operating data are sequentially denoised and multi-modal feature fused to construct a training data set and a test data set.
[0032] The training data set is used to train the constructed hybrid model to obtain a trained steam turbine generator fault prediction model. In the training process, the backpropagation algorithm and the stochastic gradient descent (SGD) optimization algorithm are used to update the parameters of the model to minimize the cross-entropy loss function between the predicted results and the true fault types. At the same time, the early stopping method and the regularization technique (such as L2 regularization) are used to prevent model overfitting and improve the generalization ability of the model.
[0033] The steam turbine generator fault prediction model includes an input layer, a CNN feature extraction module, an LSTM time series modeling module, a fully connected layer, and an output layer. The CNN feature extraction module includes a convolutional layer and a pooling layer, and the convolutional layer is followed by a ReLU (Rectified Linear Unit) activation function.
[0034] The output layer uses a softmax activation function.
[0035] Embodiment 6 The embodiment provides a fault maintenance system of a steam turbine generator, which comprises the following steps of: The characteristic vector acquisition unit is configured to sequentially perform denoising processing and multi-modal feature fusion processing on the obtained real-time operation parameters of the steam turbine generator to be tested, so as to obtain a characteristic vector. The fault type prediction unit is configured to take the characteristic vector as an input of a pre-constructed steam turbine generator fault prediction model, so as to obtain a type of fault of the steam turbine generator to be tested in a future time period. The maintenance scheme acquisition unit is configured to combine the predicted fault type with a pre-constructed fault database, so as to obtain a current corresponding maintenance scheme of the steam turbine generator to be tested.
[0036] Embodiment 7 On the basis of the embodiment 6, the embodiment provides a fault maintenance system of a steam turbine generator, wherein the pre-constructed steam turbine generator fault prediction model comprises an input layer, a CNN feature extraction module, an LSTM time sequence modeling module, a full connection layer and an output layer.
[0037] The pre-constructed fault database comprises an equipment information table, an operation data table, a fault record table and a maintenance log table.
[0038] Embodiment 8 The embodiment further provides a computing device. The computing device comprises a bus, a processor, a memory and a communication interface. The processor, the memory and the communication interface communicate through the bus. The computing device can be a server or a terminal device. It should be understood that the number of processors and memories in the computing device is not limited in the present application.
[0039] The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus can include a channel for transmitting information between various components (for example, the memory, the processor, the communication interface) of the computing device.
[0040] The processor can include any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a Tensor Processing Unit (TPU), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), a microprocessor (MP), or a digital signal processor (DSP), etc.
[0041] The memory can include a volatile memory, such as a random access memory (RAM). The processor can also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD).
[0042] The memory stores executable program codes, and the processor executes the executable program codes to respectively implement the functions of the aforementioned various units, thereby implementing, for example, the method described in Embodiment 1, etc. That is, the memory can store instructions for the methods and functions of the computing device involved in any of the above embodiments.
[0043] The communication interface uses a transceiving module such as, but not limited to, a network interface card, a transceiver, etc., to implement the communication between the computing device and other devices or communication networks.
[0044] Embodiment 9 The embodiments also provide a computer-readable storage medium having stored thereon computer instructions, which, when executed by a processor, cause the processor to perform the methods and functions of the computing device involved in any of the above embodiments.
[0045] In general, the various embodiments of the present disclosure can be implemented in hardware or special-purpose circuits, software, logic or any combination thereof. Some aspects can be implemented in hardware, while other aspects can be implemented in firmware or software which can be executed by a controller, microprocessor or other computing device. While various aspects of embodiments of the present disclosure are illustrated and described as block diagrams, flow charts, or using some other pictorial representation, it will be understood that the blocks, apparatus, systems, techniques or methods described herein can be implemented in, as non-limiting examples, hardware, software, firmware, special-purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.
[0046] Embodiment 10 This embodiment provides at least one computer program product which is tangibly stored on a non-transitory computer readable storage medium. The computer program product includes computer executable instructions, for example, instructions included in program modules, executed by devices at a target real or virtual processor to perform processes / methods as described above with reference to the accompanying figures. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform particular tasks or implement particular abstract data types. In various embodiments, the functionality of program modules can be combined or split between program modules as desired. Machine executable instructions for program modules can be executed within a local or distributed device. In a distributed device, program modules can be located in local and remote memory storage devices.
[0047] Computer program code for carrying out operations of the present disclosure can be written in one or more programming languages. These computer program codes can be provided to a processor of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus to cause the program codes to be executed by the computer or other programmable data processing apparatus, such that the functions / operations specified in the flowcharts and / or block diagrams are performed. The program codes can be executed entirely on the computer, partially on the computer, as a stand-alone software package, partially on the computer and partially on a remote computer, or entirely on a remote computer or server.
[0048] In the context of the present disclosure, computer program code or related data can be carried by any suitable carrier, to enable the device, apparatus or processor to perform the various processes and operations described above. Examples of carriers include signals, computer readable media, and the like. Examples of signals can include electrical, optical, radio, sound or other forms of propagated signals, such as carrier waves, infrared signals, and the like.
[0049] A computer readable medium can be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the computer readable storage medium include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), and a digital versatile disc (DVD), or any suitable combination of the foregoing.
[0050] The above-described embodiments are merely intended to illustrate the technical solutions of the present application, but not to limit the same; even though the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A method for troubleshooting a steam turbine generator, characterized in that: The following steps are involved: The obtained real-time operating parameters of the steam turbine generator to be tested are subjected to denoising and multi-modal feature fusion processing in sequence to obtain a feature vector; The characteristic vector is used as the input of the pre-built turbine generator fault prediction model to obtain the type of fault that will occur in the turbine generator under test in the future time period; The predicted fault type is combined with the pre-built fault database to obtain the current maintenance plan corresponding to the steam turbine generator under test.
2. A method for troubleshooting a steam turbine generator according to claim 1, characterized in that: The obtained real-time operating parameters of the steam turbine generator to be tested are subjected to denoising and multi-modal feature fusion processing in turn to obtain a feature vector. The specific method is: The wavelet threshold denoising method is used to denoise the real-time operating parameters of the steam turbine generator to obtain the denoised operating parameters of different modes. Feature extraction is performed on the denoised different modal operating parameters, and the extracted different modal features are fused to construct a comprehensive feature vector.
3. A method for troubleshooting a steam turbine generator according to claim 1, characterized in that: The pre-built turbine generator fault prediction model includes an input layer, a CNN feature extraction module, an LSTM time series modeling module, a fully connected layer, and an output layer.
4. A method for troubleshooting a steam turbine generator according to claim 1, characterized in that: Pre-built fault database, specifically: Obtain historical operating data, fault records and maintenance logs of steam turbine generators; The historical operation data, fault records and maintenance logs are cleaned and standardized to obtain processed operation data, fault records and maintenance logs; The processed operation data, fault records and maintenance logs are associated by time or equipment ID to form data information; The fault tree analysis method is used to combine data information and expert knowledge to establish a fault database, which includes equipment information table, operation data table, fault record table and maintenance log table.
5. A fault repair system for a steam turbine generator, characterized in that: include: A characteristic vector acquisition unit is used to sequentially perform denoising and multi-modal feature fusion processing on the acquired real-time operating parameters of the steam turbine generator to obtain a characteristic vector; A fault type prediction unit is used to use the feature vector as an input of a pre-built steam turbine generator fault prediction model to obtain the type of fault that may occur in the steam turbine generator to be tested in a future time period; The maintenance plan acquisition unit is used to combine the predicted fault type with the pre-built fault database to obtain the current maintenance plan corresponding to the steam turbine generator to be tested.
6. A steam turbine generator fault repair system according to claim 5, characterized in that: The pre-built turbine generator fault prediction model includes an input layer, a CNN feature extraction module, an LSTM time series modeling module, a fully connected layer, and an output layer.
7. A steam turbine generator fault repair system according to claim 5, characterized in that: The pre-built fault database includes equipment information table, operation data table, fault record table and maintenance log table.
8. An electronic device, characterized in that: The electronic device comprises a processor and a memory, wherein computer instructions are stored in the memory. When the computer instructions are executed by the processor, the electronic device executes the method according to any one of claims 1 to 5.
9. A computer program product, characterized in that The computer program product contains computer-executable instructions, which implement the method according to any one of claims 1 to 5 when executed.
10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which implement the method according to any one of claims 1 to 5 when executed by a processor.