Wind power operation and maintenance fault diagnosis method and system

Abstract

The invention provides a wind power operation and maintenance fault diagnosis method and system, and belongs to the technical field of wind power operation and maintenance. The method comprises the following steps: constructing a wind power operation and maintenance knowledge base based on multi-source historical operation data of target wind power equipment and operation knowledge experience data of the target wind power equipment; training based on a wind power operation and maintenance knowledge base to obtain a fault diagnosis model; wherein the fault diagnosis model comprises a plurality of feature extraction combination layers used for extracting data features of corresponding layers in the multi-source historical operation data of the target wind power equipment; inputting the multi-source real-time operation data, collected in real time, of the target wind power equipment into the fault diagnosis model, so as to extract the data features of the corresponding layers in the multi-source real-time operation data of the target wind power equipment based on each feature extraction combination layer, and extracting the multi-source real-time operation data of the target wind power equipment based on the data features of each layer; and determining a wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment. Intelligent operation and maintenance management of the wind power equipment is realized, the operation and maintenance efficiency and quality are improved, and the operation and maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power operation and maintenance, and in particular to a wind power operation and maintenance fault diagnosis method, a wind power operation and maintenance fault diagnosis system, a machine-readable storage medium and an electronic device. Background Art

[0002] As the global demand for clean energy continues to grow, wind power generation has been widely used as a renewable energy technology. However, wind farms are usually located in remote areas with harsh environments and difficult equipment maintenance. Traditional wind power operation and maintenance mainly rely on manual inspections and experience judgment, which has problems such as low efficiency, poor accuracy, and high cost.

[0003] The prior art provides a maintenance knowledge aggregation method based on data statistics, which collects historical operation data and historical processing experience of wind turbines and uses mathematical statistical methods to establish a relationship model between these historical operation data and historical processing experience and equipment failures or performance changes. Although data statistics can analyze some common operation and maintenance problems and form operation and maintenance knowledge, for some complex failures, such as simultaneous failures of multiple components and intermittent failures, methods based on data statistics may be difficult to accurately analyze the specific root causes.

[0004] Therefore, how to realize intelligent operation and maintenance management of wind power equipment, improve operation and maintenance efficiency and quality, and reduce operation and maintenance costs is an urgent problem that needs to be solved. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide a wind power operation and maintenance fault diagnosis method and system to at least solve the above-mentioned problem of how to realize intelligent operation and maintenance management of wind power equipment, improve operation and maintenance efficiency and quality, and reduce operation and maintenance costs.

[0006] In order to achieve the above-mentioned object, the first aspect of the present invention provides a wind power operation and maintenance fault diagnosis method, comprising: Construct a wind power operation and maintenance knowledge base based on multi-source historical operation data of target wind power equipment and operation knowledge and experience data of target wind power equipment; A fault diagnosis model is obtained by training based on a wind power operation and maintenance knowledge base; wherein the fault diagnosis model includes a plurality of feature extraction combination layers for extracting data features of corresponding levels in multi-source historical operation data of target wind power equipment; The multi-source real-time operation data of the target wind power equipment collected in real time is input into the fault diagnosis model to extract the data features of the corresponding levels in the multi-source real-time operation data of the target wind power equipment based on each feature extraction combination layer, and based on the data features of each level, determine the wind power operation and maintenance fault diagnosis results corresponding to the target wind power equipment.

[0007] Optionally, the multi-source historical operation data of the target wind power equipment includes operation data, environmental data and maintenance record data of the target wind power equipment; The rules for obtaining multi-source historical operation data of target wind power equipment include: The operation data, environmental data and maintenance record data of the target wind power equipment are obtained through various sensors deployed on the target wind power equipment and the data interface connected to the industrial control system of the target wind power equipment.

[0008] Optionally, after acquiring multi-source historical operation data of the target wind power equipment, the method further includes: Preprocess the multi-source historical operation data of the target wind power equipment; The preprocessing rules include: The multi-source historical operation data of the target wind power equipment are sequentially cleaned, denoised and normalized.

[0009] Optionally, the construction rules of the wind power operation and maintenance knowledge base include: Based on the multi-source historical operation data of the target wind power equipment, a corresponding relational database and / or non-relational database is constructed to obtain a wind power operation and maintenance database; The wind power operation and maintenance database and the operating knowledge and experience data of the target wind power equipment are integrated by using the knowledge graph method to construct a wind power operation and maintenance knowledge base.

[0010] Optionally, the training rules of the above fault diagnosis model include: Perform fault state labeling on the operation fault data of the target wind power equipment in the wind power operation and maintenance knowledge base to obtain training samples, and divide the training samples into a training set and a test set; Obtaining an initial model, initializing the weights and biases of the initial model, training the initial model using a training set, and iteratively updating the weights and biases during the training process until the model converges or the number of training times reaches a preset number of training times, thereby obtaining an initial model of the fault diagnosis model; Using the test set to test the initial model of the fault diagnosis model, based on the test results, calculate the performance index value corresponding to the initial model of the fault diagnosis model; Based on the performance indicator values ​​corresponding to the initial model of the fault diagnosis model, the initial model of the fault diagnosis model is adjusted to obtain a trained fault diagnosis model; wherein the method of adjusting the initial model of the fault diagnosis model at least includes increasing the number of layers of the fault diagnosis model and adjusting the number of neurons of the fault diagnosis model.

[0011] Optionally, after obtaining the fault diagnosis model, the method further includes: Optimize the parameter combination of multi-source real-time operation data input to the fault diagnosis model; Among them, the optimization rules for parameter combinations of multi-source real-time operation data include: With the goal of maximizing the power generation efficiency of the target wind power equipment, the operating data and environmental data of the target wind power equipment are taken as the optimization objects, and the parameter combination of the optimization object is optimized through the genetic algorithm.

[0012] Optionally, the above rules for iteratively updating weights and biases include: Use a small batch stochastic gradient descent algorithm to iteratively update weights and biases; Among them, the specific contents of the small batch stochastic gradient descent algorithm include: A preset number of training data are randomly selected from the training set to calculate the gradient, and the weights and biases are updated based on the gradient calculation results.

[0013] Optionally, after obtaining the wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment, the method further includes: When it is determined that the wind power operation and maintenance fault diagnosis result is that the target wind power equipment currently has a wind power operation and maintenance fault, fault alarm information is generated based on the wind power operation and maintenance fault diagnosis result; wherein the fault alarm information at least includes fault type information, fault location information and fault severity information of the target wind power equipment; When it is determined that the wind power operation and maintenance fault diagnosis result indicates that the target wind power equipment will have a wind power operation and maintenance fault at a future moment, a wind power equipment maintenance plan is generated based on the wind power operation and maintenance fault diagnosis result, and after completing the task allocation of the wind power equipment maintenance plan, the execution data of the wind power equipment maintenance plan is obtained in real time.

[0014] A second aspect of the present invention provides a wind power operation and maintenance fault diagnosis system, comprising: A wind power operation and maintenance knowledge base construction module is used to construct a wind power operation and maintenance knowledge base based on multi-source historical operation data of target wind power equipment and operation knowledge experience data of target wind power equipment; A fault diagnosis model training module is used to obtain a fault diagnosis model based on a wind power operation and maintenance knowledge base; wherein the fault diagnosis model includes a plurality of feature extraction combination layers for extracting data features of corresponding levels in multi-source historical operation data of target wind power equipment; The fault diagnosis module is used to input the multi-source real-time operation data of the target wind power equipment collected in real time into the fault diagnosis model, so as to extract the data features of the corresponding levels in the multi-source real-time operation data of the target wind power equipment based on each feature extraction combination layer, and determine the wind power operation and maintenance fault diagnosis results corresponding to the target wind power equipment based on the data features of each level.

[0015] In a third aspect of the present invention, a machine-readable storage medium is provided, on which instructions are stored, and when the instructions are executed by a processor, the processor is configured to execute the wind power operation and maintenance fault diagnosis method.

[0016] In a fourth aspect of the present invention, an electronic device is provided. The electronic device includes 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 wind power operation and maintenance fault diagnosis method is implemented.

[0017] Through the above technical scheme, a wind power operation and maintenance fault diagnosis method and system are provided to construct a wind power operation and maintenance knowledge base based on the multi-source historical operation data of the target wind power equipment and the operation knowledge and experience data of the target wind power equipment. By integrating the multi-source historical operation data of the target wind power equipment, more comprehensive and accurate wind turbine operation status information can be obtained. A fault diagnosis model is obtained by training based on the wind power operation and maintenance knowledge base; wherein the fault diagnosis model includes a plurality of feature extraction combination layers for extracting data features of corresponding levels in the multi-source historical operation data of the target wind power equipment. The multi-source real-time operation data of the target wind power equipment is input into the trained fault diagnosis model, and the trained fault diagnosis model extracts the data features of corresponding levels in the multi-source real-time operation data of the target wind power equipment through each feature extraction combination layer, and determines the wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment based on the data features of each level. The method and system use artificial intelligence technology to analyze and learn massive amounts of wind power operation and maintenance knowledge (i.e., operating knowledge and experience data of wind power equipment), process and analyze complex multi-source historical operating data of wind turbines, and can accurately extract deep-level features of wind power equipment operating data after training, thereby improving the accuracy and generalization ability of fault diagnosis, and improving operation and maintenance efficiency. This realizes intelligent analysis and management of wind power operation and maintenance faults, and thus realizes intelligent operation and maintenance management of wind power equipment, improves operation and maintenance efficiency and quality, and reduces operation and maintenance costs.

[0018] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings: Figure 1 It is a flow chart of a wind power operation and maintenance fault diagnosis method provided by an embodiment of the present invention; Figure 2 It is a block diagram of a wind power operation and maintenance fault diagnosis system provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of the structure of an electronic device provided by a preferred embodiment of the present invention.

[0020] Description of Reference Numerals 10 - electronic device, 100 - processor, 101 - memory, 102 - computer program. DETAILED DESCRIPTION

[0021] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0022] Figure 1 FIG. 1 is a flow chart of a wind power operation and maintenance fault diagnosis method provided by an embodiment of the present invention. Figure 1 As shown, an embodiment of the present invention provides a wind power operation and maintenance fault diagnosis method, comprising: S110: constructing a wind power operation and maintenance knowledge base based on multi-source historical operation data of the target wind power equipment and operation knowledge and experience data of the target wind power equipment; In some implementations of this embodiment, the multi-source historical operation data of the above-mentioned target wind power equipment includes the operation data, environmental data and maintenance record data of the target wind power equipment; the acquisition rules of the multi-source historical operation data of the target wind power equipment include: acquiring the operation data, environmental data and maintenance record data of the target wind power equipment through various sensors deployed on the target wind power equipment and the data interface connected to the industrial control system of the target wind power equipment.

[0023] Specifically, the method can collect operating data, environmental data, maintenance record data, etc. of the target wind power equipment through sensors, monitoring equipment, and data interfaces.

[0024] Among them, the methods of collecting multi-source historical operation data of target wind power equipment include: 1. Collect multi-source historical operating data of the target wind power equipment by installing sensors and monitoring equipment: Install temperature sensors, pressure sensors, vibration sensors, wind speed sensors, etc. on the target wind power equipment to collect the operating status and environmental data of the target wind power equipment in real time. 2. Obtain multi-source historical operating data of the target wind power equipment through the data interface: Connect the data interface with the industrial control system of the target wind power equipment (such as the control system, SCADA system, etc.) to obtain the operating parameters, fault records and other information of the target wind power equipment. In some implementations of this embodiment, after acquiring the multi-source historical operation data of the target wind power equipment, the method also includes: preprocessing the multi-source historical operation data of the target wind power equipment; wherein the preprocessing rules include: sequentially performing data cleaning, denoising and normalization processing on the multi-source historical operation data of the target wind power equipment.

[0025] Specifically, by sequentially performing data cleaning, denoising and normalization on the collected multi-source historical operation data of the target wind power equipment, the quality and availability of the multi-source historical operation data of the target wind power equipment can be improved. Among them, data cleaning includes removing outliers, noise and duplicate data to ensure the accuracy and integrity of the data. Denoising is to denoise the data using a filtering algorithm to improve the quality of the data. The denoised data is normalized to map the data to the [0,1] interval for subsequent data analysis and processing.

[0026] In some implementations of this embodiment, the construction rules of the above-mentioned wind power operation and maintenance knowledge base include: based on the multi-source historical operation data of the target wind power equipment, constructing a corresponding relational database and / or non-relational database to obtain a wind power operation and maintenance database; using a knowledge graph to integrate the wind power operation and maintenance database and the operation knowledge and experience data of the target wind power equipment to construct a wind power operation and maintenance knowledge base.

[0027] Specifically, database technology and knowledge graph technology can be used to integrate and store the pre-processed multi-source historical operation data of the target wind power equipment and related operation and maintenance knowledge, experience and other information (i.e., the operation knowledge and experience data of the target wind power equipment) to build a wind power operation and maintenance knowledge base. By integrating the multi-source historical operation data of the target wind power equipment, more comprehensive and accurate wind turbine operation status information can be obtained, and appropriate knowledge representation methods (i.e., database technology and knowledge graph technology) can be used to convert the integrated multi-source historical operation data of the target wind power equipment and the operation knowledge and experience data of the target wind power equipment into a form that can be processed by a computer, i.e., a wind power operation and maintenance knowledge base.

[0028] The specific construction process of the wind power operation and maintenance knowledge base is as follows: 1. Database design: Use relational database and non-relational database to design wind power operation and maintenance database to store equipment information, operation data, maintenance records, fault cases and other information.

[0029] 2. Knowledge graph construction: Use knowledge graph technology to construct a wind power operation and maintenance knowledge graph (i.e., wind power operation and maintenance knowledge base) to associate and integrate equipment information, component information, fault information, and maintenance measures information to facilitate knowledge query and reasoning.

[0030] S120: Obtaining a fault diagnosis model based on wind power operation and maintenance knowledge base training; wherein the fault diagnosis model includes a plurality of feature extraction combination layers for extracting data features of corresponding levels in multi-source historical operation data of target wind power equipment; In some implementations of this embodiment, a convolutional neural network (CNN) in a deep learning algorithm is used to construct a fault diagnosis model. For example, the vibration signal data of the wind turbine is input into the CNN model. After training, the CNN model can automatically identify different types of faults, such as bearing faults, gear faults, etc. When new vibration data is input, the CNN model can quickly and accurately diagnose the fault type and fault location. Among them, deep learning technology processes and analyzes the complex data of wind turbines, and after training, it can accurately extract the deep-level features of the data, thereby improving the accuracy and generalization ability of fault diagnosis and drawing correct conclusions.

[0031] In some implementations of this embodiment, a CNN model is used to design different combinations of convolutional layers, pooling layers, and fully connected layers to obtain multiple feature extraction combination layers (wherein each feature extraction combination layer includes a corresponding convolutional layer, pooling layer, and fully connected layer) to extract data features of corresponding levels in the multi-source historical operation data of the target wind power equipment.

[0032] In some implementations of this embodiment, the training rules of the above fault diagnosis model include: Perform fault state labeling on the operation fault data of the target wind power equipment in the wind power operation and maintenance knowledge base to obtain training samples, and divide the training samples into a training set and a test set; Specifically, the data set in the annotated wind power operation and maintenance knowledge base is divided into a training set and a test set. The training set is used to train the deep learning model (i.e., the pre-built fault diagnosis model), and the test set is used to evaluate the performance of the initial model of the fault diagnosis model. 70% of the data in the annotated data set in the wind power operation and maintenance knowledge base is used as the training set, and 30% of the data is used as the test set.

[0033] Among them, the two steps of cleaning and denoising the collected multi-source historical operation data of the target wind turbine equipment, and labeling and classifying the cleaned multi-source historical operation data of the target wind turbine equipment are crucial to ensuring the quality and reliability of the multi-source historical operation data, and effectively avoid the possible erroneous analysis results caused by inaccurate multi-source historical operation data.

[0034] After obtaining the initial model and initializing the weights and biases of the initial model, specifically, initializing the constructed deep learning model (i.e., the pre-constructed fault diagnosis model) includes setting the initial weights and biases (it should be noted that the initialized neural network parameters mainly include weights and biases. The weights are the parameters connecting neurons, and the bias term is used to adjust the activation threshold of neurons). Using the pre-trained model weights (i.e., the initialized weights and biases) to initialize the fault diagnosis model can speed up the convergence of the fault diagnosis model.

[0035] The initial model is trained using the training set, and during the training process, the weights and biases are iteratively updated until the model converges or the number of training times reaches a preset number of training times, thereby obtaining an initial model of the fault diagnosis model; In some implementations of this embodiment, the above-mentioned rules for iteratively updating weights and biases include: using a small batch stochastic gradient descent algorithm to iteratively update weights and biases; wherein the specific content of the small batch stochastic gradient descent algorithm includes: randomly selecting a preset number of training data from a training set to calculate the gradient, and updating the weights and biases based on the gradient calculation results.

[0036] Specifically, in the process of training the initialized fault diagnosis model using the training set, the optimization algorithm can use stochastic gradient descent (SGD), and the loss function can select mean square error (MSE). The deep learning model (i.e., the initialized fault diagnosis model) is trained using the training set. During the training process, the output of the fault diagnosis model is made as close to the true value as possible by continuously adjusting the weights and biases of the fault diagnosis model. The rules for adjusting the weights and biases of the fault diagnosis model include using a small batch stochastic gradient descent algorithm, randomly selecting a small batch of data from the training set each time to calculate the gradient, and updating the weights and biases. The training process can last for multiple epochs until the fault diagnosis model converges or the number of training times reaches a preset number of training times.

[0037] It should be noted that model convergence means that in the process of machine learning and deep learning, the output of the model gradually approaches a stable state and no longer changes significantly. This usually means that the learning process of the model has reached a balance, can fit the training data well, and can generalize to unseen data to a certain extent. Among them, the judgment criteria for model convergence include: 1. Changes in the value of the loss function: During the training process, if the value of the loss function gradually decreases and tends to stabilize, it means that the model is converging; 2. Performance of the training and validation sets: The performance of the model on the training set gradually improves, and the performance on the validation set also remains stable or improves, indicating that the model is not overfitting and is effectively converging; 3. The amplitude of parameter updates: The amplitude of model parameter updates gradually decreases, indicating that the model is close to the optimal solution and is in a convergence state.

[0038] Using the test set to test the initial model of the fault diagnosis model, based on the test results, calculate the performance index value corresponding to the initial model of the fault diagnosis model; Specifically, the test set is used to test the initial model of the fault diagnosis model, and based on the test results, the performance indicator values ​​(such as accuracy, recall rate, and F1 value, etc.) corresponding to the initial model of the fault diagnosis model are calculated.

[0039] Based on the performance indicator values ​​corresponding to the initial model of the fault diagnosis model, the initial model of the fault diagnosis model is adjusted to obtain a trained fault diagnosis model; wherein the method of adjusting the initial model of the fault diagnosis model at least includes an adjustment method of increasing the number of layers of the fault diagnosis model and an adjustment method of adjusting the number of neurons of the fault diagnosis model.

[0040] Specifically, according to the performance indicator values ​​corresponding to the initial model of the fault diagnosis model, the initial model of the fault diagnosis model is tuned to adjust the parameters and architecture of the fault diagnosis model, thereby improving the performance of the fault diagnosis model. The tuning is performed by increasing the number of layers of the fault diagnosis model, adjusting the number of neurons of the fault diagnosis model, changing the parameters of the optimization algorithm, etc. In some implementations of this embodiment, cross-validation and other techniques can also be used to avoid overfitting.

[0041] In some implementations of this embodiment, after obtaining the trained fault diagnosis model, the method also includes: optimizing the parameter combination of multi-source real-time operation data input into the fault diagnosis model; wherein, the optimization rules for the parameter combination of multi-source real-time operation data include: taking maximizing the power generation efficiency of the target wind power equipment as the goal, taking the operation data and environmental data of the target wind power equipment as the optimization objects, and optimizing the parameter combination of the optimization object through a genetic algorithm.

[0042] Specifically, a genetic algorithm is used to optimize the parameter combination of multi-source real-time operating data input to the fault diagnosis model. With the goal of maximizing the power generation efficiency of the target wind power equipment, environmental data such as wind speed, wind direction, temperature, and the operating data of the target wind power equipment are used as input, and the operating parameter combination is continuously optimized through a genetic algorithm. For example, the pitch angle, speed and other parameters of the wind turbine are adjusted to adapt to different wind conditions and improve power generation efficiency.

[0043] In the above implementation process, the method uses artificial intelligence technologies such as machine learning and deep learning to analyze and mine the data in the wind power operation and maintenance knowledge base, realizing functions such as fault diagnosis, predictive maintenance, and optimized operation. Artificial intelligence technology can be used to analyze and mine a large amount of wind power equipment operation data, realize functions such as fault diagnosis, predictive maintenance, and optimized operation, and improve the efficiency and quality of wind power operation and maintenance.

[0044] S130: Input the multi-source real-time operation data of the target wind power equipment collected in real time into the fault diagnosis model, extract the data features of the corresponding levels in the multi-source real-time operation data of the target wind power equipment based on each feature extraction combination layer, and determine the wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment based on the data features of each level.

[0045] Specifically, the method constructs a wind power operation and maintenance knowledge base based on the multi-source historical operation data of the target wind power equipment and the operation knowledge and experience data of the target wind power equipment. By integrating the multi-source historical operation data of the target wind power equipment, more comprehensive and accurate wind turbine operation status information can be obtained. A fault diagnosis model is obtained by training based on the wind power operation and maintenance knowledge base; wherein the fault diagnosis model includes a plurality of feature extraction combination layers for extracting data features of corresponding levels in the multi-source historical operation data of the target wind power equipment. The multi-source real-time operation data of the target wind power equipment is input into the trained fault diagnosis model, and the trained fault diagnosis model extracts data features of corresponding levels in the multi-source real-time operation data of the target wind power equipment through each feature extraction combination layer, and determines the wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment based on the data features of each level. This method uses artificial intelligence technology to analyze and learn massive amounts of wind power operation and maintenance knowledge (i.e., the operating knowledge and experience data of wind power equipment), and processes and analyzes the complex multi-source historical operation data of wind turbines. After training, it can accurately extract the deep-level characteristics of the operating data of wind power equipment, improve the accuracy and generalization ability of fault diagnosis, improve operation and maintenance efficiency, and realize intelligent analysis and management of wind power operation and maintenance faults, thereby realizing intelligent operation and maintenance management of wind power equipment, improving operation and maintenance efficiency and quality, and reducing operation and maintenance costs.

[0046] In some implementations of the present embodiment, after obtaining the wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment, the method also includes: when it is determined that the wind power operation and maintenance fault diagnosis result is that the target wind power equipment currently has a wind power operation and maintenance fault, generating fault alarm information based on the wind power operation and maintenance fault diagnosis result; wherein the fault alarm information includes at least fault type information, fault location information and fault severity information of the target wind power equipment; when it is determined that the wind power operation and maintenance fault diagnosis result is that the target wind power equipment will have a wind power operation and maintenance fault at a future time, generating a wind power equipment maintenance plan based on the wind power operation and maintenance fault diagnosis result, and after completing the task allocation of the wind power equipment maintenance plan, obtaining the execution data of the wind power equipment maintenance plan in real time.

[0047] Specifically, according to the wind power operation and maintenance fault diagnosis results corresponding to the target wind power equipment, decision support and task management are provided to the operation and maintenance personnel to realize the intelligent operation and maintenance management of the wind power equipment.

[0048] Among them, the intelligent operation and maintenance management of wind power equipment includes functions such as fault alarm, maintenance plan formulation and task allocation, as follows: 1. Fault alarm: When a target wind power equipment fails, an alarm message is automatically issued to notify the operation and maintenance personnel to handle it. For example, the fault alarm message is sent to the operation and maintenance personnel through mobile phone text messages, emails, etc., and the fault type, fault location, fault severity and other information are displayed on the system interface; 2. Maintenance plan formulation: According to the wind power operation and maintenance failures that may occur in the target wind power equipment in the future (i.e., wind power operation and maintenance failure prediction results), formulate a maintenance plan for the target wind power equipment and arrange maintenance time and resources reasonably. For example, if it is predicted that a certain wind turbine may fail in the next week, this method will automatically generate a maintenance plan and arrange operation and maintenance personnel to conduct inspections and maintenance at the appropriate time. 3. Task allocation: Assign maintenance tasks to the corresponding operation and maintenance personnel and track the execution of tasks. For example, tasks can be assigned to the most suitable personnel based on factors such as the operation and maintenance personnel's skill level, work experience, and geographical location. At the same time, this method can track the progress of tasks in real time to ensure that tasks are completed on time.

[0049] In the above implementation process, the method has the following effects: 1. Improve operation and maintenance efficiency: Use artificial intelligence technology to realize intelligent operation and maintenance management of wind power equipment, which can quickly analyze massive operation and maintenance knowledge, accurately diagnose faults, predict maintenance needs, and improve operation and maintenance efficiency; 2. Improve the quality of operation and maintenance: By analyzing and mining a large amount of wind power equipment operation data, it is possible to monitor and evaluate the equipment status in real time, discover potential problems in time, and improve the quality of operation and maintenance; 3. Reduce operation and maintenance costs: Intelligent operation and maintenance management can reduce the workload of manual inspection and maintenance, and reduce operation and maintenance costs. Figure 2 FIG. 1 is a block diagram of a wind power operation and maintenance fault diagnosis system provided by an embodiment of the present invention. Figure 2 As shown, an embodiment of the present invention provides a wind power operation and maintenance fault diagnosis system, comprising: A wind power operation and maintenance knowledge base construction module is used to construct a wind power operation and maintenance knowledge base based on multi-source historical operation data of target wind power equipment and operation knowledge experience data of target wind power equipment; A fault diagnosis model training module is used to obtain a fault diagnosis model based on a wind power operation and maintenance knowledge base; wherein the fault diagnosis model includes a plurality of feature extraction combination layers for extracting data features of corresponding levels in multi-source historical operation data of target wind power equipment; The fault diagnosis module is used to input the multi-source real-time operation data of the target wind power equipment collected in real time into the fault diagnosis model, so as to extract the data features of the corresponding levels in the multi-source real-time operation data of the target wind power equipment based on each feature extraction combination layer, and determine the wind power operation and maintenance fault diagnosis results corresponding to the target wind power equipment based on the data features of each level.

[0050] Specifically, the system constructs a wind power operation and maintenance knowledge base based on the multi-source historical operation data of the target wind power equipment and the operation knowledge and experience data of the target wind power equipment. By integrating the multi-source historical operation data of the target wind power equipment, more comprehensive and accurate wind turbine operation status information can be obtained. A fault diagnosis model is obtained by training based on the wind power operation and maintenance knowledge base; wherein the fault diagnosis model includes a plurality of feature extraction combination layers for extracting data features of corresponding levels in the multi-source historical operation data of the target wind power equipment. The multi-source real-time operation data of the target wind power equipment is input into the trained fault diagnosis model, and the trained fault diagnosis model extracts data features of corresponding levels in the multi-source real-time operation data of the target wind power equipment through each feature extraction combination layer, and determines the wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment based on the data features of each level. The system uses artificial intelligence technology to analyze and learn massive amounts of wind power operation and maintenance knowledge (i.e., operating knowledge and experience data of wind power equipment), and processes and analyzes complex multi-source historical operating data of wind turbines. After training, it can accurately extract the deep-level characteristics of the operating data of wind power equipment, improve the accuracy and generalization ability of fault diagnosis, improve operation and maintenance efficiency, and realize intelligent analysis and management of wind power operation and maintenance faults, thereby realizing intelligent operation and maintenance management of wind power equipment, improving operation and maintenance efficiency and quality, and reducing operation and maintenance costs.

[0051] The embodiment of the present invention further provides a machine-readable storage medium, on which instructions are stored. When the instructions are executed by the processor 100, the processor 100 is configured to execute the above-mentioned wind power operation and maintenance fault diagnosis method.

[0052] Machine-readable storage media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0053] An embodiment of the present invention further provides an electronic device 10, which includes a memory 101, a processor 100, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, the above-mentioned wind power operation and maintenance fault diagnosis method is implemented.

[0054] like Figure 3 FIG. 1 is a schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, the electronic device 10 of this embodiment includes: a processor 100, a memory 101, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, the steps in the above method embodiment are implemented. Alternatively, when the processor 100 executes the computer program 102, the functions of each module / unit in the above device embodiment are implemented.

[0055] Exemplarily, the computer program 102 may be divided into one or more modules / units, one or more modules / units are stored in the memory 101, and are executed by the processor 100 to complete the present invention. One or more modules / units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program 102 in the electronic device 10. For example, the computer program 102 may be divided into a wind power operation and maintenance knowledge base construction module, a fault diagnosis model training module, and a fault diagnosis module.

[0056] The electronic device 10 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device 10 may include, but is not limited to, a processor 100 and a memory 101. Those skilled in the art will appreciate that Figure 3 It is only an example of the electronic device 10 and does not constitute a limitation of the electronic device 10. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0057] The processor 100 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0058] The memory 101 may be an internal storage unit of the electronic device 10, such as a hard disk or memory of the electronic device 10. The memory 101 may also be an external storage device of the electronic device 10, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (FlashCard), etc. equipped on the electronic device 10. Further, the memory 101 may also include both an internal storage unit of the electronic device 10 and an external storage device. The memory 101 is used to store computer programs and other programs and data required by the electronic device 10. The memory 101 may also be used to temporarily store data that has been output or is to be output.

[0059] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be 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. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0060] It should be understood by those skilled in the art that the embodiments of the present application may be provided as methods, systems, or computer program 102 products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program 102 product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0061] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program 102 products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by the computer program 102 instructions. These computer program 102 instructions can be provided to a processor 100 of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor 100 of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0062] These computer program 102 instructions may also be stored in a computer readable memory 101 that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory 101 produce an article of manufacture including an instruction device that implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0063] These computer program 102 instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0064] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0065] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A wind power operation and maintenance fault diagnosis method, characterized in that: include: Construct a wind power operation and maintenance knowledge base based on multi-source historical operation data of target wind power equipment and operation knowledge and experience data of target wind power equipment; A fault diagnosis model is obtained by training based on a wind power operation and maintenance knowledge base; wherein the fault diagnosis model comprises a plurality of feature extraction combination layers for extracting data features of corresponding levels in multi-source historical operation data of target wind power equipment; The multi-source real-time operation data of the target wind power equipment collected in real time is input into the fault diagnosis model to extract the data features of the corresponding levels in the multi-source real-time operation data of the target wind power equipment based on each feature extraction combination layer, and based on the data features of each level, determine the wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment.

2. The wind power operation and maintenance fault diagnosis method according to claim 1, characterized in that: The multi-source historical operation data of the target wind power equipment includes operation data, environmental data and maintenance record data of the target wind power equipment; The acquisition rules of the multi-source historical operation data of the target wind power equipment include: The operation data, environmental data and maintenance record data of the target wind power equipment are obtained through various sensors deployed on the target wind power equipment and the data interface connected to the industrial control system of the target wind power equipment.

3. The wind power operation and maintenance fault diagnosis method according to claim 2, characterized in that: After acquiring multi-source historical operation data of the target wind power equipment, the method further includes: Preprocess the multi-source historical operation data of the target wind power equipment; The pre-processing rules include: The multi-source historical operation data of the target wind power equipment are sequentially cleaned, denoised and normalized.

4. The wind power operation and maintenance fault diagnosis method according to claim 1, characterized in that: The construction rules of the wind power operation and maintenance knowledge base include: Based on the multi-source historical operation data of the target wind power equipment, a corresponding relational database and / or non-relational database is constructed to obtain a wind power operation and maintenance database; The wind power operation and maintenance database and the operating knowledge and experience data of the target wind power equipment are integrated by using the knowledge graph method to construct a wind power operation and maintenance knowledge base.

5. The wind power operation and maintenance fault diagnosis method according to claim 1, characterized in that: The training rules of the fault diagnosis model include: Perform fault state labeling on the operation fault data of the target wind power equipment in the wind power operation and maintenance knowledge base to obtain training samples, and divide the training samples into a training set and a test set; Obtaining an initial model, initializing the weights and biases of the initial model, training the initial model using a training set, and iteratively updating the weights and biases during the training process until the model converges or the number of training times reaches a preset number of training times, thereby obtaining an initial model of the fault diagnosis model; Using the test set to test the initial model of the fault diagnosis model, based on the test results, calculate the performance index value corresponding to the initial model of the fault diagnosis model; Based on the performance indicator values ​​corresponding to the initial model of the fault diagnosis model, the initial model of the fault diagnosis model is adjusted to obtain a trained fault diagnosis model; wherein the method of adjusting the initial model of the fault diagnosis model at least includes increasing the number of layers of the fault diagnosis model and adjusting the number of neurons of the fault diagnosis model.

6. The wind power operation and maintenance fault diagnosis method according to claim 5, characterized in that: After obtaining the fault diagnosis model, the method further includes: Optimize the parameter combination of multi-source real-time operation data input to the fault diagnosis model; The optimization rules for the parameter combination of the multi-source real-time operation data include: With the goal of maximizing the power generation efficiency of the target wind power equipment, the operating data and environmental data of the target wind power equipment are taken as the optimization objects, and the parameter combination of the optimization object is optimized through the genetic algorithm.

7. The wind power operation and maintenance fault diagnosis method according to claim 5, characterized in that: The rules for iteratively updating weights and biases include: Use a small batch stochastic gradient descent algorithm to iteratively update weights and biases; The specific contents of the small batch stochastic gradient descent algorithm include: A preset number of training data are randomly selected from the training set to calculate the gradient, and the weights and biases are updated based on the gradient calculation results.

8. The wind power operation and maintenance fault diagnosis method according to claim 1, characterized in that: After obtaining the wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment, the method further includes: When it is determined that the wind power operation and maintenance fault diagnosis result is that the target wind power equipment currently has a wind power operation and maintenance fault, fault alarm information is generated based on the wind power operation and maintenance fault diagnosis result; wherein the fault alarm information at least includes fault type information, fault location information and fault severity information of the target wind power equipment; When it is determined that the wind power operation and maintenance fault diagnosis result indicates that the target wind power equipment will have a wind power operation and maintenance fault at a future moment, a wind power equipment maintenance plan is generated based on the wind power operation and maintenance fault diagnosis result, and after completing the task allocation of the wind power equipment maintenance plan, the execution data of the wind power equipment maintenance plan is obtained in real time.

9. A wind power operation and maintenance fault diagnosis system, characterized in that: include: A wind power operation and maintenance knowledge base construction module is used to construct a wind power operation and maintenance knowledge base based on multi-source historical operation data of target wind power equipment and operation knowledge experience data of target wind power equipment; A fault diagnosis model training module is used to obtain a fault diagnosis model based on a wind power operation and maintenance knowledge base; wherein the fault diagnosis model includes a plurality of feature extraction combination layers for extracting data features of corresponding levels in multi-source historical operation data of target wind power equipment; The fault diagnosis module is used to input the multi-source real-time operation data of the target wind power equipment collected in real time into the fault diagnosis model, so as to extract the data features of the corresponding levels in the multi-source real-time operation data of the target wind power equipment based on each feature extraction combination layer, and determine the wind power operation and maintenance fault diagnosis result corresponding to the target wind power equipment based on the data features of each level.

10. A machine-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by a processor, the processor is configured to execute the wind power operation and maintenance fault diagnosis method as described in any one of claims 1 to 8.

11. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the wind power operation and maintenance fault diagnosis method according to any one of claims 1 to 8 is implemented.

Images