Wind power bolt axial force real-time monitoring method, system and electronic equipment

Abstract

The application discloses a wind power bolt axial force real-time monitoring method and system and electronic equipment, and relates to the technical field of axial force monitoring.The method comprises the following steps: obtaining a bolt assembly and a connecting shaft assembly of a target wind turbine generator, and constructing a stress nephogram;building a potential field decision maker and embedding it in a wind power monitoring system, uploading a wind power working condition based on a data port, determining a stress potential field by initializing the stress nephogram, performing dislocation simulation, generating a force field monitoring strategy;an auxiliary communication protocol is used to disassemble the force field monitoring strategy, perform sensing multithreading and monitoring driving, determine force sensing data, and perform real-time stress potential field conversion, and based on the stress coding sequence, a judgment is made based on macroscopic deviation and microscopic dislocation, a bolt axial force map is generated, and a pop-up window is displayed on the system interface of the wind power monitoring system.The technical problems of bolt axial force monitoring lag and low precision in the prior art are solved, and the technical effects of realizing wind power bolt axial force real-time monitoring and improving monitoring precision are achieved.

Description

Technical Field

[0001] This invention relates to the field of axial force monitoring technology, specifically to a method, system, and electronic equipment for real-time monitoring of axial force in wind turbine bolts. Background Technology

[0002] Bolts are crucial components connecting key parts of wind turbine generators, and their axial force status directly affects the overall structural stability and safety. Currently, bolt axial force monitoring relies heavily on periodic manual maintenance or single-point sensor data collection. This not only results in delayed response but also makes it difficult to accurately identify the stress state during operation. Consequently, monitoring is untimely, coverage is insufficient, and data accuracy is low, failing to meet the all-weather, all-condition safe operation requirements of wind power equipment. Summary of the Invention

[0003] This application provides a method, system, and electronic equipment for real-time monitoring of axial force in wind turbine bolts, which solves the technical problems of lagging and low accuracy in bolt axial force monitoring in the prior art.

[0004] The first aspect of this application provides a method for real-time monitoring of axial force on wind turbine bolts, the method comprising:

[0005] The bolt assembly and connecting shaft assembly of the target wind turbine are acquired, and a stress cloud map is constructed. Each stress point cloud is identified by a stress coding sequence based on stress characteristics. A potential field decision-maker is built based on the stress cloud map and embedded in the wind power monitoring system. The wind power operating conditions are uploaded based on the system data port. By initializing the stress cloud map, the stress potential field is determined and misalignment simulation is performed to generate a force field monitoring strategy. The force field monitoring strategy is a macro-micro coupling strategy for the force sensor array. An auxiliary communication protocol is used to determine the force sensor data and perform real-time stress potential field conversion by decomposing the force field monitoring strategy and performing multi-threaded sensor deployment and monitoring drive. Based on the stress coding sequence, a judgment based on macroscopic deviation and microscopic dislocation is performed to generate a bolt axial force spectrum, which is displayed in a pop-up window on the system interface of the wind power monitoring system.

[0006] A second aspect of this application provides a real-time monitoring system for axial force of wind turbine bolts, the system comprising:

[0007] The system comprises the following modules: **Stress Cloud Map Construction Module:** This module acquires the bolt and connecting shaft components of the target wind turbine and constructs a stress cloud map. Each stress point cloud is identified by a stress coding sequence based on stress characteristics. **Strategy Generation Module:** This module builds a potential field decision-maker based on the stress cloud map and embeds it into the wind power monitoring system. It uploads wind power operating conditions through the system data port, initializes the stress cloud map, determines the stress potential field, performs misalignment simulation, and generates a force field monitoring strategy. This strategy is a macro-micro coupling strategy for the force sensor array. **Monitoring Module:** This module assists in communication by disassembling the force field monitoring strategy, performing multi-threaded sensor deployment and monitoring drive, determining force sensor data, and performing real-time stress potential field conversion. Using the stress coding sequence as a condition, it performs judgment based on macroscopic deviations and microscopic dislocations, generating a bolt axial force spectrum, which is displayed as a pop-up on the system interface of the wind power monitoring system.

[0008] A third aspect of this application provides an electronic device, comprising: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the real-time monitoring method for axial force of wind turbine bolts provided in this application.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] First, the bolt assembly and connecting shaft assembly of the target wind turbine are acquired, and a stress cloud map is constructed. Each stress point cloud is identified by a stress coding sequence based on stress characteristics. Then, a potential field decision-maker is built based on the stress cloud map and embedded in the wind power monitoring system. Wind power operating conditions are uploaded via the system data port. By initializing the stress cloud map, the stress potential field is determined, and misalignment simulation is performed to generate a force field monitoring strategy. This strategy is a macro-micro coupling strategy for the force sensor array. Finally, an auxiliary communication protocol is used. By decomposing the force field monitoring strategy and implementing multi-threaded sensor deployment and monitoring drive, force sensor data is determined, and real-time stress potential field conversion is performed. Using the stress coding sequence as a condition, a judgment based on macroscopic deviation and microscopic dislocation is made to generate a bolt axial force spectrum, which is displayed as a pop-up window on the system interface of the wind power monitoring system. This solves the technical problems of lagging and low accuracy in bolt axial force monitoring in existing technologies, achieving the technical effect of real-time monitoring of wind turbine bolt axial force and improving monitoring accuracy. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1This is a schematic diagram of the real-time monitoring method for axial force of wind turbine bolts provided in the embodiments of this application;

[0013] Figure 2 This is a schematic diagram of the structure of the real-time monitoring system for axial force of wind turbine bolts provided in the embodiments of this application;

[0014] Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application.

[0015] Explanation of reference numerals in the attached diagram: Cloud map construction module 11, strategy generation module 12, monitoring module 13, processor 21, memory 22, input device 23, output device 24. Detailed Implementation

[0016] This application solves the technical problems of lagging and low accuracy in bolt axial force monitoring in the prior art by providing a method, system and electronic equipment for real-time monitoring of axial force of wind turbine bolts.

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0018] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0019] Example 1, as Figure 1 As shown in the embodiment of this application, a method for real-time monitoring of axial force of wind turbine bolts is provided, wherein the method includes:

[0020] Obtain the bolt assembly and connecting shaft assembly of the target wind turbine and construct a stress cloud map, in which each stress point cloud is identified by a stress coding sequence based on stress characteristics.

[0021] In this embodiment of the application, by performing mechanical-electrical integrated modeling of the wind turbine structure, various bolt assemblies used for connecting key parts such as impeller, tower, nacelle, flange, and bearing are identified, and the spatial structure information of the connecting shaft assembly that mates with it is determined.

[0022] Based on the geometric positions and structural parameters of the bolt assemblies and connecting shaft assemblies, a three-dimensional point cloud dataset including bolt assembly points and axial connection points is established. Furthermore, using a digital twin platform, the stress response under wind power operating conditions is simulated and solved to obtain stress response data for each key connection point under different load conditions, constructing a stress cloud map covering the overall structural connection relationships. Each stress point in the stress cloud map corresponds to a physical spatial location and possesses physical quantity attributes such as stress intensity and stress direction. To achieve structural feature identification and subsequent processing, a stress characteristic-based coding mechanism is introduced for each stress point cloud. A stress characteristic classification coding method is used to identify its stress coding sequence. This coding sequence can uniquely characterize the stress state characteristics of a specific connection structure under specific operating conditions, and is used for subsequent potential field analysis, risk identification, and axial force spectrum generation.

[0023] Furthermore, constructing a stress cloud map includes:

[0024] For the mechanical and electrical structure of the target wind turbine, key point clouds are determined based on the assembly positions of bolt assemblies; the connecting shaft assemblies of each bolt assembly are located to determine axial point clouds; the key point clouds and the axial point clouds are spatially structurally phase-stitched, and stress coding sequence identifiers based on stress characteristics are introduced to form the stress cloud map.

[0025] Preferably, based on the mechanical and electrical structure drawings and modeling data of the target wind turbine, the assembly positions of various bolt components in the wind turbine are extracted, focusing on key parts such as tower flange bolts, nacelle foundation connecting bolts, main bearing connecting bolts, gearbox fixing bolts, and impeller connecting bolts. A corresponding set of key connection points is established, forming the first type of point cloud, i.e., the key point cloud. Through structural analysis and topology matching, the shaft components connected to each bolt component, such as the main shaft, drive shaft, and connecting shaft, are further identified, and their axial arrangement and force transmission direction are obtained, constructing the second type of point cloud, i.e., the axial point cloud. Then, the key point cloud and the axial point cloud are further processed... Phase stitching of the spatial structure is performed, and the association and reconstruction of each connection point in the structural space are realized through three-dimensional coordinate fitting and spatial matching algorithms. In the reconstructed point cloud data, for each pair of key connection points and their corresponding axial force transmission paths, the stress distribution pattern is identified based on historical stress data and mechanical models, and each set of point clouds is assigned a stress coding sequence. This stress coding sequence can reflect the main stress types and distribution characteristics of the structure under operating conditions, including bending stress, shear stress, fretting wear stress, fatigue stress and external disturbance stress, etc., thereby constructing a stress cloud map with spatial structural characteristics and stress distribution characteristics.

[0026] Furthermore, a stress-encoded sequence identifier based on stress characteristics is introduced, including:

[0027] Based on the mechanical and electrical structure, stress characteristics are classified. By introducing a coding mode based on stress characteristics, the key point cloud is identified by stress coding sequence. The stress characteristics are at least divided into a first bending stress, a second fretting wear stress, a third fatigue stress, and a fourth external stress. The first bending stress includes at least the uneven preload of flange bolts, the second fretting wear includes at least the alternating load of fastening bolts, the third fatigue stress includes at least the oscillation of bearing bolts, and the fourth external stress includes at least the external environmental corrosion and erosion effect of the foundation anchor bolt group.

[0028] Based on the mechanical structure layout and electrical coupling characteristics of wind turbine units, the stress types borne by different connection positions are systematically classified. Based on engineering mechanics principles and actual operating data, typical stress types are divided into four categories: First, bending stress, mainly caused by inconsistent deformation and uneven distribution of preload at the connection of structural components; second, fretting wear stress, mainly generated by relative slippage of bolt connection surfaces under small alternating loads; third, fatigue stress, caused by the accumulation of stress and performance degradation of bolt structures due to long-term cyclic loads; and fourth, external stress, usually caused by the long-term effects of natural factors such as wind, sand, salt spray, damp heat, and low temperature.

[0029] Based on the above classification, a stress coding pattern with identification capabilities is introduced. This coding pattern can adopt a multi-dimensional vector combination method, corresponding to different stress types and their structural sources. For example, the coding of the first bending stress includes the non-uniform preload distribution originating from the flange bolt assembly, recording its preload matrix and lateral offset amplitude; the coding of the second fretting wear stress includes the alternating load intensity and frequency characteristics borne by the fastening bolt assembly during operation; the coding of the third fatigue stress includes the stress cycle number and peak distribution of the bearing bolt assembly caused by the spindle oscillation; and the coding of the fourth external stress includes the corrosion level, scour path, and climate disturbance index of the foundation anchor group in marine, plateau, or dusty environments. Through the above classification mechanism and feature extraction method, a mapping relationship is established between each key point cloud and its corresponding structural stress information, forming a unique stress coding sequence identifier.

[0030] A potential field decision-maker is built based on the stress cloud map and embedded in the wind power monitoring system. The wind power operating conditions are uploaded based on the system data port. By initializing the stress cloud map, the stress potential field is determined and misalignment simulation is performed to generate a force field monitoring strategy. The force field monitoring strategy is a macro-micro coupling strategy for the force sensing array.

[0031] Based on stress cloud maps, a potential field-driven decision-making and control logic is established, and a potential field decision-maker is built and integrated into the wind power monitoring system. Specifically, based on the spatial distribution of each stress point cloud and the stress coding sequence in the stress cloud map, a high-dimensional stress field data model is constructed. Multi-scale tensor expansion is performed on the stress information to extract the stress evolution laws of each structural connection region. This stress field data is embedded into the data processing architecture of the wind power monitoring system. Operating parameters of the wind turbine, including wind speed, rotational speed, vibration frequency, temperature, and power load, are received in real time through the system data port as external driving inputs to initialize and reconstruct the stress cloud map. During initialization, a correlation mapping between stress, structure, and operating conditions is established to calculate the current stress potential field distribution. Furthermore, a misalignment simulation mechanism is introduced to simulate local misalignment behavior that may occur under conditions such as insufficient structural preload, thermal expansion, and dynamic load disturbances. Based on this, a force field monitoring strategy for force sensing arrays is constructed and generated. This strategy employs a macro-micro coupling mechanism: at the macro level, it focuses on the axial force variation trend and stress redistribution law of the overall structure; at the micro level, it performs fine-grained discrimination on the instantaneous response, material damage evolution, and dislocation characteristics of specific sensing points. The force field monitoring strategy is adaptable to various types of force sensing arrays (such as piezoelectric, magnetoelastic, and surface acoustic wave arrays) and provides dual support for parameter-driven and model-based reasoning, providing a decision-making basis for real-time stress monitoring and bolt axial force spectrum generation.

[0032] Furthermore, a potential field decision-maker is constructed based on the stress cloud map, including:

[0033] A potential field region is deployed and the stress cloud map is built in. Guided by the logic of the first initialization and the second simulation, a first potential field region is constructed. Taking the potential field state as input, the force sensing array as condition, and the monitoring strategy as output, a second decision region is deployed. The second decision region is a generation architecture determined based on adversarial network training, including a first generation branch and a second generation branch in parallel. The first generation branch takes the stress potential field as input, and the second generation branch takes the misaligned potential field as input. The first potential field region and the second decision region are cascaded to construct the potential field decision-maker.

[0034] In the target wind power monitoring system, a potential field region is deployed as a processing unit for stress field evolution. The stress cloud map is embedded in this region as a structural input, and the execution strategy is guided by a first initialization and a second simulation. The first initialization is used to load and correct the stress point distribution in the stress cloud map based on the known current operating conditions of the wind turbine, thus completing the construction of the static stress potential field. The second simulation, based on the structural disturbance model and the prediction of operating condition changes, performs dynamic disturbance and misalignment inference on the stress cloud map, generating a misalignment potential field model for prediction and judgment, forming the basic region for dynamic stress situation perception, namely the first potential field region.

[0035] Secondly, a second decision region is constructed for generating monitoring strategies. This region takes the potential field state as the input signal and a multi-type force sensor array deployed at the wind turbine bolts as the physical constraint. A multi-dimensional monitoring strategy is output through an intelligent decision model. The second decision region is built on an adversarial neural network architecture and employs a generator-discriminator joint optimization mechanism to achieve a highly robust response to the potential field state and potential misalignment risks. Specifically, the second decision region includes a first generation branch and a second generation branch set in parallel: the first generation branch takes the initialized stress potential field as input to simulate the force sensing response characteristics under normal connection conditions; the second generation branch takes the misaligned potential field constructed through simulation as input to generate monitoring response characteristics under abnormal structural stress conditions. Both branches are trained adversarially with the discriminator through a tensor fusion module to enhance the generalization ability and robustness of the strategy output.

[0036] By cascading the first potential field region and the second decision region, a complete potential field decision-maker is constructed.

[0037] Furthermore, determining the force field monitoring strategy includes:

[0038] The wind power operating conditions are acquired, and the load transfer is analyzed to determine the stress transfer field. The stress cloud map is initialized based on the stress transfer field to determine the stress potential field. The stress potential field is simulated for misalignment and a stress monitoring decision based on macro-micro is made to determine the force field monitoring strategy.

[0039] First, real-time operating data of the target wind turbine is acquired. Wind turbine operating data includes parameters such as rotor load, tower vibration, nacelle yaw status, temperature changes, and wind speed fluctuations. Through dynamic analysis of this operating data, key load paths are identified based on load transfer relationships, and the stress transfer field acting on the bolt assemblies and their connecting shaft assemblies is determined. For example, the transfer path from the rotor load to the main shaft bearing, then to the gearbox, and finally to the tower bolt assembly constitutes a component-based transfer path. The stress transfer field reflects the load flow and stress migration characteristics caused by structural connections, motion interference, and external force disturbances between components during wind turbine operation. Subsequently, the previously constructed stress cloud map is initialized based on the stress transfer field, mapping the transfer field parameters to the point cloud space. Stress state assignment and updates are performed through coupling mapping relationships to obtain a stress potential field reflecting the current operating conditions. This stress potential field describes the stress gradient and regional response characteristics of the wind turbine bolt assemblies under different loads in the form of a multidimensional tensor.

[0040] Furthermore, the stress potential field is subjected to misalignment simulation to construct potential structural anomaly evolution scenarios. Misalignment simulation includes macroscopic simulations of structural displacement at connection points and preload failure trends, as well as microscopic simulations of fine-grained stress responses such as wear at connection surfaces, fatigue crack initiation, or dislocation slip. By incorporating multi-temporal and multi-spatial scale response data from the simulation process into the monitoring strategy decision-making, an integrated macro-micro coupled stress monitoring decision-making process is executed. Finally, by integrating macroscopic stress deviations and microscopic dislocation response results, a force field monitoring strategy adapted to the current operating conditions is generated.

[0041] Furthermore, stress monitoring decision-making, determining force field monitoring strategies, includes:

[0042] For the target wind turbine, a force sensing array is deployed, comprising a first array based on piezoelectric thin film sensing, a second array based on magnetoelastic sensing, and a third array based on surface acoustic waves; wherein the first array focuses on axial stress, the second array focuses on preload, and the third array focuses on stress compensation based on temperature drift; using the force sensing array as the front-end monitoring device, the force field monitoring strategy is determined.

[0043] For key connection points of wind turbine units, multiple types of force sensor arrays are configured, encompassing sensing technologies based on different physical mechanisms to meet multi-dimensional stress monitoring needs. First, a first force sensor array based on piezoelectric thin-film sensing technology is deployed. This array primarily focuses on real-time acquisition of axial stress, utilizing the piezoelectric effect to highly sensitively sense dynamic changes in the tensile or compressive state of bolts. Second, a second force sensor array based on magnetoelastic sensing principles is deployed. This array specifically monitors bolt preload, reflecting the stability and fluctuation characteristics of preload by detecting changes in the permeability of magnetic materials under stress. Next, a third force sensor array based on surface acoustic wave (SAW) technology is deployed, focusing on addressing the impact of environmental factors such as temperature drift on sensing data. Through modulation of sound wave propagation characteristics, temperature compensation is implemented for stress signals, ensuring the accuracy and stability of monitoring data. These force sensor arrays, serving as front-end monitoring devices for the axial force of wind turbine bolts, integrate the advantages of different sensing technologies to form a comprehensive sensing network covering axial stress, preload, and environmental compensation.

[0044] Furthermore, using the stress coding sequence as a constraint and the force sensing array as a front-end monitoring device, a first macroscopic strategy is determined; by simulating misalignment to locate misalignment risks, and using the force sensing array as a front-end monitoring device, a second microscopic strategy is determined; and by using spatiotemporal code constraints to associate the first macroscopic strategy and the second microscopic strategy, the force field monitoring strategy is determined.

[0045] Constraints and feature filtering are performed on the data collected by the force sensing array based on the stress coding sequence to determine the first macro strategy. The first macro strategy focuses on the stress distribution trend and macro force field changes of the overall structure of the wind turbine. By performing statistical analysis and time series modeling on key parameters such as axial stress and preload of the force sensing array, the load change law and potential abnormal trends of the overall structure are identified, forming a stress state monitoring scheme covering the whole.

[0046] By combining misalignment simulation technology, and based on in-depth analysis of the potential local misalignment risks at structural connection points, a second microscopic strategy is determined. This second microscopic strategy captures minute displacement, wear, and fatigue damage signals at local connection points to achieve fine-grained monitoring and anomaly warning of the bolt contact surface, thereby enhancing the real-time tracking capability of the microscopic damage evolution process.

[0047] By utilizing spatiotemporal codes as correlation constraints, the first macroscopic strategy and the second microscopic strategy are fused and coupled at the spatiotemporal level. The spatiotemporal codes encode the spatial location and time-series information of each monitoring point, ensuring the spatiotemporal consistency and complementarity of the macroscopic and microscopic strategy data, thus achieving cross-scale, multi-dimensional collaborative monitoring. Finally, by synthesizing the correlation results, a unified force field monitoring strategy is generated.

[0048] The auxiliary communication protocol, by disassembling the force field monitoring strategy and performing multi-threaded sensor deployment and monitoring drive, determines the force sensing data and performs real-time stress potential field conversion. Based on the stress coding sequence, it performs judgment based on macroscopic deviation and microscopic dislocation, generates bolt axial force spectrum, and displays it in a pop-up window on the system interface of the wind power monitoring system.

[0049] An auxiliary communication protocol is designed to enable efficient scheduling and data coordination of force field monitoring strategies across various sensor array types. Specifically, the force field monitoring strategy is logically decomposed according to multiple dimensions such as stress type, array location, and sampling frequency, forming a set of parallel sub-strategies adaptable to various sensors. For the set of parallel sub-strategies, the protocol control module performs a multi-threaded deployment operation, that is, different sub-strategies are allocated in parallel to the corresponding sensor nodes through a thread scheduling mechanism, driving each front-end sensor to perform real-time monitoring.

[0050] Based on the collected force sensing data, the data is transmitted in real time to the core processing unit of the wind power monitoring system via a communication protocol for dynamic transformation of the stress potential field. This transformation, conditioned by a preset stress coding sequence, maps the sensing data back to the corresponding spatial stress point cloud, completing the judgment and analysis of macroscopic deviations and microscopic dislocations. Macroscopic deviation judgment is used to identify abnormal trends in stress distribution within the overall structure, while microscopic dislocation judgment focuses on subtle misalignments or damage characteristics at local connection points. Based on the judgment results, a bolt axial force map is generated, comprehensively reflecting the real-time axial force status and potential risk points of the wind turbine's bolt connections. Simultaneously, the bolt axial force map is displayed in real time via a pop-up window through the wind power monitoring system's user interface, providing intuitive risk warnings and status feedback, supporting rapid decision-making and on-site maintenance.

[0051] Furthermore, based on the determination of macroscopic deviations and microscopic dislocations, a bolt axial force spectrum is generated, including:

[0052] By identifying the stress coding sequence, a first judgment condition is determined; based on the first judgment condition, a macroscopic deviation judgment is performed on the real-time stress potential field to determine a first-step judgment result, and a microscopic misalignment judgment is performed to determine a second-step judgment result; the first-step judgment result and the second-step judgment result are combined and structurally distributed using spatial position codes to determine the bolt axial force spectrum.

[0053] By decoding and identifying the real-time input stress encoding sequence, information including stress type, loading method, application time, and spatial location is extracted to construct the first judgment condition. This first judgment condition serves as a classification basis, distinguishing the stress mode of the current monitoring data from possible anomaly types.

[0054] Based on the first judgment condition, the real-time stress potential field is judged in two steps: In the first step, based on the overall stress distribution trend of the structure, methods such as statistical offset detection and full-field anomaly fitting are used to identify macroscopic deviations and output macroscopic judgment results; In the second step, the focus is on the detailed response of local connection areas, and algorithms such as stress gradient mutation analysis and point mismatch comparison are used to identify possible microscopic misalignment phenomena and generate microscopic judgment results. After obtaining the above two types of judgment results, the spatial location code is used as the mapping and association basis to spatially combine the first step macroscopic judgment results and the second step microscopic judgment results and reconstruct the structural distribution to form a bolt axial force map. The bolt axial force map graphically describes the magnitude, variation trend and potential risk areas of the axial force at various key points of the bolted connection structure.

[0055] In summary, the embodiments of this application have at least the following technical effects:

[0056] First, the bolt assembly and connecting shaft assembly of the target wind turbine are acquired, and a stress cloud map is constructed. Each stress point cloud is identified by a stress coding sequence based on stress characteristics. Then, a potential field decision-maker is built based on the stress cloud map and embedded in the wind power monitoring system. Wind power operating conditions are uploaded via the system data port. By initializing the stress cloud map, the stress potential field is determined, and misalignment simulation is performed to generate a force field monitoring strategy. This strategy is a macro-micro coupling strategy for the force sensor array. Finally, an auxiliary communication protocol is used. By decomposing the force field monitoring strategy and implementing multi-threaded sensor deployment and monitoring drive, force sensor data is determined, and real-time stress potential field conversion is performed. Using the stress coding sequence as a condition, a judgment based on macroscopic deviation and microscopic dislocation is made to generate a bolt axial force spectrum, which is displayed as a pop-up window on the system interface of the wind power monitoring system. This solves the technical problems of lagging and low accuracy in bolt axial force monitoring in existing technologies, achieving the technical effect of real-time monitoring of wind turbine bolt axial force and improving monitoring accuracy.

[0057] Example 2, based on the same inventive concept as the wind turbine bolt axial force real-time monitoring method in the aforementioned examples, such as... Figure 2 As shown, this application provides a real-time monitoring system for axial force of wind turbine bolts, wherein the system includes:

[0058] Stress cloud map construction module 11: Acquires the bolt assembly and connecting shaft assembly of the target wind turbine, constructs a stress cloud map, wherein each stress point cloud is identified by a stress coding sequence based on stress characteristics; Strategy generation module 12: Builds a potential field decision-maker based on the stress cloud map and embeds it into the wind power monitoring system, uploads wind power conditions based on the system data port, determines the stress potential field by initializing the stress cloud map and performs misalignment simulation, and generates a force field monitoring strategy, wherein the force field monitoring strategy is a macro-micro coupling strategy for the force sensor array; Monitoring module 13: Assists the communication protocol, determines the force sensor data and performs real-time stress potential field conversion by disassembling the force field monitoring strategy and performing multi-threaded sensor deployment and monitoring drive, performs judgment based on macro-device deviation and micro-dislocation using the stress coding sequence as a condition, generates a bolt axial force spectrum, and displays it in a pop-up window on the system interface of the wind power monitoring system.

[0059] Furthermore, the cloud map construction module 11 is used to perform the following methods:

[0060] For the mechanical and electrical structure of the target wind turbine, key point clouds are determined based on the assembly positions of bolt assemblies; the connecting shaft assemblies of each bolt assembly are located to determine axial point clouds; the key point clouds and the axial point clouds are spatially structurally phase-stitched, and stress coding sequence identifiers based on stress characteristics are introduced to form the stress cloud map.

[0061] Furthermore, the cloud map construction module 11 is used to perform the following methods:

[0062] Based on the mechanical and electrical structure, stress characteristics are classified. By introducing a coding mode based on stress characteristics, the key point cloud is identified by stress coding sequence. The stress characteristics are at least divided into a first bending stress, a second fretting wear stress, a third fatigue stress, and a fourth external stress. The first bending stress includes at least the uneven preload of flange bolts, the second fretting wear includes at least the alternating load of fastening bolts, the third fatigue stress includes at least the oscillation of bearing bolts, and the fourth external stress includes at least the external environmental corrosion and erosion effect of the foundation anchor bolt group.

[0063] Furthermore, the strategy generation module 12 is used to perform the following method:

[0064] A potential field region is deployed and the stress cloud map is built in. Guided by the logic of the first initialization and the second simulation, a first potential field region is constructed. Taking the potential field state as input, the force sensing array as condition, and the monitoring strategy as output, a second decision region is deployed. The second decision region is a generation architecture determined based on adversarial network training, including a first generation branch and a second generation branch in parallel. The first generation branch takes the stress potential field as input, and the second generation branch takes the misaligned potential field as input. The first potential field region and the second decision region are cascaded to construct the potential field decision-maker.

[0065] Furthermore, the strategy generation module 12 is used to perform the following method:

[0066] The wind power operating conditions are acquired, and the load transfer is analyzed to determine the stress transfer field. The stress cloud map is initialized based on the stress transfer field to determine the stress potential field. The stress potential field is simulated for misalignment and a stress monitoring decision based on macro-micro is made to determine the force field monitoring strategy.

[0067] Furthermore, the strategy generation module 12 is used to perform the following method:

[0068] For the target wind turbine, a force sensing array is deployed, comprising a first array based on piezoelectric thin film sensing, a second array based on magnetoelastic sensing, and a third array based on surface acoustic waves; wherein the first array focuses on axial stress, the second array focuses on preload, and the third array focuses on stress compensation based on temperature drift; using the force sensing array as the front-end monitoring device, the force field monitoring strategy is determined.

[0069] Furthermore, the strategy generation module 12 is used to perform the following method:

[0070] Based on the stress potential field, using the stress coding sequence as a constraint and the force sensing array as a front-end monitoring device, a first macroscopic strategy is determined; by locating the misalignment risk through misalignment simulation, and using the force sensing array as a front-end monitoring device, a second microscopic strategy is determined; and by associating the first macroscopic strategy and the second microscopic strategy with spatiotemporal code constraints, the force field monitoring strategy is determined.

[0071] Furthermore, the monitoring module 13 is used to perform the following methods:

[0072] By identifying the stress coding sequence, a first judgment condition is determined; based on the first judgment condition, a macroscopic deviation judgment is performed on the real-time stress potential field to determine a first-step judgment result, and a microscopic misalignment judgment is performed to determine a second-step judgment result; the first-step judgment result and the second-step judgment result are combined and structurally distributed using spatial position codes to determine the bolt axial force spectrum.

[0073] Example 3, Figure 3 This is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present invention. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality or scope of the embodiments of the present invention. Figure 3 As shown, the electronic device includes a processor 21, a memory 22, an input device 23, and an output device 24; the number of processors 21 in the electronic device can be one or more. Figure 3 Taking a processor 21 as an example, the processor 21, memory 22, input device 23, and output device 24 in an electronic device can be connected via a bus or other means. Figure 3 Taking the example of a connection between China and Israel via a bus.

[0074] The memory 22, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the real-time monitoring method for axial force of wind turbine bolts in this embodiment of the invention. The processor 21 executes various functional applications and data processing of the electronic device by running the software programs, instructions, and modules stored in the memory 22, thereby realizing the aforementioned real-time monitoring method for axial force of wind turbine bolts.

[0075] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0076] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0077] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A wind turbine bolt axial force real-time monitoring method, characterized in that, The method comprises: Obtaining the bolt assembly and the connecting shaft assembly of the target wind turbine, and constructing a stress cloud map, wherein each stress point cloud is marked with a stress coding sequence based on stress characteristics; According to the stress cloud map, a potential field decision maker is built and embedded in the wind power monitoring system, the wind power working condition is uploaded based on the system data port, the stress cloud map is initialized, the stress potential field is determined and dislocation simulation is performed, and a force field monitoring strategy is generated, wherein the force field monitoring strategy is a macro-micro coupling strategy for a force sensing array; An auxiliary communication protocol is used to determine force sensing data and perform real-time stress potential field conversion by disassembling the force field monitoring strategy and performing sensing multithreading distribution and monitoring driving, and to generate a bolt shaft force map based on macroscopic deviation and microscopic dislocation based on the stress coding sequence, and to perform pop-up display on the system interface of the wind power monitoring system; Wherein, according to the stress cloud map, a potential field decision maker is built and embedded in the wind power monitoring system, the wind power working condition is uploaded based on the system data port, the stress cloud map is initialized, the stress potential field is determined and dislocation simulation is performed, and a force field monitoring strategy is generated, wherein the force field monitoring strategy is a macro-micro coupling strategy for a force sensing array; Wherein, according to the stress cloud map, a potential field decision maker is built and embedded in the wind power monitoring system, the wind power working condition is uploaded based on the system data port, the stress cloud map is initialized, the stress potential field is determined and dislocation simulation is performed, and a force field monitoring strategy is generated, wherein the force field monitoring strategy is a macro-micro coupling strategy for a force sensing array; Cascading the first potential field area and the second decision area to build the potential field decision maker; Wherein, determining the force field monitoring strategy comprises: Obtaining the wind power working condition, analyzing the wind power working condition based on load transfer, and determining the stress transfer field; Initializing the stress cloud map based on the stress transfer field to determine the stress potential field; Dislocation simulation and macro-micro stress monitoring decision are performed on the stress potential field to determine the force field monitoring strategy; Wherein, determining the force field monitoring strategy comprises: Deploying a force sensing array for the target wind turbine, wherein the force sensing array includes a first array based on piezoelectric film sensing, a second array based on magnetoelastic sensing, and a third array based on surface acoustic wave sensing; Wherein, the first array focuses on axial stress, the second array focuses on pre-tightening force, and the third array focuses on stress compensation based on temperature drift; The force sensing array is used as a front-end monitoring device to determine the force field monitoring strategy. Constructing a stress cloud map comprises:

2. The wind power bolt axial force real-time monitoring method of claim 1, wherein, For the mechanical and electrical structure of the target wind turbine, determining the key point cloud based on the assembly position of the bolt assembly; Positioning the connecting shaft assembly of each bolt assembly to determine the axial point cloud; Spatial structure phase splicing is performed on the key point cloud and the axial point cloud, stress coding sequence identification based on stress characteristics is introduced, and the stress cloud map is formed. Introducing stress coding sequence identification based on stress characteristics comprises:

3. The wind power bolt shaft force real-time monitoring method of claim 2, wherein, According to the mechanical and electrical structure, stress characteristics are classified, stress coding sequence identification is performed on the key point cloud by introducing coding modes based on stress characteristics, and the stress characteristics are at least divided into first bending stress, second fretting wear stress, third fatigue stress, and fourth external stress; ​ The first bending stress at least includes uneven pre-tightening based on flange bolts, the second fretting wear at least includes alternating load based on fastening bolts, the third fatigue stress at least includes hunting based on bearing bolts, and the fourth external stress at least includes external environmental corrosion scouring effect based on foundation anchor bolt groups.

4. The wind power bolt shaft force real-time monitoring method of claim 1, wherein, According to the stress potential field, a first macroscopic strategy is determined by taking the stress coding sequence as a constraint and the force sensing array as a front-end monitoring device. A second microscopic strategy is determined by taking the force sensing array as a front-end monitoring device through dislocation simulation positioning dislocation risk. The first macroscopic strategy and the second microscopic strategy are associated by taking a space-time code as a constraint to determine the force field monitoring strategy.

5. The wind power bolt shaft force real-time monitoring method of claim 1, wherein, Based on macroscopic deviation and microscopic dislocation, a bolt axial force atlas is generated by taking the stress coding sequence as a condition, including: A first determination condition is determined by identifying the stress coding sequence. Based on the first determination condition, a one-step determination result is determined by performing macroscopic deviation determination on the real-time stress potential field, and a two-step determination result is determined by performing microscopic dislocation determination. The one-step determination result and the two-step determination result are combined and structurally distributed by taking a spatial position code to determine the bolt axial force atlas.

6. A wind turbine bolt axial force real-time monitoring system, characterized in that, The system for implementing the wind power bolt axial force real-time monitoring method of any one of claims 1-5, the system comprising: a cloud map construction module: obtaining a bolt assembly and a connecting shaft assembly of a target wind turbine, and constructing a stress cloud map, wherein each stress point cloud is marked with a stress coding sequence based on stress characteristics; a strategy generation module: building a potential field decision maker according to the stress cloud map and embedding it in a wind power monitoring system, uploading wind power working conditions based on a system data port, determining a stress potential field by initializing the stress cloud map and performing dislocation simulation, and generating a force field monitoring strategy, wherein the force field monitoring strategy is a macro-micro coupled strategy for a force sensing array; a monitoring module: auxiliary communication protocol, the force field monitoring strategy is disassembled and monitored by driving sensing multithreading and determining force sensing data and real-time stress potential field conversion, based on macroscopic deviation and microscopic dislocation, a bolt axial force atlas is generated by taking the stress coding sequence as a condition, and a pop-up window is displayed on the system interface of the wind power monitoring system.

7. An electronic device, comprising: The electronic device comprises: a memory for storing executable instructions; a processor for executing the executable instructions stored in the memory to implement the wind power bolt axial force real-time monitoring method of any one of claims 1-5.

Images