A fan unit blade root bolt axial force monitoring system

Abstract

The application discloses a kind of fan unit blade root bolt axial force monitoring systems, it is related to fan unit monitoring technical field, the system includes following component parts: data acquisition module, data transmission module, data analysis processing center and maintenance early warning module, the axial force data of blade root bolt is collected in real time by high-precision axial force sensor in the application, and data is transmitted to data analysis processing center using wireless communication mode, data analysis processing center utilizes big data analysis technique and axial force trend prediction algorithm, accurately predicts the future change trend of axial force in combination with the operating condition of fan, this function greatly improves the safety and reliability of fan unit operation, by early warning potential axial force abnormal condition, bolt loosening, fracture failure can be effectively avoided, to prolong the service life of fan unit, reduce maintenance cost and production interruption risk.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fan unit monitoring, in particular to a fan unit blade root bolt axial force monitoring system. BACKGROUND

[0002] As the core component of the wind power generation system, the operation state of the fan unit is directly related to the power generation efficiency and safety of the entire wind farm. The blade root bolt, as a key component connecting the fan blade and the hub, bears a huge axial force and complex and variable working conditions. Its performance state directly affects the stability and service life of the fan unit. However, during actual operation, the fan unit is usually installed in remote areas with harsh environmental conditions, and it has been subjected to alternating loads for a long time. The blade root bolt is prone to looseness and fracture failure, which seriously threatens the safe operation of the fan unit.

[0003] During the operation of the fan unit, the axial force state of the blade root bolt is crucial for the safe and stable operation of the fan. The traditional monitoring method can only perform simple periodic inspection or rough measurement of the current axial force, and cannot accurately grasp the dynamic changes of the axial force in real time, let alone predict its future trend. Therefore, it is particularly important to develop a fan unit blade root bolt axial force monitoring system. SUMMARY

[0004] The purpose of the present application is to make up for the shortcomings of the prior art and provide a fan unit blade root bolt axial force monitoring system. The system can predict the future trend of the axial force by real-time acquisition of the blade root bolt axial force data, big data analysis technology and axial force trend prediction algorithm, and combination with the fan operating conditions. The system can predict the possible axial force abnormality in advance, effectively improve the safety of the fan unit operation, reduce the risk of blade loosening and falling accidents caused by abnormal axial force, and reduce the operation and maintenance cost of the fan unit.

[0005] To solve the above technical problems, the present application provides the following technical solution: a fan unit blade root bolt axial force monitoring system, which comprises a data acquisition module, a data transmission module, a data analysis and processing center and a maintenance and warning module.

[0006] The data acquisition module: real-time acquisition of the axial force data of the blade root bolt, high-precision axial force sensors are installed at the blade root bolt part, and the type and parameters of the sensors are determined according to the specifications and stress range factors of the blade root bolt during installation, and calibration and debugging are performed.

[0007] The data transmission module transmits the collected axial force data to the data analysis and processing center in real time. The data transmission is carried out wirelessly. The communication method must be determined based on the actual layout of the wind turbine unit and the site environment, and must be configured and debugged to ensure the timeliness, stability and integrity of the data transmission.

[0008] The data analysis and processing center receives the axial force data, preprocesses it, performs in-depth analysis using big data analytics, and predicts the future trend of axial force changes based on the operating conditions of the wind turbine. An axial force trend prediction algorithm is employed, as detailed below:

[0009] Suppose the sequence of axial force data collected from the blade root bolts is F = {F1, F2, F3, ..., F...} n}, where F i This represents the axial force value collected in the i-th sampling, where n is the number of samplings. The operating condition data of the wind turbine is obtained, including the wind speed sequence W = {W1, W2, W3, ..., W...}. n The rotational speed sequence R = {R1, R2, R3, ..., R} n The load sequence L = {L1, L2, L3, ..., L} n}, here W i R i L i Let the wind speed, rotational speed, and load values ​​collected in the i-th data acquisition be represented respectively. The axial force data undergoes preprocessing, including data cleaning to remove outliers and noise reduction. Let the preprocessed axial force data sequence be:

[0010] F′={F1′,F2′,F3′,…,F n ′}

[0011] A comprehensive impact factor C is constructed, and its calculation formula is as follows:

[0012]

[0013] in, The mean of the preprocessed axial force data, s F′ The standard deviation of the preprocessed axial force data. Let s be the mean of the wind speed data. W The standard deviation of the wind speed data. The mean of the rotational speed data, s R The standard deviation of the rotational speed data. s is the mean of the load data. L Let be the standard deviation of the load data, and α, β, γ, and δ be the weighting coefficients.

[0014] A model for predicting axial force trends is constructed based on a comprehensive influencing factor C. Let the predicted future axial force value be F. n+mThe prediction formula is:

[0015]

[0016] Wherein, C i is the value of the comprehensive influence factor C at the ith time, is the mean value of the comprehensive influence factor C, and i is another set of weight coefficients, and the value of i is adjusted constantly until a combination of values is found that can make the prediction accuracy reach a high level under a plurality of different working conditions, and the value of i is determined.

[0017] The maintenance warning module: when the data analysis and processing center predicts that the shaft force may exceed the safety threshold or appear abnormal change trend, the maintenance warning signal is triggered in advance, and the warning mode includes displaying warning information on the monitoring terminal, sending short message and email to notify the relevant maintenance personnel.

[0018] Further, the shaft force sensor in the data acquisition module adopts special strain gage material and corresponding process, in the manufacturing process, first use nanotechnology to introduce nanoparticles, including nanometer carbon tube or nanometer ceramic particle, in the special doping treatment of the base material, refine the microstructure, enhance the stability of the crystal to improve the sensitivity and stability, so that the strain gage has better electrical performance and smaller shaft force measurement fluctuation in complex environment, at the same time, the sensor is integrated with intelligent self-calibration and compensation module, through microprocessor and high precision reference element, real-time monitoring, comparison and error compensation, regular self-calibration to ensure high precision measurement, in addition, high strength, corrosion resistant protective coating is applied to the strain gage structure, including special organic polymer or ceramic coating, which forms a dense layer through process, blocks impurities from entering, resists wear and tear, prolongs the service life in harsh environment, improves the installation structure and process, designs high precision installation clamp, uses special fastening parts with thread locking agent to prevent loosening, optimizes the contact interface design to reduce stress concentration, ensures accurate and stable transmission of shaft force, and introduces wireless sensor network technology to build a distributed data acquisition system, multiple sensor nodes self-organize networking, with independent acquisition processing and collaborative interaction ability, uses routing protocol and synchronization algorithm to guarantee data transmission, improves monitoring accuracy and reliability through data fusion collaborative processing, and provides comprehensive and accurate data support for safe operation of the fan unit.

[0019] Further, the hardware platform of the data analysis processing center module adopts a distributed computing architecture, edge computing devices are deployed on site for fan units, and preliminary screening, preprocessing and classification marking are performed after data collection, reducing invalid data transmission and assisting the distributed computing architecture in focusing on deep tasks. Heterogeneous computing is introduced in the design of the computing node, equipped with CPU, GPU or FPGA chips, and the computing characteristics are coordinated to accelerate data processing, improve computing efficiency to meet real-time monitoring requirements, optimize data sharding and load balancing strategies, dynamically adjust according to shaft force data characteristics and node state to ensure efficient and stable operation of each node, set up data caching and prefetching mechanism, cache frequently used data and prefetch computing needs, speed up data access and processing speed, enhance fault tolerance and reliability, provide redundant backup and transaction processing mechanism for nodes, regularly check and diagnose to ensure continuous and accurate data processing, and lay a solid foundation for safe operation of fan units.

[0020] Further, in addition to the conventional data cleaning and noise reduction processing, the data analysis processing center also adopts a feature extraction method based on wavelet transform when preprocessing data, the specific operation is as follows:

[0021] The collected shaft force data sequence F = {F1, F2, F3, …, F n} is subjected to wavelet transform and decomposed into wavelet coefficients at different scales. By analyzing these wavelet coefficients, the characteristic information of the shaft force data at different frequency bands can be extracted. According to the extracted characteristic information, the shaft force data is further screened and optimized to remove some noise components and irrelevant information that have little effect on shaft trend prediction, making the preprocessed shaft force data F' more representative and regular, thereby improving the accuracy and reliability of subsequent prediction based on the comprehensive influence factor and shaft trend prediction model. When performing wavelet transform, a specific wavelet basis function is selected according to the characteristics of the shaft force data and the prediction requirements. After multiple experiments and comparisons, it is determined that the wavelet basis function can better capture the internal variation law of the shaft force data when processing the shaft force data.

[0022] Further, when determining the weight coefficients a, β, γ, δ of the comprehensive influence factor C, the data analysis and processing center performs simulation experiments through a large amount of actual operation data of the fan unit, analyzes the correlation between the shaft force change and each operating condition factor under different operating conditions, selects a plurality of groups of shaft force data and corresponding operating condition data of different types of fans under different wind speeds, rotating speeds and load conditions, respectively calculates the sensitivity of the shaft force change relative to the change of each operating condition factor under different operating conditions, and finds, through multiple experiments and statistical analysis, that when the average sensitivity of the shaft force change to the wind speed change is k1, the average sensitivity of the shaft force change to the rotating speed change is k2, and the average sensitivity of the shaft force change to the load change is k3, the value range of the weight coefficients a, β, γ, δ can be preliminarily determined, and the values of a, β, γ, δ are determined according to the design parameters and operating environment characteristic factors of the specific fan unit.

[0023] The structural parameter information of the fan unit is also introduced. For different types and structures of fan units, the force distribution of the blade root bolt, the length and weight of the blade, and the size of the hub structure parameter will affect the relationship between the shaft force and the operating condition. The shaft force of the fan unit with longer and heavier blades may be larger under the same wind speed, and the influence degree of the shaft force on the rotating speed and the load may also be different. When determining the weight coefficients, the fan unit is first classified according to the structural parameters, and then the simulation experiments and statistical analysis described above are performed for each type of fan unit to obtain more accurate weight coefficient values, so that the comprehensive influence factor C more accurately reflects the real relationship between the shaft force and the operating condition of the fan unit with different structures, and the accuracy and applicability of the shaft force trend prediction model are improved.

[0024] Further, when constructing the shaft force trend prediction model, in order to improve the robustness and adaptability of the model, an adaptive adjustment mechanism is introduced in the prediction formula After each shaft force trend prediction, the prediction result is compared and analyzed with the actual subsequent collected shaft force data, and if the deviation between the prediction result and the actual data exceeds a certain threshold, it means that the prediction model may have deviation or be not adaptive to the new operating condition change. According to the size and direction of the deviation, the weight coefficients θ i are adaptively adjusted. If the prediction result is too large, the values of some θ i need to be appropriately reduced, and if the prediction result is too small, the values of some θ i need to be increased. Through this adaptive adjustment mechanism, the shaft force trend prediction model can continuously adapt to various operating condition changes in the operation process of the fan unit, always maintains high prediction accuracy, effectively guarantees the accurate prediction of the future change trend of the shaft force, and provides a more reliable basis for maintenance and early warning.

[0025] Further, when the maintenance warning module triggers a maintenance warning signal, in addition to displaying warning information on the monitoring terminal, sending short messages or emails to notify relevant maintenance personnel, it is also equipped with a local audible and visual alarm installed in a conspicuous position near the fan unit. When the data analysis and processing center predicts that the shaft force may exceed the safety threshold, in addition to the remote notification method, the local audible and visual alarm will also start to send out strong sound and light signals, so that the on-site staff can know the situation in the first time. Even if the remote notification fails to arrive in time due to network failure, the on-site staff can still be informed of the abnormal shaft force in time, so that appropriate measures can be taken, further improving the reliability and timeliness of maintenance warning.

[0026] Further, when the data analysis and processing center establishes a data interface with other monitoring systems of the fan unit, including but not limited to wind speed monitoring system, rotating speed monitoring system, load monitoring system, it adopts a standardized and extensible interface protocol based on the general data transmission standard.

[0027] Compared with the prior art, a fan unit blade root bolt shaft force monitoring system has the following beneficial effects:

[0028] I. The system collects real-time shaft force data of the blade root bolt through high-precision shaft force sensors and transmits the data to the data analysis and processing center using wireless communication. The data analysis and processing center uses big data analysis technology and shaft force trend prediction algorithm to accurately predict the future change trend of the shaft force in combination with the operating conditions of the fan. This function greatly improves the safety and reliability of the fan unit operation. By early warning of potential shaft force abnormalities, bolt loosening and breaking failures can be effectively avoided, thereby prolonging the service life of the fan unit, reducing maintenance costs and production interruption risks.

[0029] II. The system introduces the structural parameter information of the fan unit and performs customized analysis for different types and structures of fan units, making the comprehensive influence factor and shaft force trend prediction model more accurately reflect the real relationship between shaft force and operating conditions of different structural fan units. In addition, the data analysis and processing center uses a distributed computing architecture to efficiently process massive shaft force data and operating condition data, ensuring accurate prediction of future shaft force trends in a short time. These features make the system have wide applicability and flexibility, meeting the monitoring needs of different fan units and providing strong support for intelligent operation and maintenance of the wind power industry. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0031] Figure 1 A fan unit blade root bolt axial force monitoring system operation flow chart. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present application will be described clearly and completely below. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0033] Embodiment one

[0034] This embodiment describes that in a certain large-scale land-based wind farm in Inner Mongolia, due to the large and frequent changes of wind speed in this area, the fan unit is often in a high load operation state, which puts high requirements on the reliability and durability of the fan parts. The wind farm has a total of 50 wind turbines, each unit is 80 meters high, the blade length is 40 meters, and the fixed pitch regulation technology is adopted. In order to ensure the stable operation of the fan unit and prolong the service life, the wind farm decides to introduce a fan unit blade root bolt axial force monitoring system.

[0035] At the blade root bolt part of each fan, high-precision axial force sensors are carefully selected and installed. These sensors are made of special strain gauge material, have high sensitivity, high stability and good fatigue resistance, and can accurately sense the small changes of axial force. When installing, the type and parameters of the sensor need to be determined according to the specifications and stress range of the blade root bolt, and strict calibration and debugging are required to ensure the accuracy of the data. The sensor uses advanced wireless transmission technology to send the collected axial force data to the nearby data acquisition base station in real time. This wireless transmission method not only improves the real-time performance of the data, but also avoids the problems of line aging and damage that may be caused by wired transmission. The data acquisition base station is responsible for receiving the axial force data from each fan and transmitting it to the data analysis and processing center through a wired network. In order to ensure the integrity and security of the data, data compression and encryption technology is used during transmission. According to the actual layout and field environment of the wind farm, the communication network is reasonably configured and optimized to ensure the timeliness, stability and integrity of data transmission. At the same time, a data backup and recovery mechanism is established to prevent data loss or damage.

[0036] The data analysis processing center receives and processes the axial force data, first performs data cleaning and noise reduction processing to remove outliers and noise components, and further adopts a feature extraction method based on wavelet transform to extract the characteristic information of the axial force data in different frequency bands, making the data more representative and regular, and constructing a comprehensive influence factor,

[0037] wherein, is the mean of the preprocessed axial force data, s F′ is the standard deviation of the preprocessed axial force data, is the mean of the wind speed data, s W is the standard deviation of the wind speed data, is the mean of the rotational speed data, s R is the standard deviation of the rotational speed data, is the mean of the load data, s L is the standard deviation of the load data, and α, β, γ, δ are weight coefficients. This factor comprehensively considers the influence of wind speed, rotational speed, load, and structural parameter factors of the fan unit on the axial force. Through a large number of simulation experiments and statistical analysis of actual operation data of the fan unit, the weight coefficients of the comprehensive influence factor are determined, making the prediction model more accurate and applicable. Deep mining and axial force trend prediction are performed using big data analysis technology. The prediction model is based on the comprehensive influence factor and the historical data of the axial force. Let F n+m be the predicted future mth axial force value, then the prediction formula is:

[0038]

[0039] wherein, C i is the value of the comprehensive influence factor C at the ith time, is the mean of the comprehensive influence factor C, and θ i is another set of weight coefficients. The value of θ i is adjusted constantly until the combination of values that can achieve high prediction accuracy under various different working conditions is found, and the value of θ i is finally determined. The future axial force trend is predicted. In order to improve the robustness and adaptability of the model, an adaptive adjustment mechanism is also introduced to adjust the weight coefficients adaptively according to the deviation between the prediction results and the actual data.

[0040] When the data analysis center predicts that the shaft force may exceed the safety threshold or show abnormal trend, it immediately triggers a maintenance warning signal. The warning methods are various, including displaying warning information on the wind farm monitoring center, sending messages or emails to notify maintenance personnel, and setting up remote sound and light alarms near the wind turbine units. When the warning signal is triggered, the alarm will emit strong sound and light signals to ensure that the on-site staff can know the situation in the first time and take appropriate measures. This local warning method further improves the reliability and timeliness of the warning. According to the warning signal and the trend of shaft force data, a detailed maintenance plan is developed and implemented, including inspection, tightening or replacement of blade root bolts, to ensure the stable operation of the wind turbine unit and prolong its service life.

[0041] Embodiment Two

[0042] This embodiment describes that in a vast and complex offshore wind farm, dozens or even hundreds of wind turbine units stand on the turbulent sea and capture wind energy day after day and convert it into clean electricity. However, the harsh environment on the sea, such as strong wind, huge waves and seawater corrosion, brings great challenges to the stable operation of the wind turbine, especially the blade root bolt, which is a key component connecting the blade and the hub. The state of its shaft force is directly related to the safety and stability of the wind turbine.

[0043] On each wind turbine in the offshore wind farm, we carefully install high-precision shaft force sensors. These sensors use a special strain gauge material that has been specially treated to have high sensitivity, high stability and excellent fatigue resistance. Even in harsh marine environments, these sensors can operate stably for a long time, collecting real-time shaft force data of the blade root bolt. The accuracy and stability of the data collection are greatly guaranteed, providing a solid foundation for subsequent data analysis.

[0044] Due to the special environment of offshore wind farms, wiring is difficult and costly, so we use wireless communication methods such as 4G / 5G or satellite communication to transmit shaft force data in real time to the data analysis center on land. When choosing the communication method, we fully consider the actual layout of the wind farm and the marine environment, and conduct detailed configuration and debugging to ensure the timeliness, stability and integrity of data transmission.

[0045] Receive shaft force data from each wind turbine and use distributed computing architecture to efficiently process large amounts of data. First, pre-process the shaft force data, including data cleaning, noise reduction processing, and feature extraction based on wavelet transform to extract feature information of the shaft force data in different frequency bands. Then, combined with wind speed, rotational speed and load operating condition data, a comprehensive influence factor is constructed,

[0046] wherein, is the mean of the pretreated axial force data, s F′ is the standard deviation of the pretreated axial force data, is the mean of the wind speed data, s W is the standard deviation of the wind speed data, is the mean of the rotating speed data, s R is the standard deviation of the rotating speed data, is the mean of the load data, s L is the standard deviation of the load data, α, β, γ, δ are weight coefficients, and a shaft force trend prediction model is constructed based on the factors, and let the predicted future mth shaft force value be F n+m The prediction formula is:

[0047]

[0048] wherein, C i is the value of the comprehensive influence factor C at the ith time, is the mean of the comprehensive influence factor C, θ i is another set of weight coefficients, and the value of θ i is adjusted constantly until a combination of values is found that can make the prediction accuracy reach a high level under various different working conditions, and finally the value of θ i is determined, and the prediction model has a self-adaptive adjustment mechanism that can continuously optimize the prediction accuracy according to actual data feedback. Through deep learning algorithms and big data analysis techniques, we can achieve accurate prediction of the future change trend of the blade root bolt axial force.

[0049] When the data analysis and processing center predicts that the shaft force may exceed the safety threshold or show an abnormal change trend, the maintenance warning module will immediately trigger a warning signal. The warning methods are various, including displaying warning information on the monitoring terminal, sending short message or email notifications to relevant maintenance personnel, and starting a local sound and light alarm. The local sound and light alarm is installed in a conspicuous position near the fan unit. Once triggered, it will emit strong sound and light signals so that the on-site staff can know the situation in the first time and take appropriate measures. This multi-level warning mechanism ensures the timeliness and accuracy of the warning information, providing a strong guarantee for the safe operation of the fan.

[0050] It will be apparent to those skilled in the art that the application is not limited to the details of the above-exemplified embodiments and that the present application can be implemented in other particular forms without departing from the spirit or essential characteristics of the present application. The embodiments should therefore be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the above description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. No reference signs in the claims should be considered as limiting the scope of the claims with respect to the figures of the patent document.

Claims

1. A system for monitoring the axial force of wind turbine blade root bolts, characterized in that, The system includes a data acquisition module, a data transmission module, a data analysis and processing center, and a maintenance early warning module; The data acquisition module collects the axial force data of the blade root bolt in real time. A high-precision axial force sensor is installed at the blade root bolt. During installation, the sensor model and parameters need to be determined according to the specifications of the blade root bolt and the force range factors, and calibration and debugging are required. The data transmission module transmits the collected axial force data to the data analysis and processing center in real time. The data transmission is carried out wirelessly. When selecting the communication method, it is necessary to determine it according to the actual layout of the wind turbine unit and the site environment, and to configure and debug it to ensure the timeliness, stability and integrity of the data transmission. The data analysis and processing center receives the axial force data, preprocesses it, performs in-depth analysis using big data analytics, and predicts the future trend of axial force changes based on the operating conditions of the wind turbine. An axial force trend prediction algorithm is employed, as detailed below: Let the collected blade root bolt axial force data sequence be as follows: ,in Indicates the first The axial force value collected this time. To collect data on the number of data collection sessions, we need to obtain the operating condition data of the wind turbine, including wind speed sequences. Rotational speed sequence Load sequence , here , , They represent the first The collected wind speed, rotational speed, and load values ​​are used to preprocess the axial force data, including data cleaning to remove outliers and noise reduction. Let the preprocessed axial force data sequence be: ； Construct a comprehensive impact factor The calculation formula is as follows: ； in, This represents the mean of the preprocessed axial force data. The standard deviation of the preprocessed axial force data. This is the average of the wind speed data. The standard deviation of the wind speed data. This is the average of the rotational speed data. The standard deviation of the rotational speed data. The mean of the load data. The standard deviation of the load data. These are the weighting coefficients; Based on comprehensive impact factors Construct an axial force trend prediction model, assuming the predicted future... The secondary axial force is The prediction formula is: ； in, Comprehensive Influence Factor In the The value at this time, Comprehensive Influence Factor The mean, Another set of weighting coefficients is continuously adjusted. The value is determined until a combination of values ​​that achieves a high level of prediction accuracy under various different operating conditions is found. The value; The maintenance early warning module: When the data analysis and processing center predicts that the axial force exceeds the safety threshold or shows an abnormal trend, it will trigger a maintenance early warning signal in advance. The early warning methods include displaying early warning information on the monitoring terminal, sending text messages and emails to notify maintenance personnel. In addition to conventional data cleaning and noise reduction, the data analysis and processing center also employs a feature extraction method based on wavelet transform during data preprocessing. The specific operations are as follows: The acquired axial force data sequence Wavelet transform is performed to decompose the axial force data into wavelet coefficients at different scales. By analyzing these wavelet coefficients, the characteristic information of the axial force data in different frequency bands is extracted. Based on the extracted characteristic information, the axial force data is further filtered and optimized to remove noise components and irrelevant information that have little impact on the prediction of axial force trends. When performing wavelet transform, a wavelet basis function is selected. Based on the characteristics of the axial force data and the prediction requirements, after multiple experiments and comparisons, a specific wavelet basis function is determined to be used. This wavelet basis function can better capture the inherent variation law when processing axial force data.

2. The wind turbine blade root bolt axial force monitoring system according to claim 1, characterized in that, The axial force sensor in the data acquisition module uses special strain gauge materials and corresponding processes. During manufacturing, nanotechnology is first used to introduce nanoparticles during the special doping treatment of the base material. At the same time, the sensor integrates an intelligent self-calibration and compensation module. Through a microprocessor and high-precision reference elements, it monitors, compares, and compensates for errors in real time. Regular self-calibration ensures high-precision measurement. In addition, a high-strength, corrosion-resistant protective coating is applied to the strain gauge structure to form a dense layer, which prevents impurities from entering, resists wear, and extends service life in harsh environments. The installation structure and process are improved, a high-precision installation fixture is designed, and special fastening components are used in conjunction with thread locking agents to prevent loosening. The contact interface design is optimized to reduce stress concentration. Furthermore, wireless sensor network technology is introduced to build a distributed data acquisition system. Multiple sensor nodes self-organize and network, possessing independent acquisition and processing capabilities as well as collaborative interaction capabilities. Routing protocols and synchronization algorithms are used to ensure data transmission. Data fusion and collaborative processing improve the accuracy and reliability of monitoring.

3. The wind turbine blade root bolt axial force monitoring system according to claim 1, characterized in that, The hardware platform of the data analysis and processing center adopts a distributed computing architecture. Edge computing devices are deployed at the wind turbine site. After data collection, preliminary screening, preprocessing, and classification are performed first to reduce invalid data transmission and help the distributed computing architecture focus on deep tasks. The computing node design introduces heterogeneous computing, which accelerates data processing collaboratively according to computing characteristics, improves computing efficiency to meet real-time monitoring requirements, optimizes data sharding and load balancing strategies, and dynamically adjusts them according to the characteristics of shaft force data and node status. Data caching and prefetching mechanisms are set up to cache frequently used data and prefetch the data required for calculation, speeding up data access and processing, enhancing fault tolerance and reliability, and setting up redundant backup and transaction processing mechanisms for nodes, and performing regular checks and diagnoses.

4. The wind turbine blade root bolt axial force monitoring system according to claim 1, characterized in that, The data analysis and processing center determines the comprehensive influencing factors. Weighting coefficients , , , Simulation experiments were conducted using actual operating data of the wind turbine units to analyze the correlation between axial force changes and various operating condition factors under different operating conditions. Multiple sets of axial force data from different wind turbine models under different wind speeds, rotational speeds, and load conditions, along with corresponding operating condition data, were selected. The sensitivity of axial force changes to individual changes in each operating condition factor under different operating conditions was calculated. After multiple experiments and statistical analyses, it was found that the average sensitivity of axial force changes to wind speed changes was [missing information]. At that time, the weighting coefficients can be preliminarily determined. , , , The range of values ​​is then fine-tuned based on the specific design parameters and operating environment characteristics of the wind turbine unit to determine the final value. The value; The structural parameters of the wind turbine units were also incorporated. For different models and structures of wind turbine units, the force distribution on the blade root bolts, the length and weight of the blades, and the size and structural parameters of the hub will affect the relationship between axial force and operating conditions. Wind turbine units with longer and heavier blades will experience greater axial force on the blade root bolts at the same wind speed, and the degree to which axial force affects speed and load will also differ. When determining the weighting coefficients, the above simulation experiments and statistical analyses were conducted separately for each type of wind turbine unit based on the structural parameters of the wind turbine units, so as to comprehensively influence the factors. It more accurately reflects the true relationship between axial force and operating conditions of wind turbine units with different structures.

5. The wind turbine blade root bolt axial force monitoring system according to claim 1, characterized in that, In constructing the axial force trend prediction model, the data analysis and processing center, in order to improve the robustness and adaptability of the model, modified the prediction formula... An adaptive adjustment mechanism is introduced. After each axial force trend prediction, the prediction result is compared and analyzed with the actual axial force data collected subsequently. If the deviation between the prediction result and the actual data exceeds a threshold, it indicates that the current prediction model has a bias or is not adapted to the new working condition changes. Based on the magnitude and direction of the deviation, the weighting coefficients are adjusted. Make adaptive adjustments.

6. The wind turbine blade root bolt axial force monitoring system according to claim 1, characterized in that, When a maintenance early warning module triggers a maintenance early warning signal, in addition to displaying warning information on the monitoring terminal and sending SMS and email notifications to relevant maintenance personnel, it is also equipped with a local audible and visual alarm. The local audible and visual alarm is installed in a conspicuous location near the wind turbine unit. When the data analysis and processing center predicts that the axial force exceeds the safety threshold, in addition to remote notification, the local audible and visual alarm will also be activated simultaneously, emitting a strong audible and visual signal. Even if the remote notification fails to be delivered in time due to network failure, it can ensure that on-site personnel are aware of the abnormal axial force situation in a timely manner and take appropriate measures.

7. The wind turbine blade root bolt axial force monitoring system according to claim 1, characterized in that, When establishing data interfaces with other monitoring systems of the wind turbine unit, including wind speed monitoring system, rotational speed monitoring system, and load monitoring system, the data analysis and processing center adopted a standardized and scalable interface protocol. The interface protocol was customized based on the general data transmission standard.

Images