Real-time monitoring method and system for perpendicularity of mixed-tower wind power tower

Abstract

The invention relates to the technical field of mixed tower wind power tower monitoring, and discloses a real-time monitoring method and system for the perpendicularity of a mixed tower wind power tower. The system comprises a data sensing layer, a communication control layer and a perpendicularity analysis layer, wherein the data sensing layer realizes multi-point synchronous measurement and dynamic calibration of tower deformation through a displacement sensor array, a dipmeter and the like; the communication control layer completes multi-source data time synchronization, format standardization and fusion transmission through edge calculation, protocol conversion and space-time registration technologies; the perpendicularity analysis layer combines design parameters and real-time data to construct a holographic tower space attitude model, and deformation prediction, benchmark correction and abnormity warning are achieved through a dynamic weighting algorithm. The method comprises the steps of data acquisition and calibration, processing and transmission, model construction analysis and result output. The method achieves the real-time precise monitoring of the perpendicularity of the tower, improves the operation safety, and is suitable for the state monitoring of the mixed-tower wind power tower.

Description

Technical Field

[0001] The present invention relates to the technical field of hybrid wind turbine tower monitoring, and in particular to a real-time monitoring method and system for the verticality of a hybrid wind turbine tower. Background Art

[0002] With the growing global demand for clean energy, wind power generation has been widely promoted as an important form of renewable energy. As the key supporting structure of a wind turbine generator set, the verticality of the hybrid wind turbine tower is directly related to the operational safety, stability, and power generation efficiency of the unit. During actual operation, the hybrid wind turbine tower is affected by a variety of factors, such as strong wind loads, temperature changes, mechanical vibrations, and foundation settlement. These factors may cause the tower to undergo structural deformations such as axial displacement, radial offset, and torsional deformation, thereby causing vertical deviations. Failure to monitor the verticality of the tower in a timely and accurate manner may lead to increased wear of unit components, reduced power generation efficiency, and even serious safety accidents, resulting in huge economic losses and social impacts.

[0003] Traditional tower verticality monitoring methods primarily rely on regular manual inspections and single-point measurement equipment, such as total stations and theodolites. These methods have significant limitations: manual inspections are inefficient and time-consuming, making real-time monitoring impossible and difficult to detect dynamic tower deformations. Single-point measurement equipment can only capture local data from limited locations, failing to fully reflect the tower's overall deformation characteristics, resulting in low monitoring accuracy and reliability. Furthermore, traditional methods often lack the effective integration and analysis of multi-source monitoring data, making it difficult to establish an accurate spatial posture model of the tower, scientifically predict and warn of deformation trends, and dynamically calibrate and compensate the monitoring system.

[0004] The rapid development of intelligent sensing, Internet of Things, big data analysis, and digital twin technologies has provided new technical solutions and approaches for real-time monitoring of the verticality of hybrid wind turbine towers. Key technical challenges in hybrid wind turbine tower monitoring include leveraging advanced sensor technology to achieve multi-point, simultaneous measurement of tower structural deformation; utilizing communication technologies to achieve real-time transmission and fusion processing of multi-source monitoring data; building a holographic spatial posture model of the tower using spatial geometry analysis and dynamic modeling techniques; and employing intelligent algorithms to predict deformation trends, correct baseline deviations, and generate verticality anomaly alerts. Summary of the Invention

[0005] The object of the present invention is to provide a method and system for real-time monitoring of the verticality of a hybrid wind turbine tower, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a real-time monitoring system for the verticality of a hybrid wind turbine tower, the system comprising:

[0007] Data perception layer, communication control layer and verticality analysis layer;

[0008] The data perception layer includes a displacement sensor array, an inclinometer, a dynamic calibration device and a reference positioning module, which are used to perform multi-point synchronous measurement of tower structure deformation, reference coordinate calibration and dynamic calibration compensation operations based on feedback instructions;

[0009] The communication control layer includes edge computing nodes, protocol converters, and a cloud processing platform. It also deploys a time synchronization mechanism and a multimodal data fusion strategy to establish redundant transmission links, achieve time alignment and spatial registration of multi-source monitoring data, and use an adaptive compression algorithm to extract features and transmit hierarchical data according to data accuracy requirements and transmission bandwidth. It also implements dynamic interaction between physical measurement data and 3D modeling data by configuring a multi-source data interface.

[0010] The verticality analysis layer is used to dynamically calibrate the three-dimensional deformation model of the tower using spatial geometric analysis technology combined with structural design parameters and real-time monitoring data sets, and then construct a holographic mapping tower spatial posture model. Based on this posture model, a dynamic weighted algorithm is used to predict deformation trends, correct benchmark deviations, and issue verticality anomaly alarms.

[0011] Preferably, the displacement sensor array includes at least a laser ranging unit, a fiber grating sensor, an ultrasonic echo detector and a millimeter wave radar, which are used to obtain the displacement distribution characteristics of the tower section; the dynamic calibration device includes at least a temperature compensation module, a vibration suppressor, a coordinate correction unit and a data verifier, which are used to eliminate the environmental errors of the measurement system; the reference positioning module includes at least a Beidou positioning terminal, a total station reference point, and an inertial navigation component, which is used to execute the coordinate calibration instructions issued by the verticality resolution layer.

[0012] Preferably, in the communication control layer, the data collected by the displacement sensor array is transmitted by the edge computing node to a protocol converter after being timestamped. The protocol converter then performs data format standardization processing and transmits the data together with the spatial posture information recorded by the inclinometer to a cloud processing platform for modeling and calculation. The data processing uses spatiotemporal registration technology to input the collected displacement data into three-dimensional grid models of different resolutions for joint analysis.

[0013] The method of realizing dynamic interaction between physical measurement data and three-dimensional modeling data by configuring a multi-source data interface includes: realizing dynamic interaction between physical measurement data and three-dimensional modeling data by configuring a multi-source data interface, exchanging real-time monitoring data of tower deformation, and jointly analyzing the acquired spatial coordinates, realizing dynamic conversion of coordinate systems, completing data verification, model updating, and parameter distribution, and issuing calibration control instructions to the data perception layer.

[0014] Preferably, the construction of the holographically mapped tower spatial posture model includes:

[0015] Establish a parameter correlation channel and bidirectional calibration mechanism between physical structure deformation and digital modeling space;

[0016] The actual measurement data is spatially interpolated and transformed into coordinates. Based on the axial displacement, radial offset, and torsion angle obtained by the displacement sensor array, a reference posture library and deformation feature library of the tower structure are constructed. Model parameter iteration and dynamic characteristic simulation are performed based on real-time monitoring data to convert the actual structural deformation into a high-precision digital twin model.

[0017] Optimize the parameters of the spatial posture model, input the real-time monitored deformation data into the established posture model, and use the residual correction algorithm to dynamically compensate the output of the model to obtain the optimized tower spatial posture model;

[0018] The tower spatial posture model includes a physical measurement space, a digital modeling space, a structural database, and an interaction mechanism between modules;

[0019] The physical measurement space is the data source of the posture model and contains the original deformation characteristics of the tower structure; the digital modeling space forms a mapping relationship with the physical measurement space, and geometrically represents the tower posture characteristics through multi-dimensional space modeling; the structural database integrates design parameters and real-time monitoring information, and provides a basic data set including a material property library, a load distribution library, and a connection node library; the interaction mechanism realizes data communication between modules, parameter collection and model update are realized between the physical measurement space and the structural database through a standardized interface, parameter transfer is carried out between the physical measurement space and the digital modeling space through a data channel, and information interaction is realized between the digital modeling space and the structural database through middleware.

[0020] Preferably, the deformation trend prediction using a dynamic weighted algorithm based on the posture model includes:

[0021] Based on the tower spatial posture model, historical operation posture data, environmental load records, and deformation event feature data are obtained to construct a deformation sample set;

[0022] After feature enhancement processing, the deformation sample set is divided into a training set and a test set;

[0023] Establish an ARIMA-LSTM-Prophet hybrid forecasting architecture, set model initialization parameters, input the training set into the hybrid model for joint training, use the ARIMA model to perform trend decomposition, use the LSTM network to extract time series features, and use the Prophet algorithm to identify periodic patterns. Use an adaptive weighting mechanism to balance the prediction errors of different algorithms under specific working conditions until the model fit reaches the set threshold or the scheduled training rounds are completed;

[0024] Input the test set into the trained hybrid model, calculate the comprehensive performance index of the model, and select the optimal deformation prediction model;

[0025] Based on the optimal deformation prediction model, the spatial distribution trend of the tower deformation is output, and the verticality deviation evolution path is determined in combination with the material mechanical parameters.

[0026] The training process of the hybrid prediction architecture includes:

[0027] The Boosting method is used to generate training sequences for multiple base models, and the prediction performance of each base model is evaluated through sliding window verification.

[0028] The gradient descent algorithm is used to calculate the dynamic fusion coefficient of each base model, and the weighted average mechanism is adopted to comprehensively predict the output results of the base models.

[0029] Preferably, the reference deviation correction is performed using a dynamic weighted algorithm based on the posture model, including:

[0030] Based on the tower spatial posture model, the axial deviation, radial offset, and torsional deformation parameters of the structural deformation are extracted, and a deviation feature vector set is constructed.

[0031] Independent component analysis is used to reduce the noise of the deviation feature vector to obtain the core deviation feature component;

[0032] A deviation classification model based on Gaussian mixture model was established, and the optimal number of classifications was determined by the BIC criterion;

[0033] The classified deviation features are input into the preset correction strategy library to match the optimal correction scheme and generate a directional benchmark correction parameter set.

[0034] Preferably, the verticality abnormality alarm is performed using a dynamic weighted algorithm based on the posture model, including:

[0035] Based on the tower spatial posture model, the structural vibration characteristics, wind speed load spectrum, and foundation settlement data are collected to build an operating status feature library;

[0036] Perform overlapping segmentation processing on the data in the operating status feature library to generate a spatial distribution sample set;

[0037] Establish a convolutional neural network model, set the convolution kernel size and pooling parameters, and obtain the deep expression of spatial features through backpropagation calculation;

[0038] The spatial features are input into the classifier for state pattern recognition, and the discrimination results between normal operation state and verticality abnormal state are output.

[0039] Preferably, the step of constructing a holographically mapped tower spatial posture model further includes:

[0040] The sliding spatial window mechanism is used to divide the continuous monitoring area into blocks, and the data blocks in each spatial window are independently subjected to feature extraction;

[0041] Establish a correlation matrix between data block features and structural loads, record the deformation distribution patterns corresponding to different working conditions, incrementally update the posture model through an online learning algorithm, and trigger the model topology reconstruction mechanism when an unrecorded deformation pattern is detected.

[0042] Preferably, the method for generating the reference correction parameter set includes:

[0043] Establish a correspondence table between deviation types and correction methods, including correction methods for axial deviation corresponding to prestress adjustment and radial offset corresponding to counterweight optimization;

[0044] A genetic algorithm is used to search for the optimal correction parameter combination, including adjustment amplitude, action position and correction timing. The correction effect is evaluated in real time through a double closed-loop control mechanism. When the residual deviation exceeds the threshold, the parameter re-optimization process is triggered.

[0045] Preferably, the present invention further includes a real-time monitoring method for the verticality of a hybrid wind turbine tower, comprising the following steps:

[0046] The tower structure is synchronously collected at multiple points using a displacement sensor array and an inclinometer to obtain axial displacement, radial offset, and torsion angle measurements. Based on the coordinate calibration results of the reference positioning module, a dynamic calibration device is used to compensate for temperature drift and eliminate vibration interference.

[0047] The collected heterogeneous monitoring data is timestamped through edge computing nodes, and a multimodal data fusion strategy is used to implement time synchronization and spatial registration of displacement data and inclination data. An adaptive compression algorithm is used to extract displacement distribution characteristics based on transmission bandwidth constraints, and a protocol converter is used to convert physical measurement data into a standardized data format compatible with the 3D modeling space.

[0048] The tower structure design parameters are loaded into the cloud processing platform, and the pre-built three-dimensional deformation model is dynamically calibrated in combination with the real-time monitoring data set. The residual correction algorithm is used to optimize the model parameters and then generate a holographic mapping tower spatial posture model. Based on this model, the axial deformation gradient, radial offset trend and torsion angle change are fused through a dynamic weighted algorithm to output the deformation trend prediction results, benchmark deviation correction amount and verticality anomaly alarm signal.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] In terms of data acquisition and calibration, the data perception layer integrates a displacement sensor array, an inclinometer, a dynamic calibration device, and a reference positioning module, enabling simultaneous multi-point measurement of tower structural deformation. The displacement sensor array, which includes a variety of sensors such as laser ranging units and fiber Bragg grating sensors, can capture the displacement distribution characteristics of the tower's cross-section. The dynamic calibration device effectively eliminates environmental errors through components such as temperature compensation modules and vibration suppressors. The reference positioning module utilizes Beidou positioning terminals for coordinate calibration, ensuring the high accuracy and reliability of the original monitoring data and providing a solid foundation for subsequent analysis.

[0051] The communication control layer establishes an efficient data transmission and processing system. Edge computing nodes timestamp data, and protocol converters standardize data formats. Combined with time synchronization mechanisms and multimodal data fusion strategies, this ensures temporal and spatial alignment of multi-source data. An adaptive compression algorithm extracts features and performs layered transmission based on data accuracy and transmission bandwidth, ensuring data integrity and improving transmission efficiency. A multi-source data interface enables dynamic interaction between physical measurement data and 3D modeling data, promoting the deep integration of monitoring data and models.

[0052] The verticality resolution layer achieves precise analysis of the tower's state through spatial geometric analysis and intelligent algorithms. The constructed holographic mapping tower spatial posture model establishes a parameter association and bidirectional calibration mechanism between the physical structure and the digital modeling space, and combines spatial interpolation, coordinate transformation, and residual correction algorithms to transform actual deformations into a high-precision digital twin model, comprehensively and in real time reflecting the tower's spatial posture. Based on this model's dynamic weighted algorithm, for deformation trend prediction, an ARIMA-LSTM-Prophet hybrid prediction architecture is employed, integrating the advantages of multiple algorithms to achieve scientific predictions of the spatial distribution trend of tower deformation and the evolution path of verticality deviation. For baseline deviation correction, independent component analysis and Gaussian mixture models are used to reduce the noise, classify, and match the optimal correction scheme for deviation features, improving the relevance and effectiveness of baseline correction. For verticality anomaly alarms, a convolutional neural network model is used to perform in-depth analysis and pattern recognition of operating status characteristics, enabling timely and accurate identification of abnormal conditions and providing reliable assurance for the tower's safe operation.

[0053] Furthermore, the system's sliding spatial window mechanism, online learning algorithm, and model topology reconstruction mechanism enable incremental updates and adaptive adjustments to the posture model based on real-time monitoring data, improving the model's adaptability to complex working conditions and novel deformation patterns. The application of a genetic algorithm and a dual closed-loop control mechanism to generate baseline correction parameters enables optimization of correction parameters and real-time evaluation of correction effects, further enhancing the system's intelligence and automation. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a working principle diagram of the real-time monitoring system for the verticality of a hybrid wind turbine tower according to the present invention;

[0055] Figure 2 Flowchart of data processing and interaction of the communication control layer;

[0056] Figure 3 Flowchart of tower deformation trend prediction based on hybrid algorithm. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] See also Figure 1-Figure 3 The present invention relates to a real-time monitoring system for the verticality of a hybrid wind turbine tower. The core architecture of the system consists of a data perception layer, a communication control layer, and a verticality analysis layer. The specific implementation steps are as follows:

[0059] The data perception layer utilizes a displacement sensor array, inclinometer, dynamic calibration device, and reference positioning module to achieve multi-point synchronous measurement of tower structural deformation, reference coordinate calibration, and dynamic calibration compensation. The displacement sensor array covers key sections of the tower, collecting multi-dimensional data such as axial displacement and radial offset in real time. The inclinometer simultaneously records the tower's spatial attitude parameters. The dynamic calibration device compensates for environmental errors (such as temperature drift and vibration interference) in real time based on feedback instructions. The reference positioning module uses Beidou positioning terminals and total station reference points to obtain absolute coordinate references, providing an initial reference for verticality calculations.

[0060] The communication control layer builds redundant transmission links through edge computing nodes, protocol converters, and a cloud-based processing platform to achieve temporal and spatial registration of multi-source data. Edge computing nodes add timestamps to sensor data, protocol converters standardize data formats, and the cloud-based processing platform employs a multimodal data fusion strategy combined with an adaptive compression algorithm to extract features and transmit them in layers from heterogeneous data. Dynamic interaction between physical measurement data and 3D modeling data is achieved through a multi-source data interface.

[0061] The verticality analysis layer, based on spatial geometric analysis techniques, combines structural design parameters with real-time monitoring data to dynamically calibrate the tower's three-dimensional deformation model and construct a holographically mapped tower spatial posture model. Based on this model, a dynamic weighted algorithm is used to predict deformation trends, correct baseline deviations, and generate verticality anomaly alerts, providing real-time decision support for safe tower operation.

[0062] The technical solution of the present invention is further described in detail below with reference to specific embodiments.

[0063] Example 1:

[0064] The data perception layer serves as the data acquisition core of the hybrid wind turbine tower verticality real-time monitoring system. Through the collaborative work of a displacement sensor array, inclinometer, dynamic calibration device, and reference positioning module, it enables multi-point synchronous measurement of tower structural deformation, reference coordinate calibration, and dynamic calibration compensation. This layer's design must meet high-precision, anti-interference, and strong real-time requirements to ensure reliable raw data for subsequent data processing and analysis.

[0065] 1. Composition and function realization of displacement sensor array

[0066] The displacement sensor array utilizes a three-dimensional layout combining multiple sensor types, including at least laser ranging units, fiber Bragg grating (FBG) sensors, ultrasonic echo detectors, and millimeter-wave radars. These sensors are distributed at key locations across the tower at different heights, based on their measurement principles and applicable scenarios. The laser ranging unit emits a laser beam and receives reflected light from the measured surface. Using the time-of-flight method, it calculates the distance change between the sensor and the tower surface, achieving high-precision radial displacement measurements with millimeter-level accuracy. This makes it suitable for locations prone to significant deformation, such as the tower top and segment interfaces. The fiber Bragg grating (FBG) sensor is affixed to the tower's steel structure by gluing or embedding. Utilizing the strain- and temperature-dependent Bragg wavelength of the FBG, it simultaneously monitors axial strain and ambient temperature. When the tower undergoes axial tension or compression, the FBG pitch changes, resulting in a shift in the wavelength of the reflected light. A demodulator analyzes this wavelength change to obtain strain data. A temperature compensation algorithm is also employed to mitigate the effects of temperature drift on the measurement results. Ultrasonic echo detectors emit ultrasonic pulses and receive reflected echoes from the tower's internal structure. Based on the time delay and amplitude characteristics of the echo signals, they analyze the displacement distribution and internal defects of the tower's cross section. This makes them suitable for detecting deformation characteristics of complex structures such as flange joints. Millimeter-wave radars, on the other hand, emit broadband millimeter-wave signals and receive scattered echoes from the tower's surface. Using synthetic aperture radar (SAR) technology, they generate two-dimensional or three-dimensional point cloud data of the tower, enabling non-contact, full-area monitoring of the tower's dynamic deformation. This makes them particularly suitable for continuous monitoring in harsh environments such as strong winds, rain, and snow.

[0067] Various sensors are distributed in an array across horizontal sections at varying heights within the tower. Within each section, sensors are evenly spaced circumferentially, forming a multi-layered, ring-shaped monitoring network. For example, 3-5 layers of monitoring sections are located at the base, middle, and top of the tower, with 8-12 sensors per layer. This ensures radial, axial, and torsional deformation data collection covers the entire height of the tower. A synchronized triggering mechanism is used between sensors to synchronize multi-point data collection, with synchronization errors controlled to the microsecond level to ensure temporal and spatial consistency in subsequent data fusion.

[0068] 2. Layout and data collection of inclinometer

[0069] The inclinometer is used to monitor the spatial attitude parameters of the tower in real time, including pitch, yaw and roll angles. The device uses a high-precision MEMS inertial measurement unit (IMU), which integrates a three-axis accelerometer, a three-axis gyroscope and a three-axis magnetometer. It calculates the real-time attitude angle of the tower through a multi-sensor fusion algorithm. The inclinometer is installed at the center of the top of the tower and the center of mass of each segment to obtain attitude change data of the tower as a whole and each segment. The top inclinometer is mainly used to monitor the vertical deviation of the tower as a whole, while the inclinometers of each segment are used to analyze the impact of local deformation of the tower on the overall attitude. The measurement data is transmitted to the communication control layer in real time through a wired or wireless communication link, and is synchronized in time and spatially aligned with the displacement sensor data.

[0070] 3. Error Elimination Mechanism of Dynamic Calibration Device

[0071] The dynamic calibration device is the core component of the data perception layer that eliminates environmental errors. It integrates a temperature compensation module, a vibration suppressor, a coordinate correction unit, and a data verifier. The temperature compensation module establishes a temperature-deformation error model to perform real-time corrections for sensor drift caused by ambient temperature changes. This is achieved by deploying temperature sensors near temperature-sensitive devices such as fiber Bragg grating sensors to monitor ambient temperature changes in real time. A polynomial function relationship between temperature and measurement error is established through historical data fitting. For the axial strain measurement error of a fiber Bragg grating sensor, the error can be expressed as:

[0072] Δε=a·T 2 +b·T+c

[0073] Where T is the temperature, and a, b, and c are fitting coefficients. During data acquisition, an error compensation is calculated based on the real-time temperature value and added to the original measurement data to dynamically correct for temperature drift. The vibration suppressor utilizes inertial damping technology. By placing a damping mass on the sensor mounting base, it reduces interference from external excitations such as wind load and mechanical vibration on the measurement signal. The mass and stiffness parameters of the damping mass are optimized based on the natural vibration frequency of the tower, creating a resonance suppression effect for vibrations of specific frequencies, thereby reducing the impact of vibration noise on displacement and inclination measurement data.

[0074] The coordinate correction unit dynamically adjusts the sensor coordinate system based on the coordinate calibration instructions issued by the verticality resolution layer. These calibration instructions are generated based on the coordinate calibration results of the reference positioning module. For example, if the Beidou positioning terminal detects that the tower foundation has settled, causing the global coordinate system to shift, the coordinate correction unit uses translation and rotation transformation matrices to convert the sensor's local coordinate system data into absolute coordinate values ​​in the global coordinate system, ensuring spatial reference consistency across different sensor data.

[0075] The data verifier verifies the validity of collected data using a redundant verification algorithm. For the displacement sensor array, a cross-comparison method is used to compare data from multiple sensors within the same cross-section. When the deviation of a sensor's measurement value from the mean of other sensors within the same cross-section exceeds a preset threshold (e.g., three standard deviations), the data is identified as an outlier and removed, triggering a sensor fault alarm. Inclination measurement data is verified using the multi-sensor fusion results of the inertial navigation unit to ensure the reliability of the attitude angle data.

[0076] 4. Coordinate calibration process of the reference positioning module

[0077] The reference positioning module, consisting of a Beidou positioning terminal, a total station reference point, and an inertial navigation unit, provides an absolute coordinate reference and attitude reference for the entire monitoring system. The Beidou positioning terminal utilizes a multi-mode satellite receiver capable of simultaneously receiving signals from multiple satellite systems, including Beidou, GPS, and GLONASS. Using real-time kinematic (RTK) technology, it obtains the three-dimensional absolute coordinates of the tower base, achieving centimeter-level positioning accuracy. The Beidou positioning terminal's antenna is deployed in an open area near the tower base to avoid signal obstruction and multipath effects that could affect signal quality.

[0078] The total station datum point establishes the initial reference plane for tower verticality monitoring through optical measurement. The specific operation is as follows: Set up two or three total station measuring stations on stable ground around the tower, with good visibility between them. Use the total station to measure multiple characteristic points on the datum plane at the tower base to establish the datum plane equation in the local coordinate system. This datum plane serves as the initial reference for tower verticality calculations, and all subsequent monitoring data deviations are calculated based on this plane.

[0079] The inertial navigation component, integrated within the inclinometer, uses real-time measurement data from the gyroscope and accelerometer to calculate the tower's attitude angle change rate, enabling high-frequency monitoring of the tower's dynamic attitude (sampling frequency can reach over 100Hz). Inertial navigation data is integrated with Beidou positioning data and total station measurement data through a Kalman filter algorithm to generate a globally consistent coordinate reference and attitude parameters, providing high-precision benchmark data for the verticality resolution layer.

[0080] The data perception layer, through the collaborative work of the aforementioned components, forms a complete process from data acquisition, error calibration, to benchmark calibration. The displacement sensor array enables multi-point synchronous acquisition of multi-dimensional deformation data, the inclinometer provides attitude parameters, the dynamic calibration device eliminates environmental errors, and the benchmark positioning module establishes a coordinate reference. Together, these ensure that the data input to the communication control layer is highly precise, spatially and temporally consistent, and physically accurate, laying the foundation for subsequent real-time resolution and intelligent analysis of tower verticality. The design of this layer fully considers the structural characteristics and complex operating environment of hybrid wind turbine towers. Through technical means such as multi-sensor fusion, dynamic calibration, and benchmark calibration, it solves the problems of insufficient accuracy of traditional single-point monitoring and the significant impact of environmental interference, achieving comprehensive, real-time, and reliable monitoring of tower deformation.

[0081] Example 2:

[0082] The communications control layer is the data transmission and processing hub of the hybrid wind turbine tower verticality real-time monitoring system. Through the collaborative operation of edge computing nodes, protocol converters, and cloud processing platforms, it enables time synchronization, spatial registration, feature extraction, and model interaction for multi-source heterogeneous data. This layer's design must meet the requirements of high reliability, low latency, and strong compatibility to ensure the accuracy, integrity, and timeliness of data during transmission and processing.

[0083] Edge computing nodes, as near-end data processing units, are deployed at the base of the tower or in nearby cabinets. They directly interface with various sensors in the data perception layer and perform raw data preprocessing and transmission control. Their core functions include timestamping, noise filtering, and data encapsulation. During timestamping, edge computing nodes use built-in high-precision clock modules (such as those based on the Global Positioning System (GPS)) to add microsecond-accurate timestamps to each set of data collected by the displacement sensor array. This ensures strict temporal alignment of multi-source data (such as laser ranging data, fiber Bragg grating (FBG) strain data, and millimeter-wave radar point cloud data). This timing synchronization mechanism is fundamental to subsequent data fusion and modeling, avoiding spatial registration errors caused by time skew. Regarding noise filtering, edge computing nodes employ a sliding window filtering algorithm to suppress high-frequency vibration noise. For example, to address high-frequency tower vibration caused by wind excitation, a 50ms sliding window is set, and the displacement data within the window is subjected to median or mean filtering to eliminate anomalous fluctuations while retaining the low-frequency trend components that reflect the actual tower deformation. During the data encapsulation process, the edge computing node packages the processed sensor data of various types according to the preset data frame format. Each data frame contains fields such as sensor number, timestamp, measurement value, check code, etc., and is sent to the protocol converter through redundant transmission links (such as industrial Ethernet and 5G wireless communication in parallel) to improve the reliability of data transmission. When the main link (such as industrial Ethernet) is interrupted due to a fault, the system automatically switches to the backup link (such as 5G network) to continue transmitting data, ensuring the continuity of the monitoring process.

[0084] As a core component for heterogeneous data format conversion, the protocol converter features a built-in multi-protocol parsing engine, supporting the parsing and conversion of various industrial communication protocols (such as Modbus, OPC UA, and MQTT) as well as sensor-specific data formats. The processing process consists of two steps. First, the protocol converter converts various types of raw data, such as ASCII distance values ​​from laser ranging units, wavelength drift from fiber Bragg grating sensors, and binary point cloud data from millimeter-wave radar, into uniform physical units (such as millimeters, microstrain, and meters) through corresponding parsing modules and populates them into standardized data structures. Second, the spatial attitude information (pitch, yaw, and roll angles in degrees) recorded by the inclinometer is timestamped with the displacement data. Time synchronization algorithms (such as clock calibration based on the NTP protocol) ensure that the timestamp error between the two data types does not exceed 10 microseconds, resulting in a spatiotemporally consistent dataset containing spatial position (X, Y, and Z displacement) and attitude parameters. After format conversion and time alignment, the protocol converter transmits the data via TCP / IP to a cloud processing platform for subsequent modeling and computation. In addition, the protocol converter also supports data compression. Based on the transmission bandwidth constraints and data accuracy requirements, it uses adaptive compression algorithms (such as compression algorithms based on wavelet transform) to perform layered compression on large-capacity data such as millimeter-wave radar point cloud data. High-resolution details are retained for point cloud data in key areas (such as tower flange connection nodes), and low-resolution compression is performed on data in non-critical areas, reducing transmission bandwidth occupancy while ensuring data availability.

[0085] The cloud-based processing platform is the core of data fusion and model interaction at the communication and control layer. Built on a distributed computing architecture, it boasts powerful data storage, processing, and 3D modeling capabilities. The platform first uses spatiotemporal registration technology to jointly analyze multi-source monitoring data with the tower's 3D model. Specifically, the tower structure is divided into 3D mesh models of varying resolutions. High-precision meshes (10-50 mm) are used for areas prone to local deformation, such as flange connections and welds, while low-precision meshes (500-1000 mm) are used for regular structures, such as the tower body. Based on the spatial layout of the displacement sensors, the displacement data collected by each sensor is mapped to the corresponding grid cell. For example, the radial offset data from the laser ranging unit at the top of the tower is mapped to the Y-axis coordinate of the top grid cell, and the axial strain data from the fiber Bragg grating sensor is mapped to the Z-axis strain parameter of the grid cell in the section where it is located. This multi-resolution modeling and data mapping approach enables a coordinated analysis of the tower's overall deformation trends and local detailed features.

[0086] In terms of the dynamic interaction between physical measurement data and three-dimensional modeling data, the cloud processing platform establishes a two-way data transmission channel by configuring multi-source data interfaces (such as RESTAPI and WebSocket interfaces). On the one hand, the platform inputs the real-time monitored deformation data (such as the axial displacement, radial offset, and torsion angle of each section) into the three-dimensional modeling space through the interface, triggering the iterative update of the model parameters. For example, when the radial offset of a certain section is detected to exceed the design threshold, the platform automatically adjusts the stiffness parameters of the section in the three-dimensional model and recalculates the overall force distribution of the tower to reflect the performance changes caused by material fatigue or structural damage; on the other hand, the theoretical parameters of the three-dimensional modeling space (such as design load, material elastic modulus, and preset deformation threshold) are fed back to the data perception layer through the interface to optimize the sensor layout and dynamic calibration parameters. For example, based on the simulation results of the three-dimensional model, it is determined that the sensor deployment density needs to be increased in high-stress areas, or the temperature compensation coefficient of the dynamic calibration device needs to be adjusted to match the temperature-deformation relationship predicted by the model. During the interaction process, the system automatically completes the dynamic conversion of the coordinate system. Through the coordinate transformation matrix (translation, rotation, scaling), the measurement data in the local coordinate system of the sensor is converted into the absolute coordinates in the global coordinate system of the tower. At the same time, the theoretical coordinates of the three-dimensional model are converted into local coordinates that can be recognized by the sensor to ensure the consistency of the data in different spaces.

[0087] During the data processing process, the cloud processing platform also adopts a multimodal data fusion strategy to perform multi-level fusion of displacement data, inclination data, and environmental load data (such as wind speed and temperature). In the data layer fusion stage, various types of data are integrated into a unified data set through spatiotemporal registration and format standardization; in the feature layer fusion stage, dimensionality reduction algorithms such as principal component analysis (PCA) are used to extract core features that characterize tower deformation (such as the main displacement direction and the maximum strain area) from multidimensional data; in the decision layer fusion stage, structural design specifications and historical operation data are combined to generate comprehensive assessment results (such as verticality deviation level and deformation trend warning). This multimodal fusion method can fully utilize the complementarity of different types of data to improve the accuracy and reliability of monitoring results.

[0088] The communication control layer, through a three-level architecture consisting of edge computing nodes, protocol converters, and a cloud-based processing platform, enables full-process control from data acquisition to model analysis. Edge computing nodes address the real-time preprocessing and reliable transmission of near-end data, protocol converters eliminate format barriers for heterogeneous data, and the cloud-based processing platform leverages powerful computing and modeling capabilities to enable deep data mining and dynamic interaction. This layer not only ensures the spatiotemporal consistency and physical accuracy of multi-source data but also provides efficient data support for model construction and dynamic weighting algorithms in the verticality analysis layer. It is a key component in achieving intelligent, real-time performance in the entire monitoring system. Its design fully considers the complexity of hybrid wind turbine tower monitoring scenarios. Through redundant transmission links, adaptive compression algorithms, and multimodal data fusion, it effectively addresses challenges such as large data volumes, demanding real-time transmission, and strong environmental interference, laying a solid foundation for accurate tower verticality monitoring and safety assessment.

[0089] Example 3:

[0090] The verticality analysis layer dynamically calibrates and accurately characterizes the tower's three-dimensional deformation by constructing a holographically mapped spatial posture model of the tower. Its core process involves parameter correlation between physical and digital spaces, model construction, parameter optimization, and interaction mechanism design. This layer utilizes multi-dimensional modeling and data fusion techniques to transform actual monitoring data into a digital twin model suitable for deformation analysis, providing foundational support for subsequent dynamic weighting algorithms.

[0091] The system first establishes a parameter association channel and a bidirectional calibration mechanism between the physical structural deformation and the digital modeling space. Raw data from the physical measurement space (such as axial strain collected by displacement sensors and yaw angles from inclinometers) is converted into regular grid data using spatial interpolation algorithms (such as inverse distance weighted interpolation and kriging interpolation) to match the three-dimensional grid structure of the digital modeling space. The coordinate conversion phase utilizes a seven-parameter transformation method. Using translation parameters (ΔX, ΔY, ΔZ), rotation parameters (ωx, ωy, ωz), and a scale parameter (k), the measurement data in the sensor's local coordinate system is converted into absolute coordinates in the tower's global coordinate system, ensuring consistency with the spatial reference of the digital model. The bidirectional calibration mechanism allows theoretical parameters from the digital modeling space (such as design loads and Poisson's ratio of materials) to be transferred back to the physical measurement space. For example, when the digital model simulation indicates that stress in a certain area exceeds the material's yield strength, the system automatically adjusts the sampling frequency or measurement range of the sensors in that area to obtain more intensive monitoring data.

[0092] Based on basic data such as axial displacement, radial offset, and torsion angle acquired by the displacement sensor array, the system constructs a baseline pose library and a deformation feature library for the tower structure. The baseline pose library stores the ideal geometric parameters of the tower in its unloaded state, including the design coordinates of each section, perpendicularity tolerance, and material elastic modulus, serving as a baseline reference for deformation analysis. The deformation feature library, based on historical monitoring data, records deformation patterns under different operating conditions (such as average wind speed of 5 m / s, strong wind speed of 25 m / s, and earthquake conditions), including displacement distribution curves, strain peak locations, and torsion angle variation ranges. For example, under strong wind conditions, the deformation feature library records the nonlinear relationship between radial offset and wind speed at the tower top, as well as the axial strain distribution patterns at each segment interface. The baseline pose library and deformation feature library are iteratively updated using real-time monitoring data. The system employs an incremental learning algorithm, automatically updating the corresponding statistical parameters (such as mean and variance) in the library with each new data acquisition, ensuring that the model always reflects the latest state of the tower.

[0093] During the model construction process, the system uses displacement sensor data as a basis, combined with structural design parameters (such as tower wall thickness and flange connection stiffness), to generate an initial three-dimensional deformation model of the tower through finite element analysis. The model adopts a hybrid modeling method of shell units and beam units. The tower body uses shell units to simulate thin-walled structures, and the flange connection nodes use a combination of beam units and spring units to simulate complex force characteristics. Subsequently, through spatial interpolation and coordinate transformation, the actual measurement data is mapped to the corresponding nodes of the model, forming a preliminary mapping relationship between the physical measurement space and the digital modeling space. For example, the axial strain value measured by the fiber grating sensor of a certain section is applied to the corresponding unit of the model as a boundary condition, driving the model to calculate the theoretical displacement value of the section, which is compared with the measured value of the laser ranging unit to verify the initial accuracy of the model.

[0094] The parameter optimization phase uses a residual correction algorithm to dynamically compensate the model output results. The specific steps are as follows: First, the real-time monitored deformation data (such as the measured radial offset of the tower center at a certain moment) is compared with the predicted value of the digital model to calculate the residual (measured value - predicted value); then, the spatial distribution of the residual is fitted using the least squares method to generate a residual correction matrix, which contains the residual correction coefficients for each model node; finally, the residual correction matrix is ​​embedded in the model calculation process to compensate subsequent prediction results node by node. For example, if the model predicts a radial offset of 20mm for a node, but the measured value is 22mm, the correction coefficient for that node is set to +2mm, and the calculated result for that node is automatically increased by 2mm in subsequent predictions. In addition, the system uses a sliding spatial window mechanism to block the continuous monitoring area. The data blocks within each spatial window (such as a cylindrical area with a height of 5 meters and a circumference of 30 degrees) are independently feature extracted. By calculating parameters such as the displacement gradient and strain mean of the area, a correlation matrix between the data block features and the structural load is established. For example, when the displacement gradient within a certain spatial window suddenly increases, the correlation matrix indicates that there may be local load anomalies in this area, which requires further analysis.

[0095] The tower spatial posture model includes four major modules: physical measurement space, digital modeling space, structural database and interaction mechanism. The physical measurement space, serving as the data source, collects the tower's raw deformation characteristics in real time through sensors and continuously inputs them into the model in the form of a data stream. The digital modeling space uses multi-dimensional geometric modeling (such as 3D meshes and stress contours) to visualize the tower's posture and output deformation predictions. The structural database integrates design parameters with real-time monitoring information. The material property library stores data such as steel grade, yield strength, and elastic modulus; the load distribution library records historical wind speed and temperature loads; and the connection node library contains parameters such as flange bolt torque and weld strength. The interaction mechanism enables data integration between modules through standardized interfaces. The physical measurement space and the structural database utilize APIs for parameter acquisition and model updates. For example, the structural database regularly obtains sensor calibration parameters from the physical measurement space and updates the temperature correction coefficients in the material property library. Real-time parameter transfer between the physical measurement space and the digital modeling space occurs through data channels, ensuring synchronization between measurement data and model calculations. Information exchange between the digital modeling space and the structural database is achieved through middleware. For example, when calculating stress distribution, the digital model retrieves material elastic modulus and load distribution data from the structural database, generates stress contours, and then feeds these back to the structural database for storage.

[0096] When a new deformation pattern is detected (such as an unrecorded torsion angle combination), the system triggers the model topology reconstruction mechanism. The specific process is as follows: first, by comparing the data block features with the correlation matrix, the spatial window corresponding to the abnormal deformation pattern is identified; then, the model mesh within the spatial window is encrypted (such as reducing the mesh size from 100mm to 20mm), and the unit type is re-divided (such as converting shell units to solid units); finally, the stress and strain distribution of the area is recalculated based on the encrypted mesh, the deformation feature library is updated, and the model parameters are adjusted. This dynamic reconstruction mechanism enables the model to adapt to the complex changes of the tower structure and improve the ability to capture sudden deformations.

[0097] The verticality analysis layer, through the aforementioned process, constructs a holographically mapped tower spatial posture model, achieving precise alignment between the physical and digital worlds. This model not only reflects the tower's current deformation state in real time but also continuously improves prediction accuracy through parameter iteration and dynamic compensation mechanisms. This provides a reliable spatial benchmark and data support for deformation trend prediction, benchmark deviation correction, and verticality anomaly alerts based on dynamic weighting algorithms. Its design fully integrates spatial geometry analysis, digital twins, and real-time data processing technologies, addressing the limited adaptability of traditional monitoring models to complex operating conditions and providing an intelligent analysis tool for the safe operation of hybrid wind turbine towers.

[0098] Example 4:

[0099] The verticality analysis layer, based on the tower's spatial posture model, uses a dynamic weighted algorithm to predict deformation trends, correct baseline deviations, and generate verticality anomaly alerts. Its core process encompasses data sample construction, hybrid model training, deviation feature processing, and the application of state recognition algorithms. Through multi-algorithm fusion and intelligent analysis, this layer provides decision support for real-time tower verticality assessment and fault warning.

[0100] Deformation trend prediction constructs a sample set based on historical data. The system extracts historical operating attitude data (such as axial displacement, radial offset, and torsion angle at each moment), environmental load records (such as wind speed, wind direction, and temperature at the corresponding moment), and deformation event characteristic data (such as the operating parameters when the historical maximum offset occurred) from the tower's spatial attitude model, forming a deformation sample set containing multi-dimensional information in time, space, and the environment. To improve model training efficiency, the sample set undergoes feature enhancement processing, including data normalization (mapping displacement values ​​to the [-1, 1] interval), missing value filling (using linear interpolation between adjacent moments), and outlier removal (based on the 3σ principle). The processed samples are divided into a training set (70%) and a test set (30%), used for model training and performance verification, respectively.

[0101] The hybrid prediction model adopts the ARIMA-LSTM-Prophet architecture, combining the advantages of three algorithms to achieve multi-scale trend analysis. Key parameters are set during the model initialization phase, such as the difference order of the ARIMA model (d=1), the number of hidden layer neurons in the LSTM network (128), and the seasonal cycle parameters of the Prophet algorithm (annual cycle, monthly cycle, and daily cycle). During the training process, the ARIMA model first decomposes the time series data of the training set into trend terms, separating long-term trend terms (such as the increasing trend of axial displacement of the tower due to foundation settlement), seasonal cycle terms (such as the thermal expansion and contraction cycle caused by temperature changes), and random fluctuation terms; the LSTM network receives the decomposed random fluctuation terms and extracts nonlinear features in the time series data through a gating mechanism (such as the mutation pattern of displacement under strong wind conditions); the Prophet algorithm models the long-term trend terms and seasonal cycle terms, identifying periodic patterns in the data (such as the periodic changes in radial offset caused by the daily temperature difference between morning and evening). The outputs of the three algorithms are fused through an adaptive weighting mechanism, with the weight coefficients dynamically adjusted based on the current operating conditions. For example, in strong wind conditions, the LSTM network is given a higher weight because it excels at capturing nonlinear features. Under normal operating conditions, the weights of the three algorithms tend to be balanced. The training process uses the Boosting method to generate training sequences for multiple base models. The prediction performance of each base model in different time intervals is evaluated through sliding window verification (window size of 100 samples). The gradient descent algorithm is used to optimize the dynamic fusion coefficients of the base models until the model fit (such as the root mean square error (RMSE)) reaches a set threshold (such as ≤5mm) or the predetermined number of training rounds (such as 500 rounds) is completed. After training is completed, the test set is input into the hybrid model, and comprehensive performance indicators (such as RMSE and mean absolute error (MAE)) are calculated. The model with the highest prediction accuracy is selected as the optimal deformation prediction model. Based on the spatial distribution trend of tower deformation output by the model (such as the predicted value sequence of radial offset of each section in the next 48 hours), combined with the material mechanical parameters (such as steel elastic modulus and allowable stress), the deviation evolution path is simulated through finite element analysis method to determine the development trend of verticality deviation and potential risk areas.

[0102] The baseline deviation correction process begins with feature extraction and noise reduction. The system extracts real-time parameters of axial deviation, radial offset, and torsional deformation from the tower's spatial posture model, constructing a feature vector containing multidimensional deviation values ​​(e.g., a vector at a certain moment is [ΔZ = 30 mm, ΔR = 25 mm, θ = 0.8°]). Because the monitoring data may be contaminated by environmental noise, independent component analysis (ICA) is used to reduce noise on the feature vectors. By maximizing non-Gaussianity, the core deviation feature components are isolated, while irrelevant components such as sensor noise and electromagnetic interference are removed. Subsequently, a deviation classification model is established based on the Gaussian mixture model (GMM). The Bayesian Information Criterion (BIC) is used to determine the optimal number of categories (e.g., classifying deviations into "mild," "moderate," and "severe"). Each category corresponds to a different deviation range and physical meaning—for example, "mild" deviation corresponds to normal deformation during daily operation, while "severe" deviation may indicate structural damage. The classified deviation features are then fed into a pre-defined correction strategy library, which contains a mapping table for correction methods such as axial deviation and prestress adjustment, and radial offset and counterweight optimization. Taking radial offset as an example, the system matches the weight and installation position of the counterweight according to the size of the offset: when the offset is ≤50mm, the first-level counterweight solution is triggered (adding a 50kg adjustable mass block); when the offset is >50mm and ≤100mm, the second-level solution is triggered (adding a 100kg mass block and adjusting the position). The correction parameters (adjustment amplitude, action position, correction timing) are searched for the optimal combination through a genetic algorithm. The algorithm uses the minimization of residual deviation as the objective function, generates multiple generations of candidate solutions through selection, crossover, and mutation operations, and finally determines the optimal parameter combination. A dual closed-loop control mechanism is adopted in the correction process: the inner loop calculates the residual deviation through real-time monitoring data and evaluates the correction effect; the outer loop adjusts the genetic algorithm parameters according to the feedback results of the inner loop. If the residual deviation exceeds the threshold (such as 10mm), the parameter re-optimization process is triggered until the deviation is controlled within a safe range.

[0103] The verticality anomaly alarm utilizes a convolutional neural network (CNN) model to implement state pattern recognition. The system first collects structural vibration characteristics (such as acceleration time history curves and frequency response functions), wind speed load spectra, and foundation settlement data to construct an operational state feature library containing samples of normal and abnormal conditions. To improve the model's ability to capture time series features, the data in the feature library is segmented into overlapping segments. For example, a continuous one-hour vibration data set is segmented into 500 one-minute samples with a 50% overlap, generating a spatially distributed sample set containing time series information. The CNN model architecture consists of an input layer, a convolutional layer, a pooling layer, and a classification layer. The input layer receives normalized sample data (with dimensions of 1×60×3, corresponding to the time point and feature dimension, respectively). The convolutional layer uses a 3×3 convolution kernel to extract local features (such as sudden changes in vibration amplitude within a time window) and introduces nonlinearity through the ReLU activation function. The pooling layer uses a max pooling operation (with a pooling window of 2×2) to reduce feature dimensionality and retain key information. The classification layer outputs probabilities of "normal operation" and "verticality anomaly" using the Softmax function. Model training uses a backpropagation algorithm to optimize weight parameters, improving classification accuracy by minimizing the cross-entropy loss function. When the model's predicted probability of "vertical anomaly" for a particular sample exceeds a threshold (e.g., 90%), the tower is deemed abnormal. The system automatically triggers an alarm mechanism, sending warnings to maintenance personnel via audio and visual signals and text messages. The system also records the time, location, and characteristic parameters of the anomaly (e.g., wind speed of 28 m / s and radial offset of 80 mm at the top at the time of the anomaly), providing detailed records for fault diagnosis.

[0104] The dynamic weighted algorithm achieves multi-level analysis of tower deformation through multi-model fusion and hierarchical processing: a hybrid prediction model predicts trends based on temporal and spatial features, a baseline deviation correction process provides closed-loop control for real-time deviations, and a CNN model identifies abnormal patterns from vibration and load data. This algorithmic system fully utilizes the multi-dimensional data of the tower's spatial posture model. By combining data-driven and physical models, it improves the monitoring system's intelligence and decision-making reliability, enabling timely identification of potential risks and providing targeted correction strategies, providing strong technical support for the safe operation of hybrid wind turbine towers.

[0105] Example 5:

[0106] When constructing the tower spatial posture model for holographic mapping, the verticality analysis layer uses a sliding spatial window mechanism and an online learning algorithm to achieve dynamic optimization and adaptive updating of the model. The specific implementation is as follows:

[0107] The system uses a sliding spatial window mechanism to perform block processing on the continuous monitoring area of ​​the tower, dividing the tower into multiple vertical cylindrical areas along the height direction, and each cylindrical area is further divided into several fan-shaped spatial windows on the horizontal section (for example, every 2 meters in the height direction is a cylindrical area, and every 45 degrees in the horizontal section is a fan-shaped spatial window). Each spatial window corresponds to a local area of ​​the tower, such as a specific orientation of a certain section of the tower. For the monitoring data in each spatial window (such as the axial displacement and radial offset data of the displacement sensor array in the area), the system independently performs feature extraction and calculates parameters such as the displacement gradient, strain mean, and torsion angle change rate of the area. For example, within a certain spatial window, the uniformity of the deformation of the area is evaluated by calculating the standard deviation of the radial offset data. If the standard deviation suddenly increases, it indicates that there may be local damage or load abnormality in the area.

[0108] Based on feature extraction, the system establishes an association matrix between data block features and structural loads. The rows of the association matrix represent different spatial windows, and the columns represent load types (such as static loads, wind loads, temperature loads) and corresponding characteristic parameters (such as wind speed, temperature change, and axial force). Through historical data statistics, the deformation distribution patterns of each spatial window under different working conditions (such as the startup phase, full-power operation phase, and shutdown maintenance phase) are recorded. For example, under full-power operation and a wind speed of 15m / s, the association matrix records that the average radial offset of a spatial window in the middle of the tower is 30mm, and the average strain is 80μe, which corresponds to the wind speed load at the same moment. This association matrix provides a data basis for the online learning of the model, enabling the system to identify typical deformation patterns under different load conditions.

[0109] The online learning algorithm optimizes the posture model through an incremental update mechanism. When new monitoring data is input, the system first assigns the data to the corresponding spatial window, calculates the real-time feature parameters of the spatial window, and then compares them with the historical patterns recorded in the association matrix. If the real-time feature matches a historical pattern (e.g., with a similarity exceeding 85%), the model parameters are adjusted according to the preset update rules. For example, the stiffness parameters of the mesh elements corresponding to the spatial window are updated to reflect the gradual changes in material properties. If the real-time feature does not match any historical patterns (i.e., an unrecorded deformation pattern is detected, such as an abnormally large torsion angle of a spatial window at low wind speeds), the model topology reconstruction mechanism is triggered. The topology reconstruction process includes: refining the model mesh in the area where the spatial window is located (e.g., reducing the original mesh size from 200mm to 50mm), increasing the number of mesh elements to improve local modeling accuracy; redefining the mesh element type (e.g., converting shell elements to solid elements to more accurately simulate complex loads); recalculating the stress and strain distribution based on the refining mesh, and recording the new deformation pattern in the association matrix to expand the scope of the model.

[0110] During the iteration of model parameters, the system dynamically compensates the output results through the residual correction algorithm, and combines the characteristic analysis results of the sliding spatial window to achieve fine-tuning of the model. For example, if the residual between the measured value of the real-time radial offset of a spatial window and the model prediction value continues to exceed 10mm, the system will determine that there is a deviation in the model parameters in this area, and gradually reduce the residual by adjusting the elastic modulus parameters of the material corresponding to the spatial window. In addition, basic data sets such as the material property library and load distribution library in the structural database are also synchronized and adjusted according to the characteristic updates of the spatial window. For example, when a certain area has multiple abnormal deformations, the material property library automatically marks the steel in the area as likely to have fatigue damage, adjusts its elastic modulus and yield strength values, and provides more accurate parameter support for subsequent model calculations.

[0111] The interaction mechanism plays a key role in spatial window processing and model updating. The characteristic parameters of the spatial window are transmitted in real time between the physical measurement space and the digital modeling space via a data channel, ensuring that the digital model can promptly reflect the latest status of the physical structure. The digital modeling space and the structural database interact through middleware. For example, when the model topology reconstruction requires new material parameters, the middleware retrieves the relevant data from the structural database and feeds the new parameters back to the database storage after the model update is completed. The physical measurement space and the structural database implement parameter collection and model updates through standardized interfaces. For example, the structural database regularly obtains the calibration parameters of each spatial window from the physical measurement space to update the environmental load and deformation correlation model in the load distribution library.

[0112] By combining a sliding spatial window mechanism with an online learning algorithm, the system achieves refined local analysis of tower deformation and dynamic optimization of the overall model. This design not only improves the model's adaptability to complex working conditions but also promptly captures abnormal changes in the tower structure, providing a more accurate model foundation for subsequent functions such as deformation trend prediction and baseline deviation correction. Furthermore, the model topology reconstruction mechanism ensures the system's autonomous evolution in response to new deformation patterns, avoiding monitoring blind spots caused by model rigidity, thereby enhancing the reliability and intelligence of the entire monitoring system.

[0113] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0114] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A real-time monitoring system for the verticality of a hybrid wind turbine tower, characterized in that: It includes: Data perception layer, communication control layer and verticality analysis layer; The data perception layer includes a displacement sensor array, an inclinometer, a dynamic calibration device and a reference positioning module, which are used to perform multi-point synchronous measurement of tower structure deformation, reference coordinate calibration and dynamic calibration compensation operations based on feedback instructions; The communication control layer includes edge computing nodes, protocol converters, and a cloud processing platform. It also deploys a time synchronization mechanism and a multimodal data fusion strategy to establish redundant transmission links, achieve time alignment and spatial registration of multi-source monitoring data, and use an adaptive compression algorithm to extract features and transmit hierarchical data according to data accuracy requirements and transmission bandwidth. It also implements dynamic interaction between physical measurement data and 3D modeling data by configuring a multi-source data interface. The verticality analysis layer is used to dynamically calibrate the three-dimensional deformation model of the tower using spatial geometric analysis technology combined with structural design parameters and real-time monitoring data sets, and then construct a holographic mapping tower spatial posture model. Based on this posture model, a dynamic weighted algorithm is used to predict deformation trends, correct benchmark deviations, and issue verticality anomaly alarms.

2. A real-time monitoring system for verticality of a hybrid wind turbine tower according to claim 1, characterized in that: The displacement sensor array includes at least a laser ranging unit, a fiber grating sensor, an ultrasonic echo detector and a millimeter wave radar, and is used to obtain the displacement distribution characteristics of the tower cross section; The dynamic calibration device includes at least a temperature compensation module, a vibration suppressor, a coordinate correction unit and a data verifier, which are used to eliminate the environmental errors of the measurement system; the reference positioning module includes at least a Beidou positioning terminal, a total station reference point, and an inertial navigation component, which is used to execute the coordinate calibration instructions issued by the verticality resolution layer.

3. A real-time monitoring system for verticality of a hybrid wind turbine tower according to claim 1, characterized in that: In the communication control layer, the data collected by the displacement sensor array is transmitted by the edge computing node after being timestamped to the protocol converter. The protocol converter then performs data format standardization processing and transmits it together with the spatial posture information recorded by the inclinometer to the cloud processing platform for modeling and calculation. The data processing uses spatiotemporal registration technology to input the collected displacement data into three-dimensional grid models of different resolutions for joint analysis. The method of realizing dynamic interaction between physical measurement data and three-dimensional modeling data by configuring a multi-source data interface includes: realizing dynamic interaction between physical measurement data and three-dimensional modeling data by configuring a multi-source data interface, exchanging real-time monitoring data of tower deformation, and jointly analyzing the acquired spatial coordinates, realizing dynamic conversion of coordinate systems, completing data verification, model updating, and parameter distribution, and issuing calibration control instructions to the data perception layer.

4. A real-time monitoring system for verticality of a hybrid wind turbine tower according to claim 1, characterized in that: The construction of the holographically mapped tower spatial posture model comprises: Establish a parameter correlation channel and bidirectional calibration mechanism between physical structure deformation and digital modeling space; The actual measurement data is spatially interpolated and transformed into coordinates. Based on the axial displacement, radial offset, and torsion angle obtained by the displacement sensor array, a reference posture library and deformation feature library of the tower structure are constructed. Model parameter iteration and dynamic characteristic simulation are performed based on real-time monitoring data to convert the actual structural deformation into a high-precision digital twin model. Optimize the parameters of the spatial posture model, input the real-time monitored deformation data into the established posture model, and use the residual correction algorithm to dynamically compensate the output of the model to obtain the optimized tower spatial posture model; The tower spatial posture model includes a physical measurement space, a digital modeling space, a structural database, and an interaction mechanism between modules; The physical measurement space is the data source of the posture model and contains the original deformation characteristics of the tower structure; the digital modeling space forms a mapping relationship with the physical measurement space, and geometrically represents the tower posture characteristics through multi-dimensional space modeling; the structural database integrates design parameters and real-time monitoring information, and provides a basic data set including a material property library, a load distribution library, and a connection node library; the interaction mechanism realizes data communication between modules, parameter collection and model update are realized between the physical measurement space and the structural database through a standardized interface, parameter transfer is carried out between the physical measurement space and the digital modeling space through a data channel, and information interaction is realized between the digital modeling space and the structural database through middleware.

5. The real-time monitoring system for verticality of a hybrid wind turbine tower according to claim 1, characterized in that: The deformation trend prediction using a dynamic weighted algorithm based on the posture model includes: Based on the tower spatial posture model, historical operation posture data, environmental load records, and deformation event feature data are obtained to construct a deformation sample set; After feature enhancement processing, the deformation sample set is divided into a training set and a test set; Establish an ARIMA-LSTM-Prophet hybrid forecasting architecture, set model initialization parameters, input the training set into the hybrid model for joint training, use the ARIMA model to perform trend decomposition, use the LSTM network to extract time series features, and use the Prophet algorithm to identify periodic patterns. Use an adaptive weighting mechanism to balance the prediction errors of different algorithms under specific working conditions until the model fit reaches the set threshold or the scheduled training rounds are completed; Input the test set into the trained hybrid model, calculate the comprehensive performance index of the model, and select the optimal deformation prediction model; Based on the optimal deformation prediction model, the spatial distribution trend of the tower deformation is output, and the verticality deviation evolution path is determined in combination with the material mechanical parameters. The training process of the hybrid prediction architecture includes: The Boosting method is used to generate training sequences for multiple base models, and the prediction performance of each base model is evaluated through sliding window verification. The gradient descent algorithm is used to calculate the dynamic fusion coefficient of each base model, and the weighted average mechanism is adopted to comprehensively predict the output results of the base models.

6. A real-time monitoring system for verticality of a hybrid wind turbine tower according to claim 1, characterized in that: The method of correcting the reference deviation using a dynamic weighted algorithm based on the posture model includes: Based on the tower spatial posture model, the axial deviation, radial offset, and torsional deformation parameters of the structural deformation are extracted, and a deviation feature vector set is constructed. Independent component analysis is used to reduce the noise of the deviation feature vector to obtain the core deviation feature component; A deviation classification model based on Gaussian mixture model was established, and the optimal number of classifications was determined by the BIC criterion; The classified deviation features are input into the preset correction strategy library to match the optimal correction scheme and generate a directional benchmark correction parameter set.

7. A real-time monitoring system for verticality of a hybrid wind turbine tower according to claim 1, characterized in that: The verticality abnormality alarm is performed using a dynamic weighted algorithm based on the posture model, including: Based on the tower spatial posture model, the structural vibration characteristics, wind speed load spectrum, and foundation settlement data are collected to build an operating status feature library; Perform overlapping segmentation processing on the data in the operating status feature library to generate a spatial distribution sample set; Establish a convolutional neural network model, set the convolution kernel size and pooling parameters, and obtain the deep expression of spatial features through backpropagation calculation; The spatial features are input into the classifier for state pattern recognition, and the discrimination results between normal operation state and verticality abnormal state are output.

8. A method and system for real-time monitoring of verticality of a hybrid wind turbine tower according to claim 1, characterized in that: The construction of the holographically mapped tower spatial posture model further includes: The sliding spatial window mechanism is used to divide the continuous monitoring area into blocks, and the data blocks in each spatial window are independently subjected to feature extraction; Establish a correlation matrix between data block features and structural loads, record the deformation distribution patterns corresponding to different working conditions, incrementally update the posture model through an online learning algorithm, and trigger the model topology reconstruction mechanism when an unrecorded deformation pattern is detected.

9. The real-time monitoring system for verticality of a hybrid wind turbine tower according to claim 1, characterized in that: The method for generating the benchmark correction parameter set includes: Establish a correspondence table between deviation types and correction methods, including correction methods for axial deviation corresponding to prestress adjustment and radial offset corresponding to counterweight optimization; A genetic algorithm is used to search for the optimal correction parameter combination, including adjustment amplitude, action position and correction timing. The correction effect is evaluated in real time through a double closed-loop control mechanism. When the residual deviation exceeds the threshold, the parameter re-optimization process is triggered.

10. A real-time monitoring method for the verticality of a hybrid wind turbine tower, characterized in that: The following steps are involved: The tower structure is synchronously collected at multiple points using a displacement sensor array and an inclinometer to obtain axial displacement, radial offset, and torsion angle measurements. Based on the coordinate calibration results of the reference positioning module, a dynamic calibration device is used to compensate for temperature drift and eliminate vibration interference. The collected heterogeneous monitoring data is timestamped through edge computing nodes, and a multimodal data fusion strategy is used to implement time synchronization and spatial registration of displacement data and inclination data. An adaptive compression algorithm is used to extract displacement distribution characteristics based on transmission bandwidth constraints, and a protocol converter is used to convert physical measurement data into a standardized data format compatible with the 3D modeling space. The tower structure design parameters are loaded into the cloud processing platform, and the pre-built three-dimensional deformation model is dynamically calibrated in combination with the real-time monitoring data set. The residual correction algorithm is used to optimize the model parameters and generate a holographic mapping tower spatial posture model. Based on this model, the axial deformation gradient, radial offset trend and torsion angle change are fused through a dynamic weighted algorithm to output the deformation trend prediction results, benchmark deviation correction amount and verticality anomaly alarm signal.

Images