A method for identifying dynamic stress field of steel-concrete transition zone of hybrid tower wind turbine based on digital-analog cooperative driving

Abstract

The application discloses a kind of based on digital-analog collaborative driving's dynamic stress field identification method of steel-concrete transition zone of mixed tower wind turbine unit, comprising: creating macroscopic wind turbine cabin mixed tower foundation coupling dynamics numerical analysis model, the calculated displacement of steel-concrete transition zone of mixed tower wind turbine unit multibody dynamics analysis model under the action load of shutdown operating condition is obtained;In the steel-concrete connection transition section of mixed tower wind turbine tower drum, the complementary layout of displacement sensor and accelerometer of LVDT electromagnetic induction principle is designed, and the predicted displacement is obtained;The contact stiffness and interface friction coefficient of the contact element of steel-concrete transition zone microscopic slip and debonding behavior of wind turbine multibody dynamics analysis model are inverted, and the dynamic stress field of steel-concrete transition zone under the action of tower top limit load is calculated.The application realizes the cross-scale perception of steel-concrete interface complex stress field and the real-time early warning of hidden damage by heterogeneous sensor network, nonlinear multibody dynamics modeling and data model two-way coupling mechanism.

Description

Technical Field

[0001] This invention relates to the field of structural health monitoring and safety assessment of wind turbine generator sets, specifically to a method for identifying the dynamic stress field in the steel-concrete transition zone of a hybrid tower wind turbine generator set based on digital-analog collaborative driving. Background Technology

[0002] Concrete-steel composite towers for wind turbines are widely used due to their high load-bearing capacity and economy. However, the steel-concrete transition section is prone to interface slippage, debonding, and stress concentration due to abrupt changes in stiffness, alternating loads, and differences in material properties. Existing technologies suffer from the following problems:

[0003] 1. Monitoring limitations: Traditional local strain gauge methods cannot capture the discontinuous slip stress field at the steel-concrete interface;

[0004] 2. Model distortion: Conventional finite element models ignore microscale contact nonlinear behavior, resulting in excessive stress prediction errors;

[0005] 3. Insufficient real-time performance: Full-condition numerical simulation takes too long and makes it difficult to achieve rapid feedback of dynamic stress field. Summary of the Invention

[0006] To address the three major technical bottlenecks in monitoring the interface slippage and dynamic stress field of the steel-concrete transition section of hybrid tower wind turbines—namely, monitoring blind spots, modeling distortion, and computational inefficiency—this invention provides a method for identifying the dynamic stress field in the steel-concrete transition zone of hybrid tower wind turbines based on a digital-model collaborative driving approach. Through a heterogeneous sensor network, nonlinear multibody dynamics modeling, and a bidirectional coupling mechanism with the data model, this method achieves cross-scale perception of the complex stress field at the steel-concrete interface and real-time early warning of hidden damage. It is applicable to dynamic stress monitoring, damage identification, and life assessment of the steel-concrete connection transition section of large hybrid tower wind turbines.

[0007] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows:

[0008] A method for identifying the dynamic stress field in the steel-concrete transition zone of a hybrid tower wind turbine based on digital-analog collaborative driving, the method comprising the following steps:

[0009] S1. Construct a multibody dynamic model of the hybrid tower with nonlinear interface contact, embed contact elements of micro-slippage and debonding behavior in the steel-concrete transition zone, and create a coupled dynamic numerical analysis model of the macroscopic wind turbine nacelle hybrid tower foundation. Calculate the displacement of the steel-concrete transition zone of the multibody dynamic analysis model of the hybrid tower wind turbine under the load of shutdown conditions.

[0010] S2, in the steel-concrete connection transition section of the hybrid tower wind turbine, a complementary layout of displacement sensor and accelerometer based on LVDT electromagnetic induction principle is designed to synchronously collect real-time signals of multi-source data from displacement monitoring and vibration monitoring, and optimize the time synchronization strategy of multi-source sensors; the digital-analog collaborative driving mechanism captures and corrects the dynamic stress field of the steel-concrete transition zone of the hybrid tower, preprocesses the acceleration to convert it into displacement, and fuses it with the displacement data monitored by the LVDT displacement sensor to obtain the predicted displacement.

[0011] S3. By comparing the predicted displacement and the calculated displacement, the contact stiffness and interface friction coefficient of the contact element in the steel-concrete transition zone of the multibody dynamics analysis model of the wind turbine are inverted to determine the micro-slippage and debonding behavior. Based on the inverted contact stiffness, the dynamic stress field of the steel-concrete transition zone under the action of the ultimate load at the top of the tower is calculated using the multibody dynamics analysis model of the hybrid tower wind turbine considering different working conditions and wind conditions.

[0012] Step S1 further includes:

[0013] A multibody dynamics analysis model of a wind turbine, including wind turbine blades, nacelle, steel-concrete tower, and foundation, was established using the ABAQUS / Explicit module. Rigid connections or spring damping systems were adopted between the blade-nacelle, nacelle-tower, tower connection section, and tower-to-foundation. The blades were connected by MPC, with the connection point located at the hub center. The hub weight and the weight of the nacelle and drive train were considered as point masses. The wind turbine, consisting of three blades, was connected to the tower top by MPC at the hub center. A multibody dynamics model of a hybrid tower wind turbine considering geometric nonlinearity was established.

[0014] Contact elements representing the microscopic slip and debonding behavior of the steel-concrete transition zone are embedded in the steel-concrete transition section, and microscopic slip and debonding elements implemented by custom user subroutines are embedded in the steel-concrete interface. Their constitutive equations are coupled with the normal crushing criterion and the tangential friction energy dissipation mechanism to define the contact relationship.

[0015] Furthermore, in step S2, designing a complementary layout of displacement sensors and accelerometers based on the LVDT electromagnetic induction principle in the steel-concrete connection transition section of the hybrid wind turbine tower includes the following steps:

[0016] LVDT displacement sensors and triaxial accelerometers are arranged in the steel-concrete connection transition zone of the hybrid tower wind turbine tower, using a circumferential orthogonal staggered layout. A multi-channel synchronous acquisition device is used as the core controller, and all sensor signal lines are uniformly connected to the device to receive its output TTL trigger pulses, ensuring that the synchronization delay of each channel is less than the preset delay threshold. The sensor cables use equal-length wiring and are equipped with RC delay matching circuits to compensate for transmission time difference to the 0.1μs level. The trigger signal is transmitted using twisted-pair shielded wire, with the shielding layer grounded at a single point and combined with a ferrite magnetic ring filter to suppress interference.

[0017] Furthermore, in step S2, the arrangement rules for the LVDT displacement sensor and accelerometer are as follows:

[0018] The LVDT displacement sensor is installed on the outer surface of the concrete 15cm axially from the lower end of the steel section flange, and is distributed circumferentially at the two orthogonal diameter endpoints. The positioning direction of the orthogonal diameter is deviated by 45° from the main wind direction axis of the tower, so that the LVDT displacement sensor covers the maximum deformation zone under the combined action of aerodynamic load and gravity. The accelerometer is arranged at symmetrical points on the inner side of the steel section, forming an inner and outer ring interlaced monitoring network with the LVDT displacement sensor. The sensor point coordinates are matched with the preset coupling node group in the ABAQUS finite element model.

[0019] Furthermore, the LVDT displacement sensor and accelerometer are integrated into a single package housing, and the installation angle is adjusted by a universal joint to adapt to the curvature of the flange surface.

[0020] Furthermore, in step S2, the process of capturing and correcting the dynamic stress field of the steel-concrete transition zone of the hybrid tower using a digital-analog collaborative driving mechanism, preprocessing the acceleration to convert it into displacement, and fusing it with the displacement data monitored by the LVDT displacement sensor to obtain the predicted displacement includes the following steps:

[0021] Considering the monitoring data of the hybrid-tower wind turbine under shutdown conditions where it is not generating electricity, the acceleration signal is processed by double integration. A displacement field reconstruction method based on adaptive weighted fusion is adopted, which integrates the input acceleration with the displacement u. acc (t) and LVDT measured displacement u LVDT (t) Perform GPS clock synchronization to ensure that the accelerometer and LVDT displacement sensor time scales are aligned before data fusion. This fusion is used to generate the predicted time-domain displacement field value from the LVDT measured displacement. The fusion formula is as follows:

[0022] u pred (t)=w1·[u LVDT,0° (t)+u LVDT,180° [(t)] / 2+w2·u acc (t)

[0023] The weighting strategy is as follows: when the vibration frequency is less than 10Hz, the weight ω1 corresponding to LVDT is larger; when the vibration frequency is greater than or equal to 10Hz, the weight ω2 corresponding to the acceleration dynamic response is larger; if the LVDT signal is lost at a certain moment, linear extrapolation is performed based on the acceleration integral displacement of the adjacent 4-second window.

[0024] Furthermore, in step S2, the process of optimizing the time synchronization strategy for multiple sensor sources includes the following steps:

[0025] Redundant signal calibration technology is used for a limited number of sensor data. In the preprocessing stage, multiple equivalent extended load conditions are input, covering a yaw angle range of ±120°. An interpolation fitting factor library is generated to compensate for the modal truncation error caused by low-density point placement.

[0026] Further, in step S3, the predicted displacement and the calculated displacement are compared. The predicted displacement is used as the target for comparison with the ABAQUS / Explicit model. Through parameter optimization, the interface contact stiffness and friction coefficient of the contact element in the steel-concrete transition zone micro-slip and debonding behavior are inverted. Specifically, this includes the following steps:

[0027] Establish an error optimization function:

[0028]

[0029] Where u pred To predict displacement, u FEM Calculate the displacement for ABAQUS, where t is the time interval, t = 1, 2, ..., T;

[0030] The interface contact stiffness K is iteratively updated using a gradient optimization algorithm. n And the friction coefficient μ, the convergence condition is ΔJ / J < 1%; where ΔJ represents the relative percentage change of the error function J in two adjacent iterations, that is:

[0031]

[0032] For missing node displacement data, the complete displacement field is reconstructed using radial basis function interpolation based on adjacent LVDT monitoring values. (k) and J (k+1) These are the error functions for the k-th and (k+1)-th iterations, respectively.

[0033] Step S3 further includes:

[0034] Based on the modified model, a multi-condition wind load spectrum is applied, and the stress tensor of the steel-concrete interface element is extracted and mapped to the global coordinate system.

[0035] Time-varying Mises stress cloud maps and interface slip distributions are generated using Field Output, and out-of-limit regions are marked.

[0036] Further, in step S3, the time-varying Mises stress cloud map is a time-varying equivalent stress distribution cloud map of each unit in the steel-concrete transition zone, and a red warning area is marked where the stress exceeds 80% of the steel yield strength.

[0037] The interface slip distribution includes dynamic safety margin analysis; wherein, the interface slip volumetric damage index is D. s =∑s p / sc , where s c To calibrate the critical slip threshold, s p The value represents the plastic slip; the dynamic safety margin analysis is based on the safety margin of each element calculated using field superposition techniques. Where σ yield σ is the yield strength of steel. eq The equivalent stress of the unit.

[0038] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0039] The present invention provides a method for identifying the dynamic stress field in the steel-concrete transition zone of a hybrid tower wind turbine based on digital-analog collaborative driving. This method achieves accurate monitoring of the dynamic stress field at the steel-concrete interface of the hybrid tower, fundamentally solving the safety risks caused by monitoring blind spots, modeling distortion, and computational inefficiency in the prior art. Testing has demonstrated the following technical advantages of this invention:

[0040] (1) Improved modeling accuracy: The micro-contact model reduces the slip prediction error from 30% to 5%;

[0041] (2) Breakthrough in monitoring efficiency: The number of sensors is reduced by 80%, and the computing efficiency is improved to the real-time level (response delay <30 seconds);

[0042] (3) Full coverage of engineering applicability: compatible with 5MW to 15MW units, with an error of less than 8% for both onshore and offshore operating conditions. Attached Figure Description

[0043] Figure 1 This is a flowchart of the method for identifying the dynamic stress field in the steel-concrete transition zone of a hybrid tower wind turbine based on digital-analog collaborative driving according to the present invention.

[0044] Figure 2 This is a numerical analysis model diagram of the coupled dynamics of a hybrid tower wind turbine.

[0045] Figure 3 This is a numerical model diagram of the steel-concrete transition zone; where (a) represents the detailed drawing of segment 46, (b) represents the prestressing layout diagram of segments 1 to 46, and (c) represents the fully assembled model diagram of the concrete tower.

[0046] Figure 4 This is a schematic diagram of the sensor layout in the steel-concrete transition section; where (a) represents the front view of the sensor layout in the transition section, and (b) represents the top view of the sensor layout in the transition section.

[0047] Figure 5 This is a calculated displacement diagram of the steel-concrete transition zone under the load of the shutdown condition;

[0048] Figure 6 This is a schematic diagram comparing and approximating the predicted displacement with the calculated displacement. Detailed Implementation

[0049] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0050] This invention discloses a method for identifying the dynamic stress field in the steel-concrete transition zone of a hybrid tower wind turbine based on digital-analog collaborative driving. The method includes the following steps:

[0051] S1. Construct a multibody dynamic model of the hybrid tower with nonlinear interface contact, embed contact elements of micro-slippage and debonding behavior in the steel-concrete transition zone, and create a coupled dynamic numerical analysis model of the macroscopic wind turbine nacelle hybrid tower foundation. Calculate the displacement of the steel-concrete transition zone of the multibody dynamic analysis model of the hybrid tower wind turbine under the load of shutdown conditions.

[0052] S2, in the steel-concrete connection transition section of the hybrid tower wind turbine, a complementary layout of displacement sensor and accelerometer based on LVDT electromagnetic induction principle is designed to synchronously collect real-time signals of multi-source data from displacement monitoring and vibration monitoring, and optimize the time synchronization strategy of multi-source sensors; the digital-analog collaborative driving mechanism captures and corrects the dynamic stress field of the steel-concrete transition zone of the hybrid tower, preprocesses the acceleration to convert it into displacement, and fuses it with the displacement data monitored by the LVDT displacement sensor to obtain the predicted displacement.

[0053] S3. By comparing the predicted displacement and the calculated displacement, the contact stiffness and interface friction coefficient of the contact element in the steel-concrete transition zone of the multibody dynamics analysis model of the wind turbine are inverted to determine the micro-slippage and debonding behavior. Based on the inverted contact stiffness, the dynamic stress field of the steel-concrete transition zone under the action of the ultimate load at the top of the tower is calculated using the multibody dynamics analysis model of the hybrid tower wind turbine considering different working conditions and wind conditions.

[0054] The specific process of this invention is as follows: Figure 1 As shown.

[0055] Step 1: First, an ABAQUS multibody dynamics model of the hybrid tower is constructed, specifically including the wind turbine blades, nacelle, steel-concrete tower, and foundation. The connections between components in the model are clearly defined. The blades and nacelle are connected to the hub center via MPCs. A spring-damping system is installed at the junction of the nacelle and tower. The steel-concrete transition zone between tower sections uses a rigid connection. Fixed constraints are applied to the tower base to simulate the foundation boundary conditions. Addressing the deficiency of traditional models that neglect micro-contact nonlinearity, this step embeds micro-slip-debonding contact elements and defines a dynamic stiffness reduction criterion for hard contact superposition in their normal behavior, expressed as:

[0056]

[0057] In the formula K n0 The initial interfacial contact stiffness is given by α, which is the crushing damage factor; the tangential behavior is based on a friction model, introducing shear stress τ.

[0058]

[0059] In the formula, μ is the interfacial friction coefficient to be inverted, and γ p This represents the cumulative amount of plastic slip. During implementation, the wind load spectrum under shutdown conditions (according to load generation rules) is applied, and the calculated displacement of the steel-concrete transition zone is output for subsequent parameter inversion (corresponding to...). Figure 5 (Verification data). This step requires overall element mesh generation, selection of appropriate element mesh type, and refinement of interface mesh size; determination of aerodynamic loads based on IEC Class IIA wind spectrum generated by Bladed, mapping to the equivalent bending moment at the top of the tower; application of gravity field and boundary constraints (such as support fixing constraints).

[0060] Multi-field coupled dynamic modeling:

[0061] Assembly Model: Create a fully assembled model of the hybrid tower in ABAQUS. Detailed drawings of the steel-concrete transition section are shown below. Figure 3 As shown in (a) above, the prestressing layout diagram is visible. Figure 3 (b) In the diagram, the fully assembled model of the hybrid tower is detailed below. Figure 3 In (c), the modal synthesis method is used to reduce the number of degrees of freedom to 5% of the original model, while retaining 99% of the effective modal energy; boundary loads: dynamic wind spectrum loads (10-minute time history) are generated according to IEC 614001 standard and mapped to the tower top center point through the substructure coupling method, reducing the aerodynamic load calculation time by 45%.

[0062] Step 2: Sensor deployment and signal synchronization to resolve monitoring blind spots. A front view of the sensor deployment in the transition section is shown below. Figure 4 As shown in (a) above, the top view is as follows: Figure 4 As shown in (b) in the figure. The LVDT sensor is installed at the outer surface of the concrete 15cm below the flange of the steel section (at the 0° and 180° positions), in the axial direction, to cover the main deformation direction; the triaxial accelerometer is installed at the bolt holes on the inner wall of the steel section (at the 90° and 270° positions), in the radial direction, to form an inner and outer ring interlaced monitoring network with the LVDT.

[0063] (2) Signal synchronization must ensure time synchronization: GPS timing module is used to ensure that the clock reference of all sensors is consistent and the synchronization error is less than 50ns, eliminating time delay interference; the data acquisition card adopts rising edge triggering mode with a delay of less than 50ns to ensure phase consistency.

[0064] (3) Preprocessing

[0065] ①Acceleration signal noise reduction:

[0066] Joint filtering: First, noise and useful information in the signal are separated by empirical mode decomposition, and then wavelet threshold filtering is performed on the high-frequency components.

[0067] Filtering effect: Signal-to-noise ratio improved to 60dB, displacement integral error less than 0.02mm.

[0068] ②LVDT drift correction:

[0069] Baseline correction: A piecewise polynomial fitting method is used to eliminate low-frequency baseline drift.

[0070] Dynamic adjustment: The correction window is adaptively adjusted according to the real-time signal characteristics.

[0071] Step 3: Data fusion and model parameter inversion to address the problem of insufficient data-driven generalization ability.

[0072] (1) The purpose of data fusion is to combine the dynamic displacement of the accelerometer with the measured displacement of the LVDT to generate a high-precision full-field displacement prediction value.

[0073] ①Acceleration signal processing:

[0074] Baseline correction: The baseline of the acceleration signal is fitted using the least squares method to eliminate low-frequency drift.

[0075] Frequency domain integration: Perform a fast Fourier transform on the acceleration signal, divide it twice by the frequency operator, and then perform an inverse fast Fourier transform to generate displacement data.

[0076] ②LVDT signal processing:

[0077] Baseline correction: Low-frequency baseline drift is eliminated by piecewise polynomial fitting with a sliding window length of 5 seconds.

[0078] ③ Displacement field fusion:

[0079] Weighted average: Combining acceleration integral displacement and LVDT measured displacement, the weights are dynamically adjusted: Low frequency band (<10Hz): LVDT weight is higher (0.8), acceleration weight is lower (0.2); High frequency band (≥10Hz): Acceleration weight is higher (0.7), LVDT weight is lower (0.3).

[0080] Combined LVDT direct measurement value u LVDT (t) and acceleration double integral displacement u acc (t), the full-field displacement prediction formula is generated by the following formula:

[0081]

[0082] The system is dynamically adjusted based on the real-time vibration frequency to ensure that the root mean square error of the displacement after fusion is less than 0.05 mm, thus guaranteeing the accuracy requirements.

[0083] (2) Model parameter inversion

[0084] The goal is to compare the predicted displacement with the simulated calculated displacement, such as Figure 6 As shown, the contact stiffness and friction coefficient in the optimization model are optimized.

[0085] Inversion process:

[0086] ① Initialization parameters: Set the initial contact stiffness K n The range and initial value of the friction coefficient μ.

[0087] ② Transient simulation: Run transient dynamic simulation using ABAQUS / Explicit to generate simulated displacement u under shutdown conditions. FEM (K n ,μ,t), such as Figure 5 As shown.

[0088] ③ Displacement comparison calculation and fusion displacement u pred (t) and simulated displacement u FEM (K n Error of μ,t):

[0089] Error = u pred (t)-u FEM (K n ,μ,t);

[0090] With the goal of minimizing the error, define the optimization function as follows:

[0091]

[0092] ④ Parameter Optimization: An improved Levenberg-Marquardt algorithm is used to iteratively update the contact stiffness and friction coefficient. Finite difference automatic step size adjustment is introduced into the Jacobian matrix calculation, with an initial step size of 1%, reduced to 0.1% after convergence. After 6 iterations, the residual decrease rate is less than 1%, and the optimal parameters are output.

[0093] ⑤ Missing data processing: For displacement data of missing nodes, the complete displacement field is reconstructed by radial basis function interpolation based on adjacent LVDT monitoring values.

[0094] Step 4: Dynamic Stress Field Reconstruction and Output

[0095] 4.1 Rapid Calculation of Stress Tensor

[0096] Loading under different load conditions: Apply multi-load wind load spectrum (IEC614001 standard turbulence model) to the modified model, extract the stress tensor of the steel-concrete interface element and map it to the global coordinate system;

[0097] Stress field output: Time-varying Mises stress cloud map and interface slip distribution are generated through Field Output, and the over-limit area is marked (stress threshold is 80% of the steel yield strength).

[0098] 4.2 Real-time Early Warning Rule Generation

[0099] Yellow alert: Triggered when the cumulative amount of interface sliding exceeds 50% of the critical sliding amount.

[0100] Red Alert: Stop the machine immediately when the cumulative amount of interface slip exceeds 90% of the critical slip or the local stress ratio exceeds 80% of the yield strength of the steel.

[0101] Cloud computing acceleration: By deploying GPU parallel solvers on wind farm edge servers, the dynamic stress field update frequency reaches 10Hz, and the early warning delay is less than 30 seconds.

[0102] Step 5: Model self-calibration and long-term reliability assurance to solve the problem of time-varying effects in practical engineering.

[0103] Dynamic calibration of health status begins with benchmark experiments. Specifically, a static load test (500kN radial pressure) is conducted on the tower every quarter. The friction coefficient μ and stiffness reduction factor are then corrected based on the measured displacement. Then, the automatic learning and updating continues. Specifically, by combining historical operation and maintenance data, a Bayesian network is used to update the initial values ​​of parameters, so that the model can adapt to the material aging effect.

[0104] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0105] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for identifying the dynamic stress field of the steel-concrete transition zone of a hybrid tower wind turbine based on digital-analog collaborative driving, characterized in that, The method comprises the following steps: S1, a multi-body dynamics model of a mixed tower containing an interface contact nonlinearity is constructed, a contact element of micro-slip and debonding behavior of a steel-concrete transition zone is embedded, a coupled dynamics numerical analysis model of a macro wind turbine cabin mixed tower foundation is created, and a calculated displacement of the steel-concrete transition zone of the multi-body dynamics analysis model of the mixed tower wind turbine under the action of a load in a shutdown working condition is obtained; S2, a complementary layout of a displacement sensor based on an LVDT electromagnetic induction principle and an accelerometer is designed in a steel-concrete connection transition section of a mixed tower wind turbine tower, real-time signals of multi-source data of displacement monitoring and vibration monitoring are synchronously collected, a time synchronization strategy of the multi-source sensor is optimized, a dynamic stress field of the steel-concrete transition zone of the mixed tower is captured and corrected by a digital-analog collaborative driving mechanism, the acceleration is converted into displacement, and the predicted displacement is obtained by fusing the displacement data monitored by the LVDT displacement sensor; S3, the predicted displacement is compared with the calculated displacement, the contact stiffness and the interface friction coefficient of the contact element of the micro-slip and debonding behavior of the steel-concrete transition zone of the multi-body dynamics analysis model of the wind turbine are inverted, and the dynamic stress field of the steel-concrete transition zone under the action of the ultimate load at the top of the tower under different working conditions and wind conditions is calculated based on the mixed tower wind turbine multi-body dynamics analysis model after the inversion of the contact stiffness.

2. The method of claim 1, wherein the method is characterized by, Step S1 further comprises: An ABAQUS / Explicit module is used to establish a wind turbine multi-body dynamics analysis model containing a wind blade, a cabin, a steel-concrete tower drum and a foundation, rigid connection or a spring damping system is used between the blade-cabin, the cabin-tower drum, the tower drum connection section and the tower drum and the foundation, the blade is connected by MPC, the connection point is located at the hub center, the hub weight is considered as point mass, the weight of the cabin and the transmission chain is considered as point mass, a wind wheel composed of three blades is connected with the tower top by MPC, and a mixed tower wind turbine multi-body dynamics model considering geometric nonlinearity is established; A contact element of micro-slip and debonding behavior of a steel-concrete transition zone is embedded in a steel-concrete transition section, a micro-slip debonding element realized by a user-defined subroutine is embedded in a steel-concrete interface, a constitutive equation of the element is coupled with a normal crushing criterion and a tangential friction energy dissipation mechanism to define a contact relationship.

3. The method of claim 1, wherein the method is characterized by: In step S2, the complementary layout of the LVDT displacement sensor based on the electromagnetic induction principle and the accelerometer in the steel-concrete connection transition section of the mixed tower wind turbine tower comprises the following steps: An LVDT displacement sensor and a three-axis accelerometer are arranged in the steel-concrete connection transition zone of the mixed tower wind turbine tower drum, a ring-shaped orthogonal staggered layout is adopted, a multi-channel synchronous acquisition device is used as a core controller, all sensor signal lines are uniformly connected to the device and receive the TTL trigger pulse output by the device, and it is ensured that the synchronization delay of each channel is less than a preset delay threshold; the sensor cable is laid with equal length and is additionally provided with an RC delay matching circuit to compensate the transmission time difference to the level of 0.1 microseconds; the trigger signal is transmitted by a twisted pair shielded line, the shielding layer is single-point grounded, and an iron oxide magnetic ring filter is combined to suppress interference.

4. The method of claim 1, wherein the method is characterized by, In step S2, the arrangement rule of the LVDT displacement sensor and the accelerometer is: The LVDT displacement sensor is installed on the outer surface of the concrete 15 cm below the lower end of the steel segment flange, and is distributed in the ring direction at two orthogonal diameter endpoints. The positioning direction of the orthogonal diameter is deflected by 45° from the axis of the tower main wind direction, so that the LVDT displacement sensor covers the maximum deformation area under the combined action of aerodynamic load and gravity. The accelerometers are arranged symmetrically inside the steel segment, forming an inner-outer circle staggered monitoring network with the LVDT displacement sensor. The sensor point coordinates are matched with the pre-set coupling node group in the ABAQUS finite element model.

5. The method of claim 1 or 4, wherein, The LVDT displacement sensor and the accelerometer are integrated in an integrated packaging shell, and the installation angle is adjusted through a universal joint to adapt to the curvature of the flange surface.

6. The method of claim 1, wherein the method further comprises: In step S2, the digital-analog cooperative driving mechanism captures and corrects the dynamic stress field of the steel-concrete transition zone of the tower, pre-processes the acceleration to convert it into displacement, and fuses the displacement data obtained by the LVDT displacement sensor to obtain the predicted displacement process, which includes the following steps: Considering the monitoring data of the shutdown condition of the mixed tower wind turbine, the double integral processing of the acceleration signal is carried out, the displacement field reconstruction method based on adaptive weighted fusion is adopted, and the input acceleration integral displacement and the measured displacement of LVDT GPS clock synchronization is performed to ensure that the accelerometer and the LVDT displacement sensor are time-aligned, and then the data fusion is performed to generate the time-domain displacement field prediction value by fusing the measured displacement of LVDT; wherein the fusion formula is: ； The weight distribution strategy is: when the vibration frequency is less than 10 Hz, the weight corresponding to the LVDT is greater ; when the vibration frequency is greater than or equal to 10 Hz, the weight corresponding to the acceleration dynamic response is greater ; if the LVDT signal is lost at a certain moment, the linear extrapolation is performed based on the acceleration integral displacement of the adjacent 4-second window.

7. The method of claim 1, wherein the method further comprises: In step S2, the process of optimizing the time synchronization strategy of the multi-source sensor includes the following steps: For a limited number of sensor data, a redundant signal calibration technology is used. In the pre-processing stage, input equivalent multiple sets of extended load working conditions, which cover a yaw angle of ±120°, generate an interpolation fitting factor library, and compensate for the modal truncation error caused by low-density point distribution.

8. The method of claim 1, wherein the method is characterized by: In step S3, by comparing the predicted displacement and the calculated displacement, the predicted displacement is taken as the target and compared with the ABAQUS / Explicit model, and the interface contact stiffness and friction coefficient of the contact element of the steel-concrete transition zone micro-slip and debonding behavior are inverted through parameter optimization; Specifically includes the following steps: An error optimization function is established: ； wherein is the predicted displacement, is the ABAQUS calculated displacement, t is the time, ; The interface contact stiffness is iteratively updated using a gradient optimization algorithm and the coefficient of friction , the convergence condition being ; wherein ΔJ represents the relative percentage change in the error function J between two successive iterations, i.e.: ； For the missing node displacement data, the complete displacement field is reconstructed by radial basis function interpolation based on the neighboring LVDT monitoring values, and are the error functions of the kth and k+1th iterations, respectively.

9. The method of claim 1, wherein the method is characterized by, Step S3 further includes: Based on the corrected model, apply multi-condition wind load spectrum, extract stress tensor of steel-concrete interface element and map to global coordinate system; Generate time-varying Mises stress cloud and interface slip distribution through Field Output, and mark the over-limit area.

10. The method of claim 9, wherein the method is characterized by: In step S3, the time-varying Mises stress cloud is a time-varying equivalent stress distribution cloud of each element in the steel-concrete transition zone, and a red warning area is marked when the stress exceeds 80% of the yield strength of steel. The interface slip amount distribution comprises an interface slip system accumulated damage index and dynamic safety margin analysis; wherein the interface slip system accumulated damage index is wherein is a calibrated critical slip threshold value, is a plastic slip amount; the dynamic safety margin analysis is a safety margin of each unit calculated based on field superposition technology wherein is a yield strength of steel, is an equivalent stress of the unit.

Images