Nonlinear Constraint Simulation System and Method for Dynamic Load Balancing of Wind Turbine Blades under Variable Operating Conditions

Abstract

This invention discloses a nonlinear constraint simulation system and method for dynamic load balance of wind turbine blades under varying operating conditions, relating to the field of dynamic load technology for wind turbine blades under varying operating conditions. Specifically, it includes the following steps: extracting the stress response changes of blade nodes within each directional reversal segment based on intermediate simulation data; constructing the mapping relationship of residual disturbance effects in the structural path; generating a directional reversal trace distribution map for spatial offset analysis of the reversing structure; performing trajectory fitting on the residual disturbance in the directional reversal trace distribution map; extracting the spatial offset of the trajectory in the axial direction of the path; and calculating the trajectory deflection index and duration weight function based on the duration of the disturbance effect within the reversal segment; and determining whether to perform a breakpoint reconstruction operation based on the reconstruction judgment model jointly constructed by the two. This invention solves the problem of reconstructing multi-segment reversal disturbance paths, achieving adaptive correction of the simulation path and accurate modeling of stress transfer.

Description

Technical Field

[0001] This invention relates to the field of dynamic load technology for wind turbine blades under varying operating conditions, and specifically to a simulation system and method for nonlinear constraint balancing dynamic loads of wind turbine blades under varying operating conditions. Background Technology

[0002] In wind power systems, wind turbine blades, as crucial components directly subjected to wind loads, have a critical impact on the overall safety and efficiency of the turbine under complex wind conditions. In actual operation, wind turbine blades face the combined effects of various variable operating conditions, including structural deviations of the turbine itself (such as blade zero-degree error, aerodynamic shape changes, pitch control errors, and drivetrain alignment deviations) and environmental disturbances (such as wind shear, tower shadow effect, and turbulence). To accurately grasp the true aerodynamic state of the blades during operation, it is necessary to construct a simulation model that integrates multiple variables, coupling mechanisms, and constraints to systematically simulate the dynamic response of wind turbine blades under variable operating conditions. The so-called nonlinear constraint simulation of dynamic load balance under variable operating conditions for wind turbine blades refers to establishing a mathematical model considering nonlinear coupling constraints (including material nonlinearity, geometric nonlinearity, and external disturbances) based on dynamic wind condition changes and blade structural response mechanisms. By simulating changes in blade tip clearance, aerodynamic shape changes, and load transfer characteristics, the simulation modeling and computational analysis of the aerodynamic balance behavior of wind turbine blades under dynamic load conditions are achieved. This type of simulation system can combine wind turbine operation monitoring data to build a hybrid simulation architecture that integrates data-driven and physical model constraints. It is used to realistically reflect the mechanical and aerodynamic coupling state of wind turbine blades under complex conditions and to provide a theoretical basis for operational stability assessment and structural optimization.

[0003] Existing nonlinear constraint simulation techniques for dynamic load balance of wind turbine blades under varying operating conditions typically achieve simulation objectives by establishing multi-physics coupled models of the wind turbine blade under different operating states. This type of technology uses wind conditions as the driving parameter, combining blade structural parameters, aerodynamic characteristics, and operating condition information. First, an aerodynamic simulation module simulates the instantaneous aerodynamic forces acting on the blade under environmental variables such as wind speed, wind direction, and turbulence intensity. Then, the aerodynamic forces are input into a structural response analysis module, using the finite element method to solve for the blade's deformation, stress, and vibration response under dynamic loads. This process incorporates geometric nonlinearity, material nonlinearity, and boundary constraints to construct a complete nonlinear coupled simulation framework. Next, an aerodynamic-structural coupling feedback mechanism dynamically corrects for aerodynamic shape changes caused by blade deformation, achieving closed-loop linkage between aerodynamics and structure. Furthermore, a disturbance modeling module is included to simulate the effects of uncertain disturbances such as tower shadows and wind shear, further incorporating dynamic disturbance correction terms to improve simulation accuracy. Finally, the simulation calculates and outputs the blade's headroom changes, aerodynamic balance offset, and overall turbine response parameters under different operating conditions. The overall process covers multiple stages, including wind condition modeling, aerodynamic simulation, structural response analysis, nonlinear coupling solution, disturbance modeling and feedback correction, forming an integrated dynamic load balance simulation system for wind turbine blades driven by multi-source information and under nonlinear dynamic constraint control.

[0004] The existing technology has the following shortcomings:

[0005] In the simulation of dynamic loads under varying operating conditions for wind turbine blades, when multiple disturbances, such as wind shear disturbances and pitch command switching, are continuously superimposed on the blade's operating condition within the same simulation cycle, the load direction may reverse multiple times in a short period. In such cases, the system needs to accurately transmit this reverse-switching dynamic load information to the blade root for stress response calculation. Since simulation methods typically construct dynamic load solution paths based on continuous splicing of the operating condition trend direction, the system defaults to generating load path nodes in a single main direction, thus ignoring the consistency requirements of the path logic for reverse load switching. This leads to contradictions in the force transmission relationship between nodes in the reverse segment, resulting in path reversal conflicts. Furthermore, existing nonlinear constraint simulation technology for dynamic load balance under varying operating conditions for wind turbine blades cannot determine whether to reconstruct the dynamic load path based on the duration of load action under multiple directional reversals in the load path. This causes the simulation system to continue executing the solution using a logically flawed path structure, leading to load direction conflicts or breaks at intermediate nodes, resulting in problems such as solution non-convergence and boundary condition mismatch. This not only causes the blade stress results to deviate from the actual physical state, but also interferes with the judgment of the load distribution law of the wind turbine, misleads the structural design parameters and scheduling control strategies, and in severe cases will directly affect the credibility of the simulation system in high-precision structural verification and operation evaluation.

[0006] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0007] The purpose of this invention is to provide a nonlinear constraint simulation system and method for dynamic load balance of wind turbine blades under varying operating conditions, so as to solve the problems in the background art mentioned above.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions, specifically including the following steps:

[0009] The target simulation period is divided into multiple load action segments. A node sequence is generated according to the disturbance trend, and a connection relationship is established based on the load direction between nodes to construct the dynamic load transfer path.

[0010] Perform directional change analysis on the nodes in the dynamic load transmission path to identify whether there are two or more directional reversal segments, and mark the start and end positions of each directional reversal segment.

[0011] Given that two or more reversal sections have been identified, the stress response changes of blade nodes in each reversal section are extracted based on intermediate simulation data. The mapping relationship of residual disturbance effects in the structural path is constructed, and a reversal trace distribution map is generated for spatial offset analysis of the reversal structure.

[0012] Trajectory fitting is performed on the residual disturbance in the direction reversal trace distribution map to extract the spatial offset of the trajectory in the path axial direction. Combined with the duration of the disturbance in the reversal section, the trajectory deflection index and duration weight function are calculated. Based on the reconstruction judgment model composed of the two, it is determined whether to perform the breakpoint reconstruction operation.

[0013] When the trajectory deflection index meets the path breakpoint reconstruction conditions, breakpoints and mirror nodes are inserted in the direction reversal section, the path node arrangement order is adjusted, the path reconstruction operation is completed, and the trajectory deflection index and corresponding path information are written into the simulation dataset.

[0014] Preferably, the target simulation period is divided into multiple load action segments, a node sequence is generated according to the disturbance trend, and a connection relationship is established between the nodes based on the load direction to construct the dynamic load transfer path. Specifically, this includes the following steps:

[0015] Based on wind shear disturbance, pitch control commands and changes in blade operating status, the time series of disturbance behavior within the target simulation period is extracted.

[0016] Based on the changes in the frequency and amplitude of disturbances in the time series of disturbance behavior, the target simulation period is divided into multiple disturbance-dominated load action segments.

[0017] In each load application zone, nodes representing the location and direction of the disturbance input are generated sequentially, and arranged into a node sequence according to the temporal relationship of the disturbance influence between nodes.

[0018] Based on the load direction relationship between adjacent nodes, directional connections are established between nodes to represent the transmission path of dynamic load under the disturbed structure, and all nodes are sequentially combined to form the initial dynamic load transmission path.

[0019] Preferably, the direction change analysis of the nodes in the dynamic load transmission path is performed to identify whether there are two or more direction reversal segments, and the start and end positions of each direction reversal segment are marked. Specifically, this includes the following steps:

[0020] Extract the direction vectors between adjacent nodes in the dynamic load transfer path, and calculate the angle between the direction vectors of each pair of adjacent nodes;

[0021] Based on the angle calculation results, adjacent node pairs with an angle greater than 180 degrees are marked as having reversed direction, and their position indices are recorded in the path sequence.

[0022] Traverse the entire path direction sequence, group and cluster all marked direction reversal nodes, and identify whether there are two or more direction reversal segments.

[0023] For each direction reversal segment, extract the index positions of the starting and ending nodes of its continuous reversals, and establish a direction reversal segment boundary index table for subsequent path stability assessment and structural reconstruction judgment.

[0024] Preferably, given that two or more directional reversal sections have been identified, the stress response changes of blade nodes within each directional reversal section are extracted based on intermediate simulation data. A mapping relationship of residual disturbance effects in the structural path is constructed, and a directional reversal trace distribution map is generated for spatial offset analysis of the reversing structure. Specifically, this includes the following steps:

[0025] Given that two or more directional reversal sections have been identified, the blade node stress response time series in the intermediate simulation data is obtained, and the node subset corresponding to each directional reversal section is extracted from it.

[0026] For each subset of nodes in the direction reversal section, calculate the relative gradient value of stress change between nodes and mark the node index where the change amplitude is greater than a set threshold.

[0027] Each marked node is mapped one-to-one with its path position, and a disturbance residual mapping array is established in the path coordinate system.

[0028] The node positions in the disturbance residual mapping array are mapped to the stress gradient magnitude as a set of two-dimensional graphical trajectory points, and then connected sequentially in chronological order to form a trajectory curve;

[0029] The generated trajectory curve is projected along the path coordinate axis to construct a distribution map of direction reversal traces, and this distribution map is used as the input image data for the calculation of spatial offset of the reverse structure.

[0030] Preferably, a perturbation residual mapping array in the path coordinate system is established by mapping each marked node to its path position one-to-one, specifically as follows:

[0031] A unified path axis coordinate system is set up, and the three-dimensional position information of the nodes in each direction reversal section in the simulation space is mapped to the linear axial distance scale under the path coordinate system by projection transformation.

[0032] The stress gradient value of each marked node is used as the magnitude reference value, and a pair of data sets is formed with the corresponding axial distance of the path. A two-dimensional mapping array is established using the node number as the index.

[0033] For continuous data point segments in the two-dimensional mapping array with abrupt gradient changes, a smoothing function is set to perform gradient curve interpolation fitting to ensure the continuity and resolvability of the perturbation residual mapping on the path structure.

[0034] Preferably, the node positions in the disturbance residual mapping array are mapped to the stress gradient magnitudes as a set of two-dimensional graphical trajectory points, and then connected sequentially in chronological order to form a trajectory curve, specifically:

[0035] Using each data point in the perturbation residual mapping array as input, extract its path axial position and stress gradient magnitude as horizontal and vertical coordinate values ​​to generate a two-dimensional coordinate point set.

[0036] The generated trajectory point set is sorted by timestamp, and the point set index is rearranged according to the simulation time order of the nodes to ensure that the direction of the trajectory curve is consistent with the order of disturbance propagation.

[0037] Based on a time-ordered set of points, all point pairs are connected using a curve fitting method based on cubic spline interpolation to generate a continuous and smooth perturbation trajectory line, which is used as the input for subsequent trajectory offset calculation.

[0038] Preferably, the residual disturbance in the direction reversal trace distribution map is subjected to trajectory fitting, the spatial offset of the trajectory in the path axial direction is extracted, and the trajectory deflection index and duration weight function are calculated based on the duration of the disturbance in the reversal section. The reconstruction judgment model based on the two is used to determine whether to perform the breakpoint reconstruction operation. Specifically, the following steps are included:

[0039] Project each disturbance trajectory in the direction reversal trace distribution map along the axial direction, extract the maximum and average offset values ​​of the trajectory curve in the main axis direction of the path, and calculate the trajectory deflection index.

[0040] Extract the disturbance time series corresponding to the current direction reversal segment from the intermediate simulation data, calculate the time span from the start to the end of the disturbance segment, and construct the relative proportion function of the time span in the whole simulation cycle as the duration weight function;

[0041] After normalizing the trajectory deflection index and the duration weight function respectively, they are input into the two-parameter judgment formula to construct a joint evaluation expression. The weight ratio factor of the two is set in the form of exponential decay.

[0042] The output of the two-parameter evaluation expression is compared with the preset breakpoint reconstruction judgment threshold range. If the output is within the judgment threshold range, the current direction reversal segment is determined to trigger the breakpoint reconstruction logic.

[0043] Preferably, the trajectory deflection index is calculated as follows:

[0044] For each direction of the reverse trajectory curve, position sampling is performed in the axial coordinate of the path, the offset values ​​of all sampling points in the axial direction are extracted, and their mean and standard deviation are calculated.

[0045] The axial distance between the maximum offset point of the trajectory curve and the starting point of the trajectory is quantified and multiplied by the standard deviation ratio to obtain an offset weight index that reflects the degree of offset concentration and extreme value deviation.

[0046] The ratio of the offset weight index to the aforementioned standard deviation is used as the trajectory deflection index to quantify the disturbance dispersion and spatial deformation trend of the direction reversal segment in the path structure.

[0047] Preferably, the structure and application of the two-parameter decision model include the following steps:

[0048] Set a normalization factor for the trajectory deflection index and the duration weighting function, and process them into standardized parameter pairs with values ​​ranging from 0 to 1;

[0049] A two-input multiplicative coupled model is constructed based on adjustable parameters α and β, where α is multiplied by the trajectory deflection exponent and β is multiplied by the duration weight function. The sum of the two products is then input into the Sigmoid function to transform it into a continuous evaluation value.

[0050] The evaluation value is compared with the lower and upper limits of the set breakpoint reconstruction trigger interval. If the result falls within the range, the direction reversal segment is marked as a candidate segment for breakpoint reconstruction and the path reconstruction process begins.

[0051] Preferably, the nonlinear constraint simulation system for dynamic load balance of wind turbine blades under varying operating conditions includes a disturbance driving path construction module, a direction reversal identification and analysis module, a disturbance residual mapping extraction module, a dual-parameter determination and modeling module, and a path reconstruction execution and writing module.

[0052] The disturbance-driven path construction module divides the target simulation period into multiple load action segments, generates a node sequence according to the disturbance trend, and establishes a connection relationship between nodes based on the load direction to construct the dynamic load transfer path.

[0053] The direction reversal identification and analysis module analyzes the direction change of nodes in the dynamic load transmission path, identifies whether there are two or more direction reversal segments, and marks the start and end positions of each direction reversal segment.

[0054] The disturbance residual mapping extraction module, under the premise that two or more directional reversal sections have been identified, extracts the stress response changes of blade nodes in each directional reversal section based on intermediate simulation data, constructs the mapping relationship of disturbance residual influence in the structural path, and generates a directional reversal trace distribution map for spatial offset analysis of the reversing structure.

[0055] The dual-parameter judgment modeling module performs trajectory fitting on the residual disturbance in the direction reversal trace distribution map, extracts the spatial offset of the trajectory in the path axial direction, and calculates the trajectory deflection index and duration weight function based on the duration of the disturbance in the reversal section. Based on the reconstruction judgment model composed of the two, it determines whether to perform the breakpoint reconstruction operation.

[0056] The path reconstruction execution and writing module inserts breakpoints and mirror nodes in the direction reversal section when the trajectory deflection index meets the path breakpoint reconstruction conditions, adjusts the path node arrangement order, completes the path reconstruction operation, and writes the trajectory deflection index and corresponding path information into the simulation dataset.

[0057] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0058] 1. This invention subdivides the target simulation period into multiple disturbance-dominated load action segments and generates node sequences and directional connections based on disturbance trends, enabling the dynamic construction of dynamic load transfer paths that conform to the logic of multiple disturbance inputs. Especially under conditions of overlapping disturbances such as wind shear disturbances and pitch control commands, traditional simulation path construction methods typically assume a single load transfer direction, thus ignoring the reversal characteristics of actual physical disturbances. This invention, by accurately calibrating the reversal direction and constructing a multi-segment path reversal structure, achieves higher logical consistency in the construction of dynamic load paths, significantly reducing simulation errors caused by node direction conflicts within the path, and laying the foundation for high-precision blade simulation.

[0059] 2. This invention introduces a residual disturbance mapping mechanism and trajectory distribution map generation process. Utilizing the changing trends of node stress response in intermediate simulation data, stress gradient analysis and trajectory fitting are performed on the direction reversal segment. Compared to traditional methods that rely solely on path structure analysis to determine reconstruction, this method jointly models the spatial offset and temporal persistence of the disturbance residue, forming a trajectory deflection exponent and duration weighting function with high interpretability and quantifiable evaluation capabilities. This approach not only effectively avoids the subjectivity of path structure reconstruction but also improves the modeling depth and accuracy of the simulation system for complex disturbance responses, ensuring that the reversal disturbance "leaves a trace" in the path and is traceable, thus enhancing the physical consistency of the system.

[0060] 3. This invention constructs a dual-parameter judgment model that integrates the trajectory deflection index and the duration of the disturbance. It then uses exponential decay weighting and a Sigmoid function mapping to form a continuous evaluation expression, enabling dynamic determination of whether to perform path breakpoint reconstruction in the direction reversal section. This mechanism significantly enhances the system's adaptive adjustment capability to abnormal disturbance structures, achieving a dynamic balance between path continuity and physical rationality. Finally, by inserting breakpoints and mirror nodes in the reversal section and writing them into the simulation dataset, the closed-loop data transmission link and path rationality in the simulation process are ensured. This fundamentally solves problems such as node stress conflicts, boundary condition mismatches, and solution non-convergence caused by reverse disturbances, significantly enhancing the stability and reliability of the wind turbine blade simulation system in structural verification and operational evaluation. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0062] Figure 1 This is a flowchart illustrating the nonlinear constraint simulation system and method for dynamic load balance of wind turbine blades under varying operating conditions, as described in this invention.

[0063] Figure 2 This is a schematic diagram of the module of the nonlinear constraint simulation system and method for dynamic load balance of wind turbine blades under varying operating conditions according to the present invention. Detailed Implementation

[0064] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0065] This invention provides, for example Figure 1 The nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions, as shown, specifically includes the following steps:

[0066] The target simulation period is divided into multiple load action segments. A node sequence is generated according to the disturbance trend, and a connection relationship is established based on the load direction between nodes to construct the dynamic load transfer path.

[0067] In this embodiment, the target simulation period is divided into multiple load action segments, a node sequence is generated according to the disturbance trend, and a connection relationship is established based on the load direction between nodes to construct a dynamic load transfer path. Specifically, the steps include:

[0068] Based on wind shear disturbance, pitch control commands and changes in blade operating status, the time series of disturbance behavior within the target simulation period is extracted.

[0069] The extraction of disturbance behavior time series can be achieved by performing multi-dimensional data fusion and behavior extraction on historical monitoring data during wind turbine operation. Specifically, during the simulation initialization phase, wind shear disturbance records, pitch command execution data, and blade angular velocity and angular displacement change data are first retrieved. These data originate from wind farm sensors, control system logs, and blade sensor points, respectively. Subsequently, through timestamp-aligned synchronization processing, these three types of data are mapped onto a unified time axis, forming a complete multi-source disturbance input data set. Then, a disturbance event identification algorithm is used to automatically label the wind speed gradient change rate, pitch command edge jump points, and blade attitude change points, thereby constructing a disturbance behavior label set. These disturbance labels are further arranged in chronological order to eliminate overlapping interference and invalid intervals. Valid disturbance event intervals are filtered through preset disturbance persistence rules, ultimately generating a disturbance behavior time series covering the entire simulation cycle. The entire process is implemented based on algorithm-driven automatic extraction logic, requiring no manual intervention, and is suitable for batch data processing and structured disturbance modeling tasks in software environments.

[0070] Based on the changes in the frequency and amplitude of disturbances in the time series of disturbance behavior, the target simulation period is divided into multiple disturbance-dominated load action segments.

[0071] By constructing a disturbance feature window scanning mechanism combined with sliding statistical analysis, segmentation of disturbance behavior time series and division of load action zones can be achieved in software. Specifically, firstly, a disturbance analysis window based on a time axis sliding is set, and the disturbance occurrence frequency (number of events per unit time) and disturbance amplitude changes (such as wind shear rate of change and pitch angle displacement gradient) within the local time segment are extracted for each disturbance event in the disturbance behavior time series. Then, a statistical feature aggregation algorithm is used to normalize and score the disturbance features within continuous time slices, forming a disturbance intensity index sequence. This index sequence serves as the core basis for judging the changes in time segments. Subsequently, by setting a disturbance intensity threshold change boundary, when the index shows a significant jump or a stable trend change, it is marked as a time segment division point. Further, based on these division points, the entire simulation cycle is divided into multiple disturbance-dominated time segments, and each segment is labeled as a relatively independent load action unit for subsequent path node construction and reversal behavior analysis. This processing method is entirely executed by software algorithms, possessing repeatability, no human intervention characteristics, and is suitable for large-scale wind condition simulation data processing scenarios.

[0072] In each load application zone, nodes representing the location and direction of the disturbance input are generated sequentially, and arranged into a node sequence according to the temporal relationship of the disturbance influence between nodes.

[0073] By employing disturbance impact factor analysis and direction encoding strategies, nodes representing the location and direction of disturbance inputs can be automatically generated in the software, and a node sequence can be constructed based on the disturbance propagation timeline. Specifically, within each defined load action segment, the system first retrieves disturbance behavior records within the segment, identifies wind shear position change points, pitch control initiation angles, and blade response start points, using these disturbance occurrence points as spatial location input parameters for the nodes. Subsequently, based on the disturbance propagation path on the wind turbine blades and the corresponding time delay characteristics, combined with data such as wind direction change rate and pitch angle change direction, the directional trend of each disturbance's effect on the blades is calculated, and a direction vector label is assigned to each node. Next, a time-series sorting algorithm based on the disturbance sequence is used to sort the nodes according to the actual time sequence of the disturbance's impact, constructing a node sequence arranged along the load transmission path direction. This node sequence serves as the skeleton structure for subsequent path construction, ensuring the accurate propagation of load information in the simulation. The entire process relies on structured data recognition, vector construction, and temporal logic organization, and is executed automatically by the software without manual intervention, making it suitable for large-scale path model generation under high-frequency disturbances.

[0074] Based on the load direction relationship between adjacent nodes, directional connections are established between nodes to represent the transmission path of dynamic load under the disturbed structure, and all nodes are sequentially combined to form the initial dynamic load transmission path.

[0075] The system can implement directional connections between nodes and construct an initial dynamic load transfer path in the software by using a load direction consistency discrimination algorithm and a path chain splicing strategy. Specifically, the system first traverses the node sequence generated in the previous step, extracts the direction vector labels of each pair of adjacent nodes, and calculates the trend of the angle change between them to determine whether the load direction relationship between adjacent nodes has continuity or reversal characteristics. If the direction change is within a set threshold range, it is considered that the directions are consistent, and the software connects the two nodes with a directed edge, indicating that the load can be directly transferred in this segment; if the direction change exceeds the threshold but does not constitute a reverse conflict, a direction transition connection mark is introduced to indicate that there is a micro-directional drift in the load propagation. After completing the connection of all nodes, the system constructs a complete, directed, and ordered initial dynamic load transfer path according to the node time sequence and connection relationship order. This path not only retains the directional information of the disturbance influence, but also provides a structural basis for subsequent judgment of whether there is a directional conflict in the path. The entire process relies on vector analysis, graph structure splicing, and rule-driven logic judgment, and is fully automated in the software environment.

[0076] The reason for dividing the target simulation period into multiple load action segments, generating node sequences according to disturbance trends, and establishing connections between nodes based on load directions to construct dynamic load transfer paths is that, under varying operating conditions of wind turbine blades, disturbance factors such as wind shear changes and pitch control exhibit strong instability and temporal coupling characteristics. Different disturbances have independent start and end points and influence continuity on the time axis, directly affecting the blade's stress mode and path distribution. If the simulation period is not segmented and all disturbances are treated as a unified background, temporal confusion between disturbance behaviors can easily occur, leading to logically chaotic connections or force direction breaks in the simulation path, severely impacting the accuracy and continuity of load transfer during the simulation. However, by segmenting and constructing nodes, disturbance behaviors can be clearly located, and the specific transfer paths of disturbances in the blade structure can be extracted. This gives the dynamic load path structural directionality, staged nature, and response correspondence, providing a stable, controllable, and structurally reasonable load modeling foundation for subsequent judgment of path reversal and identification of force closure problems.

[0077] Perform directional change analysis on the nodes in the dynamic load transmission path to identify whether there are two or more directional reversal segments, and mark the start and end positions of each directional reversal segment.

[0078] In this embodiment, the direction change analysis of the nodes in the dynamic load transmission path is performed to identify whether there are two or more direction reversal segments, and the start and end positions of each direction reversal segment are marked. Specifically, the steps include:

[0079] Extract the direction vectors between adjacent nodes in the dynamic load transfer path, and calculate the angle between the direction vectors of each pair of adjacent nodes;

[0080] The extraction of direction vectors and calculation of angles between adjacent nodes in a dynamic load transfer path can be implemented in software using vector geometric analysis and data structure traversal algorithms. Specifically, after the load path is constructed, the system sequentially reads the spatial coordinates and direction attributes of each node in the path. The direction attributes can be determined by the direction of disturbance influence or the load propagation trend. Then, for each pair of adjacent nodes in the path, two direction vectors are constructed based on their direction attributes, and the angle between them is calculated using the vector dot product formula. To ensure accuracy, the system normalizes each vector to eliminate the influence of magnitude on the angle value and stores the angle values ​​between each pair of nodes as a direction change matrix for subsequent reversal segment identification and analysis. The entire process is executed through array loop traversal and mathematical operations, exhibiting high efficiency and repeatability. It is suitable for batch direction difference evaluation in large-scale node chains, ensuring that the directional logical relationships between all nodes in the path can be accurately extracted and quantified.

[0081] Based on the angle calculation results, adjacent node pairs with an angle greater than 180 degrees are marked as having reversed direction, and their position indices are recorded in the path sequence.

[0082] Automatic identification and indexing of direction-reversed node pairs can be achieved in the software through threshold comparison and sequence labeling mechanisms. Specifically, after calculating the angle between adjacent nodes in the dynamic load transfer path, the system screens the angle results one by one and sets an identification threshold for direction reversal, typically a directional change trend with an angle greater than 180 degrees. When the angle between a pair of adjacent nodes exceeds this threshold, it can be determined that the node pair exhibits direction reversal behavior in the path. The software marks the starting node position of the node pair as the "direction reversal start point" and the ending node position as the "direction reversal end point," and simultaneously writes the index position of the node pair into the path sequence index table. This table serves as the basic data structure for subsequent segment clustering and path reconstruction judgment, enabling precise location of direction reversal behavior in the path structure. This processing is completed at the data level through Boolean judgment and array indexing, possessing automatic identification, unbiased judgment, and high compatibility features, making it suitable for applications requiring rapid extraction of potential reversal logic in complex path structures.

[0083] Traverse the entire path direction sequence, group and cluster all marked direction reversal nodes, and identify whether there are two or more direction reversal segments.

[0084] Clustering analysis based on positional continuity and inversion density can be used in software to group and cluster labeled direction-reversed nodes and identify the number of direction-reversed segments. Specifically, the software first reads the path direction sequence that has been determined and labeled with angles, and extracts the index positions of all nodes marked as "direction-reversal start point" or "direction-reversal end point." Then, by constructing a sliding window or interval scanning mechanism, these node indices are traversed according to their order in the path, and the distance between adjacent inverted nodes is calculated. When the distance between multiple inverted nodes is within a set continuity threshold, such as when they are separated by no more than a preset number of nodes in the path structure, the system groups these nodes into a continuous direction-reversed segment. If two or more sets of inverted nodes satisfying the clustering conditions can be identified in the entire path, multiple direction-reversed segments are considered to exist. This process is fully automated through segment merging, positional clustering, and quantity determination algorithms, without relying on manual input. It can adapt to path structures with different perturbation frequencies and node distribution densities, ensuring the accuracy and robustness of direction-reversed segment division.

[0085] For each direction reversal segment, extract the index positions of the starting and ending nodes of its continuous reversals, and establish a direction reversal segment boundary index table for subsequent path stability assessment and structural reconstruction judgment.

[0086] By employing segment boundary positioning and index structure generation strategies, the software can extract the start and end node indices of each direction-reversed segment and establish a complete boundary index table for these segments. Specifically, after clustering the direction-reversed nodes, the system automatically identifies the position of the first direction-reversed node in each cluster as the start node index for that segment, and the position of the last direction-reversed node in that cluster as the end node index. Subsequently, the system writes the start and end indices of each direction-reversed segment into a memory-based boundary index table in the form of key-value pairs or a two-dimensional array, adding a unique identifier to each pair for subsequent fast retrieval. This boundary index table not only accurately describes the logical boundaries of each direction-reversed segment in the path structure but also provides a basis for judgment in the next step of path reconstruction, such as determining whether a breakpoint insertion or mirror rearrangement is needed for the reversed segment. The entire process is automatically extracted and generated by the software based on the clustering results, without relying on manual rule configuration. It possesses high versatility and path structure adaptability, effectively supporting the dynamic path reconstruction needs in complex simulation environments.

[0087] The reason for analyzing the directional changes of nodes in the dynamic load transfer path, identifying the existence of two or more directional reversal segments, and marking the start and end positions of each directional reversal segment is that when wind turbine blades experience complex variable operating conditions, such as wind shear combined with pitch switching, multiple reversals of load direction can easily occur within a short period. If these reversals are not clearly identified and their boundaries are not located, the simulation system will assume the load path is a continuous structure in a single direction during the solution process, ignoring the load flow fracture and force chain reconstruction requirements caused by direction switching, leading to logical errors or non-convergence in the subsequent stress transfer model. By accurately analyzing directional changes and locating the start and end nodes of the reversal segments, a clear data foundation can be provided for subsequent path structural stability assessment, trajectory offset calculation, and breakpoint reconstruction judgment. This ensures that the dynamic load simulation path remains reasonable and accurate under multiple disturbance coupling conditions, effectively supporting the physical reliability and engineering applicability of the nonlinear simulation results.

[0088] Given that two or more reversal sections have been identified, the stress response changes of blade nodes in each reversal section are extracted based on intermediate simulation data. The mapping relationship of residual disturbance effects in the structural path is constructed, and a reversal trace distribution map is generated for spatial offset analysis of the reversal structure.

[0089] In this embodiment, given that two or more directional reversal sections have been identified, the stress response changes of blade nodes within each directional reversal section are extracted based on intermediate simulation data. A mapping relationship of residual disturbance effects in the structural path is constructed, and a directional reversal trace distribution map is generated for spatial offset analysis of the reversing structure. Specifically, the following steps are included:

[0090] Given that two or more directional reversal sections have been identified, the blade node stress response time series in the intermediate simulation data is obtained, and the node subset corresponding to each directional reversal section is extracted from it.

[0091] The following process can be implemented in software: First, retrieve the full-node stress response dataset of the blades recorded within the current simulation cycle from the simulation system. This dataset is typically indexed by time, with each time step containing the stress values ​​of all blade nodes. Then, based on the start and end position indices of the previously identified direction reversal segments, locate the corresponding time range of these reversal segments and extract stress response slices within that time range. Further, based on the path node sequence within the reversal segment, filter out the set of blade node indices participating in the directional change, and extract stress response curves matching these node numbers from the time slices, thus forming the node subset data corresponding to the direction reversal segment. This is necessary because the disturbance to the structural path during direction reversal is mainly reflected in the stress response behavior of the nodes involved in the structural turning point. Only by focusing on these nodes and extracting their stress change process can we provide accurate and targeted input data for subsequent judgment of disturbance trajectories, analysis of spatial offsets, and evaluation of path reconstruction. The entire process can be efficiently implemented in a software environment using methods such as data index matching, time window filtering, and node filtering.

[0092] For each subset of nodes in the direction reversal section, calculate the relative gradient value of stress change between nodes and mark the node index where the change amplitude is greater than a set threshold.

[0093] This can be achieved as follows: First, for each extracted subset of nodes in the direction reversal segment, the stress response time series is arranged into a two-dimensional array, where each row corresponds to the continuous stress value of a node, and each column corresponds to a time step. Then, the stress value of each node is differentially analyzed in the time dimension to obtain its stress change rate. Next, the stress change rates of adjacent nodes are differentially analyzed in the spatial dimension to obtain the relative gradient values ​​of stress response between adjacent nodes in the path structure. These gradient values ​​reflect the degree of stress inconsistency between different nodes during disturbance propagation and are important indicators for identifying residual disturbance accumulation. Afterward, based on a preset gradient change amplitude threshold (which can be dynamically set by combining statistical distribution or historical disturbance behavior), nodes exceeding the threshold are marked, and their indices in the path structure are recorded. This approach is used because during direction reversal, the disturbance load is usually not uniformly distributed during path propagation, but rather forms localized stress accumulation or rapid release phenomena. Only through relative gradient values ​​can the location of these drastic changes be revealed, thus providing a key input for structural sensitivity in subsequent disturbance trajectory identification and path reconstruction. The entire processing can be achieved using typical software data analysis methods such as matrix differencing, threshold comparison, and index mapping, offering advantages such as high efficiency, scalability, and adaptability to various working conditions.

[0094] Each marked node is mapped one-to-one with its path position, and a disturbance residual mapping array is established in the path coordinate system.

[0095] The node positions in the disturbance residual mapping array are mapped to the stress gradient magnitude as a set of two-dimensional graphical trajectory points, and then connected sequentially in chronological order to form a trajectory curve;

[0096] The generated trajectory curve is projected along the path coordinate axis to construct a distribution map of direction reversal traces, and this distribution map is used as the input image data for the calculation of spatial offset of the reverse structure.

[0097] This can be achieved through the following process: First, given a defined path coordinate axis (usually the blade axial direction in the simulation model), each point in the trajectory curve is projected onto this coordinate axis using a projection transformation method (such as orthogonal projection or linear mapping) according to its three-dimensional spatial coordinates, preserving the principal component information of the point in the axial direction. Then, the projected trajectory point set is constructed into a two-dimensional image, where the horizontal axis represents the axial distance of the path, and the vertical axis represents the trajectory point offset or stress gradient intensity. These points are then connected using interpolation or continuous polylines to generate a clear "direction reversal trace distribution map." This image can be represented as a pixel matrix for input into image processing algorithms or used as a function curve for subsequent path offset analysis models. The reason for projecting the coordinate axis is that path reconstruction judgment focuses on the spatial deformation trend of disturbances along the main transmission direction. Retaining only information in this direction can maximize the removal of irrelevant disturbance components, highlighting the main offset behavior of the reversal structure and improving the accuracy and stability of the judgment. Furthermore, constructing it as image data provides structured input for subsequent trajectory offset calculations based on pattern recognition or numerical statistics, serving as a crucial bridge for path rationality assessment and breakpoint reconstruction judgment. The entire process can be efficiently completed through processing mechanisms such as matrix projection operations, graphic mapping, and image data generation in the software, making it suitable for structural pattern analysis of large-scale simulation data.

[0098] In this embodiment, a disturbance residual mapping array is established in the path coordinate system by mapping each marked node to its path position one by one. Specifically:

[0099] A unified path axis coordinate system is set up, and the three-dimensional position information of the nodes in each direction reversal section in the simulation space is mapped to the linear axial distance scale under the path coordinate system by projection transformation.

[0100] This can be achieved through the following steps: First, select a reference axis representing the main direction of load transmission in the entire simulation model. Typically, the geometric center axis of the blade or the line connecting the blade root and the blade tip is selected as the path axial reference, and it is defined as the principal axis of the path axial coordinate system (e.g., the X-axis). Next, for each node in the direction reversal section, extract its three-dimensional coordinates (X, Y, Z) in the simulation space, and project the position vector of each node onto the path principal axis using the vector projection formula. This yields the linear distance of the node relative to the axis starting point, which serves as its axial scale value in the path coordinate system. These scale values ​​are used to uniformly represent the relative positions of different nodes in the structural path, forming a standardized path positioning basis that can be used for subsequent stress gradient mapping and trajectory construction. The reason for this projection transformation is that the nodal response distribution of wind turbine blades after being disturbed in three-dimensional space exhibits spatial discreteness. Directly processing the data using three-dimensional coordinates would result in a lack of continuous axial logic in the path information, making it difficult to extract the disturbance propagation pattern. By uniformly mapping to the path axial coordinate system, not only can the dimensionality be compressed and data aggregation enhanced, but the transmission behavior of the disturbance along the main direction can also be more accurately reflected. This is a crucial preliminary operation for realizing path structure visualization, disturbance trend quantification, and offset trajectory extraction. The entire process can be efficiently completed using software methods such as vector operations, coordinate transformation, and distance calculation, exhibiting high adaptability and stability.

[0101] The stress gradient value of each marked node is used as the magnitude reference value, and a pair of data sets is formed with the corresponding axial distance of the path. A two-dimensional mapping array is established using the node number as the index.

[0102] This can be achieved as follows: First, extract the stress gradient value corresponding to each node in the simulation time series from the set of labeled nodes obtained in the previous steps, and record its corresponding axial distance along the path, i.e., the position scale of the node in the unified path coordinate system. Then, combine these two parameters as a data pair to form a two-dimensional array with "axial position - stress gradient value" as the basic structure, where each item is indexed by the node number, achieving data structure retrieval and clarity. This mapping array not only retains the information of disturbance intensity (through stress gradient) and disturbance location (through axial distance), but also seamlessly integrates with subsequent trajectory fitting processing. The reason for adopting this method is that when wind turbine blades are affected by multiple disturbances, the response differences exhibited by nodes at different locations often determine whether the path needs to be reconstructed. The two-dimensional mapping array can clearly express the logical relationship of disturbance "intensity-location" in a highly integrated form, enabling the software to accurately locate the "anomaly point" of stress reversal on the spatial axis, while quantifying its disturbance amplitude, providing a standard input format for graphical reconstruction of disturbance trajectories and path offset analysis. This operation can be efficiently implemented in the software through techniques such as array construction, index binding, and structured data assembly, and has good applicability and scalability.

[0103] For continuous data point segments in the two-dimensional mapping array with abrupt gradient changes, a smoothing function is set to perform gradient curve interpolation fitting to ensure the continuity and resolvability of the perturbation residual mapping on the path structure.

[0104] To process continuous data segments with abrupt gradient changes in a two-dimensional mapping array, a smoothing function can be set in software and gradient curve interpolation fitting can be performed. The specific implementation process is as follows: First, scan the stress gradient values ​​of adjacent data points in the two-dimensional mapping array to identify continuous data segments with a rate of change greater than a preset abrupt change threshold, and mark the start and end indices of these data segments. Then, apply the smoothing function to these data segments to denoise and buffer their gradient data. Common methods include moving average, weighted sliding window, or variable weighting based on the Bessel kernel function to reduce the influence of high-frequency disturbances or local outliers in the data. Next, perform interpolation fitting on the smoothed gradient values, such as using cubic spline interpolation or least squares curve fitting, to establish a continuous, differentiable, and differentiable fitting curve on the path axial coordinates, ultimately generating a stable gradient curve model. The purpose of this process is to transform the local abrupt changes in the original gradient data into a smooth and analytically sound function curve, so that the mapping of the disturbance residue in the path structure exhibits a physically continuous propagation trend, facilitating subsequent graph generation and trajectory analysis.

[0105] The term "smoothing function" refers to a mathematical tool used in numerical computation to reduce data fluctuations and suppress noise. Its core purpose is to improve the continuity and trend readability of data sequences. Common forms include moving averages, exponential smoothing, or Gaussian filtering. "Gradient curve interpolation fitting," on the other hand, constructs a continuous function curve based on existing discrete data points using mathematical methods to approximate the original data trend. Its core is to select a function form suitable for the characteristics of disturbance propagation (such as polynomial, spline, or exponential functions) and use interpolation algorithms to estimate the curve's values ​​at unsampled points. The combined use of these two methods can correct non-physical jumps in the path caused by simulation disturbances and restore the continuous propagation process of disturbances in the structure. This is a key intermediate processing step in wind turbine blade path reconstruction simulation to achieve data-to-model conversion, and can be fully implemented in software and integrated into the simulation post-processing algorithm.

[0106] In this embodiment, the node positions in the disturbance residual mapping array are mapped to the stress gradient magnitudes as a set of two-dimensional graphical trajectory points, and then connected sequentially in chronological order to form a trajectory curve, specifically:

[0107] Using each data point in the perturbation residual mapping array as input, extract its path axial position and stress gradient magnitude as horizontal and vertical coordinate values ​​to generate a two-dimensional coordinate point set.

[0108] This can be implemented step-by-step using software. First, in the existing two-dimensional mapping array, each data point contains the axial distance of the node's path (i.e., its linear position in a unified path coordinate system) and the stress gradient magnitude of that node. These two values ​​are used as the X and Y coordinates of the point, respectively. Then, the entire array is traversed, and each pair (axial distance, stress gradient) is read and organized into a two-dimensional coordinate point set, forming a set with the structure [(x1, y1), (x2, y2), ..., (x n , y n The coordinate data structure is as follows: the point set can be stored as a list, matrix, or dictionary structure, facilitating subsequent interpolation and trajectory fitting operations. This approach is adopted because a two-dimensional coordinate point set can intuitively represent the distribution pattern of disturbance residues in the path structure. The X-axis reflects the spatial position of the disturbance along the blade path, while the Y-axis reflects the intensity of the disturbance's impact. This "position-intensity" two-dimensional relationship is the core foundation for subsequently identifying the disturbance propagation trajectory and assessing abnormal path change trends. The entire process requires only simple data traversal and field extraction operations, combined with basic coordinate data structure construction, and can be efficiently implemented in software, ensuring logical consistency and computational continuity in subsequent graphical analysis.

[0109] The generated trajectory point set is sorted by timestamp, and the point set index is rearranged according to the simulation time order of the nodes to ensure that the direction of the trajectory curve is consistent with the order of disturbance propagation.

[0110] The generated trajectory point set is timestamped and the point set index is rearranged according to the simulation time order of the nodes. This can be achieved through the following software method: First, when constructing the trajectory point set, in addition to recording the axial position and stress gradient magnitude of each point, the simulation timestamp corresponding to that node must also be extracted as the third attribute value. Then, all trajectory points are organized into a triplet structure containing timestamps (e.g., (axial position, stress gradient value, timestamp)), forming a sortable list or array. Next, a stable sorting algorithm (such as quicksort or mergesort) is used to sort the list in ascending order with the timestamp as the primary key, ensuring that the order of each point strictly reflects the order in which it was triggered by the disturbance during the simulation. After sorting, the point set index is reset according to the new order to generate a trajectory path that reflects the temporal logic of the disturbance propagation. This sorting operation is necessary because, under the condition of multiple disturbances input concurrently or direction reversal, the trajectory points may have jumps or mixed arrangements in spatial distribution. If they are not reordered according to time, the direction of the fitted curve will be disordered, and the dynamic path of disturbance propagation cannot be accurately restored, thus interfering with the offset analysis and path reconstruction judgment. This process relies on standard data sorting and structure reconstruction techniques and can be fully automated through software workflows, ensuring the consistency of the trajectory line's directionality and disturbance timing, thus laying an accurate temporal foundation for subsequent trajectory fitting and spatial offset calculations.

[0111] Based on a time-ordered set of points, all point pairs are connected using a curve fitting method based on cubic spline interpolation to generate a continuous and smooth perturbation trajectory line, which is used as the input for subsequent trajectory offset calculation.

[0112] Based on a time-ordered set of points, a curve fitting method using cubic spline interpolation is employed to connect all point pairs. This can be achieved through the following software process: First, the time-ordered two-dimensional trajectory point set is split into two numerical sequences: the path axial position sequence and the corresponding stress gradient value sequence. Then, a piecewise polynomial model is constructed on these sequences using a cubic spline interpolation function. This ensures that each interpolation curve not only passes through the corresponding trajectory point but also has continuous first and second derivatives between segments, achieving overall curve smoothness and physical plausibility. Through this fitting process, several intermediate points can be generated between two trajectory points, expanding the original discrete trajectory points into a fitted curve with a continuous path. This method is chosen because perturbation trajectories typically exhibit nonlinear propagation characteristics. Using simple straight lines or linear interpolation is insufficient to reflect the turning trend of the actual perturbation curve along the path. Cubic splines not only accurately reconstruct the spatial position of the original points but also provide a smooth transition without introducing significant oscillations, making the final trajectory line closer to the actual perturbation path. This provides a precise basis for subsequent spatial offset extraction and path plausibility assessment. The entire process can be automatically completed by numerical computing libraries such as SciPy and NumPy by calling interpolation functions, making it perfectly suitable for software implementation environments.

[0113] Cubic spline interpolation is a numerical computation method. Its core idea is to divide the entire point set into several continuous small intervals, and within each interval, use a cubic polynomial (i.e., a polynomial with a highest degree of three) as the fitting function. This constructs a smooth curve that is continuous overall and exhibits continuity in its first and second derivatives at the interval connections. This method is particularly suitable for describing datasets with strong nonlinear trends but requiring smooth curve transitions. Compared to high-order polynomial fitting, it does not induce the Runge phenomenon (endpoint oscillations), and therefore is widely used in high-precision scenarios such as simulation modeling and path reconstruction. In the simulation of dynamic load paths for wind turbine blades, cubic spline interpolation can effectively fit the stress gradient propagation trajectory, constructing disturbance lines that closely match the true distribution of disturbance behavior. It is an important mathematical foundation for realizing path offset analysis and graphical derivation modeling.

[0114] Trajectory fitting is performed on the residual disturbance in the direction reversal trace distribution map to extract the spatial offset of the trajectory in the path axial direction. Combined with the duration of the disturbance in the reversal section, the trajectory deflection index and duration weight function are calculated. Based on the reconstruction judgment model composed of the two, it is determined whether to perform the breakpoint reconstruction operation.

[0115] In this embodiment, the residual disturbance in the direction reversal trace distribution map is fitted with a trajectory, the spatial offset of the trajectory in the path axial direction is extracted, and the trajectory deflection index and duration weight function are calculated based on the duration of the disturbance in the reversal section. The reconstruction judgment model based on the two is used to determine whether to perform the breakpoint reconstruction operation. The specific steps include:

[0116] Project each disturbance trajectory in the direction reversal trace distribution map along the axial direction, extract the maximum and average offset values ​​of the trajectory curve in the main axis direction of the path, and calculate the trajectory deflection index.

[0117] Extract the disturbance time series corresponding to the current direction reversal segment from the intermediate simulation data, calculate the time span from the start to the end of the disturbance segment, and construct the relative proportion function of the time span in the whole simulation cycle as the duration weight function;

[0118] In the simulation data, the start and end nodes of each reversal segment on the time axis can be located by retrieving the node index range corresponding to the reversal segment. Each node in the simulation data is accompanied by a timestamp, so the time span of the reversal segment can be calculated by extracting the timestamp values ​​of the start and end nodes. The ratio of this span to the total time length of the complete simulation cycle yields the proportion of the disturbance duration of the reversal segment. To avoid abnormal scaling due to different simulation time bases, the time unit and sampling precision need to be standardized before constructing the ratio. This proportion is then used as an input parameter for subsequent function fitting, and a "duration weight function" is constructed through exponential amplification, interval normalization, or a custom function mapping. In this way, the intensity characteristics of the disturbance effect of different reversal segments in the time dimension can be quantified, serving as a time weighting factor for path reconstruction judgment.

[0119] The core of "constructing a relative proportion function of the time span within the entire simulation cycle as a duration weight function" lies in transforming the "existence time" of the disturbance's impact into a "weight for decision-making." For example, if a simulation cycle is 100 seconds and the disturbance duration of the direction reversal segment is 20 seconds, then its time proportion is 0.2; in this case, a monotonically increasing weight function can be set, such as the weight function... ,in For time percentages, This is an adjustment factor that allows longer-lasting disturbances to have a greater impact on the evaluation. Based on the example above, if k=3, then... This indicates that approximately 45% of the time dimension in this segment exhibits a tendency to trigger reconstruction. This weighting function can be selected as a linear, exponential, or sigmoid function according to simulation requirements, allowing it to flexibly adapt to different simulation rhythms and disturbance levels, thus achieving a quantitative expression of the time dimension's participation in path judgment logic.

[0120] After normalizing the trajectory deflection index and the duration weight function respectively, they are input into the two-parameter judgment formula to construct a joint evaluation expression. The weight ratio factor of the two is set in the form of exponential decay.

[0121] To ensure that the trajectory deflection index and duration weighting function participate in the joint evaluation under the same scale, these two parameters need to be normalized first. Common normalization methods include min-max scaling or Z-score standardization. The min-max method is recommended, mapping the two parameters to the interval [0,1] to ensure their comparability in the evaluation function. The two normalized values ​​are input into a predefined two-parameter decision formula, which combines these two parameters to quantify the disturbance intensity and time span of the direction reversal segment. Normalization is necessary to prevent one parameter from "masking" the role of the other parameter in the evaluation expression due to its large numerical scale, thus affecting the fairness and robustness of the evaluation. In addition, normalization allows for flexible use of different forms of combination functions (such as linear superposition, product models, nonlinear coupling, etc.) for path reconstruction determination.

[0122] The phrase "constructing a joint evaluation expression and setting the weighting factors for both using an exponential decay form" refers to using an exponential function to weight the two normalized indices in the joint decision function, in order to reflect the non-linear variation of the influence of trajectory offset and disturbance duration on reconstruction decision as their values ​​change. For example, the following evaluation function can be constructed: ,in, This is the normalized trajectory deflection index. The normalized duration weight function, and To adjust the weighting coefficients of the two influences (satisfying) ),and , Control the response curvature (exponential decay rate) of each item. For example, when , At that time, the first item This indicates that even if the trajectory deviation is small, reconstruction may be triggered if the weighting coefficient is set high. The advantages of this method are: strong adjustability, smooth curve changes, and better characterization of the relative importance of different disturbance features, thereby improving the sensitivity and scientific rigor of the judgment. Finally, the function output is compared with a set threshold interval to determine whether the breakpoint reconstruction conditions are met.

[0123] The output of the two-parameter evaluation expression is compared with the preset breakpoint reconstruction judgment threshold range. If the output is within the judgment threshold range, the current direction reversal segment is determined to trigger the breakpoint reconstruction logic.

[0124] The purpose of this approach is to automate, quantify, and refine the determination of whether path structure requires breakpoint reconstruction during complex wind turbine blade simulations. Since not all disturbances involved in direction reversal sections possess the same destructive power—some reversals are short-lived and have small offset amplitudes—performing breakpoint reconstruction for all such instances would not only increase simulation complexity but also potentially introduce unnecessary path disturbances, affecting solution stability. Therefore, by introducing the output of a two-parameter evaluation expression and comparing it with a preset breakpoint reconstruction threshold range, the path structure adjustment logic is ensured only when both the "trajectory deflection" and "disturbance duration" parameters simultaneously reach certain intensity and duration thresholds. This mechanism effectively avoids overreacting to minor disturbances and ensures timely intervention and reconstruction in high-risk path segments, thereby guaranteeing the accuracy, continuity, and efficiency of the simulation process. It represents a precise control method for determining structural stability in nonlinear simulation paths.

[0125] In this embodiment, the trajectory deflection index is calculated as follows:

[0126] For each direction of the reverse trajectory curve, position sampling is performed in the axial coordinate of the path, the offset values ​​of all sampling points in the axial direction are extracted, and their mean and standard deviation are calculated.

[0127] This process can be implemented in software using simulation data array manipulation and statistical analysis methods, specifically as follows: First, position sampling is performed on the reverse trajectory curve in each direction, that is, several discrete sampling points are extracted at equal intervals or in a keypoint manner on its path axial coordinate (usually the projected axis from the blade root to the tip). For each sampling point, its spatial position value in that axial coordinate direction (i.e., the offset distance relative to the starting point of the path) is extracted, forming a one-dimensional numerical array. Next, using conventional statistical analysis methods, the arithmetic mean (to reflect the overall level of offset) and standard deviation (to measure the dispersion or fluctuation of the offset) of all offset values ​​in the array are calculated. This process can be implemented in software using common numerical computing libraries such as NumPy in Python and the mean and std functions in MATLAB, ensuring efficiency and scalability. The core purpose of this step is to provide a statistical basis for the subsequent quantitative analysis of trajectory offset, so that the assessment of the degree of offset has a clear data basis.

[0128] The axial distance between the maximum offset point of the trajectory curve and the starting point of the trajectory is quantified and multiplied by the standard deviation ratio to obtain an offset weight index that reflects the degree of offset concentration and extreme value deviation.

[0129] This step can be implemented in software using mathematical methods of spatial geometric analysis and normalization. First, in the obtained direction-reversal trajectory curve, the maximum offset point is determined, i.e., the trajectory sampling point with the maximum offset value in the axial coordinate system is found. Next, the axial distance between this maximum offset point and the trajectory starting point is calculated as a quantitative indicator reflecting the distance the offset propagates along the path. This distance can be directly obtained from the absolute value of the coordinate difference between the two points. Then, this axial distance is compared with the previously calculated standard deviation. This ratio reflects the proportion of the maximum offset relative to the dispersion of the entire trajectory offset. If the ratio is high, it indicates that the offset is mainly concentrated near the trajectory starting point, and the offset structure is more concentrated; if it is low, the offset distribution is more dispersed. To further quantify the significance of this structural offset, this ratio is multiplied by the standard deviation itself to obtain the offset weight index. This index simultaneously integrates two dimensions: the degree of offset concentration (through the relative distance between the maximum point and the starting point) and the offset amplitude characteristics (through the standard deviation), thus providing crucial support for the subsequent construction of the trajectory deflection index. At the software implementation level, the above calculation process can be embedded into the trajectory analysis process through matrix processing and numerical operation functions to achieve efficient calculation and repeatable application.

[0130] The ratio of the offset weight index to the aforementioned standard deviation is used as the trajectory deflection index to quantify the disturbance dispersion and spatial deformation trend of the direction reversal segment in the path structure.

[0131] This step can be implemented in software through mathematical expression calculation and data normalization. First, after obtaining the offset weight index (i.e., the axial distance between the maximum offset point and the starting point multiplied by the standard deviation) and the standard deviation of the offset values ​​of the trajectory sampling points, the offset weight index is directly divided by the standard deviation to obtain the trajectory deflection index. This trajectory deflection index actually reflects the spatial propagation characteristics of the offset amplitude under a unit degree of offset dispersion, i.e., whether the offset is significant and concentrated. Its physical meaning is: if the offset weight index is much greater than the standard deviation, it indicates that the disturbance on the path is not only high in amplitude but also concentrated at a specific location, and the structural deformation trend is clear; conversely, the offset is discrete, the fluctuation is weak, and the deformation trend is not obvious. In the implementation process, all calculation formulas can be encapsulated into a function module, which automatically extracts parameters and calculates results by traversing the reverse trajectory curve in each direction, and then uses this result for subsequent breakpoint reconstruction determination. This quantification method can significantly improve the accuracy of disturbance feature identification, providing a more discriminative evaluation basis for the feasibility analysis of path reconstruction.

[0132] In this embodiment, the structure and application of the two-parameter decision model include the following steps:

[0133] Set a normalization factor for the trajectory deflection index and the duration weighting function, and process them into standardized parameter pairs with values ​​ranging from 0 to 1;

[0134] This step can be implemented in software using standard normalization methods, commonly including Min-Max Normalization or Z-score standardization. For the two original values ​​of the trajectory deflection exponent and the duration weight function, first, their minimum and maximum values ​​are calculated across all simulation samples or historical records. Then, the normalization formula (current value - minimum value) / (maximum value - minimum value) is used to convert them into standardized values ​​between 0 and 1. If the sample values ​​are anomalies or skewed, Z-score normalization can also be used. The purpose of this normalization operation is to transform input parameters with different dimensions and amplitude levels to the same evaluation scale, enabling them to be uniformly input into the two-parameter decision model for joint evaluation. This processing can be achieved through a built-in data standardization module in the program logic, ensuring the effectiveness and robustness of subsequent model calculations. This normalization method also improves the generalization ability of the decision model and avoids evaluation bias caused by differences in parameter numerical magnitudes.

[0135] A two-input multiplicative coupled model is constructed based on adjustable parameters α and β, where α is multiplied by the trajectory deflection exponent and β is multiplied by the duration weight function. The sum of the two products is then input into the Sigmoid function to transform it into a continuous evaluation value.

[0136] This step can be implemented in software through mathematical modeling and numerical function calls. First, two adjustable parameters, α and β, are preset in the system or set by the user. These parameters assign different weights to the trajectory deflection index and the duration weight function, reflecting their relative importance in the judgment. Then, the normalized trajectory deflection index and duration weight function are multiplied by α and β respectively, resulting in two weighted terms. These two weighted results are summed to form a joint linear evaluation value. To enhance the nonlinear boundary capability of this evaluation value in judgment, it can be input into a Sigmoid function (e.g., σ(x) = 1 / (1 + e^(-x))) for nonlinear mapping, ultimately outputting a continuous value between 0 and 1. This method not only enhances the model's ability to distinguish small perturbations but also facilitates the subsequent setting of a clear judgment interval, thereby achieving precise control over whether to trigger breakpoint reconstruction operations.

[0137] The evaluation value is compared with the lower and upper limits of the set breakpoint reconstruction trigger interval. If the result falls within the range, the direction reversal segment is marked as a candidate segment for breakpoint reconstruction and the path reconstruction process begins.

[0138] "Adjustable parameters α and β" refer to the numerical coefficients used to weight the two input variables (trajectory deflection exponent and duration weight function). These can be flexibly adjusted according to different simulation strategies to optimize the model's judgment sensitivity. The "dual-input multiplicative coupling model" refers to a combined structure that simultaneously introduces two variables, multiplies them by their respective weights, and then sums them. This approach allows the influence of the two parameters on the output to be coupled, thereby achieving more refined control logic. The "Sigmoid function" is a commonly used nonlinear activation function with an S-shaped shape. It can map any real value to a probability value between (0,1), smoothly transforming joint evaluation results into a unified evaluation interval. This helps to construct a stable and continuous judgment mechanism, and is particularly suitable for threshold control and classification boundary discrimination in decision-making problems.

[0139] When the trajectory deflection index meets the path breakpoint reconstruction conditions, breakpoints and mirror nodes are inserted in the direction reversal section, the path node arrangement order is adjusted, the path reconstruction operation is completed, and the trajectory deflection index and corresponding path information are written into the simulation dataset.

[0140] This step can be implemented in software using graph structure reconstruction and simulation path rearrangement algorithms. First, when the trajectory deflection index reaches or falls into the set path breakpoint reconstruction judgment interval, the system inserts a "breakpoint" within the marked direction reversal segment boundary index, thus breaking the original continuous connection relationship in the path structure. Simultaneously, to maintain the continuity and directional consistency of force transmission, a "mirror node" with the opposite direction vector to the node preceding the breakpoint is generated at the breakpoint location. Starting from this new node, the connection sequence after the reversal segment is reconstructed. This operation ensures that the path logic achieves structural closure in the reverse force segment, resolving the mechanical inconsistency caused by directional conflicts in the original path.

[0141] Secondly, after path reconstruction, the node sequence needs to be reordered. Specifically, a topology traversal can be used, arranging nodes sequentially according to the newly established connections, starting from the initial perturbation node, to form a new path sequence without logical conflicts and with consistent direction. During this process, path reachability checks can be performed to confirm that each node is in a valid path, eliminating isolated nodes or redundant path segments caused by reconstruction. Simultaneously, to ensure the temporal order of load solutions, the new path sequence should be aligned with the perturbation temporal distribution on the simulation timeline to ensure that the solution engine correctly parses the load transfer links in subsequent calculations.

[0142] Finally, the trajectory deflection index and new path information need to be written into the simulation dataset for subsequent analysis and model training. Specifically, the simulation cycle index, trajectory deflection index value, and new path node number sequence at the current breakpoint reconstruction time are encapsulated into a data structure and appended to a reserved structural path change record table in the simulation dataset. This processing not only allows for the construction of a dynamic path adjustment history but also provides feature samples for pattern recognition or machine learning models under subsequent multiple perturbations. This step is crucial because it ensures the simulation platform has "memory" and "feedback" capabilities, laying the foundation for the linkage between structural simulation and scheduling strategies.

[0143] like Figure 2 The wind turbine blade dynamic load balance nonlinear constraint simulation system shown includes a disturbance driving path construction module, a direction reversal identification and analysis module, a disturbance residual mapping extraction module, a dual-parameter determination and modeling module, and a path reconstruction execution and writing module.

[0144] The disturbance-driven path construction module divides the target simulation period into multiple load action segments, generates a node sequence according to the disturbance trend, and establishes a connection relationship between nodes based on the load direction to construct the dynamic load transfer path.

[0145] The direction reversal identification and analysis module analyzes the direction change of nodes in the dynamic load transmission path, identifies whether there are two or more direction reversal segments, and marks the start and end positions of each direction reversal segment.

[0146] The disturbance residual mapping extraction module, under the premise that two or more directional reversal sections have been identified, extracts the stress response changes of blade nodes in each directional reversal section based on intermediate simulation data, constructs the mapping relationship of disturbance residual influence in the structural path, and generates a directional reversal trace distribution map for spatial offset analysis of the reversing structure.

[0147] The dual-parameter judgment modeling module performs trajectory fitting on the residual disturbance in the direction reversal trace distribution map, extracts the spatial offset of the trajectory in the path axial direction, and calculates the trajectory deflection index and duration weight function based on the duration of the disturbance in the reversal section. Based on the reconstruction judgment model composed of the two, it determines whether to perform the breakpoint reconstruction operation.

[0148] The path reconstruction execution and writing module inserts breakpoints and mirror nodes in the direction reversal section when the trajectory deflection index meets the path breakpoint reconstruction conditions, adjusts the path node arrangement order, completes the path reconstruction operation, and writes the trajectory deflection index and corresponding path information into the simulation dataset.

[0149] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0150] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0151] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0152] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0153] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0154] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0155] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions, characterized in that, Specifically, the following steps are included: The target simulation period is divided into multiple load action segments. A node sequence is generated according to the disturbance trend, and a connection relationship is established based on the load direction between nodes to construct the dynamic load transfer path. Perform directional change analysis on the nodes in the dynamic load transmission path to identify whether there are two or more directional reversal segments, and mark the start and end positions of each directional reversal segment. Given that two or more reversal sections have been identified, the stress response changes of blade nodes in each reversal section are extracted based on intermediate simulation data. The mapping relationship of residual disturbance effects in the structural path is constructed, and a reversal trace distribution map is generated for spatial offset analysis of the reversal structure. Trajectory fitting is performed on the residual disturbance in the direction reversal trace distribution map to extract the spatial offset of the trajectory in the path axial direction. Combined with the duration of the disturbance in the reversal section, the trajectory deflection index and duration weight function are calculated. Based on the reconstruction judgment model composed of the two, it is determined whether to perform the breakpoint reconstruction operation. When the trajectory deflection index meets the path breakpoint reconstruction conditions, breakpoints and mirror nodes are inserted in the direction reversal section, the path node arrangement order is adjusted, the path reconstruction operation is completed, and the trajectory deflection index and corresponding path information are written into the simulation dataset.

2. The nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions according to claim 1, characterized in that, The target simulation period is divided into multiple load application segments. A node sequence is generated according to the disturbance trend, and a connection relationship is established between the nodes based on the load direction to construct the dynamic load transfer path. The specific steps include: Based on wind shear disturbance, pitch control commands and changes in blade operating status, the time series of disturbance behavior within the target simulation period is extracted. Based on the changes in the frequency and amplitude of disturbances in the time series of disturbance behavior, the target simulation period is divided into multiple disturbance-dominated load action segments. In each load application zone, nodes representing the location and direction of the disturbance input are generated sequentially, and arranged into a node sequence according to the temporal relationship of the disturbance influence between nodes. Based on the load direction relationship between adjacent nodes, directional connections are established between nodes to represent the transmission path of dynamic load under the disturbed structure, and all nodes are sequentially combined to form the initial dynamic load transmission path.

3. The nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions according to claim 2, characterized in that, Perform directional change analysis on the nodes in the dynamic load transfer path to identify whether there are two or more directional reversal segments, and mark the start and end positions of each directional reversal segment. The specific steps include: Extract the direction vectors between adjacent nodes in the dynamic load transfer path, and calculate the angle between the direction vectors of each pair of adjacent nodes; Based on the angle calculation results, adjacent node pairs with an angle greater than 180 degrees are marked as having reversed direction, and their position indices are recorded in the path sequence. Traverse the entire path direction sequence, group and cluster all marked direction reversal nodes, and identify whether there are two or more direction reversal segments. For each direction reversal segment, extract the index positions of the starting and ending nodes of its continuous reversals, and establish a direction reversal segment boundary index table for subsequent path stability assessment and structural reconstruction judgment.

4. The nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions according to claim 3, characterized in that, Given that two or more directional reversal sections have been identified, the stress response changes of blade nodes within each directional reversal section are extracted based on intermediate simulation data. A mapping relationship of residual disturbance effects along the structural path is constructed, and a directional reversal trace distribution map is generated for spatial offset analysis of the reversing structure. The specific steps include: Given that two or more directional reversal sections have been identified, the blade node stress response time series in the intermediate simulation data is obtained, and the node subset corresponding to each directional reversal section is extracted from it. For each subset of nodes in the direction reversal section, calculate the relative gradient value of stress change between nodes and mark the node index where the change amplitude is greater than a set threshold. Each marked node is mapped one-to-one with its path position, and a disturbance residual mapping array is established in the path coordinate system. The node positions in the disturbance residual mapping array are mapped to the stress gradient magnitude as a set of two-dimensional graphical trajectory points, and then connected sequentially in chronological order to form a trajectory curve; The generated trajectory curve is projected along the path coordinate axis to construct a distribution map of direction reversal traces, and this distribution map is used as the input image data for the calculation of spatial offset of the reverse structure.

5. The nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions according to claim 4, characterized in that, By mapping each marked node to its path position, a perturbation residual mapping array in the path coordinate system is established, specifically as follows: A unified path axis coordinate system is set up, and the three-dimensional position information of the nodes in each direction reversal section in the simulation space is mapped to the linear axial distance scale under the path coordinate system by projection transformation. The stress gradient value of each marked node is used as the magnitude reference value, and a pair of data sets is formed with the corresponding axial distance of the path. A two-dimensional mapping array is established using the node number as the index. For continuous data point segments in the two-dimensional mapping array with abrupt gradient changes, a smoothing function is set to perform gradient curve interpolation fitting to ensure the continuity and resolvability of the perturbation residual mapping on the path structure.

6. The nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions according to claim 5, characterized in that, The node positions in the perturbation residual mapping array are mapped to stress gradient magnitudes as a set of two-dimensional graphical trajectory points, and then connected sequentially in chronological order to form a trajectory curve, specifically: Using each data point in the perturbation residual mapping array as input, extract its path axial position and stress gradient magnitude as horizontal and vertical coordinate values ​​to generate a two-dimensional coordinate point set. The generated trajectory point set is sorted by timestamp, and the point set index is rearranged according to the simulation time order of the nodes to ensure that the direction of the trajectory curve is consistent with the order of disturbance propagation. Based on a time-ordered set of points, all point pairs are connected using a curve fitting method based on cubic spline interpolation to generate a continuous and smooth perturbation trajectory line, which is used as the input for subsequent trajectory offset calculation.

7. The nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions according to claim 6, characterized in that, Trajectory fitting is performed on the residual disturbances in the direction reversal trace distribution map to extract the spatial offset of the trajectory in the path axial direction. Combined with the duration of the disturbance within the reversal segment, the trajectory deflection exponent and duration weighting function are calculated. Based on the reconstruction judgment model jointly constructed by these two factors, it is determined whether to perform a breakpoint reconstruction operation. Specifically, the following steps are included: Project each disturbance trajectory in the direction reversal trace distribution map along the axial direction, extract the maximum and average offset values ​​of the trajectory curve in the main axis direction of the path, and calculate the trajectory deflection index. Extract the disturbance time series corresponding to the current direction reversal segment from the intermediate simulation data, calculate the time span from the start to the end of the disturbance segment, and construct the relative proportion function of the time span in the whole simulation cycle as the duration weight function; After normalizing the trajectory deflection index and the duration weight function respectively, they are input into the two-parameter judgment formula to construct a joint evaluation expression. The weight ratio factor of the two is set in the form of exponential decay. The output of the two-parameter evaluation expression is compared with the preset breakpoint reconstruction judgment threshold range. If the output is within the judgment threshold range, the current direction reversal segment is determined to trigger the breakpoint reconstruction logic.

8. The nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions according to claim 7, characterized in that, The trajectory deflection index is calculated as follows: For each direction of the reverse trajectory curve, position sampling is performed in the axial coordinate of the path, the offset values ​​of all sampling points in the axial direction are extracted, and their mean and standard deviation are calculated. The axial distance between the maximum offset point of the trajectory curve and the starting point of the trajectory is quantified and multiplied by the standard deviation ratio to obtain an offset weight index that reflects the degree of offset concentration and extreme value deviation. The ratio of the offset weight index to the aforementioned standard deviation is used as the trajectory deflection index to quantify the disturbance dispersion and spatial deformation trend of the direction reversal segment in the path structure.

9. The nonlinear constraint simulation method for dynamic load balance of wind turbine blades under varying operating conditions according to claim 8, characterized in that, The structure and application of the two-parameter decision model include the following steps: Set a normalization factor for the trajectory deflection index and the duration weighting function, and process them into standardized parameter pairs with values ​​ranging from 0 to 1; A two-input multiplicative coupled model is constructed based on adjustable parameters α and β, where α is multiplied by the trajectory deflection exponent and β is multiplied by the duration weight function. The sum of the two products is then input into the Sigmoid function to transform it into a continuous evaluation value. The evaluation value is compared with the lower and upper limits of the set breakpoint reconstruction trigger interval. If the result falls within the range, the direction reversal segment is marked as a candidate segment for breakpoint reconstruction and the path reconstruction process begins.

10. A nonlinear constraint simulation system for dynamic load balance under varying operating conditions of wind turbine blades, used to implement the nonlinear constraint simulation method for dynamic load balance under varying operating conditions of wind turbine blades as described in any one of claims 1-9, characterized in that, It includes a disturbance-driven path construction module, a direction reversal identification and parsing module, a disturbance residual mapping extraction module, a dual-parameter determination and modeling module, and a path reconstruction execution and writing module; The disturbance-driven path construction module divides the target simulation period into multiple load action segments, generates a node sequence according to the disturbance trend, and establishes a connection relationship between nodes based on the load direction to construct the dynamic load transfer path. The direction reversal identification and analysis module analyzes the direction change of nodes in the dynamic load transmission path, identifies whether there are two or more direction reversal segments, and marks the start and end positions of each direction reversal segment. The disturbance residual mapping extraction module, under the premise that two or more directional reversal sections have been identified, extracts the stress response changes of blade nodes in each directional reversal section based on intermediate simulation data, constructs the mapping relationship of disturbance residual influence in the structural path, and generates a directional reversal trace distribution map for spatial offset analysis of the reversing structure. The dual-parameter judgment modeling module performs trajectory fitting on the residual disturbance in the direction reversal trace distribution map, extracts the spatial offset of the trajectory in the path axial direction, and calculates the trajectory deflection index and duration weight function based on the duration of the disturbance in the reversal section. Based on the reconstruction judgment model composed of the two, it determines whether to perform the breakpoint reconstruction operation. The path reconstruction execution and writing module inserts breakpoints and mirror nodes in the direction reversal section when the trajectory deflection index meets the path breakpoint reconstruction conditions, adjusts the path node arrangement order, completes the path reconstruction operation, and writes the trajectory deflection index and corresponding path information into the simulation dataset.

Images