A segmented wind turbine blade manufacturing method and system based on intelligent assembly control

Abstract

The application discloses a segmented wind power blade manufacturing method and system based on intelligent assembly control. The method comprises the following steps: acquiring segmented component design data, generating assembly path planning information based on a dynamic assembly optimization model; generating intelligent assembly control instructions through real-time monitoring by multi-sensor fusion technology; verifying assembly accuracy in combination with a self-adaptive deformation compensation mechanism to generate blade assembly records; and storing the records to a manufacturing data system and completing manufacturing. The deformation compensation mechanism is based on finite element analysis and real-time data, optimizes pose deviation and interface stress, and comprehensively considers the effects of gravity and temperature. The system comprises data acquisition, instruction generation, accuracy verification and storage execution modules. The application realizes millimeter-level accurate butt joint and bonding quality control of large-size blades, and improves assembly efficiency and reliability.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing technology for wind power equipment, specifically relating to a method and system for manufacturing segmented wind turbine blades based on intelligent assembly control. Background Technology

[0002] As a key component of wind turbine generators, wind turbine blades are becoming increasingly larger. To meet the demands of long-distance transportation and on-site installation, segmented design and manufacturing have become the mainstream technology in the industry. Traditional segmented blade assembly mainly relies on manual assistance and mechanical positioning, which suffers from problems such as low docking accuracy, uneven stress distribution at the bonding interface, and difficulty in real-time deformation compensation. Especially during the docking of large-sized components, the sag caused by gravity, thermal deformation caused by temperature gradients, and the uncertainty of on-site wind loads further amplify the difficulty of high-precision control, easily leading to quality hazards such as interface cracking and aerodynamic performance degradation.

[0003] In existing technologies, some solutions use laser trackers combined with fixed fixtures for positioning, but lack dynamic evaluation of the bonding quality of composite material interfaces. Other solutions introduce industrial robots for assembly, but fail to establish adaptive path planning and error compensation mechanisms for environmental factors, resulting in poor assembly consistency and high rework rates. Existing patents mostly focus on single-sensor monitoring or offline simulation, making it difficult to achieve closed-loop coordinated control of pose, stress, and deformation during assembly, thus failing to meet the high-precision and high-reliability manufacturing requirements of large wind turbine blades. Therefore, there is an urgent need for an intelligent assembly control method integrating multi-source sensing, real-time optimization, and process linkage to overcome the technical bottlenecks in the manufacturing of segmented wind turbine blades. Summary of the Invention

[0004] The purpose of this invention is to address the technical problems in existing segmented wind turbine blade assembly, such as low docking accuracy, uneven stress distribution at the bonding interface, difficulty in real-time deformation compensation, and significant impact from environmental uncertainties. This invention provides a segmented wind turbine blade manufacturing method and system based on intelligent assembly control. Through integrated control of multi-source sensing, real-time optimization, and process linkage, it achieves high-precision docking of large-size components, dynamic quality assurance of composite material interfaces, and adaptive compensation for environmental factors. This improves assembly consistency, reduces rework rates, and enhances the overall precision and reliability of large wind turbine blade manufacturing, meeting the industrial production needs of high-performance wind power equipment.

[0005] Specifically, the present invention provides a method for manufacturing segmented wind turbine blades based on intelligent assembly control, the method comprising:

[0006] Obtain segmented component design data for wind turbine blades and generate assembly path planning information based on a dynamic assembly optimization model;

[0007] The assembly path planning information is monitored in real time using multi-sensor fusion technology to generate intelligent assembly control commands.

[0008] The accuracy of the assembly process is verified by combining the intelligent assembly control commands with the adaptive deformation compensation mechanism, and a blade assembly record is generated.

[0009] The blade assembly record is stored in the manufacturing data system, and the segmented wind turbine blade is manufactured based on the blade assembly record.

[0010] Optionally, the step of acquiring segmented component design data of the wind turbine blade and generating assembly path planning information based on the dynamic assembly optimization model includes:

[0011] Receive segmented component design data for wind turbine blades, the segmented component design data including component geometry, composite material properties, connection interface specifications, and assembly timestamp;

[0012] Multi-dimensional feature extraction is performed on the segmented component design data, including component structural stiffness, weight distribution, aerodynamic characteristics and interface compatibility, to generate feature vectors;

[0013] Based on the dynamic assembly optimization model, assembly path planning information is calculated according to the feature vector and assembly environmental factors. The dynamic assembly optimization model adaptively adjusts the path weights through historical assembly data. The assembly environmental factors include temperature field distribution, gravity load and wind load.

[0014] Optionally, the step of monitoring the assembly path planning information in real time using multi-sensor fusion technology to generate intelligent assembly control commands includes:

[0015] A multi-dimensional sensor array is deployed, which includes a laser displacement sensor, a tilt sensor, a strain gauge sensor, and a temperature sensor, to collect component pose data, force data, and environmental data in real time;

[0016] The data from the multi-dimensional sensor array is processed by a sensor fusion algorithm to generate an assembly status assessment report. The sensor fusion algorithm uses Kalman filtering or particle filtering to achieve time synchronization and spatial registration of multi-source data.

[0017] Based on the assembly status assessment report and the preset assembly accuracy threshold, the adjustment amount of the actuator is calculated by the control command generation algorithm to generate intelligent assembly control commands.

[0018] Optionally, the steps of the control command generation algorithm include:

[0019] Based on the assembly status assessment report, identify the current assembly's pose deviation and stress distribution status;

[0020] The optimal adjustment trajectory of the actuator is calculated by a predictive control model, which comprehensively considers assembly accuracy requirements, dynamic response characteristics of the actuator, and safety constraints.

[0021] Generate intelligent assembly control instructions containing six degrees of freedom adjustment parameters, including three-dimensional displacement and three-dimensional rotation angle;

[0022] The intelligent assembly control commands are sent to the assembly execution system, and the execution feedback is monitored in real time.

[0023] Optionally, the step of verifying the accuracy of the assembly process and generating a blade assembly record by combining the intelligent assembly control command with an adaptive deformation compensation mechanism includes:

[0024] The intelligent assembly control command invokes the adaptive deformation compensation mechanism, calculates the accuracy deviation based on assembly path planning information and real-time sensor feedback, and verifies the accuracy of the assembly process.

[0025] After successful verification, a blade assembly record is generated, which includes component identification, assembly path, accuracy deviation, stress distribution, and assembly status. The blade assembly record is associated with the assembly path planning information.

[0026] The blade assembly records are subjected to data integrity verification and timestamp marking to ensure the traceability and immutability of the records.

[0027] Optionally, the step of storing the blade assembly record to the manufacturing data system and completing the segmented wind turbine blade manufacturing based on the blade assembly record includes:

[0028] The blade assembly record is formatted and standardized using data processing algorithms to generate an assembly report that conforms to industry standards.

[0029] The assembly report is stored in the manufacturing execution system database, and an associated index is established with the product serial number, batch information, and quality file.

[0030] Based on the assembly status in the blade assembly record, the subsequent manufacturing process is triggered, which includes bonding and curing, flaw detection and surface treatment.

[0031] A process coordination mechanism ensures the transmission of parameters and synchronization of status in each process step.

[0032] Optionally, the steps of the adaptive deformation compensation mechanism include:

[0033] An assembly deformation prediction model is established based on finite element analysis. The assembly deformation prediction model comprehensively considers the effects of gravity, temperature and historical assembly data.

[0034] The component pose and interface stress are measured in real time using multi-dimensional sensors, and the deviation between the actual measured value and the predicted value is calculated.

[0035] Based on the deviation, a deformation compensation strategy is generated, and the optimal compensation value is calculated through an iterative optimization algorithm.

[0036] Execute the optimal compensation value and determine whether the assembly process meets the preset accuracy threshold;

[0037] When the accuracy threshold is met, the assembly verification is confirmed to be successful, and a blade assembly record associated with the assembly path planning information is generated.

[0038] Optionally, before the step of obtaining the segmented component design data of the wind turbine blade, the method further includes:

[0039] The component quality inspection system based on machine vision is used to automatically inspect segmented components. The inspection includes surface defect identification, dimensional deviation measurement, interface flatness assessment and composite material lamination quality inspection.

[0040] A component quality assessment report is generated based on the test results. The quality assessment report includes a pass / fail determination, defect location marking, and geometric measurement data.

[0041] A standardized component identifier is generated for qualified components, and the standardized component identifier is associated with the component batch number, material properties and quality grade;

[0042] The measured geometric data is used as a correction input for assembly path planning to compensate for the impact of manufacturing errors on assembly accuracy.

[0043] Optionally, after the step of storing the blade assembly record to the manufacturing data system, the method further includes:

[0044] Manufacturing traceability information is generated based on the blade assembly record. The manufacturing traceability information includes component batch traceability, assembly personnel record, environmental condition log, quality parameter archive, and process change record.

[0045] The manufacturing traceability information is securely stored using data encryption technology to generate a traceability data archive;

[0046] Establish a role-based access control mechanism and provide a traceability query interface to authorized parties, including customers, quality supervision departments, and after-sales service teams;

[0047] Based on the aforementioned traceability query interface, it supports quality traceability, fault cause analysis, and responsibility identification throughout the entire manufacturing lifecycle.

[0048] This invention also discloses a segmented wind turbine blade manufacturing system based on intelligent assembly control, the system comprising:

[0049] The data acquisition module is used to acquire segmented component design data of wind turbine blades and generate assembly path planning information based on the dynamic assembly optimization model. The data acquisition module integrates a component quality inspection submodule and a design data interface submodule.

[0050] The instruction generation module is used to monitor the assembly path planning information in real time through multi-sensor fusion technology and generate intelligent assembly control instructions. The instruction generation module includes a sensor data acquisition unit, a fusion processing unit, and a control instruction calculation unit.

[0051] The accuracy verification module is used to verify the accuracy of the assembly process by combining the intelligent assembly control command with the adaptive deformation compensation mechanism, and to generate blade assembly records. The accuracy verification module integrates a finite element analysis engine and an error evaluation algorithm.

[0052] The storage execution module is used to store the blade assembly record to the manufacturing data system and trigger subsequent manufacturing processes based on the blade assembly record. The storage execution module includes a data management unit, a traceability service unit, and a process coordination unit. Attached Figure Description

[0053] Figure 1 A flowchart of a segmented wind turbine blade manufacturing method based on intelligent assembly control provided by the present invention;

[0054] Figure 2 A flowchart of the adaptive deformation compensation method provided by the present invention;

[0055] Figure 3 A schematic diagram of the segmented wind turbine blade manufacturing system based on intelligent assembly control provided by the present invention;

[0056] Reference numerals: Segmented wind turbine blade manufacturing system based on intelligent assembly control 100, data acquisition module 101, instruction generation module 102, accuracy verification module 103, storage and execution module 104. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature; in the description of this application, unless otherwise stated, "multiple" means two or more.

[0059] To more clearly illustrate the technical solution of the present invention, the present invention will be described in detail below with reference to specific embodiments, but it should not be construed as a limitation on the scope of protection of the present invention.

[0060] like Figure 1 As shown, this embodiment achieves high-precision docking and deformation compensation of segmented wind turbine blades through real-time optimization and closed-loop precision control of the assembly path, significantly improving assembly consistency and manufacturing reliability.

[0061] First, obtain the segmented component design data of the wind turbine blade, and generate assembly path planning information based on the dynamic assembly optimization model.

[0062] Specifically, the system extracts the geometric dimensions, composite material properties, and connection interface specifications of segmented components from the CAD model, and constructs a digital representation of the component's structural stiffness and weight distribution through feature vectors. Then, the dynamic assembly optimization model uses historical assembly data as training samples and employs machine learning algorithms to adaptively adjust path weights, comprehensively considering the effects of temperature field distribution, gravity load, and wind load, to generate optimal assembly path planning information containing a six-degree-of-freedom pose sequence. Understandably, this path planning information provides a precise theoretical benchmark for subsequent assembly execution, ensuring adaptability in complex environments.

[0063] Secondly, the assembly path planning information is monitored in real time using multi-sensor fusion technology to generate intelligent assembly control commands.

[0064] Specifically, an array consisting of laser displacement sensors, tilt sensors, strain gauge sensors, and temperature sensors is deployed to collect real-time data on the current pose, stress state, and environmental parameters of the components. Furthermore, a sensor fusion algorithm utilizes Kalman filtering to achieve time synchronization and spatial registration of multi-source data, generating an assembly status assessment report. Based on a comparison of the report with a preset assembly accuracy threshold, a control command generation algorithm calculates the adjustment amount of the actuator, forming intelligent assembly control commands that include three-dimensional displacement and three-dimensional rotation parameters. Understandably, this command generation process ensures the reliability of the monitoring data and the timeliness of the control response.

[0065] Furthermore, the accuracy of the assembly process is verified by combining the intelligent assembly control commands with the adaptive deformation compensation mechanism, and a blade assembly record is generated.

[0066] Specifically, intelligent assembly control commands drive the industrial robot to adjust its execution path. Simultaneously, an adaptive deformation compensation mechanism establishes a deformation prediction model based on finite element analysis, integrating the effects of gravity, temperature, and historical assembly data, and calculates the optimal compensation value through an iterative optimization algorithm. Furthermore, the component pose and interface stress are measured in real time, the deviation between the actual and predicted values ​​is calculated, and compensation is performed to determine if the accuracy threshold is met. Upon successful verification, a blade assembly record is generated, containing component identification, assembly path, accuracy deviation, stress distribution, and assembly status. Understandably, the generation of this record ensures the traceability and quality verifiability of the assembly process.

[0067] Finally, the blade assembly record is stored in the manufacturing data system, and the segmented wind turbine blade manufacturing is completed based on the blade assembly record.

[0068] Specifically, the blade assembly records are formatted and standardized to generate industry-standard assembly reports, which are then stored in the manufacturing data system database, establishing an association index with product serial numbers and batch information. This manufacturing data system is either a Manufacturing Execution System (MES) or an enterprise-level database platform, serving as the backend data service support within the overall system. Then, based on the assembly status, subsequent bonding and curing, flaw detection, and surface treatment processes are triggered. A process coordination mechanism ensures parameter transfer and status synchronization, guaranteeing a complete closed loop in the manufacturing of segmented wind turbine blades. This storage and triggering mechanism achieves data-driven process continuity and closed-loop management of the entire manufacturing process.

[0069] This embodiment acquires segmented component design data and generates assembly path planning information based on a dynamic assembly optimization model, providing a precise theoretical trajectory foundation for large-size blade docking. Real-time monitoring and intelligent assembly control commands are generated through multi-sensor fusion technology, enabling dynamic perception and immediate response during the assembly process. Accuracy is verified and blade assembly records are generated by combining intelligent assembly control commands with an adaptive deformation compensation mechanism, ensuring controllable quality at the millimeter-level docking and bonding interface. Finally, the records are stored in the manufacturing data system, and manufacturing is completed accordingly, constructing a complete closed loop from planning to execution to archiving. This significantly improves the accuracy consistency, environmental adaptability, and production traceability of segmented wind turbine blade assembly, effectively reducing rework rates and quality risks, and fully meeting the industrial manufacturing needs of high-performance wind power equipment.

[0070] In some embodiments, by refining the feature extraction of segmented component design data and optimizing the path for environmental adaptation, the dynamic and accurate generation of assembly path planning is achieved, which effectively improves the assembly adaptability and docking reliability of large-size blades in complex field environments.

[0071] Specifically, the system receives segmented component design data for wind turbine blades, which includes component geometry, composite material properties, connection interface specifications, and assembly timestamps.

[0072] The system receives complete data packets of segmented components from an enterprise resource planning system or computer-aided design software via an industrial Ethernet interface. The component geometry includes airfoil profiles, chord length distribution, and thickness variations; composite material properties include fiber layup angles, resin type, and curing process parameters; connection interface specifications define bolt hole tolerances, adhesive surface roughness, and preload requirements; and an assembly timestamp records the data generation time to ensure version consistency. The data receiving module performs format validation and redundancy removal on the input information, forming a structured component data table.

[0073] Furthermore, multi-dimensional feature extraction is performed on the segmented component design data, including component structural stiffness, weight distribution, aerodynamic characteristics, and interface compatibility, to generate feature vectors.

[0074] Specifically, the feature extraction unit first discretizes the component's geometry based on a finite element mesh, calculating the bending and torsional stiffness of each section to characterize the structural stiffness. Simultaneously, it derives the weight distribution curve along the spanwise direction using a mass integration algorithm. Aerodynamic characteristics are generated by the computational fluid dynamics preprocessing module, which includes the lift-drag coefficient and pressure center location. Interface compatibility is assessed through geometric alignment analysis to evaluate the form and position errors and gap distribution of the mating surfaces. The system normalizes these four types of features and concatenates them into a high-dimensional feature vector, which serves as the input basis for the dynamic assembly optimization model. Understandably, the construction of this feature vector ensures the model's comprehensive understanding of the component's physical properties.

[0075] Furthermore, based on the dynamic assembly optimization model, assembly path planning information is calculated according to the feature vector and assembly environmental factors. The dynamic assembly optimization model adaptively adjusts the path weights through historical assembly data. The assembly environmental factors include temperature field distribution, gravity load, and wind load influence.

[0076] Specifically, the dynamic assembly optimization model employs a deep neural network structure, using feature vectors as input and assembly environmental factors as conditional inputs. Temperature field distribution is collected in real-time by an infrared thermometer array to form a heat map; gravity load is calculated based on the cantilever state of the components and the support scheme to determine sag curves; and wind load influence is mapped as lateral disturbance force using on-site anemometer data. The model integrates an online learning mechanism, using the accuracy deviation and stress distribution from the previous assembly as feedback signals to adaptively adjust the weight coefficients of each input feature for path generation. The final output includes assembly path planning information containing the six-degree-of-freedom trajectory of the robot's end effector. This adaptive weight adjustment mechanism enables the path planning to dynamically respond to changes in the on-site environment, ensuring the robustness and optimality of the planning information.

[0077] In some embodiments, the collaborative work of a multi-dimensional sensor array and a fusion algorithm enables real-time and accurate monitoring of assembly path planning information and dynamic generation of control commands, effectively improving the response speed and accuracy stability of the segmented wind turbine blade assembly process.

[0078] Specifically, a multi-dimensional sensor array is deployed, which includes a laser displacement sensor, a tilt sensor, a strain gauge sensor, and a temperature sensor, to collect component pose data, force data, and environmental data in real time.

[0079] A laser displacement sensor array is deployed around the assembly station for high-frequency scanning of the six-DOF pose changes of the blade segment assembly; tilt sensors are installed on the assembly support frame and robot base to monitor the tilt angle caused by gravity; strain gauge sensors are attached to key areas near the connection interfaces to capture micro-strain on the bonding surfaces and bolt areas; temperature sensors are distributed in a grid pattern on the assembly surface and in the ambient space to record the real-time temperature field distribution. All sensors are synchronously triggered via an industrial fieldbus, with a data acquisition frequency of no less than 100Hz to ensure data temporal consistency.

[0080] Furthermore, the data from the multi-dimensional sensor array is processed using a sensor fusion algorithm to generate an assembly status assessment report. The sensor fusion algorithm employs Kalman filtering or particle filtering to achieve time synchronization and spatial registration of multi-source data.

[0081] Specifically, the sensor fusion algorithm first timestamps and aligns the laser displacement and tilt angle data, removes outliers, and then inputs the data into a Kalman filter to generate a real-time pose state estimate of the component. Simultaneously, strain gauge and temperature data are used as observation inputs, and an extended Kalman filter is used to correct for thermal deformation and stress coupling effects. The algorithm completes spatial registration in a unified coordinate system and outputs an assembly state assessment report containing the current pose deviation, interface stress distribution, and environmental influence factors. Understandably, this report provides a reliable multi-dimensional state basis for subsequent control command generation.

[0082] Furthermore, based on the assembly status assessment report and the preset assembly accuracy threshold, the adjustment amount of the actuator is calculated through the control instruction generation algorithm to generate intelligent assembly control instructions.

[0083] Specifically, the control command generation algorithm receives the assembly status assessment report and compares it in real time with a preset millimeter-level docking accuracy threshold to identify posture deviations and areas of abnormal stress. Based on the magnitude and trend of the deviations, it calculates the required three-dimensional displacement and three-dimensional rotation angle of the industrial robot's end effector. The algorithm integrates the dynamic response characteristics of the actuator with safety obstacle avoidance constraints to generate a smooth adjustment trajectory, which is then encapsulated as an intelligent assembly control command containing time series and velocity curves. Understandably, the generation of this command ensures the real-time performance and safety of the monitored execution loop.

[0084] In some embodiments, through precise identification of the assembly state and trajectory optimization of predictive control, accurate calculation and safe issuance of the actuator adjustment amount are achieved, effectively ensuring the smoothness and reliability of the six-degree-of-freedom adjustment during the assembly of segmented wind turbine blades.

[0085] Specifically, based on the assembly status assessment report, the current assembly's pose deviation and stress distribution status are identified.

[0086] The system analyzes the pose state estimation and interface stress distribution data in the assembly status assessment report, and uses a threshold segmentation method to identify three-dimensional displacement and three-dimensional rotation deviations that exceed a preset accuracy threshold. Simultaneously, it combines stress distribution heatmaps to mark high-stress areas on the bonding surface and areas with insufficient bolt preload, forming a joint deviation-stress identification result. Furthermore, the identification process updates the deviation and stress vectors in real time, providing structured input for subsequent trajectory calculations.

[0087] Furthermore, the optimal adjustment trajectory of the actuator is calculated through a predictive control model, which comprehensively considers assembly accuracy requirements, dynamic response characteristics of the actuator, and safety constraints.

[0088] Specifically, the predictive control model uses the deviation vector as the state input to construct dynamic constraint equations that include robot joint velocities, accelerations, and end effector loads. The model performs rolling optimization within a look-ahead time window, with the objective function balancing pose convergence speed and stress homogenization, while also incorporating the actuator's maximum stroke, joint torque limits, and component obstacle avoidance distances as hard constraints. The model employs a quadratic programming solver to output a smooth, six-DOF adjustable trajectory, ensuring trajectory continuity and dynamic reachability.

[0089] Furthermore, intelligent assembly control instructions containing six degrees of freedom adjustment parameters are generated, including three-dimensional displacement and three-dimensional rotation angle.

[0090] Specifically, the system discretizes the optimal adjustment trajectory into time-series control points. Each control point encapsulates three-dimensional displacement (X, Y, Z) and three-dimensional rotation angles (Roll, Pitch, Yaw), and adds velocity and acceleration feedforward values. Control commands are packaged in a standardized communication protocol format, including command number, execution duration, and checksum, forming intelligent assembly control commands that can be directly parsed by the assembly execution system.

[0091] Furthermore, the intelligent assembly control commands are sent to the assembly execution system, and the execution feedback is monitored in real time.

[0092] Specifically, commands are transmitted in real time to the assembly execution system via industrial Ethernet, driving the industrial robot to execute each control point sequentially. Simultaneously, a feedback channel is activated to collect the actual pose and joint status of the actuators in real time. The system compares the expected values ​​of the commands with the actual feedback values, calculates the execution error, and records it in a temporary buffer. If the error exceeds the limit, a compensatory replanning is immediately triggered to ensure the closed-loop stability of the assembly process. Understandably, this distribution and feedback mechanism guarantees efficient execution of control commands and rapid response to anomalies.

[0093] In some embodiments, through the deep collaboration between intelligent control commands and deformation compensation, closed-loop verification of assembly accuracy and holographic recording of key parameters are achieved, effectively ensuring the geometric accuracy and mechanical performance consistency of the segmented wind turbine blade docking interface.

[0094] Specifically, the intelligent assembly control command invokes the adaptive deformation compensation mechanism, calculates the accuracy deviation based on assembly path planning information and real-time sensor feedback, and verifies the accuracy of the assembly process.

[0095] Upon receiving intelligent assembly control commands, the system synchronously activates an adaptive deformation compensation mechanism, using assembly path planning information as the theoretical baseline trajectory. Real-time sensor feedback data (including laser displacement, strain, and temperature) is continuously input into the compensation mechanism. By comparing point-by-point with the baseline trajectory, the system calculates the six-degree-of-freedom accuracy deviation and interface stress deviation at the current moment, forming a dynamic deviation field.

[0096] Furthermore, after successful verification, a blade assembly record is generated, which includes component identification, assembly path, accuracy deviation, stress distribution, and assembly status, wherein the blade assembly record is associated with the assembly path planning information.

[0097] Specifically, when the accuracy deviation and stress deviation of multiple consecutive sampling periods all fall within the preset threshold range, the system determines that the verification is successful. The record generation module instantly collects the component's unique identifier, complete assembly path sequence, final accuracy deviation statistics (mean and standard deviation), interface stress distribution cloud map, and assembly status label (qualified / pending repair), and encapsulates them in the form of a structured data packet. All fields are linked to the assembly path planning information through a bidirectional index.

[0098] Furthermore, the blade assembly records are subjected to data integrity verification and timestamp marking to ensure the traceability and immutability of the records.

[0099] Specifically, the system performs hash operations on the blade assembly records to generate digital fingerprints and attaches high-precision timestamps (accurate to milliseconds). The fingerprints and timestamps are written together into the record header, forming a tamper-proof chain. Simultaneously, redundant storage in local caches and the manufacturing data system ensures data integrity and traceability during transmission and archiving. Understandably, this verification and marking mechanism provides a reliable basis for subsequent quality analysis and accountability.

[0100] In some embodiments, the generation of standardized assembly reports and seamless integration with the manufacturing data system enable precise triggering of blade assembly records for subsequent processes and transmission of parameters throughout the entire process, effectively improving the continuity and quality control of segmented wind turbine blade manufacturing.

[0101] Specifically, the blade assembly records are formatted and standardized using data processing algorithms to generate an assembly report that conforms to industry standards.

[0102] The system initiates a data processing algorithm to parse and unify the units of component identifiers, assembly paths, accuracy deviations, stress distributions, and assembly statuses in the blade assembly records. The algorithm formats and transforms the data according to wind power industry standards (such as IEC 61400-5), generating a PDF assembly report containing structured tables, stress contour maps, and path curves, ensuring the report is clear, readable, and meets quality archiving requirements.

[0103] Furthermore, the assembly report is stored in the manufacturing execution system database, and an associated index is established with the product serial number, batch information, and quality records.

[0104] Specifically, the assembly report is uploaded to the Manufacturing Execution System (MES) database via industrial Ethernet, automatically matching the product serial number and batch number of the current blade. The system creates a multi-dimensional correlation index, including serial number-report ID, batch-accuracy statistics, and component identifier-stress profile, achieving two-level data binding at the single blade level and batch level, laying the foundation for subsequent traceability and statistical analysis.

[0105] Furthermore, based on the assembly status in the blade assembly record, subsequent manufacturing processes are triggered, including bonding and curing, flaw detection, and surface treatment.

[0106] Specifically, the system analyzes the assembly status label in real time. When the label is determined to be "qualified," it automatically triggers the process flow instructions in the manufacturing execution system, sequentially initiating the bonding and curing (setting temperature-time curve), flaw detection (ultrasonic phased array scanning), and surface treatment (spraying protective layer) processes. Before each process is started, corresponding parameters are preloaded, such as curing temperature optimized based on stress distribution and flaw detection path focused on areas with accuracy deviation.

[0107] Furthermore, a process coordination mechanism is used to ensure the transmission of parameters and the synchronization of status in each process step.

[0108] Specifically, the process coordination mechanism adopts an event-driven architecture. Upon completion of each process, status codes and key parameters (such as peak curing temperature and flaw detection coordinates) are immediately transmitted back to the manufacturing execution system. The system performs consistency checks on the parameters and updates the blade's lifecycle state machine, ensuring that any abnormality in any stage can halt downstream processes and trigger an alarm. In essence, this coordination mechanism achieves closed-loop data flow from assembly to post-processing and rapid anomaly response, guaranteeing efficient collaboration across the entire manufacturing chain.

[0109] In some embodiments, such as Figure 2 As shown, through closed-loop iterative optimization of deformation prediction and real-time measurement, the optimal execution of adaptive deformation compensation is achieved, effectively eliminating the interference of environmental factors such as gravity and temperature on the assembly accuracy of large-size blades.

[0110] Specifically, an assembly deformation prediction model is established based on finite element analysis, which comprehensively considers the effects of gravity, temperature, and historical assembly data.

[0111] Before assembly begins, the system loads a finite element mesh model of the blade segment assembly, pre-applies gravity loads to simulate sag deformation under cantilever conditions, and generates a temperature deformation field by mapping the thermal expansion coefficient based on real-time temperature sensor data. This model integrates historical assembly data as empirical correction factors to fine-tune the stiffness matrix and boundary conditions, forming a deformation prediction model for the current operating conditions.

[0112] Furthermore, the component pose and interface stress are measured in real time using multi-dimensional sensors, and the deviation between the actual measured value and the predicted value is calculated.

[0113] Specifically, the laser displacement sensor and tilt sensor synchronously acquire the real-time six-degree-of-freedom pose of the components, while the strain gauge sensor continuously monitors the stress distribution at the bonding interface and bolt area. Then, the measured data are compared point by point with the theoretical output of the deformation prediction model to calculate the pose deviation vector and stress deviation scalar, generating a quantitative deviation report.

[0114] Furthermore, a deformation compensation strategy is generated based on the deviation, and the optimal compensation value is calculated through an iterative optimization algorithm.

[0115] Specifically, the system takes the deviation report as input and first generates an initial compensation strategy, including reverse pose adjustment of the robot's end effector and fine-tuning instructions for the local support mechanism. The iterative optimization algorithm uses gradient descent or model predictive control, and converges to the optimal compensation value through multiple iterations while satisfying the constraints of the actuator stroke and allowable material stress, ensuring that the deviation is minimized.

[0116] Furthermore, the optimal compensation value is executed and it is determined whether the assembly process meets the preset accuracy threshold.

[0117] Specifically, the industrial robot and the support mechanism synchronously execute the optimal compensation value, updating the component pose and interface stress in real time. Immediately after execution, the system remeasures and compares the results with an accuracy threshold. If all deviations fall within the threshold range, the assembly process is confirmed to meet the requirements.

[0118] Furthermore, when the accuracy threshold is met, the assembly verification is confirmed to be successful, and a blade assembly record associated with the assembly path planning information is generated.

[0119] Specifically, upon successful verification, the system immediately generates a blade assembly record containing the final compensation value, a comparison of deviations before and after execution, updated stress distribution, and verification status. A unique index is established between the record and the initial assembly path planning information to ensure data traceability throughout the entire process. Understandably, this compensation and verification closed-loop mechanism significantly improves assembly accuracy and environmental adaptability.

[0120] In some embodiments, through automated entry detection and precise feedback correction of measured data, early compensation for manufacturing errors of segmented components and dynamic calibration of assembly paths are achieved, effectively improving the initial accuracy of blade docking and the overall assembly consistency.

[0121] Specifically, a machine vision-based component quality inspection system is used to automatically inspect segmented components. The inspection includes surface defect identification, dimensional deviation measurement, interface flatness assessment, and composite material lamination quality inspection.

[0122] Before the components enter the assembly station, a machine vision system consisting of a high-resolution line scan camera and a structured light projector performs a full-surface scan to automatically identify surface defects such as bubbles, scratches, and fiber breaks. Simultaneously, laser triangulation is used to obtain the contour coordinates of key sections, calculating dimensional deviations from the design model. Digital image processing algorithms are employed to evaluate the flatness and roughness of the interface area, and transmitted light imaging is used to detect lamination debonding and porosity within the composite material.

[0123] Furthermore, a component quality assessment report is generated based on the test results. The quality assessment report includes a pass / fail determination, defect location marking, and geometric measurement data.

[0124] Specifically, the inspection system compares each indicator with the preset tolerance to generate a pass / fail judgment; it annotates the defective area with pixel-level coordinates and includes magnified images; and it outputs a full-section geometric measurement data table, including the actual values ​​of chord length, thickness, and torsion angle. The report is packaged in structured JSON format and includes the inspection timestamp and environmental temperature and humidity records.

[0125] Furthermore, standardized component identifiers are generated for qualified components, and these standardized component identifiers are associated with the component batch number, material properties, and quality grade.

[0126] Specifically, the system automatically generates QR codes or RFID tags for qualified components as standardized component identifiers. The tag content encodes the component batch number, fiber / resin type, curing process parameters, and quality grade (A / B / C). The identifier is permanently bound to the non-pneumatic surface of the component via laser engraving or electronic writing, ensuring traceability throughout the entire process.

[0127] Furthermore, the measured geometric data is used as a correction input for assembly path planning to compensate for the impact of manufacturing errors on assembly accuracy.

[0128] Specifically, geometric measurement data is injected into the dynamic assembly optimization model in real time as initial boundary conditions to correct local deviations in the theoretical CAD model. Based on the measured chord length and torsional differences, the model automatically adjusts the docking attitude and support point positions in the assembly path planning information. The corrected path planning information is directly sent to the assembly execution system, achieving a closed-loop feedback from manufacturing errors to assembly compensation. Understandably, this detection-correction mechanism reduces accumulated errors at the source, improving the overall aerodynamic performance and structural reliability of the blade.

[0129] In some embodiments, the structured archiving and access-controlled querying of full lifecycle traceability information enables secure storage and multi-party collaborative traceability of blade manufacturing data, effectively supporting quality dispute analysis, responsibility determination, and continuous improvement.

[0130] Specifically, manufacturing traceability information is generated based on the blade assembly records. The manufacturing traceability information includes component batch traceability, assembly personnel records, environmental condition logs, quality parameter archives, and process change records.

[0131] After the blade assembly record is confirmed, the system automatically expands to generate a manufacturing traceability information package: extracting batch numbers and material certificate links from standardized component identifiers; reading the operator's employee number and work period from the industrial control system login information; synchronously collecting environmental temperature and humidity sensor logs to form a time series; archiving accuracy deviations, stress distributions, and verification status as quality parameters; and finally adding records of any manual or automatic changes to process parameters. All entries are organized in key-value pairs to ensure a clear and scalable structure.

[0132] Furthermore, the manufacturing traceability information is securely stored using data encryption technology to generate traceability data archives.

[0133] Specifically, the system uses the AES-256 algorithm to symmetrically encrypt the manufacturing traceability information package. The key is dynamically generated by the hardware security module and rotated periodically. After encryption, the data is appended with a digital signature and integrity check code, and encapsulated into an immutable traceability data archive. The archive is stored in a distributed manner on redundant nodes of the manufacturing data system to ensure high availability and prevent single points of failure.

[0134] Furthermore, a role-based access control mechanism is established to provide traceability query interfaces to authorized parties, including customers, quality supervision departments, and after-sales service teams.

[0135] Specifically, the system is configured with a role-based access control matrix: customers have read-only quality parameters and batch information; the quality control department has access to all logs and change records; and the after-sales team has permission to query assembly and environmental data associated with defects. The query interface is implemented via HTTPS and OAuth 2.0 protocols, and authorized parties can log in with a digital certificate to obtain the traceability data archives of the corresponding view.

[0136] Furthermore, based on the aforementioned traceability query interface, it supports quality traceability, fault cause analysis, and responsibility determination throughout the entire manufacturing lifecycle.

[0137] Specifically, authorized parties can trace the entire event chain from incoming inspection and assembly to post-processing by inputting the product serial number or time range through the query interface. The system supports visualized timeline displays and key parameter curve comparisons for fault reproduction. Combined with change records and personnel logs, it automatically generates responsibility analysis reports, providing objective evidence for arbitration and improvement. Understandably, this traceability system achieves a balance between data security and business transparency, comprehensively safeguarding the quality reputation of the blade throughout its entire lifecycle.

[0138] In some embodiments, such as Figure 3As shown, a segmented wind turbine blade manufacturing system 100 based on intelligent assembly control is also provided. Through the modular system architecture and the fine coordination of sub-modules, it realizes full-link intelligent control from data acquisition to process execution, effectively improving the automation level and quality consistency of segmented wind turbine blade manufacturing.

[0139] The data acquisition module 101 is used to acquire segmented component design data of wind turbine blades and generate assembly path planning information based on the dynamic assembly optimization model. The data acquisition module 101 integrates a component quality inspection submodule and a design data interface submodule.

[0140] Specifically, the data acquisition module 101 retrieves the geometric dimensions, composite material properties, and connection interface specifications of the segmented components in real time from the enterprise product data management system through the design data interface submodule; simultaneously, the component quality inspection submodule activates the machine vision system to perform surface defect identification and dimensional measurement on the incoming components. Further, the data acquisition module 101 integrates the measured data with the design data, inputs it into the dynamic assembly optimization model, and generates assembly path planning information including a six-degree-of-freedom trajectory and environmental correction factors.

[0141] The instruction generation module 102 is used to monitor the assembly path planning information in real time through multi-sensor fusion technology and generate intelligent assembly control instructions. The instruction generation module 102 includes a sensor data acquisition unit, a fusion processing unit, and a control instruction calculation unit.

[0142] Specifically, the sensor data acquisition unit coordinates the arrays of laser, tilt, strain, and temperature sensors to upload pose, force, and environmental data in real time; the fusion processing unit performs Kalman filtering to complete time synchronization and spatial registration; and the control command calculation unit calculates the adjustment amount of the actuator based on the fusion report and accuracy threshold, and encapsulates it into intelligent assembly control commands. Furthermore, the commands are output to the assembly execution system in time-series format.

[0143] The accuracy verification module 103 is used to verify the accuracy of the assembly process by combining the intelligent assembly control command with the adaptive deformation compensation mechanism, and to generate a blade assembly record. The accuracy verification module 103 integrates a finite element analysis engine and an error evaluation algorithm.

[0144] Specifically, the accuracy verification module 103 receives intelligent assembly control commands and real-time sensor feedback, calls the finite element analysis engine to predict gravity and temperature deformation, and the error assessment algorithm iteratively calculates the optimal compensation value and drives execution. Further, after successful verification, a blade assembly record containing accuracy deviations, stress distribution, and status labels is generated, and data integrity verification is completed.

[0145] The storage execution module 104 is used to store the blade assembly record to the manufacturing data system and trigger subsequent manufacturing processes based on the blade assembly record. The storage execution module 104 includes a data management unit, a traceability service unit, and a process coordination unit.

[0146] Specifically, the data management unit formats and stores the blade assembly records into the manufacturing data system and establishes a serial number index; the traceability service unit expands and generates a full lifecycle archive; the process coordination unit analyzes the assembly status and triggers bonding curing, flaw detection, and surface treatment processes. Furthermore, the various units collaborate to achieve parameter transmission and status synchronization, ensuring a closed-loop manufacturing process. In essence, the modular system realizes a seamless, data-driven intelligent manufacturing process.

[0147] The above description is merely an exemplary embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention specification and drawings under the technical concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for manufacturing segmented wind turbine blades based on intelligent assembly control, characterized in that, The method includes: Obtain segmented component design data for wind turbine blades and generate assembly path planning information based on a dynamic assembly optimization model; The assembly path planning information is monitored in real time using multi-sensor fusion technology to generate intelligent assembly control commands. The assembly process accuracy is verified by combining the intelligent assembly control commands with an adaptive deformation compensation mechanism, and a blade assembly record is generated. The adaptive deformation compensation mechanism includes: establishing an assembly deformation prediction model based on finite element analysis, which comprehensively considers the effects of gravity, temperature, and historical assembly data; measuring the component pose and interface stress in real time using multi-dimensional sensors, and calculating the deviation between the actual measured value and the predicted value; generating a deformation compensation strategy based on the deviation, and calculating the optimal compensation value through an iterative optimization algorithm; executing the optimal compensation value and determining whether the assembly process meets a preset accuracy threshold; when the accuracy threshold is met, the assembly verification is confirmed to be successful, and a blade assembly record associated with the assembly path planning information is generated. The blade assembly record is stored in the manufacturing data system, and the segmented wind turbine blade is manufactured based on the blade assembly record.

2. The method as described in claim 1, characterized in that, The steps of acquiring segmented component design data of wind turbine blades and generating assembly path planning information based on a dynamic assembly optimization model include: Receive segmented component design data for wind turbine blades, the segmented component design data including component geometry, composite material properties, connection interface specifications, and assembly timestamp; Multi-dimensional feature extraction is performed on the segmented component design data, including component structural stiffness, weight distribution, aerodynamic characteristics and interface compatibility, to generate feature vectors; Based on the dynamic assembly optimization model, assembly path planning information is calculated according to the feature vector and assembly environmental factors. The dynamic assembly optimization model adaptively adjusts the path weights through historical assembly data. The assembly environmental factors include temperature field distribution, gravity load and wind load.

3. The method as described in claim 1, characterized in that, The step of real-time monitoring of the assembly path planning information using multi-sensor fusion technology to generate intelligent assembly control commands includes: A multi-dimensional sensor array is deployed, which includes a laser displacement sensor, a tilt sensor, a strain gauge sensor, and a temperature sensor, to collect component pose data, force data, and environmental data in real time; The data from the multi-dimensional sensor array is processed by a sensor fusion algorithm to generate an assembly status assessment report. The sensor fusion algorithm uses Kalman filtering or particle filtering to achieve time synchronization and spatial registration of multi-source data. Based on the assembly status assessment report and the preset assembly accuracy threshold, the adjustment amount of the actuator is calculated by the control command generation algorithm to generate intelligent assembly control commands.

4. The method as described in claim 3, characterized in that, The control command generation algorithm includes: Based on the assembly status assessment report, identify the current assembly's pose deviation and stress distribution status; The optimal adjustment trajectory of the actuator is calculated by a predictive control model, which comprehensively considers assembly accuracy requirements, dynamic response characteristics of the actuator, and safety constraints. Generate intelligent assembly control instructions containing six degrees of freedom adjustment parameters, including three-dimensional displacement and three-dimensional rotation angle; The intelligent assembly control commands are sent to the assembly execution system, and the execution feedback is monitored in real time.

5. The method as described in claim 1, characterized in that, The step of verifying the accuracy of the assembly process and generating a blade assembly record by combining the intelligent assembly control command with the adaptive deformation compensation mechanism includes: The intelligent assembly control command invokes the adaptive deformation compensation mechanism, calculates the accuracy deviation based on assembly path planning information and real-time sensor feedback, and verifies the accuracy of the assembly process. After successful verification, a blade assembly record is generated, which includes component identification, assembly path, accuracy deviation, stress distribution, and assembly status. The blade assembly record is associated with the assembly path planning information. The blade assembly records are subjected to data integrity verification and timestamp marking to ensure the traceability and immutability of the records.

6. The method as described in claim 1, characterized in that, The step of storing the blade assembly record to the manufacturing data system and completing the segmented wind turbine blade manufacturing based on the blade assembly record includes: The blade assembly record is formatted and standardized using data processing algorithms to generate an assembly report that conforms to industry standards. The assembly report is stored in the manufacturing execution system database, and an associated index is established with the product serial number, batch information, and quality file. Based on the assembly status in the blade assembly record, the subsequent manufacturing process is triggered, which includes bonding and curing, flaw detection and surface treatment. A process coordination mechanism ensures the transmission of parameters and synchronization of status in each process step.

7. The method as described in claim 1, characterized in that, Before the step of obtaining the segmented component design data of the wind turbine blade, the method further includes: The component quality inspection system based on machine vision is used to automatically inspect segmented components. The inspection includes surface defect identification, dimensional deviation measurement, interface flatness assessment and composite material lamination quality inspection. A component quality assessment report is generated based on the test results. The quality assessment report includes a pass / fail determination, defect location marking, and geometric measurement data. A standardized component identifier is generated for qualified components, and the standardized component identifier is associated with the component batch number, material properties and quality grade; The measured geometric data is used as a correction input for assembly path planning to compensate for the impact of manufacturing errors on assembly accuracy.

8. The method as described in claim 1, characterized in that, After the step of storing the blade assembly record to the manufacturing data system, the method further includes: Manufacturing traceability information is generated based on the blade assembly record. The manufacturing traceability information includes component batch traceability, assembly personnel record, environmental condition log, quality parameter archive, and process change record. The manufacturing traceability information is securely stored using data encryption technology to generate a traceability data archive; Establish a role-based access control mechanism and provide a traceability query interface to authorized parties, including customers, quality supervision departments, and after-sales service teams; Based on the aforementioned traceability query interface, it supports quality traceability, fault cause analysis, and responsibility identification throughout the entire manufacturing lifecycle.

9. A segmented wind turbine blade manufacturing system based on intelligent assembly control, characterized in that, The system includes: The data acquisition module is used to acquire segmented component design data of wind turbine blades and generate assembly path planning information based on the dynamic assembly optimization model. The data acquisition module integrates a component quality inspection submodule and a design data interface submodule. The instruction generation module is used to monitor the assembly path planning information in real time through multi-sensor fusion technology and generate intelligent assembly control instructions. The instruction generation module includes a sensor data acquisition unit, a fusion processing unit, and a control instruction calculation unit. The accuracy verification module is used to verify the accuracy of the assembly process through the intelligent assembly control commands combined with an adaptive deformation compensation mechanism, and to generate a blade assembly record. The adaptive deformation compensation mechanism includes: establishing an assembly deformation prediction model based on finite element analysis, which comprehensively considers the effects of gravity, temperature, and historical assembly data; measuring the component pose and interface stress in real time using multi-dimensional sensors, and calculating the deviation between the actual measured value and the predicted value; generating a deformation compensation strategy based on the deviation, and calculating the optimal compensation value through an iterative optimization algorithm; executing the optimal compensation value and determining whether the assembly process meets a preset accuracy threshold; when the accuracy threshold is met, confirming that the assembly verification is passed and generating a blade assembly record. The accuracy verification module integrates a finite element analysis engine and an error evaluation algorithm. The storage execution module is used to store the blade assembly record to the manufacturing data system and trigger subsequent manufacturing processes based on the blade assembly record. The storage execution module includes a data management unit, a traceability service unit, and a process coordination unit.

Images