Reconstruction method and device for wind turbine complete machine manufacturing system based on digital twinning

Abstract

This invention provides a reconstruction method and apparatus for a wind turbine manufacturing system based on digital twins. By acquiring historical data to construct a production prediction function, the theoretically optimal alternative process parameters are first solved mathematically. Then, the alternative parameters are dynamically simulated and corrected with high fidelity using a digital twin system. Finally, the verified reconstruction parameters are applied to the physical system. By combining data-driven global optimization capabilities with model-driven detailed verification capabilities, a seamless connection between virtual pre-simulation and physical execution of the wind turbine manufacturing process is achieved. This shortens the iteration cycle, reduces iteration costs, and enables precise adjustment of process parameters, avoiding downtime risks and production interruptions caused by direct adjustments on the physical production line. Consequently, the efficiency of wind turbine manufacturing and the stability of product quality are improved.

Description

Technical Field

[0001] This invention relates to the field of wind power equipment manufacturing technology, and in particular to a reconstruction method and apparatus for a wind turbine manufacturing system based on digital twins. Background Technology

[0002] Against the backdrop of global energy transition, wind power, with its significant advantages such as being clean and renewable, has become a key force in adjusting the energy structure. With continuous technological advancements, wind power is rapidly moving towards higher power output, lighter weight, and greater intelligence. This trend places more stringent demands on wind turbine manufacturing, making the manufacturing process for wind turbines, including core components such as blades, hubs, nacelles, and towers, increasingly complex, involving numerous challenges such as multi-process coordination, multi-equipment linkage, and multi-parameter coupling. In the manufacturing process of wind turbines, the setting and adjustment of many process parameters rely heavily on the practical production experience of engineers and front-line operators.

[0003] However, relying on actual production experience for process iteration means that continuous trial and error is required during the production process, which leads to high trial and error costs and long cycles in the wind turbine manufacturing system. Summary of the Invention

[0004] This invention provides a reconstruction method and apparatus for a wind turbine manufacturing system based on digital twins, which solves the defects of existing technologies that rely on actual generation experience for process iteration, resulting in high trial and error costs and long cycles in wind turbine manufacturing systems.

[0005] This invention provides a method for reconstructing a wind turbine manufacturing system based on digital twins, comprising: Obtain historical product indicators corresponding to historical process parameters of the wind turbine manufacturing system; Based on the historical process parameters and the historical product indicators, a production prediction function is constructed between the product indicators and the process parameters of the wind turbine manufacturing system. With the goal of optimizing the product indicators, the production forecast function is solved to obtain a set of alternative process parameters; The set of alternative process parameters is input into the digital twin system for simulation to obtain simulation results; Based on the simulation results, the candidate process parameters in the set of candidate process parameters are corrected to obtain reconstructed process parameters; the reconstructed process parameters are applied to the wind turbine manufacturing system.

[0006] According to the present invention, a method for reconstructing a wind turbine manufacturing system based on digital twins, wherein the reconstructed process parameters are obtained by correcting the candidate process parameters in the candidate process parameter set based on the simulation results, including: The simulation results are compared with preset indicators to obtain comparison results; If the comparison result shows a deviation, the candidate process parameters are corrected until the comparison result corresponding to the corrected process parameters has no deviation, and the corrected process parameters are used as the reconstructed process parameters.

[0007] According to the present invention, a method for reconstructing a wind turbine manufacturing system based on digital twins includes inputting the set of alternative process parameters into the digital twin system for simulation to obtain simulation results, comprising: Candidate process parameters are selected from the set of alternative process parameters. The candidate process parameters are input into the digital twin system. Based on the digital twin system, at least one of the following adjustments is made to the process candidate parameters corresponding to each process step: equipment layout adjustment, link logic adjustment, and scheduling rule adjustment, to obtain the simulation results.

[0008] According to the present invention, a reconstruction method for a wind turbine manufacturing system based on digital twin is provided, wherein the digital twin system includes a virtual device layer, a virtual link layer, and a virtual scheduling layer; The virtual device layer is used to map the equipment layout features of the wind turbine manufacturing system; The virtual link layer is used to simulate material transfer and signal interaction between equipment in the wind turbine manufacturing system; The virtual scheduling layer is used to map scheduling rules for personnel, materials, and equipment.

[0009] According to the present invention, a reconstruction method for a wind turbine manufacturing system based on digital twins is provided, wherein the historical process parameters include the historical process parameters of each process. The step of constructing a production prediction function between product indicators and the process parameters of the wind turbine manufacturing system based on the historical process parameters and historical product indicators includes: The production prediction function is constructed based on the historical parameters of each process under the overall process, the process measurement error of each process, the overall process measurement error, and the historical product indicators.

[0010] According to the present invention, a reconstruction method for a wind turbine manufacturing system based on digital twins is provided, wherein the product indicators include production efficiency indicators, production cost indicators, and production quality indicators.

[0011] According to the present invention, a reconstruction method for a wind turbine manufacturing system based on digital twins is provided, wherein the method involves correcting the candidate process parameters in the candidate process parameter set based on the simulation results to obtain reconstructed process parameters, and then includes: Based on the reconfigured process parameters, process adjustment instructions are constructed; The process adjustment instruction is sent to the wind turbine manufacturing system to instruct the wind turbine manufacturing system to produce according to the process adjustment instruction.

[0012] The present invention also provides a reconfiguration device for a wind turbine manufacturing system based on digital twins, comprising: The acquisition unit acquires historical product indicators corresponding to historical process parameters of the wind turbine manufacturing system. The function construction unit constructs a production prediction function between the product indicators and the process parameters of the wind turbine manufacturing system based on the historical process parameters and the historical product indicators. The optimization unit, with the goal of optimizing the product indicators, solves the production prediction function to obtain a set of alternative process parameters; The simulation unit inputs the set of alternative process parameters into the digital twin system for simulation and obtains simulation results. The reconstruction unit modifies the candidate process parameters in the candidate process parameter set based on the simulation results to obtain the reconstructed process parameters; the reconstructed process parameters are applied to the wind turbine manufacturing system.

[0013] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the reconstruction method of the wind turbine manufacturing system based on digital twin as described above.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the reconstruction method of the wind turbine manufacturing system based on digital twin as described above.

[0015] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the reconstruction method of the wind turbine manufacturing system based on digital twin as described above.

[0016] The present invention provides a reconstruction method and apparatus for a wind turbine manufacturing system based on digital twins. By acquiring historical data to construct a production prediction function, the theoretically optimal alternative process parameters are first solved mathematically. Then, the alternative parameters are dynamically simulated and corrected with high fidelity using a digital twin system. Finally, the verified reconstruction parameters are applied to the physical system. By combining data-driven global optimization capabilities with model-driven detailed verification capabilities, the virtual pre-simulation and physical execution of the wind turbine manufacturing process are seamlessly integrated. This shortens the iteration cycle, reduces iteration costs, and enables precise adjustment of process parameters, avoiding downtime risks and production interruptions caused by direct adjustments on the physical production line. Consequently, the efficiency of wind turbine manufacturing and the stability of product quality are improved. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the reconstruction method of a wind turbine manufacturing system based on digital twins provided in an embodiment of the present invention. Figure 2 This is a virtual reconfiguration system architecture diagram of the wind turbine manufacturing system based on digital twins provided by the present invention; Figure 3 This is a schematic diagram of the structure of the reconstruction device for the wind turbine manufacturing system based on digital twins provided by the present invention; Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] To address the aforementioned issues, this invention provides a reconstruction method for a wind turbine manufacturing system based on digital twins. By combining historical data-driven approaches with digital twin simulation verification, it achieves rapid, low-cost, and accurate reconstruction of process parameters. Figure 1 This is a flowchart illustrating the reconstruction method of a wind turbine manufacturing system based on digital twins provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes: Step 110: Obtain the historical product indicators corresponding to the historical process parameters of the wind turbine manufacturing system.

[0021] Here, the wind turbine manufacturing system refers to the physical manufacturing workshop or production line used to produce core components of wind turbine generators such as blades, hubs, nacelles, and towers. It includes various processing equipment, material handling equipment, and assembly stations. Historical process parameters refer to the various manufacturing conditions and control variables actually executed in past production batches. In practice, historical process parameters may include, but are not limited to, process sequences, equipment operating parameters, and material flow logic. Historical product indicators refer to the quantitative performance data of the final wind turbine product or component produced under the influence of historical process parameters.

[0022] Specifically, when adjustments to the process parameters of a wind turbine manufacturing system are needed, the first step is to use IoT data acquisition devices deployed within the system to retrieve historical production logs stored in a database. From these logs, historical process parameters and their corresponding historical product indicators can be obtained. It's understandable that wind turbine manufacturing involves multi-process collaboration; therefore, the acquired data should be multi-source, heterogeneous data with temporal correlation. For example, for the j-th process of the i-th wind turbine, the specific process execution parameters can be obtained, and this data can be correlated with quality inspection data, time records, and cost accounting data after the process is completed or the turbine rolls off the production line, thus forming a mapping dataset between process parameters and product indicators.

[0023] Step 120: Based on the historical process parameters and the historical product indicators, construct a production prediction function between the product indicators and the process parameters of the wind turbine manufacturing system.

[0024] Here, the production forecasting function refers to a mathematical model that can characterize the mapping relationship between process parameters as input variables and product indicators as output variables in the manufacturing process, so as to quantify the impact of the process on the results through mathematical means, thereby getting rid of simple human experience judgment.

[0025] Specifically, machine learning algorithms can be used to train and fit historical product indicators corresponding to the historical process parameters obtained in step 110, thereby constructing a production prediction function between the product indicators and the process parameters of the wind turbine manufacturing system. It can be understood that the constructed production prediction function establishes a mathematical mapping relationship between each stage of the wind turbine manufacturing process and the final result. For example, process parameters can be set as independent variables, with production efficiency, cost, and quality as dependent variables for modeling. It can also be understood that the production prediction function can be not only a single formula but also a function group containing multiple sub-models, each used to predict product indicators in different dimensions. It should be noted that by constructing the production prediction function, given a set of input process parameters, the theoretically expected product indicators can be calculated, providing an evaluation benchmark for subsequent parameter optimization.

[0026] Step 130: With the goal of optimizing the product indicators, the production prediction function is solved to obtain a set of alternative process parameters.

[0027] Here, the alternative process parameter set refers to a set or more combinations of process parameters that have been selected through theoretical calculations and are expected to achieve better production results, but have not yet been verified by virtual simulation.

[0028] Specifically, the production forecast function can be transformed into a multi-objective optimization problem, and constraints can be set according to the actual physical constraints of the wind turbine manufacturing system, such as upper and lower limits of equipment parameters and reasonable range of measurement errors. Then, optimization algorithms can be used to solve the production forecast function. For example, optimization algorithms such as genetic algorithms, particle swarm optimization, or community partitioning algorithms can be used. During the solution process, the optimization algorithm will search for the solution vector that makes the objective function value optimal within the feasible region. The solution vectors with the highest ranking can be selected to form a set of candidate process parameters. For example, for the production of a new type of wind turbine, the set of candidate process parameters obtained may include a specific hub welding sequence and the corresponding robotic arm movement speed.

[0029] Step 140: Input the set of alternative process parameters into the digital twin system for simulation and obtain simulation results.

[0030] Here, a digital twin system refers to a virtual system constructed in virtual space that maintains a high degree of consistency with the physical wind turbine manufacturing system in terms of geometry, physical characteristics, and behavior. Simulation results refer to the virtual production data generated after running alternative process parameter sets in the digital twin system, including but not limited to virtual production capacity, virtual assembly status, and virtual energy consumption.

[0031] Specifically, a set of alternative process parameters can be selected from the theoretically optimal set of candidate process parameters and loaded into the digital twin system. The digital twin system then simulates the real physical manufacturing process, such as different process routes, like the sequence of blade forming and the execution of wheel hub welding. During the simulation, the digital twin system can perform dynamic simulations of all elements, checking for smooth coordination between equipment, congestion in material transport, and the existence of spatial collisions—dynamic problems that are difficult to predict with simple functions in the physical model. This allows for the identification of potential problems with the alternative parameters under complex actual operating conditions.

[0032] Step 150: Based on the simulation results, the candidate process parameters in the candidate process parameter set are corrected to obtain the reconstructed process parameters.

[0033] The reconfiguration process parameters are applied to the wind turbine manufacturing system.

[0034] Here, reconfigured process parameters refer to the process parameters and system configuration commands that are finally determined after virtual simulation verification and necessary adjustments, and can be directly issued to the physical system for execution.

[0035] Specifically, the simulation results are first compared with the design standards or expected goals. If the simulation results show deviations, such as excessive dimensional deviations of the virtual product, bottlenecks in the production line, or equipment collisions, the alternative process parameters are fine-tuned according to the type of deviation. For example, the equipment layout coordinates may be adjusted, the link logic rules modified, or the scheduling strategy changed. The adjusted parameters can be simulated again until the simulation results meet the design requirements. The parameters that finally meet the requirements are the reconstructed process parameters. Subsequently, the reconstructed process parameters are sent to the physical wind turbine manufacturing system through the virtual-physical interaction interface of the digital twin system. Finally, the physical system can automatically adjust the equipment status based on these parameters, such as updating the robot's motion trajectory and adjusting the conveyor speed, thereby realizing the dynamic reconstruction of the physical manufacturing system and the implementation of the process.

[0036] The method provided in this invention constructs a production prediction function by acquiring historical data. First, it mathematically solves for the theoretically optimal alternative process parameters. Then, it uses a digital twin system to perform high-fidelity dynamic simulation and correction of the alternative parameters. Finally, it applies the verified reconstructed parameters to the physical system. By combining data-driven global optimization capabilities with model-driven detailed verification capabilities, it achieves seamless integration of virtual pre-simulation and physical execution of wind turbine manufacturing processes. This shortens the iteration cycle, reduces iteration costs, and enables precise and dynamic adjustment of process parameters. It avoids the downtime risks and production interruptions caused by direct adjustments on the physical production line, thereby improving the efficiency of wind turbine manufacturing and the stability of product quality.

[0037] Based on any of the above embodiments, step 150 includes: The simulation results are compared with preset indicators to obtain comparison results; If the comparison result shows a deviation, the candidate process parameters are corrected until the comparison result corresponding to the corrected process parameters has no deviation, and the corrected process parameters are used as the reconstructed process parameters.

[0038] Here, preset indicators refer to the technical standards or quality thresholds used to define the manufactured product, serving as a baseline for judging whether virtual production is qualified. Preset indicators can include geometric tolerances, physical performance indicators, and stability indicators of the production process. For example, in a blade production scenario, a preset indicator could be the final dimensional deviation range of the blade; in a tower welding scenario, it could be the minimum standard value for weld strength. Here, the comparison result refers to the difference between the simulation output value and the preset indicator value, or the logical judgment status, such as qualified or unqualified.

[0039] Specifically, after a full-process simulation is completed by the digital twin system, key quality data of the virtual product can be automatically extracted and used as the simulation results. These results are then matched and verified item by item against preset indicators stored in the database. For example, the degree of curing data after virtual blade forming can be extracted and compared with the curing range required by standard processes; or the amount of thermal deformation after welding of a virtual tower can be extracted and compared with the maximum allowable deformation. Understandably, through this comparison, it is possible to accurately identify which specific process steps or quality dimensions of the alternative process parameter set fail to meet design requirements.

[0040] When the comparison results show deviations, the source of the deviation in the process can be located first. For example, if the blade size deviation is found to be too large, it may be identified as insufficient holding time or uneven temperature distribution in the molding process. Next, the corresponding process parameters can be adjusted. Based on the direction and degree of the deviation, specific parameters in the set of alternative process parameters can be fine-tuned using preset process adjustment rules or expert knowledge bases, such as extending the holding time or adjusting the heating curve, to generate corrected process parameters.

[0041] Subsequently, the digital twin system can be triggered again to perform a new round of simulation using the corrected parameters, and new simulation results can be obtained for comparison. This process may be repeated multiple times until the simulation results meet the preset requirements. At this point, the parameter combination confirmed by the last iteration is locked as the reconstructed process parameters and is ready to be sent to the physical system.

[0042] It should be noted that deviation in the comparison results means that the simulation results exceed the allowable tolerance range of the preset indicators. This implies that if the current alternative process parameters are directly used for physical production, it is highly likely to produce defective products or cause production accidents. Conversely, no deviation in the comparison results corresponding to the corrected process parameters means that after one or more rounds of iterative processes of parameter adjustment, re-simulation, and result comparison, all indicators of the virtual simulation results meet the allowable range of the preset indicators.

[0043] The method provided in this invention corrects candidate process parameters when there is a deviation between the simulation results and preset indicators, until the comparison results corresponding to the corrected process parameters are without deviation. The corrected process parameters are then used as the reconstructed process parameters. This ensures that the process parameters ultimately applied to the physical system are not only theoretically optimal but also verified as feasible and meet quality standards in virtual reality. In other words, by performing high-frequency closed-loop iterative corrections in virtual space, the expensive and time-consuming trial-and-error process in physical production is replaced. This effectively solves the product quality fluctuation problem caused by the disconnect between the physical manufacturing process and virtual simulation in existing technologies, ensuring the accurate implementation of wind turbine manufacturing processes and the consistency of product quality.

[0044] Based on any of the above embodiments, step 140 includes: Candidate process parameters are selected from the set of alternative process parameters. The candidate process parameters are input into the digital twin system. Based on the digital twin system, at least one of the following adjustments is made to the process candidate parameters corresponding to each process step: equipment layout adjustment, link logic adjustment, and scheduling rule adjustment, to obtain the simulation results.

[0045] Here, candidate process parameters refer to a specific set of process routes and parameter combinations selected from the set of multiple alternative process parameters obtained in step 130 for the current simulation verification. For example, for a specific model of wind turbine, the candidate set may contain multiple process routes, such as different hub welding sequences. In this case, one set needs to be selected as a candidate for the simulation stage, either individually or according to priority.

[0046] In addition, equipment layout adjustment here refers to changing the physical location, orientation, or spatial configuration of virtual equipment in the virtual environment according to the requirements of the new process. Link logic adjustment here refers to modifying the material transfer paths, signal interaction logic, or control command flow between virtual devices. Scheduling rule adjustment here refers to reallocating the work sequence and collaboration mode of personnel, materials, and equipment based on the new process flow.

[0047] Specifically, once candidate process parameters are input, they are fed into the digital twin system, which can identify the physical environment requirements for execution based on these parameters. If the candidate process parameters involve the introduction of a new model or significant changes to the process flow, such as adding a blade production line or altering the "symmetrical welding-repair welding" process for wheel hub welding, the virtual manufacturing system can then be reconstructed automatically or with manual assistance through the digital twin system.

[0048] The refactoring process includes: For equipment layout adjustments, the relative positions of devices such as robotic welding stations and material conveyor lines can be adjusted at the virtual equipment layer to adapt to new welding trajectories and prevent motion interference. For link logic adjustments, material transfer and signal interaction between devices can be simulated at the virtual link layer, such as updating the start / stop logic of conveyor lines to match the new production cycle. For scheduling rule adjustments, intelligent algorithms can be used at the virtual scheduling layer to dynamically optimize the scheduling of personnel, materials, and equipment, generating the optimal production schedule that adapts to the current candidate process parameters.

[0049] After completing one or more adjustments in the refactoring process, the simulation is initiated to calculate the actual production capacity, efficiency, and energy consumption of the candidate process parameters under the refactored virtual system architecture, thus obtaining the simulation results. For example, process routes can be extracted from the set of candidate process parameters, and different process routes can be simulated using a digital twin model to calculate the efficiency, cost, and turbine quality of the production equipment for these process routes, thus obtaining the simulation results. In addition, for multi-model adaptation scenarios, parameters from the set of candidate process parameters can be quickly called up through the twin model, and process parameters adapted to new models can be iteratively applied.

[0050] The method provided in this invention, by introducing dynamic adjustments to equipment layout, link logic, and scheduling rules during the simulation phase, breaks the limitation of traditional simulations that only verify process parameters while ignoring the physical configuration constraints of the system. It achieves a deep integration of process optimization and system reconfiguration in wind turbine manufacturing, not only verifying the theoretical feasibility of process parameters, but also pre-simulating the physical and logical reconfiguration process required for the manufacturing system to adapt to the process in virtual space. This ensures seamless integration of the reconfiguration scheme when it is physically implemented, avoids production stoppages caused by system configuration mismatch, and greatly improves the manufacturing system's ability to quickly switch between multiple models.

[0051] Based on any of the above embodiments, the digital twin system includes a virtual device layer, a virtual link layer, and a virtual scheduling layer; The virtual device layer is used to map the equipment layout features of the wind turbine manufacturing system; The virtual link layer is used to simulate material transfer and signal interaction between equipment in the wind turbine manufacturing system; The virtual scheduling layer is used to map scheduling rules for personnel, materials, and equipment.

[0052] Here, the virtual device layer refers to the geometric shape and spatial location of various production resources in the digital twin system. The device layout features here encompass the device's geometric dimensions, physical attributes, motion behavior characteristics, and spatial location information in the shop floor coordinate system.

[0053] Specifically, in the virtual equipment layer, high-fidelity 3D geometric models can be constructed for core equipment involved in wind turbine manufacturing, such as blade molds, welding robots, and assembly platforms. This layer carries the layout information of the physical system, that is, the placement position and relative distance of each piece of equipment in the virtual workshop. Understandably, when adjusting the equipment layout, this is mainly achieved by modifying the coordinate parameters of the virtual equipment or rearranging the equipment models within this virtual equipment layer. For example, when simulating the addition of a new blade production line, its location in virtual space is planned and its layout features, such as safety distances, are verified relative to existing equipment.

[0054] Here, the virtual link layer is responsible for defining the flow paths and control logic of production elements between different devices. Material transfer here involves the flow direction of raw materials and semi-finished products between devices; signal interaction here involves the transmission logic of PLC control signals, sensor trigger signals, etc., between devices.

[0055] Specifically, the virtual link layer is built on top of the virtual device layer and defines the connection relationships between devices. For example, it simulates the logical path of a conveyor belt transporting a wheel hub from the cleaning station to the welding station, as well as the signal handshake logic that triggers the welding robot to start when the wheel hub arrives. It can be understood that adjusting the link logic involves modifying the connection topology or signal triggering conditions within this layer, such as changing the flow direction of the material conveyor or adjusting the logical sequence of multiple devices to adapt to the new process flow.

[0056] Here, the virtual scheduling layer is responsible for the dynamic allocation of production resources and task assignment. The scheduling rules here involve the algorithmic logic of how to optimally arrange personnel shifts, material delivery timing, and equipment operation queues in a complex manufacturing environment with multiple concurrent processes and multiple resource competition.

[0057] Specifically, the virtual scheduling layer integrates intelligent scheduling algorithms, such as community partitioning algorithms. This layer calculates and simulates the dynamic scheduling process of personnel, materials, and equipment based on production order demands. For example, when blade production and tower manufacturing simultaneously compete for overhead crane resources, this layer makes decisions based on preset priority rules. It can be understood that adjusting scheduling rules involves reconfiguring production resources within this layer by adjusting algorithm parameters or switching scheduling strategies, thereby verifying the production efficiency under different scheduling schemes.

[0058] The system architecture provided in this invention, by constructing a layered digital twin system comprising a virtual device layer, a virtual link layer, and a virtual scheduling layer, achieves full-element, multi-dimensional mapping of the physical manufacturing system. Specifically, the virtual device layer ensures the rationality and collision-free nature of the spatial layout; the virtual link layer ensures the connectivity and deadlock-free nature of logical control; and the virtual scheduling layer ensures the efficiency and conflict-free nature of resource allocation. Thus, through the collaborative work of this three-layer architecture, the digital twin system can comprehensively support the dynamic virtual reconstruction of manufacturing processes and systems from geometric, logical, and management dimensions, greatly improving the accuracy and executability of the reconstruction scheme.

[0059] Based on any of the above embodiments, the historical process parameters include the historical process parameters of each process step; Step 120 includes: The production prediction function is constructed based on the historical parameters of each process under the overall process, the process measurement error of each process, the overall process measurement error, and the historical product indicators.

[0060] Specifically, based on historical data and technological mechanisms, a mathematical mapping relationship can be established between each stage of the wind turbine manufacturing process and the final result. Each process stage can be represented as follows: The final results include efficiency, cost, and quality, represented by the following symbols: , , It indicates. Among them. , , , Subscript of the equals symbol Represented as the first A wind turbine unit, with the following label Indicates the first step in the entire process flow. Each process step. Therefore, with the goals of maximizing efficiency, minimizing cost, and optimizing quality, let the set of process parameters represent all processes. (The j-th process of the i-th wind turbine), the process measurement error of each process is... The overall process measurement error is The production forecast function can be expressed as: As the formula shows, this set of functions establishes a mapping relationship between process steps, errors, and the final product's production efficiency, cost, and quality. Therefore, by simultaneously optimizing efficiency, cost, and quality, the multi-objective optimal configuration of the wind turbine manufacturing process can be achieved.

[0061] The method provided in this invention considers measurement errors at both the process and system levels during function construction, making the production prediction function no longer a mathematical formula under an idealized environment, but a robust engineering model that can more realistically reflect the fluctuations in the physical manufacturing process. This ensures that the alternative process parameters subsequently solved based on the function have higher reliability and anti-interference capabilities in actual production, and avoids the failure of the theoretical optimal solution in practical applications due to ignoring errors.

[0062] Based on any of the above embodiments, product indicators include production efficiency indicators, production cost indicators, and production quality indicators.

[0063] Specifically, in constructing the production prediction function and subsequent solution process, this embodiment of the invention uses the aforementioned three indicators as core elements of multi-objective optimization. For example, it constructs a set of objective functions that include maximizing efficiency, minimizing cost, and optimizing quality. In actual wind turbine manufacturing, these three indicators are often mutually restrictive; for example, increasing welding speed may improve efficiency but reduce quality. Therefore, by clearly defining these three dimensions of indicators, the optimal balance point among them can be found during parameter optimization, i.e., the optimal solution can be obtained.

[0064] The method provided in this invention, by clearly defining product indicators including production efficiency indicators, production cost indicators, and production quality indicators, achieves a comprehensive evaluation of the wind turbine manufacturing process. This avoids the one-sided optimization of solely pursuing production capacity while neglecting quality or cost, and ensures that the final reconstructed process parameters can achieve the comprehensive optimal configuration for wind turbine manufacturing.

[0065] Based on any of the above embodiments, step 150 is followed by: Based on the reconfigured process parameters, process adjustment instructions are constructed; The process adjustment instruction is sent to the wind turbine manufacturing system to instruct the wind turbine manufacturing system to produce according to the process adjustment instruction.

[0066] Specifically, after obtaining the reconfigured process parameters, the final confirmed reconfigured process parameters can be converted into a specific sequence of control commands. For example, the adjusted blade molding temperature curve can be converted into the setpoint of the temperature control system and written into the command. Then, the process adjustment command can be sent to the execution layer of the physical manufacturing site in real time. After receiving the command, the physical wind turbine manufacturing system automatically adjusts the equipment parameters, switches the processing program, or updates the logistics scheduling logic, thereby starting production according to the new process plan.

[0067] The method provided in this invention converts the reconstructed parameters after virtual verification into executable physical instructions and automatically issues them, realizing virtual-physical linkage execution. It connects the digital twin model to the physical entity for feedback control, eliminating the need for manual input of parameters one by one at the device end. This not only greatly improves the efficiency of system reconstruction and achieves millisecond-level synchronous response, but also eliminates the risk of misoperation that may be introduced by manual operation, truly realizing intelligent closed-loop control of the wind turbine manufacturing process.

[0068] Based on any of the above embodiments Figure 2 This is a virtual reconfiguration system architecture diagram of the wind turbine manufacturing system based on digital twins provided by this invention, such as... Figure 2 As shown, the overall architecture of the system mainly consists of two parts: the real physical world and the twin space. The two interact in a closed loop through real-time data transmission and result feedback.

[0069] The real-world representation on the left side of the architecture diagram encompasses the entire operational process of the wind turbine manufacturing pipeline. This pipeline involves the manufacturing and assembly of core components of the wind turbine, such as the blades, tower, nacelle, and base shown in the diagram. At the physical manufacturing site, IoT sensing devices are deployed to perform real-time data sensing and exchange, collecting comprehensive data including equipment operating status, material transfer locations, personnel operation behaviors, and tooling fixture status. This data forms the foundation for the real-time state mapping of the physical manufacturing system.

[0070] It should be noted that the arrow in the middle of the architecture diagram illustrates the data flow of the virtual-physical interaction. The massive amounts of manufacturing data perceived by the physical world are synchronized to the twin space on the right through a real-time transmission channel with millisecond-level low latency.

[0071] In the twin space on the right side of the architecture diagram, real-time data synchronization is first performed to ensure consistency between the virtual model and the physical entity. Based on this, a twin model of the wind turbine manufacturing process pipeline is constructed. This modeling process includes the construction of a virtual equipment layer, a virtual link layer, and a virtual scheduling layer, realizing a digital mirror of the physical manufacturing process. Within the twin model, dynamic simulation can be performed based on input alternative process parameters, and simulation results can be output.

[0072] The optimization feedback in the diagram aims to achieve "maximum wind volume and strongest resistance to harsh environments" as the ultimate product performance optimization goal. This can be understood as a high standard requirement for production quality indicators, namely, ensuring the power generation efficiency and structural strength of the finished wind turbine through manufacturing process optimization. Furthermore, the simulation results are evaluated by combining production efficiency and cost indicators. If the simulation results do not meet the standards, the process parameters or system configuration are adjusted through the optimization feedback mechanism, and the simulation is performed again.

[0073] Once the simulation results in the twin space meet the preset indicators, the verified reconstructed process parameters or control commands are sent back to the real physical world through the result feedback channel. After receiving the commands, the wind turbine manufacturing process line in the physical world automatically executes process optimization actions, such as adjusting robot trajectories and conveyor speeds, thereby realizing a complete dynamic virtual reconstruction closed loop from physical data perception to virtual simulation optimization and then to physical entity execution.

[0074] Based on any of the above embodiments Figure 3 This is a schematic diagram of the structure of the reconfiguration device for a wind turbine manufacturing system based on digital twins provided by the present invention, as shown below. Figure 3 As shown, the device includes: Unit 310 acquires historical product indicators corresponding to historical process parameters of the wind turbine manufacturing system. Function construction unit 320 constructs a production prediction function between product indicators and process parameters of the wind turbine manufacturing system based on the historical process parameters and the historical product indicators. The optimization unit 330, with the goal of optimizing the product indicators, solves the production prediction function to obtain a set of alternative process parameters; Simulation unit 340 inputs the set of alternative process parameters into the digital twin system for simulation and obtains simulation results; The reconstruction unit 350 corrects the candidate process parameters in the candidate process parameter set based on the simulation results to obtain the reconstructed process parameters; the reconstructed process parameters are applied to the wind turbine manufacturing system.

[0075] The apparatus provided in this invention constructs a production prediction function by acquiring historical data. First, it mathematically solves for the theoretically optimal alternative process parameters. Then, it uses a digital twin system to perform high-fidelity dynamic simulation and correction of the alternative parameters. Finally, it applies the verified reconstructed parameters to the physical system. By combining data-driven global optimization capabilities with model-driven detailed verification capabilities, it achieves seamless integration of virtual pre-simulation and physical execution of wind turbine manufacturing processes. This shortens the iteration cycle, reduces iteration costs, and enables precise adjustment of process parameters, avoiding downtime risks and production interruptions caused by direct adjustments on the physical production line. Consequently, it improves the efficiency of wind turbine manufacturing and the stability of product quality.

[0076] Based on any of the above embodiments, the reconstruction unit is specifically used for: The simulation results are compared with preset indicators to obtain comparison results; If the comparison result shows a deviation, the candidate process parameters are corrected until the comparison result corresponding to the corrected process parameters has no deviation, and the corrected process parameters are used as the reconstructed process parameters.

[0077] Based on any of the above embodiments, the simulation unit is specifically used for: Candidate process parameters are selected from the set of alternative process parameters. The candidate process parameters are input into the digital twin system. Based on the digital twin system, at least one of the following adjustments is made to the process candidate parameters corresponding to each process step: equipment layout adjustment, link logic adjustment, and scheduling rule adjustment, to obtain the simulation results.

[0078] Based on any of the above embodiments, the digital twin system includes a virtual device layer, a virtual link layer, and a virtual scheduling layer; The virtual device layer is used to map the equipment layout features of the wind turbine manufacturing system; The virtual link layer is used to simulate material transfer and signal interaction between equipment in the wind turbine manufacturing system; The virtual scheduling layer is used to map scheduling rules for personnel, materials, and equipment.

[0079] Based on any of the above embodiments, the historical process parameters include the historical process parameters of each process step; Function building blocks are specifically used for: The production prediction function is constructed based on the historical parameters of each process under the overall process, the process measurement error of each process, the overall process measurement error, and the historical product indicators.

[0080] Based on any of the above embodiments, the product indicators include production efficiency indicators, production cost indicators, and production quality indicators.

[0081] Based on any of the above embodiments, the reconfiguration unit is followed by an instruction issuing unit, which is specifically used for: Based on the reconfigured process parameters, process adjustment instructions are constructed; The process adjustment instruction is sent to the wind turbine manufacturing system to instruct the wind turbine manufacturing system to produce according to the process adjustment instruction.

[0082] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4As shown, the electronic device may include a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a reconstruction method for a wind turbine manufacturing system based on digital twins. This method includes: obtaining historical product indicators corresponding to historical process parameters of the wind turbine manufacturing system; constructing a production prediction function between the product indicators and the process parameters of the wind turbine manufacturing system based on the historical process parameters and the historical product indicators; solving the production prediction function with the goal of optimizing the product indicators to obtain a set of candidate process parameters; inputting the set of candidate process parameters into a digital twin system for simulation to obtain simulation results; correcting the candidate process parameters in the set of candidate process parameters based on the simulation results to obtain reconstructed process parameters; and applying the reconstructed process parameters to the wind turbine manufacturing system.

[0083] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0084] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the reconstruction method for a wind turbine manufacturing system based on digital twins provided by the above methods. The method includes: obtaining historical product indicators corresponding to historical process parameters of the wind turbine manufacturing system; constructing a production prediction function between the product indicators and the process parameters of the wind turbine manufacturing system based on the historical process parameters and the historical product indicators; solving the production prediction function with the goal of optimizing the product indicators to obtain a set of candidate process parameters; inputting the set of candidate process parameters into a digital twin system for simulation to obtain simulation results; correcting the candidate process parameters in the set of candidate process parameters based on the simulation results to obtain reconstructed process parameters; and applying the reconstructed process parameters to the wind turbine manufacturing system.

[0085] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a method for reconstructing a wind turbine manufacturing system based on a digital twin, as provided by the methods described above. This method includes: obtaining historical product indicators corresponding to historical process parameters of the wind turbine manufacturing system; constructing a production prediction function between the product indicators and the process parameters of the wind turbine manufacturing system based on the historical process parameters and the historical product indicators; solving the production prediction function with the goal of optimizing the product indicators to obtain a set of candidate process parameters; inputting the set of candidate process parameters into a digital twin system for simulation to obtain simulation results; correcting the candidate process parameters in the set of candidate process parameters based on the simulation results to obtain reconstructed process parameters; and applying the reconstructed process parameters to the wind turbine manufacturing system.

[0086] The device embodiments described above are merely illustrative. 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 modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0087] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for reconstructing a wind turbine manufacturing system based on digital twins, characterized in that, include: Obtain historical product indicators corresponding to historical process parameters of the wind turbine manufacturing system; Based on the historical process parameters and the historical product indicators, a production prediction function is constructed between the product indicators and the process parameters of the wind turbine manufacturing system. With the goal of optimizing the product indicators, the production forecast function is solved to obtain a set of alternative process parameters; The set of alternative process parameters is input into the digital twin system for simulation to obtain simulation results; Based on the simulation results, the candidate process parameters in the candidate process parameter set are corrected to obtain the reconstructed process parameters; The reconfiguration process parameters are applied to the wind turbine manufacturing system.

2. The reconstruction method for a wind turbine manufacturing system based on digital twins according to claim 1, characterized in that, The step of correcting the candidate process parameters in the candidate process parameter set based on the simulation results to obtain reconstructed process parameters includes: The simulation results are compared with preset indicators to obtain comparison results; If the comparison result shows a deviation, the candidate process parameters are corrected until the comparison result corresponding to the corrected process parameters has no deviation, and the corrected process parameters are used as the reconstructed process parameters.

3. The reconstruction method for a wind turbine manufacturing system based on digital twins according to claim 1, characterized in that, The step of inputting the set of alternative process parameters into the digital twin system for simulation and obtaining simulation results includes: Candidate process parameters are selected from the set of alternative process parameters. The candidate process parameters are input into the digital twin system. Based on the digital twin system, at least one of the following adjustments is made to the process candidate parameters corresponding to each process step: equipment layout adjustment, link logic adjustment, and scheduling rule adjustment, to obtain the simulation results.

4. The reconstruction method for a wind turbine manufacturing system based on digital twins according to claim 3, characterized in that, The digital twin system includes a virtual device layer, a virtual link layer, and a virtual scheduling layer; The virtual device layer is used to map the equipment layout features of the wind turbine manufacturing system; The virtual link layer is used to simulate material transfer and signal interaction between equipment in the wind turbine manufacturing system; The virtual scheduling layer is used to map scheduling rules for personnel, materials, and equipment.

5. The reconstruction method for a wind turbine manufacturing system based on digital twins according to any one of claims 1 to 4, characterized in that, The historical process parameters include the historical process parameters for each process step; The step of constructing a production prediction function between product indicators and the process parameters of the wind turbine manufacturing system based on the historical process parameters and historical product indicators includes: The production prediction function is constructed based on the historical parameters of each process under the overall process, the process measurement error of each process, the overall process measurement error, and the historical product indicators.

6. The reconstruction method for a wind turbine manufacturing system based on digital twins according to any one of claims 1 to 4, characterized in that, The product indicators include production efficiency indicators, production cost indicators, and production quality indicators.

7. The reconstruction method for a wind turbine manufacturing system based on digital twins according to any one of claims 1 to 4, characterized in that, The process involves correcting the candidate process parameters in the candidate process parameter set based on the simulation results to obtain reconstructed process parameters, followed by: Based on the reconfigured process parameters, process adjustment instructions are constructed; The process adjustment instruction is sent to the wind turbine manufacturing system to instruct the wind turbine manufacturing system to produce according to the process adjustment instruction.

8. A reconfiguration device for a wind turbine manufacturing system based on digital twins, characterized in that, include: The acquisition unit acquires historical product indicators corresponding to historical process parameters of the wind turbine manufacturing system. The function construction unit constructs a production prediction function between the product indicators and the process parameters of the wind turbine manufacturing system based on the historical process parameters and the historical product indicators. The optimization unit, with the goal of optimizing the product indicators, solves the production prediction function to obtain a set of alternative process parameters; The simulation unit inputs the set of alternative process parameters into the digital twin system for simulation and obtains simulation results. The reconstruction unit corrects the candidate process parameters in the candidate process parameter set based on the simulation results to obtain the reconstructed process parameters; The reconfiguration process parameters are applied to the wind turbine manufacturing system.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the reconstruction method of the wind turbine manufacturing system based on digital twins as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the reconstruction method of the wind turbine manufacturing system based on digital twins as described in any one of claims 1 to 7.

Images