Wind turbine full life cycle prediction method and device based on multi-physical field coupling

Abstract

The embodiment of the application provides a wind turbine full life cycle prediction method and device based on multi-physical field coupling, which comprises the following steps: dividing the whole structure of the wind turbine into structural components and related spaces, collecting multi-physical field sensor data from the whole structure, discretizing the structural components of the wind turbine into a plurality of finite elements, defining the related spaces as nodes connected to the finite elements, constructing a corresponding wind turbine network topology, establishing corresponding physical field control equations for the finite elements based on the network topology, determining a corresponding global coupling equation set, and solving the global coupling equation set by using a numerical method to determine the corresponding physical response data. Real-time multi-physical field sensor data and real-time physical response data are input into a set hybrid learning model for unsupervised and supervised hybrid learning to determine the life cycle prediction result of the wind turbine. The application can improve the efficiency and accuracy of wind turbine fault prediction.

Description

Technical Field

[0001] This application relates to the field of data processing, specifically to a method and device for predicting the entire life cycle of wind turbines based on multi-physics coupling. Background Technology

[0002] As a crucial component of renewable energy, wind turbines are typically deployed in harsh natural environments such as mountaintops, deserts, or oceans, often operating unattended or with minimal maintenance. This environment leads to frequent damage to wind turbine components, severely impacting operational efficiency and lifespan. To effectively extend the lifespan of wind turbines and reduce maintenance costs, it is essential to detect early-stage faults as early as possible, predicting and revealing their causes, extent, location, and development trends. This provides on-site maintenance personnel with a scientific basis for online adjustments or shutdown maintenance.

[0003] Currently, fault diagnosis and remaining life prediction of wind turbines face three major technical bottlenecks:

[0004] Data silo problem: The existing monitoring system collects 16 types of physical quantities such as vibration, stress, and oil, which belong to independent subsystems. It lacks coupling analysis across physical fields. This independent monitoring method cannot fully reflect the overall status of the unit and limits the accuracy of fault diagnosis.

[0005] Early failure concealment: Traditional threshold alarm methods are slow to respond to progressive failures (such as bolt preload relaxation), with an average warning time window of less than 72 hours, making it difficult to intervene effectively in the early stages of failure.

[0006] Digital twin fault: Mainstream life prediction models only achieve component-level mapping, ignoring the cross-domain coupling effect of related structural spaces, and failing to establish a full-dimensional mapping of equipment structure-space-time.

[0007] Therefore, there is an urgent need for a wind turbine full life cycle prediction method based on multi-physics coupling, which can realize cross-space propagation modeling of fault factors and improve the efficiency and accuracy of wind turbine fault prediction. Summary of the Invention

[0008] To address the problems in the existing technology, this application provides a method and device for predicting the entire life cycle of wind turbine generators based on multi-physics coupling, which can improve the efficiency and accuracy of wind turbine generator fault prediction.

[0009] To solve at least one of the above problems, this application provides the following technical solution:

[0010] Firstly, this application provides a method for predicting the entire life cycle of wind turbines based on multi-physics coupling, including:

[0011] The overall structure of the wind turbine is divided into structural components and related spaces. Multiphysics sensor data is collected from the structural components and related spaces. The structural components include blades, hubs, nacelles and towers. The related spaces include blade-hub connection space, hub-nacelle connection space, nacelle-tower connection space and tower-blade relative space.

[0012] The structural components of the wind turbine are discretized into multiple finite elements, and the relevant space is defined as the nodes connecting the finite elements. A corresponding wind turbine network topology is constructed. Based on the network topology, in the local coordinate system, the control equations for the fluid field, structural force field, and spatial position field are established for each finite element. The control equations of all finite elements are combined to determine the corresponding global coupling equation set. According to the sensor data acquisition of each finite element, the global coupling equation set is solved using numerical methods to determine the physical response data of the wind turbine in the entire field.

[0013] Real-time multiphysics sensor data and real-time physical response data are input into a hybrid learning model for unsupervised and supervised hybrid learning to determine the life cycle prediction results of the wind turbine. The life cycle prediction results include at least one of fault diagnosis results and remaining life prediction results.

[0014] Furthermore, the acquisition of multiphysics sensor data from the structural components and the associated space includes:

[0015] Data on the blade system, hub system, nacelle drivetrain, tower system, relative space of components, and environmental and operating conditions are collected from the structural components and related spaces.

[0016] The blade system data includes internal blade vibration, stress load, temperature, blade root flange clearance, and blade root bolt preload.

[0017] The hub system data includes vibration of the pitch motor and pitch bearing, pitch motor speed and current, and hub flange clearance.

[0018] The data on the engine room drivetrain includes engine room sway, main shaft sway and vibration, gearbox sway and vibration, gearbox oil quality, generator sway and vibration, and electrical parameters;

[0019] The tower system data includes the tower sway posture, the preload of flange bolts at each level of the tower, and the preload of anchor bolts at the bottom of the tower.

[0020] The relative space data of the components includes the clearance data between the blade tip and the tower.

[0021] Furthermore, the step of discretizing the structural components of the wind turbine into multiple finite elements and defining the relevant space as nodes connecting the finite elements to construct the corresponding wind turbine network topology includes:

[0022] The continuous structures within the four structural components—blade, hub, nacelle, and tower—are each divided into multiple discrete finite element units.

[0023] The four dynamic interactive regions, namely the blade-hub connection space, the hub-nacelle connection space, the nacelle-tower connection space, and the tower-blade relative space, are defined as the boundary nodes of the finite element units connecting different structural parts.

[0024] Based on the actual physical connection and energy transfer path of the wind turbine, the connection relationship between the boundary node and the finite element unit of the adjacent structural part is established, and the corresponding wind turbine network topology is determined, wherein the finite element unit is used as the vertex of the graph, and the connection relationship is used as the edge of the graph.

[0025] Furthermore, based on the network topology, in the local coordinate system, the governing equations for the fluid field, structural force field, and spatial position field are established for each finite element, and the governing equations of all finite elements are combined to determine the corresponding global coupling equation set, including:

[0026] Based on the component or space to which the finite element in the network topology belongs, a corresponding physical field control equation is assigned to it, and the physical property parameters and material property parameters of the finite element are used as the inherent coefficients of the control equation. The types of the physical field control equation include fluid field, structural force field and spatial position field.

[0027] In the local coordinate system, spatial dimensions and time variables are set for the governing equations to characterize the distribution and evolution of physical quantities within the finite element.

[0028] Based on the node connection relationships in the network topology, physical field continuity conditions and force balance conditions are applied between adjacent finite elements as boundary constraints of the global coupling equation set. The physical field control equations of all finite elements with the applied boundary constraints are combined to determine the corresponding global coupling equation set.

[0029] Furthermore, the step of solving the globally coupled equations using numerical methods based on the sensor data acquisition volume of each finite element to determine the corresponding physical response data of the wind turbine across the entire field includes:

[0030] The multiphysics sensor data corresponding to each finite element is used as the boundary conditions, initial conditions, and external excitation loads of the corresponding control equations and substituted into the global coupling equation set.

[0031] The global coupled equations are solved by numerical iterative algorithm to determine the physical response data of the wind turbine in the entire field.

[0032] Furthermore, before inputting real-time multiphysics sensor data and real-time physical response data into the hybrid learning model for unsupervised and supervised hybrid learning, the following steps are included:

[0033] Based on historical data of wind turbines in a healthy state, a reconstruction model based on an autoencoder is trained to establish a baseline for the normal state and determine the corresponding unsupervised learning module after training.

[0034] Based on historical data containing labeled fault types, a supervised learning model is trained, and the corresponding trained supervised learning module is determined.

[0035] Based on the unsupervised learning module and the supervised learning module after training, the corresponding hybrid machine learning model is determined.

[0036] Furthermore, the step of inputting real-time multiphysics sensor data and real-time physical response data into a hybrid learning model for unsupervised and supervised hybrid learning includes:

[0037] Based on the preset variational autoencoder and the preset adversarial autoencoder, feature dimensionality reduction and noise reduction are performed on real-time multiphysics sensor data and real-time physical response data to determine the corresponding feature representation of the same magnitude.

[0038] The same-order feature representation is input into the unsupervised learning module to determine the anomaly detection result and the operating condition clustering result output by the unsupervised learning module. The anomaly detection result is used to trigger the incremental update of the supervised learning module, and the operating condition clustering result is used to input the supervised learning module for model training and prediction.

[0039] The same-scale feature representation and the results output by the unsupervised learning module are input into the supervised learning module to determine the fault diagnosis results and remaining life prediction results of the wind turbine output by the supervised learning module.

[0040] Secondly, this application provides a wind turbine full life cycle prediction device based on multi-physics coupling, comprising:

[0041] The wind turbine structure division module is used to divide the overall structure of the wind turbine into structural components and related spaces, and to collect multi-physics sensor data from the structural components and related spaces. The structural components include blades, hubs, nacelles and towers, and the related spaces include blade-hub connection space, hub-nacelle connection space, nacelle-tower connection space and tower-blade relative space.

[0042] The wind turbine coupling equation determination module is used to discretize the structural components of the wind turbine into multiple finite elements, define the relevant space as nodes connecting the finite elements, construct the corresponding wind turbine network topology, establish the control equations of fluid field, structural force field and spatial position field for each finite element in the local coordinate system, and combine the control equations of all finite elements to determine the corresponding global coupling equation set. According to the sensor data acquisition of each finite element, the global coupling equation set is solved by numerical method to determine the corresponding physical response data of the wind turbine in the whole field.

[0043] The wind turbine life cycle self-diagnosis module is used to input real-time multiphysics sensor data and real-time physical response data into a set hybrid learning model for unsupervised and supervised hybrid learning, so as to determine the life cycle prediction result of the wind turbine. The life cycle prediction result includes at least one of fault diagnosis result and remaining life prediction result.

[0044] Thirdly, this application 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 steps of the wind turbine full life cycle prediction method based on multiphysics coupling.

[0045] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the wind turbine full life cycle prediction method based on multi-physics coupling.

[0046] Fifthly, this application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the described method for predicting the entire life cycle of wind turbine generators based on multiphysics coupling.

[0047] As can be seen from the above technical solution, this application provides a method and device for predicting the entire life cycle of wind turbines based on multi-physics coupling. By dividing the overall structure of the wind turbine into structural components and related spaces, and collecting multi-physics sensor data from the overall structure, the structural components of the wind turbine are discretized into multiple finite elements, and the related spaces are defined as nodes connecting the finite elements. A corresponding wind turbine network topology is constructed, and corresponding physical field control equations are established for the finite elements based on the network topology. The corresponding global coupling equation set is determined, and the global coupling equation set is solved using numerical methods to determine the corresponding physical response data. Real-time multi-physics sensor data and real-time physical response data are input into a set hybrid learning model for unsupervised and supervised hybrid learning to determine the life cycle prediction results of the wind turbine. This can improve the efficiency and accuracy of wind turbine fault prediction. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application 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 application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is one of the flowcharts illustrating the wind turbine full life cycle prediction method based on multi-physics coupling in the embodiments of this application;

[0050] Figure 2 This is a structural diagram of the wind turbine full life cycle prediction device based on multiphysics coupling in the embodiments of this application;

[0051] Figure 3 This is a schematic diagram of the structure of the electronic device in the embodiments of this application.

[0052] Figure label:

[0053] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver storage unit 9144, antenna 9111, speaker 9131, microphone 9132. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments 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.

[0055] The acquisition, storage, use, and processing of data in this application comply with relevant laws and regulations.

[0056] Considering that wind turbine generators are typically deployed in harsh environments, their lifecycle data suffers from data silos, data gaps, and the concealment of early faults. This application provides a method and apparatus for predicting the entire lifecycle of wind turbine generators based on multiphysics coupling. The method divides the overall structure of the wind turbine generator into structural components and related spaces, collects multiphysics sensor data from the overall structure, discretizes the structural components of the wind turbine generator into multiple finite elements, and defines the related spaces as nodes connecting these finite elements, constructing a corresponding wind turbine generator network topology. Based on the network topology, corresponding physical field control equations are established for the finite elements, determining the corresponding global coupling equation set. Numerical methods are then used to solve the global coupling equation set to determine the corresponding physical response data. Real-time multiphysics sensor data and real-time physical response data are input into a hybrid learning model for unsupervised and supervised hybrid learning to determine the lifecycle prediction results of the wind turbine generator. This improves the efficiency and accuracy of wind turbine generator fault prediction.

[0057] To improve the efficiency and accuracy of wind turbine generator fault prediction, this application provides an embodiment of a wind turbine generator full lifecycle prediction method based on multiphysics coupling. See [link to embodiment]. Figure 1 The wind turbine full life cycle prediction method based on multi-physics coupling specifically includes the following:

[0058] Step S101: Divide the overall structure of the wind turbine into structural components and related spaces, and collect multi-physics sensor data from the structural components and related spaces. The structural components include blades, hubs, nacelles and towers, and the related spaces include blade-hub connection space, hub-nacelle connection space, nacelle-tower connection space and tower-blade relative space.

[0059] Optionally, in this embodiment, this step is part of the core algorithm framework and involves the wind turbine structure theory innovation system.

[0060] Specifically, the wind turbine is divided into four parts and four related spaces according to its structure.

[0061] The four parts and their detailed components are:

[0062] 1. Leaves

[0063] ① Leaf tip

[0064] ② Middle part of the leaf

[0065] ③ Leaf base

[0066] 2. Wheel hub

[0067] ① Wheel hub A-side

[0068] ② Wheel hub B side

[0069] ③ Wheel hub C-side

[0070] ④ Fairing

[0071] 3. Cabin

[0072] ① Cabin shell

[0073] ② Main spindle

[0074] ③ Gearbox

[0075] ④ Generator

[0076] 4. Tower

[0077] ① Top of the tower

[0078] ② Each layer in the middle of the tower

[0079] ③ Lower part of the tower

[0080] ④ Bottom of the tower

[0081] The four related spaces are:

[0082] 1. Blade-hub connection space

[0083] 2. Hub-engine bay connection space

[0084] 3. Nacelle-Tower Connection Space

[0085] 4. Relative space between tower and blades

[0086] Optionally, in this embodiment, after the above structural division, we deploy sensors at various key locations in each component and space of the wind turbine to collect multi-physics sensor data to reflect the true state of the wind turbine.

[0087] Specifically, the following 16 types of data will be obtained for subsequent model building and calculations:

[0088] 1. Ambient altitude (air pressure), temperature, wind speed data, spindle speed data, gearbox speed data at each stage, generator speed data, and power generation data.

[0089] Data acquisition methods: The fan comes with a thermometer, anemometer, speed encoder, power meter, SCADA system, etc.

[0090] Belonging to the component / space: natural environment and power generation operation status.

[0091] 2. Vibration data, stress load data, and temperature data within the blade.

[0092] Data acquisition methods: ultra-low frequency triaxial vibration sensor, stress sensor, load sensor, and temperature sensor.

[0093] Parts or related spaces: leaf (leaf tip - leaf middle - leaf root).

[0094] 3. Blade root flange clearance data and blade root bolt tightening force data.

[0095] Data acquisition methods: flange gap sensor, preload sensor.

[0096] Related components or spaces: blade root, blade-hub connection space.

[0097] 4. Vibration data, speed data, and current data of the in-hub pitch motor.

[0098] Data acquisition methods: vibration sensors, speed encoders, SCADA systems, etc.

[0099] Related components or spaces: hub (A, B, C surfaces), blade orientation, blade-hub connection space.

[0100] 5. Vibration data of the pitch bearing inside the wheel hub.

[0101] Data acquisition method: vibration sensor.

[0102] Related components or spaces: hub (A, B, C surfaces), blade orientation, blade-hub connection space.

[0103] 6. Hub flange clearance data.

[0104] Data acquisition method: flange gap sensor.

[0105] Related components or spaces: wheel hub, wheel hub-engine compartment connection space.

[0106] 7. Data on shaking inside the cabin.

[0107] Data acquisition method: sway sensor, etc.

[0108] Related components or spaces: nacelle shell, hub-to-nacelle connection space, nacelle-to-tower connection space.

[0109] 8. Spindle axial movement data and main bearing vibration data.

[0110] Data acquisition methods: displacement motion sensor, acceleration vibration sensor.

[0111] Part or related space: spindle.

[0112] 9. Gearbox axial movement data and vibration data.

[0113] Data acquisition methods: displacement motion sensor, acceleration vibration sensor.

[0114] Part or related space: Gearbox.

[0115] 10. Gearbox oil quality data.

[0116] Data acquisition method: oil quality sensor.

[0117] Part or related space: Gearbox.

[0118] 11. Generator oscillation data, vibration data, and electrical data.

[0119] Data acquisition methods: displacement sensors, acceleration and vibration sensors, SCADA systems, etc.

[0120] Related component or space: generator.

[0121] 12. Vibration data, speed data, and current data of the yaw motor at the bottom of the engine room.

[0122] Data acquisition methods: vibration sensors, speed encoders, SCADA systems, etc.

[0123] Related component or space: Nacelle-tower connection space.

[0124] 13. Blade tip to tower clearance data.

[0125] Data acquisition method: Clearance radar.

[0126] Related component or space: Tower-blade relative space.

[0127] 14. Tower sway posture data.

[0128] Data acquisition method: composite tilt sensor, etc.

[0129] Components or related spaces: Tower (top-middle-bottom-base).

[0130] 15. Tightening force data of flange bolts at each level of the tower.

[0131] Data acquisition method: preload sensor.

[0132] Related components or spaces: Tower (top-middle-lower part).

[0133] 16. Tower bottom anchor bolt tightening data.

[0134] Data acquisition method: preload sensor.

[0135] Part or related space: Bottom of the tower.

[0136] Through the above steps, we are able to obtain real-time information on the wind turbine's physical structure, key stress points, and operating status through various deployed sensors.

[0137] Step S102: Discretize the structural components of the wind turbine into multiple finite elements, and define the relevant space as nodes connecting the finite elements to construct a corresponding wind turbine network topology. Based on the network topology, establish control equations for the fluid field, structural force field, and spatial position field for each finite element in the local coordinate system. Combine the control equations of all finite elements to determine the corresponding global coupling equation set. Solve the global coupling equation set using numerical methods based on the sensor data acquisition of each finite element to determine the physical response data of the wind turbine in the entire field.

[0138] Optionally, in this embodiment, this step constructs a wind turbine network topology and further utilizes the geometric and physical knowledge of wind turbines to establish a solution framework, which is used to break down data silos and accurately reflect the real-time operating status of wind turbines.

[0139] After the deployment of the sensor in step S101, each part and related space has its own data acquisition device to collect relevant data. Next, based on the key nodes of the wind turbine operation, we discretize the overall structure of the wind turbine and divide the continuous structure in each part into discrete small units.

[0140] In this embodiment, more than 100 discrete small units are set on the basis of the overall structure of the wind turbine. Each small unit is a key node in the operation of the wind turbine and is used to accurately reflect the status of each component of the wind turbine. Each discrete small unit is an independent finite element, and the relevant space is used as the node between finite elements to form a network.

[0141] Within each finite element, based on different data acquisition parameters, the following parameters are used: fluid field (temperature, airflow, noise, oil, etc.), structural force field (vibration, stress, load, preload, etc.), and spatial position field (tilt, displacement, deformation, clearance, gap), all based on the local coordinate system of the finite element. Each mathematical function and spatial boundary is established, and a bidirectional failure propagation link is established in four key connection spaces.

[0142] The equations of all elements are combined into a large system of equations, taking into account boundary conditions and loads, ultimately forming a global linear system of equations. The system of equations is solved using numerical methods (such as Gaussian elimination and iterative methods) to obtain the unknowns at the nodes, and then the response of the entire system (such as stress, strain, temperature distribution, etc.) is derived, i.e., physical response data.

[0143] The following are mathematical function expressions for different physical fields within a finite element, organized by category:

[0144] 1. Mathematical functions of fluid fields

[0145] 1.1 Temperature Field (Heat Conduction Equation)

[0146]

[0147] Temperature field

[0148] Thermal conductivity; :density; Specific heat capacity

[0149] Heat source item

[0150] Application example: Gearbox overheat warning. Based on oil temperature sensor data ( ) and gear friction heat source ( Coupling can predict the risk of local overheating (e.g., in the embodiment, the overheating warning for gears and bearings is 240 hours in advance).

[0151] 1.2 Wind Field (Simplified Navier-Stokes Equations)

[0152]

[0153] Velocity field

[0154] :pressure; Kinematic viscosity; Volume force

[0155] Application example: Blade aerodynamic load analysis. Combined with anemometer data ( The blade stress field (Formula 2.2) is used to predict flutter risk (the accuracy is improved by 41% in the example).

[0156] 1.3 Noise Field (Wave Equation)

[0157]

[0158] Sound pressure field; Speed ​​of sound

[0159] 1.4 Oil Flow (Continuity Equation)

[0160]

[0161] Oil density

[0162] 2. Mathematical functions of structural force fields

[0163] 2.1 Vibration Field (Harmonic Response Equation)

[0164]

[0165] Displacement field

[0166] :quality; Damping coefficient; Stiffness

[0167] Application example: Vibration fault diagnosis. Through vibration data ( Reverse incentive force It can identify faults such as pitting or friction damage.

[0168] 2.2 Stress Field (Hooke's Law)

[0169]

[0170] Stress tensor; : Strain tensor; Elastic tensor

[0171] 2.3 Distributed load (uniformly distributed load)

[0172]

[0173] : Foundation load; Spatial distribution coefficient

[0174] 2.4 Preload (bolt preload model)

[0175]

[0176] Bolt stiffness; Pre-stretch amount

[0177] Application example: Health monitoring of tower bolts. Combined with bolt stiffness coefficient ( ) and variables ( Dynamically assess connection reliability (such as tower anchor bolt monitoring).

[0178] 3. Mathematical functions of spatial position field

[0179] 3.1 Inclination Angle (Geometric Relationship)

[0180]

[0181] : Inclined function

[0182] 3.2 Displacement (Time-varying function)

[0183]

[0184] Amplitude; :frequency; Phase

[0185] 3.3 Deformation field (strain-displacement relationship)

[0186]

[0187] Displacement field

[0188] 3.4 Clearance / Gap (Geometric Constraints)

[0189]

[0190] : Position function of two surfaces; Minimum safe distance; Maximum safe distance

[0191] Application example: Blade tip-to-tower clearance monitoring. This is achieved through clearance radar data ( Real-time calculation of clearance distance to avoid tower sweeping accidents (such as tower-blade relative space monitoring).

[0192] Formula Explanation

[0193] ① Variable range: All formulas are based on the local coordinate system of the finite element. .

[0194] ② Linear and nonlinear: Dynamic fields such as vibration and airflow need to consider nonlinear terms, such as convection terms.

[0195] ③ Boundary conditions: The equations are closed by using boundary conditions.

[0196] ④ To facilitate model calculations, the formulas have been appropriately simplified and numerically discretized without affecting the accuracy of the results.

[0197] Application scenario example:

[0198] Monitoring of blade root bolt preload loosening (a common progressive failure):

[0199] The "blade root unit" collects bolt data through a preload sensor (Formula 2.4).

[0200] The "blade-hub connection space" serves as a transmission node, and deformation is monitored by a flange gap sensor (Formula 3.4).

[0201] The vibration sensor inside the "hub" (Formula 2.1) synchronously detects abnormal vibrations.

[0202] The data from the above three aspects and the transmission link influencing factors are substituted into the global equation system for solution. The final calculation results not only comprehensively judge the degree of failure degradation and the prediction of remaining life from multiple dimensions, but also form a multi-source verification.

[0203] Step S103: Input real-time multiphysics sensor data and real-time physical response data into the set hybrid learning model to perform unsupervised and supervised hybrid learning, so as to determine the life cycle prediction result of the wind turbine, wherein the life cycle prediction result includes at least one of fault diagnosis result and remaining life prediction result.

[0204] Optionally, in this embodiment, the real-time multiphysics sensor data obtained in the preceding steps and the real-time physical response data obtained by solving them are input into a pre-built and trained hybrid learning model. Deep analysis is performed through a hybrid paradigm of unsupervised learning and supervised learning in parallel, and finally the full life cycle prediction results of the wind turbine are output. Specifically, it includes at least one of fault diagnosis (such as fault type, location, and severity) and remaining service life prediction. Among these two prediction results, the fault diagnosis result is predicted first, and then the remaining service life prediction result is calculated.

[0205] Specifically, this process relies on a sophisticated hierarchical collaborative architecture within the hybrid learning model. Real-time data streams are first fed in parallel into the unsupervised learning module and the supervised learning module.

[0206] In the unsupervised learning channel, the system employs techniques such as isolated forests and autoencoder reconstruction errors to detect anomalies in massive amounts of unlabeled data, automatically identifying abnormal points or unknown fault modes that deviate from normal patterns. Simultaneously, it uses methods such as deep embedding clustering to automatically classify operating conditions, identifying different states of the wind turbine, such as rated power generation, start-up / shutdown, or turbulent operation. The results of this unsupervised learning (such as anomaly alarms and operating condition labels) serve as crucial contextual information and are fed into the supervised learning channel in real time.

[0207] In the supervised learning channel, the system first performs deep feature extraction and fusion on the original time-series data using a one-dimensional convolutional neural network and wavelet packet transform to capture weak signs such as vibration characteristic frequencies. Subsequently, a graph neural network and a long short-term memory network are jointly used, employing the network topology built in the previous steps as a blueprint to learn the spatial connectivity and temporal dependencies of the wind turbine components, simulating the propagation path of faults within the structure. Finally, at the decision layer, high-precision fault classification is achieved first through a collaborative attention mechanism and a multilayer perceptron. Based on this classification, sequence models such as Transformer are responsible for predicting the remaining lifespan based on the performance degradation trajectory. Throughout the process, the operating condition information and abnormal signals provided by the unsupervised module dynamically guide the decisions of the supervised model; for example, calling different diagnostic models for different operating conditions or focusing more analytical attention on abnormal signals.

[0208] This example demonstrates how this embodiment creatively decomposes the overall wind turbine structure into structural units and dynamic controls, creates a discretized model, solves the problem of cross-physical field coupling, introduces a real-time joint solver for the fluid field, structural force field, and spatial position field to achieve breakthroughs in multiple couplings and break down data silos, and finally establishes a supervised-unsupervised hybrid learning model to achieve self-evolutionary diagnosis of the wind turbine life cycle based on the knowledge graph of wind turbine fault factors.

[0209] In terms of effectiveness, firstly, this embodiment significantly extends the warning window from less than 72 hours in traditional methods to over 240 hours. Secondly, by comprehensively utilizing physical laws (embedded in physical response data) and data-driven insights, this embodiment drastically reduces the remaining life prediction error from over ±15% in traditional models to within ±5%, significantly improving prediction reliability. Finally, the model in this embodiment possesses strong generalization and interpretability, adapting not only to units of different models and environments but also providing reasonable explanations for the causes of failures and degradation stages through attention mechanisms and clustering results, providing a solid basis for maintenance decisions rather than an incomprehensible black box alarm.

[0210] As described above, the wind turbine full life cycle prediction method based on multiphysics coupling provided in this application can divide the overall structure of the wind turbine into structural components and related spaces, collect multiphysics sensor data from the overall structure, discretize the structural components of the wind turbine into multiple finite elements, define the related spaces as nodes connecting the finite elements, construct the corresponding wind turbine network topology, establish corresponding physical field control equations for the finite elements based on the network topology, determine the corresponding global coupling equation set, solve the global coupling equation set using numerical methods, determine the corresponding physical response data, and input real-time multiphysics sensor data and real-time physical response data into a set hybrid learning model for unsupervised and supervised hybrid learning to determine the life cycle prediction result of the wind turbine, thereby improving the efficiency and accuracy of wind turbine fault prediction.

[0211] In one embodiment of the wind turbine full life cycle prediction method based on multiphysics coupling in this application, it may further include the following:

[0212] Step S201: Collect blade system data, hub system data, nacelle drivetrain data, tower system data, component relative space data, and environmental and operating condition data from the structural components and related spaces;

[0213] Step S202: The blade system data includes blade internal vibration, stress load, temperature, blade root flange clearance, and blade root bolt preload.

[0214] Step S203: Hub system data includes vibration of the pitch motor and pitch bearing, pitch motor speed and current, and hub flange clearance;

[0215] Step S204: The data on the engine nacelle drivetrain includes engine nacelle sway, main shaft sway and vibration, gearbox sway and vibration, gearbox oil quality, generator sway, vibration, and electrical parameters;

[0216] Step S205: Tower system data includes tower sway posture, preload of flange bolts at each level of the tower, and preload of anchor bolts at the bottom of the tower;

[0217] Step S206: The relative space data of the components includes the clearance data between the blade tip and the tower.

[0218] Optionally, in this embodiment, the data acquisition scheme of this solution constructs the data foundation required for the prediction of the entire life cycle of wind turbine units in a systematic, multi-level, and multi-physical quantity manner.

[0219] Specifically, data on the blade system, hub system, nacelle drivetrain, tower system, relative spatial data of components, and environmental and operating conditions are collected from the structural components and related spaces. This step establishes a framework for systematic data collection from the four major structural components of the wind turbine (blades, hub, nacelle, and tower) and four key connecting spaces. This overcomes the limitations of traditional monitoring systems that only focus on a single component or independent subsystem, providing a data foundation for subsequently establishing a coupled analysis model across components and spaces.

[0220] Specifically, for the blade system, ultra-low frequency three-dimensional vibration, stress, and temperature sensors were deployed at key locations at the tip, middle, and root of the blade to achieve three-dimensional monitoring of the blade's structural health. The collected vibration and stress data were used to analyze aerodynamic loads and structural dynamic response, while temperature data was used to monitor icing on the blade surface and water ingress into the interior. The data on blade root flange clearance and bolt preload directly reflected the mechanical integrity of the blade-hub connection interface. This step aims to accurately detect subtle signs of progressive failures such as structural deformation, early cracks, and loose connections in the blade.

[0221] Specifically, for the hub system, the focus is on the core components of the pitch system. By collecting multi-dimensional data such as vibration of the pitch motor and bearings, motor speed and current, a comprehensive diagnosis of the mechanical and electrical performance of the pitch mechanism can be achieved. Monitoring the hub flange clearance is used to determine the stability of the hub-spindle connection interface. This step can effectively identify potential faults such as pitch bearing pitting, abnormal motor windings, or loss of preload at the connection interface.

[0222] Specifically, a complete transmission system monitoring system, from the main bearing to the generator, was constructed for the engine nacelle drivetrain. Engine nacelle sway data reflects the overall structural stability; axial movement and vibration data of the main shaft, gearbox, and generator are crucial for diagnosing wear and misalignment of rotating components such as bearings and gears; oil quality data directly characterizes the wear condition inside the gearbox; and generator electrical parameters are related to its electromagnetic performance and insulation health. The core function of this step is to achieve early warning and precise location of the degradation process of core drivetrain components.

[0223] Specifically, for the tower system, a composite tilt sensor is used to monitor the tower's swaying posture and assess its overall stability under wind loads and turbine operating loads. Continuous monitoring of the preload of flange bolts at each level and the bottom anchor bolts aims to directly assess the connection safety and structural integrity of this critical load-bearing structure, preventing major structural accidents caused by bolt fatigue loosening.

[0224] Specifically, regarding the relative space between components, the dynamic distance between the blade tip and the tower is measured in real time using clearance radar. This data is a direct safety guarantee to prevent collisions (tower sweep) between the blade tip and the tower, and can also be used to inversely analyze the deformation coupling effect of the entire machine under complex loads.

[0225] Through step S206, this embodiment collectively constitutes a refined data acquisition network covering the entire structure, space, and physical field of the wind turbine, laying a data foundation for the subsequent construction of a high-fidelity physical field model of all components.

[0226] In one embodiment of the wind turbine full life cycle prediction method based on multiphysics coupling in this application, it may further include the following:

[0227] Step S301: Divide the continuous structure within the four structural parts—blade, hub, nacelle, and tower—into multiple discrete finite element units.

[0228] Step S302: Define the four dynamic interaction areas, namely the blade-hub connection space, the hub-nacelle connection space, the nacelle-tower connection space, and the tower-blade relative space, as boundary nodes of finite element units connecting different structural parts;

[0229] Step S303: Based on the actual physical connection and energy transfer path of the wind turbine, establish the connection relationship between the boundary node and the finite element unit of the adjacent structural part, and determine the corresponding wind turbine network topology, wherein the finite element unit is used as the vertex of the graph and the connection relationship is used as the edge of the graph.

[0230] Optionally, in this embodiment, the structural parts are first discretized, which is the basis for physical field numerical modeling.

[0231] The large, complex, continuous structure of a wind turbine (such as blades and towers) is meshed, transforming it into a set of numerous small, regularly shaped elements (such as tetrahedrons and hexahedrons). Each discretized finite element is assigned precise geometric properties (such as size and shape) and physical properties (such as material density, elastic modulus, and thermal conductivity). Its core function is to transform the partial differential equations describing the continuous medium, which cannot be directly solved, into a system of algebraic equations solvable on each small element. This discretization process makes it possible to simulate the complex physical behavior of the entire wind turbine, laying the model foundation for subsequent high-precision calculations.

[0232] Optionally, in this embodiment, the next step is to implement the key innovation of cross-component coupling analysis.

[0233] Four key connection or interaction regions (such as the blade-hub connection space) are explicitly defined as boundary nodes in the network, moving away from traditional "blank areas" or simple boundaries. These nodes are given specific physical meanings and data processing functions as logical entities, carrying data on interactions between adjacent components, such as forces, torques, heat flow, and displacements. This step overcomes the limitations of traditional component-level modeling, organically linking physically independent but functionally tightly coupled components at the data level through these nodes, creating conditions for accurately simulating the propagation paths of faults or loads between structures.

[0234] Optionally, in this embodiment, the topology graph structure is constructed to transform the physical model into a computational model that can be recognized and processed by data-driven algorithms.

[0235] Based on the actual mechanical connections and physical interactions of the wind turbine (such as force transmission and vibration propagation), a clear connection relationship (edge) is established between the finite element elements (vertices) defined in step S301 and the boundary nodes (vertices) defined in step S302. Ultimately, a complete graph data structure that can reflect the actual structure of the wind turbine is formed.

[0236] This step has two effects: First, at the physical simulation level, the graph structure defines the mathematical relationships between the variables in the global coupled equation set; second, at the machine learning diagnostic level, the graph structure is directly used as the input of the graph neural network (GNN), enabling the AI ​​model to understand the "propagation relationship" of faults in the wind turbine mechanical structure in the same way it understands social networks, thereby achieving accurate diagnosis and prediction of cross-component coupled faults.

[0237] Through step S303, this embodiment realizes the transformation from physical entity to digital network model.

[0238] In one embodiment of the wind turbine full life cycle prediction method based on multiphysics coupling in this application, it may further include the following:

[0239] Step S401: Based on the component or space to which the finite element in the network topology belongs, assign the corresponding physical field control equation to it, and use the physical property parameters and material property parameters of the finite element as the inherent coefficients of the control equation. The types of the physical field control equation include fluid field, structural force field and spatial position field.

[0240] Step S402: In the local coordinate system, set the spatial dimension and time variable for the governing equation to characterize the distribution and evolution of physical quantities within the finite element;

[0241] Step S403: Based on the node connection relationship in the network topology, apply physical field continuity conditions and force balance conditions between adjacent finite elements as boundary constraints of the global coupling equation set. Combine the physical field control equations of all finite elements with the applied boundary constraints to determine the corresponding global coupling equation set.

[0242] Optionally, in this embodiment, the core of this process lies in transforming the continuous physical entity of the wind turbine into a discretized mathematical model that can be numerically solved by a computer.

[0243] Specifically, firstly, based on the wind turbine network topology, and according to the physical characteristics of the specific component (such as blades or gearboxes) or dynamic space (such as connection space) to which the finite element belongs, a unique physical field control equation is assigned to each discretized finite element. For example, an element located within a gearbox is assigned fluid field equations involving heat conduction and oil flow, as well as structural force field equations describing gear meshing vibrations, while an element located within the tower structure mainly involves structural force field and spatial position field equations. After determining the equation type, physical and material parameters characterizing the inherent properties of the element, such as density, elastic modulus, and thermal conductivity, are substituted as constant coefficients into the equations, thereby completing the mathematical definition of a single finite element in a physical sense.

[0244] Subsequently, an independent mathematical solution environment is established for each finite element with assigned equations and parameters. By establishing a local coordinate system on each element and defining spatial dimensions and introducing time variables, the originally static equations are transformed into four-dimensional functions that dynamically describe the changes of physical quantities (such as temperature, displacement, and stress) within the element as a function of spatial location and time. This step precisely anchors the physical phenomena from a macroscopic, holistic description to each tiny discrete element, achieving a precise characterization of complex physical fields at minute details, and laying the foundation for subsequently revealing how local faults evolve.

[0245] Finally, all the discrete, independent unit equations are systematically integrated into a global, interconnected, coupled system. The link in this integration is the node connections within the network topology.

[0246] At common nodes of adjacent units, physical continuity conditions (such as temperature continuity and displacement compatibility) and force balance conditions are enforced. These conditions, as mandatory boundary constraints, are applied to the entire equation set. In this way, the physical state of a single unit is no longer isolated; its changes are transmitted to adjacent units through nodes, thus simulating the propagation path of faults or loads in the overall wind turbine structure. Finally, all the unit control equations with applied boundary constraints are mathematically combined into a unified, globally coupled equation set with all node unknowns as the solution objective. This equation set represents a digital twin of the entire wind turbine under multiphysics operations in the computer, serving as the mathematical model for subsequent real-time simulation, state deduction, and life prediction.

[0247] Through step S403, this embodiment successfully transforms the continuous physical entity of the wind turbine into a discretized mathematical model that can be solved numerically by a computer.

[0248] In one embodiment of the wind turbine full life cycle prediction method based on multiphysics coupling in this application, it may further include the following:

[0249] Step S501: Substitute the multiphysics sensor data corresponding to each finite element as the boundary conditions, initial conditions and external excitation loads of the corresponding control equations into the global coupling equation set;

[0250] Step S502: Solve the global coupled equations using a numerical iterative algorithm to determine the physical response data of the corresponding wind turbine unit across the entire field.

[0251] Optionally, in this embodiment, this step is an interface for fusing the physical model with real-world data.

[0252] Specifically, boundary conditions refer to forcibly assigning observed data from the system boundaries to the model. For example, environmental wind speed and temperature data can be set as boundary conditions for the fluid field and temperature field equations, while the fixed constraints at the bottom of the tower can be set as displacement boundary conditions for the structural field. This ensures that the solution domain of the model remains consistent with the boundaries of the real physical world.

[0253] Specifically, initial conditions refer to providing the starting state for the governing equations describing the transient process. For example, the initial temperature and initial stress distribution of each component measured when the system starts up can be used as the initial values ​​for the heat conduction equation and the structural force field equation, allowing the dynamic evolution of the model to begin from a realistic state.

[0254] Specifically, external excitation loads refer to the physical quantities that directly drive the system response, which are substituted into the source or force terms of the equations. For example, the power generation data is applied to the system as the load of the generator, the measured wind pressure on the blade surface is applied to the structural dynamics equations as an external load, and the frictional heat generated by the gearbox is substituted into the temperature field equations as a heat source term.

[0255] After this step, the theoretical mathematical model will be instantiated into a high-fidelity digital twin for the current specific operating conditions.

[0256] Optionally, in this embodiment, after the parameterization of the equation system is completed, this step solves the large, coupled global equation system using a numerical iterative algorithm.

[0257] Preferably, the numerical iterative algorithm can be the Newton-Raphson method or the conjugate gradient method in the finite element method.

[0258] The obtained physical response data across the entire field can be used to calculate physical quantities even at locations where sensors are not installed. For example, by solving the data, one can obtain the stress distribution at key points inside the blade, the contact stress on the gearbox bearing raceway, and the temperature field contour map of the entire nacelle.

[0259] Through step S502, this embodiment successfully reconstructs and extrapolates the physical state of the space field from discrete sensor data points, providing a solid data foundation for subsequent lifecycle prediction.

[0260] In one embodiment of the wind turbine full life cycle prediction method based on multiphysics coupling in this application, it may further include the following:

[0261] Step S601: Based on historical data of wind turbine units in a healthy state, train a reconstruction model based on an autoencoder to establish a baseline for the normal state and determine the corresponding unsupervised learning module after training.

[0262] Step S602: Based on historical data containing labeled fault types, train a supervised learning model and determine the corresponding trained supervised learning module;

[0263] Step S603: Determine the corresponding hybrid machine learning model based on the unsupervised learning module and the supervised learning module after training.

[0264] Optionally, in this embodiment, this step aims to construct an automated fault diagnosis and remaining life prediction model for all components of wind power equipment, which is used to predict the life cycle of wind power equipment based on the sensor data we detected and the physical field response data we calculated.

[0265] This model employs a hierarchical hybrid machine learning architecture, including both unsupervised and supervised learning, which are conducted in parallel and in combination. The training steps for both modules before practical application are as follows:

[0266] Specifically, the core of the unsupervised learning module is to enable the machine to learn and memorize the normal operating patterns of wind turbines in a "fully healthy" state. This requires collecting a large amount of historical data generated during fault-free operation, covering readings from various sensors such as vibration, temperature, and pressure. The model used for training is typically a deep autoencoder, a special type of neural network. During training, healthy, normal data is input into the autoencoder, forcing the model to learn how to first compress the data to a low-dimensional "bottleneck" layer (capturing the most essential features of the data), and then reconstruct the original data as accurately as possible from this compressed representation. Through repeated iterations, the model gradually masters the inherent structure, variation patterns, and noise levels of healthy data. After training, this autoencoder, capable of accurately reconstructing normal data, becomes the benchmark for measuring whether the state is "abnormal."

[0267] Specifically, the goal of the supervised learning module is to build an expert system capable of accurate diagnosis and prediction. The training data consists of clearly labeled historical data; for example, a vibration signal might be labeled "bearing outer race failure," and the end of a time series might be labeled "gearbox failure." The training process involves the supervised learning model (such as deep convolutional neural networks or graph neural networks) learning the complex mapping between the complex input data and these specific fault labels or remaining life values. By adjusting millions of internal parameters, the model continuously narrows the gap between its predictions and the true labels, ultimately learning to extract patterns highly correlated with specific faults or degradation stages from the data features.

[0268] After the first two steps have independently completed the specialized training of their respective modules, this step integrates them into a unified hybrid machine learning model. This step clarifies the division of labor and cooperation between the two modules in subsequent real-time operation: the unsupervised module is responsible for providing real-time anomaly alerts and operational background information, while the supervised module builds upon this to perform deeper, more targeted diagnosis and prediction.

[0269] Through step S603, this embodiment successfully combines the breadth advantage of unsupervised learning in "discovering the unknown" with the depth advantage of supervised learning in "precise localization." This architecture enables the final hybrid model to not only handle known fault types but also possess the initial ability to cope with entirely new fault modes. Furthermore, by fusing contextual information, it improves the accuracy and robustness of the supervised model's diagnostic predictions, laying a solid core algorithmic foundation for the intelligent management of wind turbine units throughout their entire lifecycle.

[0270] In one embodiment of the wind turbine full life cycle prediction method based on multiphysics coupling in this application, it may further include the following:

[0271] Step S701: Perform feature dimensionality reduction and denoising on real-time multiphysics sensor data and real-time physical response data based on preset variational autoencoders and preset adversarial autoencoders, and determine the corresponding feature representations of the same magnitude;

[0272] Step S702: Input the same-scale feature representation into the unsupervised learning module, and determine the anomaly detection result and the operating condition clustering result output by the unsupervised learning module. The anomaly detection result is used to trigger the incremental update of the supervised learning module, and the operating condition clustering result is used to input the supervised learning module for model training and prediction.

[0273] Step S703: Input the same-scale feature representation and the results output by the unsupervised learning module into the supervised learning module to determine the fault diagnosis results and remaining life prediction results of the wind turbine output by the supervised learning module.

[0274] Optionally, in this embodiment, this step is the application process of the trained hybrid model.

[0275] Specifically, before formal data input, data preprocessing is required to standardize the data volume. The system simultaneously inputs high-dimensional, multi-source real-time sensor data (such as vibration waveforms and current signals) and physical field response data (such as stress distribution and temperature field) into a variational autoencoder (VAE) and an adversarial autoencoder (AAE). The VAE maps the input data to a probabilistic latent space through its encoder and then reconstructs it through the decoder. This process forces the model to learn robust and essential feature distributions in the data, thereby filtering out noise and discarding unimportant information. At the same time, the AAE ensures that the latent space features output by the encoder conform to a specific prior distribution (such as a Gaussian distribution) through adversarial training between its discriminator and generator (encoder-decoder). This further standardizes the feature representation and improves its regularity and separability. The two autoencoders work together to ultimately output a "feature representation of the same magnitude" with uniform dimensions, significant features, and removed redundant noise.

[0276] This step transforms multimodal data from different components, with varying dimensions and sampling frequencies, into a unified low-dimensional, dense vector with equivalent feature sensitivity. This lays the foundation for subsequent model fusion and computation. Furthermore, by removing noise and extracting robust features, it establishes a data foundation for the entire system with high generalization capability and high accuracy.

[0277] Optionally, in this embodiment, unsupervised preliminary state evaluation and pattern discovery are then performed. Feature representations of the same magnitude are input in parallel into the two core parts of the unsupervised learning module.

[0278] On one hand, the module utilizes anomaly detection algorithms such as isolated forests and support vector machines to learn the boundaries or distribution patterns of data under normal operating conditions. When the feature representation of real-time data deviates from this pattern or the reconstruction error exceeds a threshold, an anomaly alarm is triggered, enabling the initial capture of unknown faults and early, weak anomalies. These output anomaly alarms are used by the system to trigger incremental updates in downstream supervised learning modules. When the system continuously detects a new, unidentified anomaly pattern, these samples can be automatically marked as high-value samples, guiding expert intervention or directly entering the model's online learning queue, thereby driving the diagnostic system to continuously evolve and possess the ability to discover and adapt to "unknown unknowns" faults.

[0279] On the other hand, the module also employs algorithms such as deep embedding clustering to perform unlabeled clustering analysis on the feature representations, automatically dividing the data into different clusters. Each cluster corresponds to a specific operating condition (such as rated power generation, start-up and shutdown, turbulent operation, etc.) and outputs an operating condition label. The operating condition clustering results provide crucial contextual information for the subsequent supervised learning module.

[0280] Optionally, in this embodiment, this step is the decision-making stage for accurate diagnosis and quantitative prediction.

[0281] Based on features of the same magnitude, the supervised learning module uses complex models (such as CNN, GNN, LSTM, Transformer, etc.) trained on labeled historical data to perform tasks.

[0282] For fault diagnosis, the model analyzes the patterns in the feature representation, accurately classifies them into predefined fault types (such as bearing inner ring faults, gear tooth breakage, etc.), and outputs specific fault classification results.

[0283] For remaining lifetime prediction, the model uses regression analysis to learn the complex mapping relationship between feature representation and equipment performance degradation trajectory, predicting the operating time required for the equipment health index to drop to the failure threshold, which is the remaining lifetime prediction result.

[0284] Through step S703, this embodiment successfully achieves keen detection of early faults in wind turbines, accurate classification of fault modes, and high-precision prediction of remaining useful life (RUL) through the collaborative learning of hybrid models.

[0285] To improve the efficiency and accuracy of wind turbine generator fault prediction, this application provides an embodiment of a wind turbine generator lifecycle prediction device based on multiphysics coupling, which implements all or part of the aforementioned wind turbine generator lifecycle prediction method based on multiphysics coupling. See [link to embodiment]. Figure 2 The wind turbine full life cycle prediction device based on multi-physics coupling specifically includes the following components:

[0286] The wind turbine structure division module 10 is used to divide the overall structure of the wind turbine into structural components and related spaces, and to collect multi-physics sensor data from the structural components and related spaces. The structural components include blades, hubs, nacelles and towers, and the related spaces include blade-hub connection space, hub-nacelle connection space, nacelle-tower connection space and tower-blade relative space.

[0287] The wind turbine coupling equation determination module 20 is used to discretize the structural components of the wind turbine into multiple finite elements, define the relevant space as nodes connecting the finite elements, construct the corresponding wind turbine network topology, establish the control equations of fluid field, structural force field and spatial position field for each finite element in the local coordinate system, combine the control equations of all finite elements to determine the corresponding global coupling equation set, and solve the global coupling equation set by numerical method according to the sensor data acquisition of each finite element to determine the corresponding physical response data of the wind turbine in the whole field range.

[0288] The wind turbine life cycle self-diagnosis module 30 is used to input real-time multiphysics sensor data and real-time physical response data into a set hybrid learning model for unsupervised and supervised hybrid learning, so as to determine the life cycle prediction result of the wind turbine. The life cycle prediction result includes at least one of fault diagnosis result and remaining life prediction result.

[0289] As described above, the wind turbine lifecycle prediction device based on multiphysics coupling provided in this application can divide the overall structure of the wind turbine into structural components and related spaces, collect multiphysics sensor data from the overall structure, discretize the structural components of the wind turbine into multiple finite elements, define the related spaces as nodes connecting the finite elements, construct the corresponding wind turbine network topology, establish corresponding physical field control equations for the finite elements based on the network topology, determine the corresponding global coupling equation set, solve the global coupling equation set using numerical methods, determine the corresponding physical response data, and input the real-time multiphysics sensor data and real-time physical response data into a set hybrid learning model for unsupervised and supervised hybrid learning to determine the lifecycle prediction result of the wind turbine, thereby improving the efficiency and accuracy of wind turbine fault prediction.

[0290] To further illustrate this solution, this application also provides a specific application example of using the above-mentioned wind turbine life cycle prediction device based on multiphysics coupling to realize the wind turbine life cycle prediction method based on multiphysics coupling, which specifically includes the following:

[0291] 1. Application Examples

[0292] Application examples of the multiphysics coupled solver (achieving joint solution of three fields) are shown in Table 1 below:

[0293]

[0294] Table 1 Multiphysics Coupled Solver

[0295] 2. Innovative Summary

[0296] The multiphysics model proposed in this application is the "skeleton," while the supplementary machine learning engine is the "brain and neural network." The combination of the two makes it a truly "deep self-diagnostic" system that not only understands physical laws but also learns from data, discovers the unknown, and continuously grows. This greatly enhances the innovation, practicality, and competitive advantage of this application.

[0297] The innovative aspects of this application are as follows:

[0298] Innovation Point 1: The wind turbine is creatively decomposed into four structural units and four dynamic spaces, breaking through the limitations of traditional overall modeling. For the first time, "related space" is defined as dynamic boundary node, establishing failure propagation links between structural units and creating a discretized model of "structural unit + dynamic space" to solve the problem of cross-physical field coupling.

[0299] Innovation Point 2: In the local coordinate system Introducing a spatiotemporal degradation factor:

[0300]

[0301] Material aging coefficient

[0302] Innovation Point 3: Breakthrough in multi-field coupling, achieving for the first time a real-time joint solver for the fluid field, structural force field, and spatial position field. (Verification of improved computational efficiency:) (20 times improvement in computational efficiency)

[0303] Innovation Point 4: Self-evolving diagnosis, establishing the industry's first knowledge graph of wind turbine fault factors, with three types of self-updating mechanisms, realizing the self-evolution of the fault factor knowledge graph, and driving predictive maintenance into a new stage:

[0304] Parameter update: The material degradation coefficient is dynamically adjusted based on SCADA data.

[0305] Topology update: Automatically identifies newly added sensor nodes.

[0306] Equation update: when residual Reconstruct the control equations in time.

[0307] From a hardware perspective, in order to improve the efficiency and accuracy of wind turbine generator fault prediction, this application provides an embodiment of an electronic device for implementing all or part of the aforementioned wind turbine generator lifecycle prediction method based on multi-physics coupling. The electronic device specifically includes the following components:

[0308] The system comprises a processor, memory, a communications interface, and a bus; wherein the processor, memory, and communications interface communicate with each other via the bus; the communications interface is used to realize information transmission between the wind turbine full life cycle prediction method based on multiphysics coupling and core business systems, user terminals, and related databases and other related devices; the logic controller can be a desktop computer, tablet computer, or mobile terminal, etc., and this embodiment is not limited to these. In this embodiment, the logic controller can be implemented with reference to the embodiments of the wind turbine full life cycle prediction method based on multiphysics coupling in the previous embodiments, and the contents of the embodiments of the wind turbine full life cycle prediction method based on multiphysics coupling are incorporated herein, and repeated parts will not be described again.

[0309] It is understood that the user terminal may include smartphones, tablet computers, network set-top boxes, portable computers, desktop computers, personal digital assistants (PDAs), in-vehicle devices, smart wearable devices, etc. Among these, the smart wearable devices may include smart glasses, smartwatches, smart bracelets, etc.

[0310] In practical applications, parts of the wind turbine lifecycle prediction method based on multiphysics coupling can be executed on the electronic device side as described above, or all operations can be completed in the client device. The choice can be made based on the processing power of the client device and the limitations of the user's usage scenario. This application does not impose any limitations on this. If all operations are completed in the client device, the client device may further include a processor.

[0311] The aforementioned client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side; in other implementation scenarios, it may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, a server cluster consisting of multiple servers, or a distributed server structure.

[0312] Figure 3 This is a schematic block diagram illustrating the system configuration of the electronic device 9600 according to an embodiment of this application. Figure 3 As shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It is worth noting that... Figure 3 This is an example; other types of structures can also be used to supplement or replace this structure to achieve telecommunications functions or other functions.

[0313] In one embodiment, the function of the wind turbine full life cycle prediction method based on multiphysics coupling can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following control:

[0314] Step S101: Divide the overall structure of the wind turbine into structural components and related spaces, and collect multi-physics sensor data from the structural components and related spaces. The structural components include blades, hubs, nacelles and towers, and the related spaces include blade-hub connection space, hub-nacelle connection space, nacelle-tower connection space and tower-blade relative space.

[0315] Step S102: Discretize the structural components of the wind turbine into multiple finite elements, and define the relevant space as nodes connecting the finite elements to construct a corresponding wind turbine network topology. Based on the network topology, establish control equations for the fluid field, structural force field, and spatial position field for each finite element in the local coordinate system. Combine the control equations of all finite elements to determine the corresponding global coupling equation set. Solve the global coupling equation set using numerical methods based on the sensor data acquisition of each finite element to determine the physical response data of the wind turbine in the entire field.

[0316] Step S103: Input real-time multiphysics sensor data and real-time physical response data into the set hybrid learning model to perform unsupervised and supervised hybrid learning, so as to determine the life cycle prediction result of the wind turbine, wherein the life cycle prediction result includes at least one of fault diagnosis result and remaining life prediction result.

[0317] As described above, the electronic device provided in this application divides the overall structure of a wind turbine into structural components and related spaces, collects multi-physics sensor data from the overall structure, discretizes the structural components of the wind turbine into multiple finite elements, defines the related spaces as nodes connecting the finite elements, constructs a corresponding wind turbine network topology, establishes corresponding physical field control equations for the finite elements based on the network topology, determines the corresponding global coupling equation set, solves the global coupling equation set using numerical methods, determines the corresponding physical response data, and inputs real-time multi-physics sensor data and real-time physical response data into a set hybrid learning model for unsupervised and supervised hybrid learning to determine the life cycle prediction results of the wind turbine, thereby improving the efficiency and accuracy of wind turbine fault prediction.

[0318] In another embodiment, the wind turbine life cycle prediction method based on multiphysics coupling can be configured separately from the central processing unit 9100. For example, the wind turbine life cycle prediction method based on multiphysics coupling can be configured as a chip connected to the central processing unit 9100, and the function of the wind turbine life cycle prediction method based on multiphysics coupling can be realized through the control of the central processing unit.

[0319] like Figure 3 As shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily need to include these components. Figure 3 All components shown; in addition, the electronic device 9600 may also include Figure 3For components not shown, please refer to existing technologies.

[0320] like Figure 3 As shown, the central processing unit 9100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device, which receives inputs and controls the operation of various components of the electronic device 9600.

[0321] The memory 9140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 9100 may execute the program stored in the memory 9140 to perform information storage or processing, etc.

[0322] Input unit 9120 provides input to central processing unit 9100. Input unit 9120 may be, for example, a keypad or touch input device. Power supply 9170 provides power to electronic device 9600. Display 9160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0323] The memory 9140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142 for storing application programs and function programs or processes for executing the operation of the electronic device 9600 via the central processing unit 9100.

[0324] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for the electronic device for communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0325] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which is the same as in a conventional mobile communication terminal.

[0326] Based on different communication technologies, multiple communication modules 9110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132, thereby realizing typical telecommunications functions. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. Furthermore, the audio processor 9130 is also coupled to a central processing unit 9100, enabling on-device recording via the microphone 9132 and on-device playback of stored sound via the speaker 9131.

[0327] Embodiments of this application also provide a computer-readable storage medium capable of implementing all steps of the wind turbine full life cycle prediction method based on multiphysics coupling, where the execution subject is a server or client, as described in the above embodiments. The computer-readable storage medium stores a computer program that, when executed by a processor, implements all steps of the wind turbine full life cycle prediction method based on multiphysics coupling, where the execution subject is a server or client, as described in the above embodiments. For example, when the processor executes the computer program, it implements the following steps:

[0328] Step S101: Divide the overall structure of the wind turbine into structural components and related spaces, and collect multi-physics sensor data from the structural components and related spaces. The structural components include blades, hubs, nacelles and towers, and the related spaces include blade-hub connection space, hub-nacelle connection space, nacelle-tower connection space and tower-blade relative space.

[0329] Step S102: Discretize the structural components of the wind turbine into multiple finite elements, and define the relevant space as nodes connecting the finite elements to construct a corresponding wind turbine network topology. Based on the network topology, establish control equations for the fluid field, structural force field, and spatial position field for each finite element in the local coordinate system. Combine the control equations of all finite elements to determine the corresponding global coupling equation set. Solve the global coupling equation set using numerical methods based on the sensor data acquisition of each finite element to determine the physical response data of the wind turbine in the entire field.

[0330] Step S103: Input real-time multiphysics sensor data and real-time physical response data into the set hybrid learning model to perform unsupervised and supervised hybrid learning, so as to determine the life cycle prediction result of the wind turbine, wherein the life cycle prediction result includes at least one of fault diagnosis result and remaining life prediction result.

[0331] As described above, the computer-readable storage medium provided in this application divides the overall structure of a wind turbine into structural components and related spaces, collects multi-physics sensor data from the overall structure, discretizes the structural components of the wind turbine into multiple finite units, defines the related spaces as nodes connecting the finite units, constructs a corresponding wind turbine network topology, establishes corresponding physical field control equations for the finite units based on the network topology, determines the corresponding global coupling equation set, solves the global coupling equation set using numerical methods, determines the corresponding physical response data, and inputs real-time multi-physics sensor data and real-time physical response data into a set hybrid learning model for unsupervised and supervised hybrid learning to determine the life cycle prediction results of the wind turbine, thereby improving the efficiency and accuracy of wind turbine fault prediction.

[0332] Embodiments of this application also provide a computer program product capable of implementing all steps in the wind turbine full life cycle prediction method based on multiphysics coupling, where the execution subject is a server or client, as described in the above embodiments. When executed by a processor, this computer program / instruction implements the steps of the wind turbine full life cycle prediction method based on multiphysics coupling. For example, the computer program / instruction implements the following steps:

[0333] Step S101: Divide the overall structure of the wind turbine into structural components and related spaces, and collect multi-physics sensor data from the structural components and related spaces. The structural components include blades, hubs, nacelles and towers, and the related spaces include blade-hub connection space, hub-nacelle connection space, nacelle-tower connection space and tower-blade relative space.

[0334] Step S102: Discretize the structural components of the wind turbine into multiple finite elements, and define the relevant space as nodes connecting the finite elements to construct a corresponding wind turbine network topology. Based on the network topology, establish control equations for the fluid field, structural force field, and spatial position field for each finite element in the local coordinate system. Combine the control equations of all finite elements to determine the corresponding global coupling equation set. Solve the global coupling equation set using numerical methods based on the sensor data acquisition of each finite element to determine the physical response data of the wind turbine in the entire field.

[0335] Step S103: Input real-time multiphysics sensor data and real-time physical response data into the set hybrid learning model to perform unsupervised and supervised hybrid learning, so as to determine the life cycle prediction result of the wind turbine, wherein the life cycle prediction result includes at least one of fault diagnosis result and remaining life prediction result.

[0336] As described above, the computer program product provided in this application divides the overall structure of a wind turbine into structural components and related spaces, collects multi-physics sensor data from the overall structure, discretizes the structural components of the wind turbine into multiple finite elements, defines the related spaces as nodes connecting the finite elements, constructs a corresponding wind turbine network topology, establishes corresponding physical field control equations for the finite elements based on the network topology, determines the corresponding global coupling equation set, solves the global coupling equation set using numerical methods, determines the corresponding physical response data, and inputs real-time multi-physics sensor data and real-time physical response data into a set hybrid learning model for unsupervised and supervised hybrid learning to determine the life cycle prediction results of the wind turbine, thereby improving the efficiency and accuracy of wind turbine fault prediction.

[0337] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0338] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0339] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0340] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0341] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for predicting the entire life cycle of wind turbine generators based on multiphysics coupling, characterized in that, The method includes: The overall structure of the wind turbine is divided into structural components and related spaces. Multiphysics sensor data is collected from the structural components and related spaces. The structural components include blades, hubs, nacelles and towers. The related spaces include blade-hub connection space, hub-nacelle connection space, nacelle-tower connection space and tower-blade relative space. The structural components of the wind turbine are discretized into multiple finite elements, and the relevant spaces are defined as nodes connecting the finite elements to construct a corresponding wind turbine network topology. Based on the components or spaces to which the finite elements in the network topology belong, corresponding physical field control equations are assigned to them. The physical property parameters and material property parameters of the finite elements are used as the intrinsic coefficients of the control equations. The types of the physical field control equations include fluid fields, structural force fields, and spatial position fields. In a local coordinate system, spatial dimensions and time variables are set for the control equations to characterize the distribution and evolution of physical quantities within the finite elements. According to the node connection relationships in the network topology, physical field continuity conditions and force balance conditions are applied between adjacent finite elements as boundary constraints of the global coupling equation set. The physical field control equations of all finite elements with the applied boundary constraints are combined to determine the corresponding global coupling equation set. Based on the sensor data acquisition of each finite element, the global coupling equation set is solved using numerical methods to determine the physical response data of the corresponding wind turbine in the entire field. Real-time multiphysics sensor data and real-time physical response data are input into a hybrid learning model for unsupervised and supervised hybrid learning to determine the life cycle prediction results of the wind turbine. The life cycle prediction results include at least one of fault diagnosis results and remaining life prediction results.

2. The method for predicting the entire life cycle of wind turbines based on multiphysics coupling according to claim 1, characterized in that, The acquisition of multiphysics sensor data from the structural components and the associated space includes: Data on the blade system, hub system, nacelle drivetrain, tower system, relative space of components, and environmental and operating conditions are collected from the structural components and related spaces. The blade system data includes internal blade vibration, stress load, temperature, blade root flange clearance, and blade root bolt preload. The hub system data includes vibration of the pitch motor and pitch bearing, pitch motor speed and current, and hub flange clearance. The data on the engine room drivetrain includes engine room sway, main shaft sway and vibration, gearbox sway and vibration, gearbox oil quality, generator sway and vibration, and electrical parameters; The tower system data includes the tower sway posture, the preload of flange bolts at each level of the tower, and the preload of anchor bolts at the bottom of the tower. The relative space data of the components includes the clearance data between the blade tip and the tower.

3. The method for predicting the entire life cycle of wind turbines based on multiphysics coupling according to claim 1, characterized in that, The step of discretizing the structural components of the wind turbine into multiple finite elements and defining the relevant space as nodes connecting the finite elements to construct the corresponding wind turbine network topology includes: The continuous structures within the four structural components—blade, hub, nacelle, and tower—are each divided into multiple discrete finite element units. The four dynamic interactive regions, namely the blade-hub connection space, the hub-nacelle connection space, the nacelle-tower connection space, and the tower-blade relative space, are defined as the boundary nodes of the finite element units connecting different structural parts. Based on the actual physical connection and energy transfer path of the wind turbine, the connection relationship between the boundary node and the finite element unit of the adjacent structural part is established, and the corresponding wind turbine network topology is determined, wherein the finite element unit is used as the vertex of the graph, and the connection relationship is used as the edge of the graph.

4. The method for predicting the entire life cycle of wind turbines based on multiphysics coupling according to claim 1, characterized in that, The step of solving the globally coupled equations using numerical methods based on the sensor data collected by each finite element to determine the corresponding physical response data of the wind turbine across the entire field includes: The multiphysics sensor data corresponding to each finite element is used as the boundary conditions, initial conditions, and external excitation loads of the corresponding control equations and substituted into the global coupling equation set. The global coupled equations are solved by numerical iterative algorithm to determine the physical response data of the wind turbine in the entire field.

5. The method for predicting the entire life cycle of wind turbines based on multiphysics coupling according to claim 1, characterized in that, Before inputting real-time multiphysics sensor data and real-time physical response data into the hybrid learning model for unsupervised and supervised hybrid learning, the following steps are included: Based on historical data of wind turbines in a healthy state, a reconstruction model based on an autoencoder is trained to establish a baseline for the normal state and determine the corresponding unsupervised learning module after training. Based on historical data containing labeled fault types, a supervised learning model is trained, and the corresponding trained supervised learning module is determined. Based on the unsupervised learning module and the supervised learning module after training, the corresponding hybrid machine learning model is determined.

6. The method for predicting the entire life cycle of wind turbines based on multiphysics coupling according to claim 1, characterized in that, The step of inputting real-time multiphysics sensor data and real-time physical response data into a hybrid learning model for unsupervised and supervised hybrid learning includes: Based on the preset variational autoencoder and the preset adversarial autoencoder, feature dimensionality reduction and noise reduction are performed on real-time multiphysics sensor data and real-time physical response data to determine the corresponding feature representation of the same magnitude. The same-order feature representation is input into the unsupervised learning module to determine the anomaly detection result and the operating condition clustering result output by the unsupervised learning module. The anomaly detection result is used to trigger the incremental update of the supervised learning module, and the operating condition clustering result is used to input the supervised learning module for model training and prediction. The same-scale feature representation and the results output by the unsupervised learning module are input into the supervised learning module to determine the fault diagnosis results and remaining life prediction results of the wind turbine output by the supervised learning module.

7. A wind turbine full life cycle prediction device based on multiphysics coupling, characterized in that, The device includes: The wind turbine structure division module is used to divide the overall structure of the wind turbine into structural components and related spaces, and to collect multi-physics sensor data from the structural components and related spaces. The structural components include blades, hubs, nacelles and towers, and the related spaces include blade-hub connection space, hub-nacelle connection space, nacelle-tower connection space and tower-blade relative space. The wind turbine coupling equation determination module is used to discretize the structural components of the wind turbine into multiple finite elements, define the relevant spaces as nodes connecting the finite elements, construct a corresponding wind turbine network topology, assign corresponding physical field control equations to the components or spaces to which the finite elements belong in the network topology, and use the physical property parameters and material property parameters of the finite elements as the intrinsic coefficients of the control equations. The types of the physical field control equations include fluid fields, structural force fields, and spatial position fields. In a local coordinate system, spatial dimensions and time variables are set for the control equations to characterize the distribution and evolution of physical quantities within the finite elements. Based on the node connection relationships in the network topology, physical field continuity conditions and force balance conditions are applied between adjacent finite elements as boundary constraints for the global coupling equation set. The physical field control equations of all finite elements with applied boundary constraints are combined to determine the corresponding global coupling equation set. Based on the sensor data acquisition volume of each finite element, numerical methods are used to solve the global coupling equation set to determine the corresponding physical response data of the wind turbine in the entire field. The wind turbine life cycle self-diagnosis module is used to input real-time multiphysics sensor data and real-time physical response data into a set hybrid learning model for unsupervised and supervised hybrid learning, so as to determine the life cycle prediction result of the wind turbine. The life cycle prediction result includes at least one of fault diagnosis result and remaining life prediction result.

8. 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 program, it implements the steps of the wind turbine whole life cycle prediction method based on multiphysics coupling as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the wind turbine whole life cycle prediction method based on multiphysics coupling as described in any one of claims 1 to 6.

Images