Digital twinborn modeling method and device for wind turbine generator control system

Abstract

The invention provides a digital twin modeling method and device for a wind turbine generator control system, and relates to the technical field of new energy, and the method comprises the steps: constructing a multi-physical field mechanism model of the wind turbine generator control system according to obtained aerodynamic characteristic parameters, mechanical characteristic parameters and electrical characteristic parameters; performing comparative analysis based on the multi-physics field mechanism model and the preprocessed operation data to generate deviation data; if the deviation data is greater than a preset first threshold value, performing deviation prediction through a pre-trained deviation prediction model based on the preprocessed operation data, and generating a deviation correction value; and correcting a simulation result of the multi-physics field mechanism model by using the deviation correction value to form a data-mechanism dual-driven digital twinborn model. According to the digital twinning modeling method and device for the wind turbine generator control system, the output precision and environmental adaptability of the wind turbine generator digital twinning system under various working conditions are effectively improved, and the fitting ability and continuous stability of the model to the actual operation state are enhanced.

Description

Technical Field

[0001] This application relates to the field of new energy technology, and in particular to a digital twin modeling method and device for wind turbine control systems. Background Technology

[0002] With the rapid development of the wind power industry, the scale and number of wind turbine units continue to grow, placing higher demands on the real-time monitoring and performance evaluation of wind turbine unit operation status. Digital twin technology, as an important means to achieve refined management and intelligent operation and maintenance of wind turbine units, has gradually become a research and application hotspot. By constructing digital twin models of wind turbine units, the operating status of the units can be predicted and faults diagnosed in real time in a virtual environment, thereby effectively improving the operating efficiency and reliability of wind farms.

[0003] Existing digital twin modeling methods for wind turbines mainly rely on traditional mechanistic modeling and training with limited measured data. However, limitations such as insufficient computing resources, limited training sample size, high complexity of mechanistic models, and difficulty in comprehensively reflecting actual operating conditions lead to common problems in existing digital twin models, including low modeling accuracy, poor real-time performance, and insufficient reliability of simulation output results. These issues restrict the widespread application of digital twin technology in the wind power field. Especially under complex and dynamic environmental changes, traditional models often fail to reflect changes in wind turbine performance in a timely manner, affecting the predictive capabilities of the twin and the effectiveness of operation and maintenance decision support.

[0004] In recent years, with the development of artificial intelligence algorithms, models relying on large-scale data training and deep learning methods have achieved remarkable results in many fields. How to effectively combine artificial intelligence algorithms with traditional wind turbine mechanism modeling, fully utilizing physical mechanism knowledge while improving the model's adaptability and generalization performance, has become an important research direction for enhancing the level of digital twins for wind turbines.

[0005] Therefore, there is an urgent need to provide a method that can take into account the advantages of mechanism modeling and data-driven approaches in order to solve the problems of insufficient accuracy, poor real-time performance, and low reliability of twin models in existing technologies. Summary of the Invention

[0006] To address the problems in the prior art, this application provides a digital twin modeling method and apparatus for wind turbine control systems, which can solve the problems existing in the prior art.

[0007] Firstly, this application provides a digital twin modeling method for a wind turbine control system, including:

[0008] A multiphysics mechanism model of the wind turbine control system is constructed based on the obtained aerodynamic, mechanical, and electrical characteristic parameters.

[0009] Based on the multiphysics mechanism model and the preprocessed operational data, a comparative analysis is performed to generate deviation data.

[0010] If the deviation data is greater than a preset first threshold, a deviation prediction is performed based on the pre-processed running data using a pre-trained deviation prediction model to generate a deviation correction value.

[0011] The simulation results of the multiphysics mechanism model are corrected using the deviation correction value to form a data-mechanism dual-driven digital twin model.

[0012] Furthermore, the construction of a multiphysics mechanism model for the wind turbine control system based on the acquired aerodynamic, mechanical, and electrical characteristic parameters includes:

[0013] Construct an aerodynamic model based on the aforementioned aerodynamic characteristic parameters;

[0014] Construct a mechanical model based on the aforementioned mechanical characteristic parameters;

[0015] Construct an electrical model based on the electrical characteristic parameters;

[0016] The multiphysics mechanism model is constructed by integrating the aerodynamic model, the mechanical model, and the electrical model.

[0017] Furthermore, the comparative analysis based on the multiphysics mechanism model and the preprocessed operational data to generate deviation data includes:

[0018] Extract operating parameters from the preprocessed operating data;

[0019] Based on the operating parameters, the multiphysics mechanism model is used to perform simulation and generate simulation results;

[0020] The difference between the simulation results and the preprocessed running data is calculated to obtain the deviation data.

[0021] Furthermore, the step of pre-training the bias prediction model includes:

[0022] Obtain the pre-generated training dataset;

[0023] The training dataset is input into a bidirectional long short-term memory neural network for model training to obtain the bias prediction model.

[0024] Furthermore, the step of pre-generating the training dataset includes:

[0025] Obtain historical operating data of wind turbine units in wind farms;

[0026] The historical operation data is cleaned, denoised, and normalized.

[0027] Based on correlation analysis, feature variables that are highly correlated with simulation deviations are extracted from the processed historical operating data.

[0028] Based on the multiphysics mechanism model and the processed historical operating data, a comparative analysis is performed to generate historical deviation data.

[0029] The training dataset is generated using the feature variables and the corresponding historical deviation data.

[0030] Furthermore, the step of pre-generating the training dataset also includes:

[0031] Statistical analysis is performed on the historical operational data to generate extreme weather data;

[0032] The extreme weather data and the acquired extreme power grid environment data are used to simulate the operating status of wind turbine units and generate simulated operating data;

[0033] The training dataset is expanded using the simulated running data.

[0034] Furthermore, the comparative analysis based on the multiphysics mechanism model and the processed historical operating data to generate historical deviation data includes:

[0035] Extract historical operating condition parameters from the processed historical operating data;

[0036] Based on the historical operating parameters, the multiphysics mechanism model is used to perform simulation and generate historical simulation results.

[0037] The historical deviation data is obtained by calculating the difference between the historical simulation results and the processed historical operating data.

[0038] Furthermore, after correcting the simulation results of the multiphysics mechanism model using the deviation correction value to form a data-mechanism dual-driven digital twin model, the method further includes:

[0039] Real-time operating data of wind turbines in the wind farm is acquired at preset time intervals;

[0040] The digital twin model is verified using the real-time running data to determine the deviation between the simulation results of the digital twin model and the real-time running data.

[0041] If the deviation is greater than a preset second threshold, the digital twin model is updated using the real-time running data.

[0042] Furthermore, before generating deviation data through comparative analysis based on the multiphysics mechanism model and preprocessed operational data, the following steps are also included:

[0043] Obtain operational data of wind turbine units in wind farms;

[0044] The operational data is preprocessed; the preprocessing includes cleaning, noise reduction, and normalization.

[0045] Secondly, this application provides a digital twin modeling device for a wind turbine control system, comprising:

[0046] The mechanism model construction unit is used to construct a multi-physics mechanism model of the wind turbine control system based on the acquired aerodynamic, mechanical and electrical characteristic parameters.

[0047] The deviation data generation unit is used to perform comparative analysis based on the multiphysics mechanism model and the preprocessed running data to generate deviation data.

[0048] The deviation correction value generation unit is used to generate a deviation correction value based on the pre-processed running data and a pre-trained deviation prediction model if the deviation data is greater than a preset first threshold.

[0049] The digital twin model generation unit is used to correct the simulation results of the multiphysics mechanism model using the deviation correction value, so as to form a data-mechanism dual-driven digital twin model.

[0050] Furthermore, the mechanism model building unit includes:

[0051] An aerodynamic model construction module is used to construct an aerodynamic model based on the aerodynamic characteristic parameters.

[0052] A mechanical model building module is used to build a mechanical model based on the mechanical characteristic parameters.

[0053] An electrical model construction module is used to construct an electrical model based on the electrical characteristic parameters;

[0054] The mechanism model construction module is used to construct the multiphysics mechanism model by integrating the aerodynamic model, the mechanical model, and the electrical model.

[0055] Furthermore, the deviation data generation unit includes:

[0056] The operating condition parameter extraction module is used to extract operating condition parameters from the preprocessed operating data.

[0057] The simulation result generation module is used to perform simulation based on the operating parameters and the multiphysics mechanism model to generate simulation results.

[0058] The deviation data generation module is used to calculate the difference between the simulation results and the preprocessed running data to obtain the deviation data.

[0059] Furthermore, it also includes:

[0060] The dataset acquisition unit is used to acquire the pre-generated training dataset;

[0061] The model training unit is used to input the training dataset into a bidirectional long short-term memory neural network for model training to obtain the bias prediction model.

[0062] Furthermore, it also includes:

[0063] The historical operation data acquisition unit is used to acquire historical operation data of wind turbines in wind farms;

[0064] The data processing unit is used to clean, denoise, and normalize the historical operating data.

[0065] The feature variable extraction unit is used to extract feature variables that are highly correlated with simulation deviations from the processed historical running data based on correlation analysis.

[0066] The historical deviation data generation unit is used to perform comparative analysis based on the multiphysics mechanism model and the processed historical operating data to generate historical deviation data.

[0067] The dataset generation unit is used to generate the training dataset using the feature variables and the corresponding historical deviation data.

[0068] Furthermore, it also includes:

[0069] An extreme weather data generation unit is used to perform statistical analysis on the historical operational data to generate extreme weather data;

[0070] The simulated operation data generation unit is used to simulate the operating status of the wind turbine using the extreme weather data and the acquired extreme power grid environment condition data, and generate simulated operation data.

[0071] A dataset expansion unit is used to expand the training dataset using the simulated running data.

[0072] Furthermore, the historical deviation data generation unit includes:

[0073] The historical operating condition parameter extraction module is used to extract historical operating condition parameters from the processed historical operating data.

[0074] The historical simulation result generation module is used to perform simulation based on the historical operating parameters and the multiphysics mechanism model to generate historical simulation results.

[0075] The historical deviation data generation module is used to calculate the difference between the historical simulation results and the processed historical running data to obtain the historical deviation data.

[0076] Furthermore, it also includes:

[0077] The real-time operation data acquisition unit is used to acquire real-time operation data of the wind turbine in the wind farm at preset time intervals.

[0078] The model verification unit is used to verify the digital twin model using the real-time running data, and to determine the deviation between the simulation results of the digital twin model and the real-time running data.

[0079] The model update unit is used to update the digital twin model using the real-time running data if the deviation is greater than a preset second threshold.

[0080] Furthermore, it also includes:

[0081] The operation data acquisition unit is used to acquire the operation data of the wind turbine in the wind farm;

[0082] A data preprocessing unit is used to preprocess the running data; the preprocessing includes cleaning, noise reduction and normalization.

[0083] Thirdly, this application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the digital twin modeling method for the wind turbine control system described in any of the above embodiments.

[0084] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the digital twin modeling method for wind turbine control systems described in any of the above embodiments.

[0085] Fifthly, this application provides a computer program product, which includes a computer program that, when executed by a processor, implements the digital twin modeling method for wind turbine control systems described in any of the above embodiments.

[0086] This application provides a digital twin modeling method and apparatus for a wind turbine control system. It constructs a multiphysics mechanism model of the wind turbine control system based on acquired aerodynamic, mechanical, and electrical characteristic parameters. A comparative analysis is performed between the multiphysics mechanism model and preprocessed operating data to generate deviation data. If the deviation data exceeds a preset first threshold, a deviation prediction model is used to predict the deviation based on the preprocessed operating data, generating a deviation correction value. The deviation correction value is then used to correct the simulation results of the multiphysics mechanism model, forming a data-mechanism dual-driven digital twin model. This effectively improves the output accuracy and environmental adaptability of the wind turbine digital twin system under various operating conditions, enhancing the model's ability to fit actual operating states and its continuous stability.

[0087] This process involves constructing a multiphysics mechanism model of the wind turbine control system based on acquired aerodynamic, mechanical, and electrical parameters. This achieves high-fidelity modeling of the wind turbine's operating mechanism, providing an accurate physical basis for subsequent simulation analysis. By comparing and analyzing the multiphysics mechanism model with preprocessed operating data, deviation data is generated, enabling the identification and quantification of differences between the model's simulation output and the actual operating state. If the deviation data exceeds a preset first threshold, a deviation prediction model is used to predict the deviation based on the preprocessed operating data, generating a deviation correction value. This achieves intelligent prediction and dynamic compensation for significant deviations, improving the model's adaptability to complex operating conditions. The deviation correction value is used to correct the simulation results of the multiphysics mechanism model, forming a data-mechanism dual-driven digital twin model. This integrates the data-driven correction mechanism with mechanism modeling, improving the accuracy and real-time consistency of the digital twin model. The data-mechanism dual-driven modeling method can identify changes in wind turbine performance, allowing for timely model adjustments and ensuring the digital twin model follows these performance changes.

[0088] This application can fully combine the advantages of artificial intelligence algorithm technology and multiphysics high-precision modeling and simulation technology to improve the reliability of sample library data, improve the accuracy and computational efficiency of digital twin models, improve the generalization ability of models, reduce the problem of decreased model credibility caused by erroneous data introduced by traditional data augmentation techniques in artificial intelligence algorithms, and improve the reliability of models. Attached Figure Description

[0089] 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 only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0090] Figure 1 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application.

[0091] Figure 2 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application.

[0092] Figure 3 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application.

[0093] Figure 4 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application.

[0094] Figure 5 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application.

[0095] Figure 6 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application.

[0096] Figure 7 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application.

[0097] Figure 8 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application.

[0098] Figure 9 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application.

[0099] Figure 10 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application;

[0100] Figure 11 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application;

[0101] Figure 12This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application;

[0102] Figure 13 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application;

[0103] Figure 14 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application;

[0104] Figure 15 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application;

[0105] Figure 16 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application;

[0106] Figure 17 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application;

[0107] Figure 18 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application;

[0108] Figure 19 This is a schematic block diagram of the system configuration of an electronic device provided in an embodiment of this application;

[0109] Figure 20 This is a schematic diagram of the architecture of a digital twin model of a wind turbine provided in an embodiment of this application. Detailed Implementation

[0110] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the embodiments of this application will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments and their descriptions are used to explain this application, but are not intended to limit this application. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other.

[0111] The following describes the specific implementation process of the digital twin modeling method for wind turbine control systems provided in this application embodiment, using a server as the execution subject as an example.

[0112] Figure 1 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 1 As shown, the digital twin modeling method for wind turbine control systems provided in this application includes:

[0113] S101: Construct a multiphysics mechanism model of the wind turbine control system based on the obtained aerodynamic, mechanical and electrical characteristic parameters;

[0114] S102: Based on the multiphysics mechanism model and the preprocessed running data, a comparative analysis is performed to generate deviation data;

[0115] S103: If the deviation data is greater than a preset first threshold, based on the preprocessed running data, deviation prediction is performed through a pre-trained deviation prediction model to generate a deviation correction value;

[0116] S104: Use the deviation correction value to correct the simulation results of the multiphysics mechanism model to form a data-mechanism dual-driven digital twin model.

[0117] from Figure 1 As shown in the flowchart, this application provides a digital twin modeling method and apparatus for wind turbine control systems. It constructs a multiphysics mechanism model of the wind turbine control system based on acquired aerodynamic, mechanical, and electrical characteristic parameters. Based on the multiphysics mechanism model and preprocessed operating data, a comparative analysis is performed to generate deviation data. If the deviation data exceeds a preset first threshold, a deviation prediction model is used to predict the deviation based on the preprocessed operating data, generating a deviation correction value. The deviation correction value is then used to correct the simulation results of the multiphysics mechanism model, forming a data-mechanism dual-driven digital twin model. This effectively improves the output accuracy and environmental adaptability of the wind turbine digital twin system under various operating conditions, enhancing the model's ability to fit actual operating states and its continuous stability.

[0118] Each step is explained in detail below.

[0119] S101: Construct a multiphysics mechanism model of the wind turbine control system based on the obtained aerodynamic, mechanical and electrical characteristic parameters;

[0120] Specifically, the server first acquires the aerodynamic, mechanical, and electrical characteristic parameters involved in the design or actual operation of the wind turbine. These parameters reflect the fundamental physical characteristics and control behavior of the wind turbine at each subsystem level. Based on these parameters, a multiphysics mechanism model encompassing aerodynamics, mechanical dynamics, and electromagnetic processes can be constructed, providing an accurate physical basis for subsequent simulation analysis.

[0121] Figure 2 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 2 As shown, S101 includes:

[0122] S201: Construct an aerodynamic model based on the aerodynamic characteristic parameters;

[0123] Specifically, in order to construct a multiphysics mechanism model of the wind turbine control system, the model is divided into three core sub-models: aerodynamic, mechanical, and electrical, based on the typical operating characteristics of the wind turbine, in order to improve the structural clarity and simulation accuracy of the model.

[0124] First, based on the aerodynamic parameters of the wind turbine, an aerodynamic model is established to describe the wind energy conversion and the wind turbine response behavior, so as to reflect the response characteristics of the unit under different wind speeds and pitch conditions.

[0125] In one embodiment, the aerodynamic model is implemented in simulation software, and the formula for obtaining wind power by the rotation of the wind turbine is as follows:

[0126]

[0127] In the formula, ρ is the air density; S is the rotor swept area; v is the wind speed; C p The power coefficient, which is related to the tip speed ratio λ and the pitch angle β, is mainly determined by the blade design and can be approximated by the following formula:

[0128]

[0129] The overall thrust and power of the blade are calculated using the blade element theory. First, the combined airflow velocity at the blade element is calculated:

[0130]

[0131] In the formula, V x0 V represents the axial wind speed component. y0 V1 is the free-flow wind speed in the windward direction of the wind turbine, a and b are the axial and tangential induction factors, respectively, Ω is the angular velocity of the wind turbine, and r is the radial position of the current blade element.

[0132] The inflow angle φ and angle of attack α at the leaf element can be expressed as:

[0133]

[0134] In the formula, θ is the installation angle (helix angle), which is determined by the blade geometry.

[0135] The aerodynamic force dFa acting on a leaf element of length dr caused by the combined airflow velocity V0 can be decomposed into a normal force dFn and a tangential force dFt, which can be expressed as:

[0136]

[0137] In the formula, ρ is the air density; c is the chord length of the leaf element profile; Cn and Ct represent the normal force coefficient and tangential force coefficient, respectively, and the calculation formulas are as follows:

[0138]

[0139] At this point, the axial force (thrust) acting on the circular ring dr of the wind turbine plane can be expressed as:

[0140]

[0141] Where B is the number of blades. The torque acting on the circular ring dr of the wind turbine plane is:

[0142]

[0143] S202: Construct a mechanical model based on the aforementioned mechanical characteristic parameters;

[0144] Specifically, referencing the mechanical structure parameters and transmission characteristics of wind turbine generators, a mechanical model is constructed to describe torsional inertia, shaft vibration, and the dynamic behavior of rotating components, in order to simulate the transfer and loss of energy in the mechanical linkage. The mechanical part of the wind turbine generator mainly includes the transmission chain model.

[0145] In one embodiment, the mechanical model is implemented in simulation software using a two-mass transmission chain model:

[0146]

[0147] In the formula, T wt T outputs mechanical torque to the wind turbine. m T represents the mechanical torque on the motor side. gen For electromagnetic torque; H wt H is the inertial time constant of the wind turbine rotor; gen ω is the time constant of motor inertia; wt ω is the angular velocity of the wind turbine. gen K represents the angular velocity of the motor. m and D m These are the shaft stiffness and damping coefficient of the mechanical coupling, respectively.

[0148] S203: Construct an electrical model based on the electrical characteristic parameters;

[0149] Specifically, by combining the electrical parameters and control characteristics of the power generation side, an electrical model is established to describe the response characteristics of the generator, power electronic converter, and control loop, thus characterizing the dynamics of power output and control. The electrical model mainly includes the generator model.

[0150] In one embodiment, the electrical model is implemented in simulation software, using a doubly-fed asynchronous generator for modeling. The flux linkage and voltage formulas in the stator synchronous rotating coordinate system are as follows:

[0151]

[0152] V sd V sq V rd V rq These represent the voltage components of the stator along the d-axis and q-axis, and the voltage components of the rotor along the d-axis and q-axis, respectively; i sd i sq i rd i rq These represent the current components of the stator along the d-axis and q-axis, and the current components of the rotor along the d-axis and q-axis, respectively; λ sd , λ sq , λ rd , λ rq These represent the flux linkage components of the stator along the d-axis and q-axis, and the flux linkage components of the rotor along the d-axis and q-axis, respectively; R s L s These are the stator resistance and self-inductance, respectively; R r L r These are the rotor's resistance and self-inductance, respectively; L m ω1 is the equivalent mutual inductance between the coaxial stator and rotor in the dq coordinate system; ω2 is the rotational angular velocity in the dq coordinate system; ω1 = ω1 - ω2. r ω represents the angular velocity of the dq coordinate system relative to the rotor. r ω is the rotor's electric angular velocity.

[0153] The electromagnetic torque equation is as follows:

[0154]

[0155] In the formula, n p It is an extreme logarithm.

[0156] The mathematical model of the rotor-side converter in the two-phase synchronous dq coordinate system is as follows:

[0157]

[0158] In the formula, s dm s qm For the generator-side voltage source converter (VSC) three-phase switching function; u dm u qm Let i be the components of the generator-side VSC three-phase voltage in the dq coordinate system; sd i sq Let u be the component of the generator-side VSC three-phase current in the dq coordinate system; sd u sq Let ω be the components of the generator's three-phase stator voltage in the dq coordinate system; mR is the rotational angular frequency; s L s These are equivalent resistance and equivalent self-inductance, respectively; i dcm This is the equivalent DC current of the rotor-side converter.

[0159] The formula for calculating the output power of the generator-side VSC is:

[0160]

[0161] In the formula, p m q represents the active power output of the generator-side VSC. m This refers to the reactive power output from the generator-side VSC.

[0162] The mathematical model of the grid-side converter in the two-phase synchronous dq coordinate system is as follows:

[0163]

[0164] In the formula, u id u iq Let i be the components of the three-phase voltage of the converter in the dq coordinate system; id i iq Let i be the components of the three-phase current on the converter side in the dq coordinate system; gd i gq Let e ​​be the component of the three-phase current on the grid side in the dq coordinate system; d e q Let u be the components of the three-phase voltage of the power grid in the dq coordinate system; cd u cq Let ω be the components of the three-phase DC voltage on the grid side in the dq coordinate system; g C is the angular frequency of the rotation of the grid voltage vector; f For the VSC side filter capacitor; C d For busbar support capacitor; R g L g These are the grid-side filter inductor and the grid-side filter resistor, respectively; u dc DC bus voltage; s d s q These are the projection coefficients of the active power direction on the d and q axes, respectively.

[0165] The formula for calculating the output power of the grid-side converter is:

[0166]

[0167] In the formula, p g q represents the active power output of the grid-side converter. g The reactive power output of the grid-side converter; u gd ugq Let be the components of the three-phase voltage on the grid side in the dq coordinate system.

[0168] The DC bus capacitor power connecting the rotor-side converter and the grid-side converter is:

[0169]

[0170] In the formula, W represents the energy stored in the capacitor.

[0171] S204: By integrating the aerodynamic model, the mechanical model, and the electrical model, the multiphysics mechanism model is constructed.

[0172] Specifically, based on the above modeling, the aerodynamic model, mechanical model and electrical model are integrated to realize multi-physics coupling modeling of the entire process of wind turbine from energy input, mechanical transmission to electrical output, as the basic model for subsequent simulation analysis and digital twin inference.

[0173] S102: Based on the multiphysics mechanism model and the preprocessed running data, a comparative analysis is performed to generate deviation data;

[0174] Specifically, the server acquires pre-processed operational data collected during the actual operation of the wind turbine. The pre-processed operational data is then compared with the simulation results obtained based on the multiphysics mechanism model. The differences between the two are analyzed to obtain deviation data describing the model error, thereby quantifying the modeling accuracy of the multiphysics mechanism model.

[0175] Figure 3 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 3 As shown, S102 includes:

[0176] S301: Extract operating parameters from the preprocessed operating data;

[0177] Specifically, to improve the accuracy of the digital twin model of the wind turbine control system, the constructed multiphysics mechanism model needs to be effectively verified. The server extracts operating parameters from the pre-processed actual operating data of the wind turbine to characterize the operating state of the wind turbine. These operating parameters include, but are not limited to, wind speed, rotational speed, pitch angle, active power command, and grid frequency, and are consistent with the input conditions required for the multiphysics mechanism model simulation.

[0178] S302: Based on the operating parameters, perform simulation using the multiphysics mechanism model to generate simulation results;

[0179] Specifically, the server uses the extracted operating parameters as input conditions and inputs them into a multiphysics mechanism model for simulation calculations. The simulation process simulates the operating behavior of the wind turbine under the current operating conditions and outputs simulation results including key operating quantities such as torque, voltage, current, and power.

[0180] S303: Calculate the difference between the simulation results and the preprocessed running data to obtain the deviation data.

[0181] Specifically, the server compares and analyzes the simulation results with the corresponding output data in the preprocessed actual operating data, calculates the numerical differences between the two, and forms deviation data. This deviation data reflects the degree of agreement between the mechanism model and the actual system under the current operating conditions, and can serve as the basis for subsequent deviation correction and model optimization.

[0182] S103: If the deviation data is greater than a preset first threshold, based on the preprocessed running data, deviation prediction is performed through a pre-trained deviation prediction model to generate a deviation correction value;

[0183] Specifically, after completing the deviation analysis, the server determines whether the deviation exceeds the acceptable range, i.e., whether it is greater than a preset first threshold. If it exceeds this threshold, it indicates that there is a significant difference between the current mechanism model output and the actual operating conditions, making it difficult to meet the accuracy requirements. In this case, a data-driven mechanism needs to be introduced for correction.

[0184] If the deviation exceeds a threshold, the pre-trained deviation prediction model is invoked based on the pre-processed operational data to predict the deviation and output a correction value for compensation. This correction value is a quantitative compensation for the mechanism modeling error based on historical experience and learning results from actual operating conditions.

[0185] In one embodiment, if the deviation between the model simulation results and the actual running data does not exceed the first threshold, but there is still a certain gap, instead of using a complex learning model (deviation prediction model), some key parameters inside the multiphysics mechanism model are fine-tuned or locally corrected.

[0186] Figure 4 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 4 As shown, the steps for pre-training the bias prediction model include:

[0187] S401: Obtain the pre-generated training dataset;

[0188] Specifically, to enable the digital twin model of the wind turbine control system to have dynamic deviation correction capabilities, the deviation prediction model needs to be pre-trained to accurately predict the deviation between the mechanism model and the actual data during operation. The server first acquires a pre-generated training dataset, which includes multiple sample data sets. Each set of sample data contains key feature inputs of the wind turbine under different operating conditions and their corresponding target deviation values. This dataset is used to characterize the model deviation behavior that the wind turbine may exhibit under different conditions.

[0189] Figure 5 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 5 As shown, the steps for pre-generating the training dataset include:

[0190] S501: Obtain historical operating data of wind turbine units in wind farms;

[0191] Specifically, to train a neural network model with deviation prediction capabilities, a training dataset containing sufficient information needs to be constructed. This training dataset should effectively reflect the correspondence between the input characteristics and model output deviations of the wind turbine control system under different operating conditions. The server collects historical operating data generated by the wind turbine during actual operation. This historical operating data includes various key parameters closely related to the system's operating state, such as wind speed, turbine speed, output power, voltage, current, and blade pitch angle. This data can cover multiple operating stages and external environmental conditions, and is highly representative.

[0192] S502: Clean, denoise, and normalize the historical operation data;

[0193] Specifically, to improve data quality and neural network training effectiveness, the acquired raw historical data undergoes preprocessing. First, missing values, outliers, or incomplete records are removed. Second, high-frequency noise in the signal is removed using filtering, smoothing, or other statistical methods. Finally, to ensure that all features have a uniform numerical scale during training, normalization techniques are used to map all input data to a fixed range.

[0194] S503: Extract feature variables that are highly correlated with simulation deviations from processed historical operating data based on correlation analysis;

[0195] Specifically, based on preprocessed historical runtime data and using correlation analysis methods, the server selects input variables that are highly correlated with modeling bias from numerous features. These variables constitute the core input dimensions of the training dataset, helping to improve the model's predictive performance and generalization ability.

[0196] In one embodiment, the deviation value of the digital twin model of the wind turbine can be expressed as:

[0197] ΔP=P ac -P di (twenty two)

[0198] Among them, P ac P represents the actual operating power of the wind turbine. di This represents the power obtained from the simulation of the digital twin model under the same operating conditions. To ensure the accuracy of the training data, historical operating data needs to be cleaned and outliers removed before analysis.

[0199] Select wind speed v, electromagnetic torque Te, rotor speed Rpm, and yaw angle γ d Variables such as pitch angle β, which are observable in actual wind turbine operation, related to wind turbine output power, and prone to errors in modeling, are analyzed for their correlation with ΔP based on the Pearson correlation method. The formula can be expressed as:

[0200]

[0201] Among them, X i ,Y i Let i be the i-th observation of two variables. These are the sample means of the two variables, respectively.

[0202] The top 5 variables with the highest correlation were selected as input parameters for subsequent training of the bias prediction model.

[0203] S504: Based on the multiphysics mechanism model and the processed historical operating data, a comparative analysis is performed to generate historical deviation data;

[0204] Specifically, the server utilizes the pre-constructed multiphysics mechanism model, inputs operating parameters corresponding to historical operating data, and obtains the simulation results of the model under historical conditions. These simulation results are compared with the corresponding actual observations in the historical data, and the difference between the two is calculated, which is the historical deviation data, reflecting the error distribution of the mechanism model under various historical operating conditions.

[0205] Figure 7 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 7 As shown, S504 includes:

[0206] S701: Extract historical operating condition parameters from the processed historical operating data;

[0207] Specifically, the server extracts key operating parameters describing the wind turbine's operating status at various historical moments from preprocessed historical operating data. These parameters may include, but are not limited to, wind speed, turbine speed, pitch angle, output active power, grid voltage, and current, and are used to characterize the wind turbine's external conditions and internal control input status.

[0208] S702: Based on the historical operating parameters, simulation is performed using the multiphysics mechanism model to generate historical simulation results;

[0209] Specifically, the server inputs the aforementioned historical operating parameters into the constructed multiphysics mechanism model, performs simulation calculations, and obtains simulation output data corresponding to the historical operating states. The simulation results reflect the expected response behavior of the mechanism model under the corresponding historical conditions, including key quantities such as torque, voltage, and current.

[0210] S703: Perform difference calculation on the historical simulation results and the processed historical running data to obtain the historical deviation data.

[0211] Specifically, the server compares the historical simulation results of the mechanistic model based on historical operating condition parameters with the actual measured values ​​at corresponding time points in the historical operating data, and calculates the numerical difference between the two. The resulting difference value is the historical deviation data, used to characterize the error features between the mechanistic model and the actual system behavior under different historical operating conditions. This deviation data serves as the target output in the training dataset, providing a supervision signal for the subsequent training of the deviation prediction model.

[0212] S505: Generate the training dataset using the feature variables and the corresponding historical deviation data.

[0213] Specifically, the server uses the relevant feature variables obtained in the above steps as input data and historical deviation data as the target output value to construct a complete training dataset. This dataset is used for supervised training of the neural network model to achieve accurate prediction of model deviations during future operation.

[0214] Figure 6 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 6 As shown, the step of pre-generating the training dataset further includes:

[0215] S601: Perform statistical analysis on the historical operational data to generate extreme weather data;

[0216] Specifically, considering the lack of operational data for special operating conditions in actual wind turbine operation, data augmentation techniques are considered for data expansion. The server constructs simulated data reflecting extreme wind conditions based on the probability distribution characteristics of meteorological variables such as wind speed and temperature extracted statistically from historical operating data. By expanding the upper and lower limits of wind speed and constructing abrupt wind speed changes, typical extreme weather condition data such as high wind speed, low wind speed, and wind shear can be generated to supplement the less naturally distributed portion of the training samples.

[0217] S602: Using the extreme weather data and the acquired extreme power grid environment operating condition data, simulate the operating status of the wind turbine generator and generate simulated operating data;

[0218] Specifically, the server collects or constructs typical operating condition data representing abnormal grid operation states, such as grid frequency deviation, voltage drop, overvoltage, and transient power outages. This data can originate from historical waveform recordings, field test records, or extreme conditions set by technical specifications, and is used to simulate the response behavior of wind turbines under grid disturbance conditions. The constructed extreme weather and extreme grid environment data are then used as input conditions to the mechanistic model or test platform to simulate the operating state of wind turbines under extreme conditions, yielding output results such as current, voltage, and power. This simulated operating data has the ability to cover extreme boundary conditions, enhancing the breadth and robustness of the training dataset.

[0219] Some operating conditions are difficult to occur in actual operation, making digital twin modeling impossible in these scenarios. Therefore, experimental methods are needed to generate data for training. First, extreme weather data is generated based on historical data statistics and then converted into wind resource data, serving as the extreme weather environmental conditions for wind turbine operation simulation. Second, statistical analysis is performed on waveform data from grid faults and site tests to extract different types of extreme grid environment conditions, which are then converted into grid frequency, voltage, and other data, serving as the extreme grid environment conditions for wind turbine operation simulation. Finally, in a laboratory environment, the operating state of wind turbines under relevant conditions is simulated through physical wind turbine tests and hardware-in-the-loop simulation platforms. The actual output data is then used as a data sample library and input into the digital twin system for model training under extreme operating conditions.

[0220] S603: Expand the training dataset using the simulated running data.

[0221] Specifically, the server integrates the generated simulated operating data with the original training dataset, adding it as new training samples to the training set of the deviation prediction model. This expansion process improves the model's ability to identify and predict abnormal conditions, ensuring that the model maintains good accuracy and stability even when facing rare or sudden operating conditions. Through this data augmentation strategy, the problem of scarce extreme condition samples in wind turbine operating data is effectively solved, providing diverse and complex input-output comparison samples for the comprehensive training of the deviation prediction model, thereby significantly improving the modeling accuracy and adaptability of the digital twin model under boundary conditions.

[0222] S402: Input the training dataset into a bidirectional long short-term memory neural network for model training to obtain the bias prediction model.

[0223] Specifically, the training dataset is used as input and fed into a constructed bidirectional long short-term memory neural network (BiLSTM) for model training. This neural network can fully explore the relationship between current data and data from past and future time points, exhibiting high prediction accuracy for time-series data. By learning the mapping relationship between input features and target deviations in the training data, the neural network can establish a non-linear prediction capability from the input operating state to the deviation prediction output. During training, a loss function can be used to measure the prediction error, and model parameters can be optimized using algorithms such as backpropagation and gradient descent until the prediction accuracy meets the preset standard.

[0224] In one embodiment, the actual operating data of the top 5 variables most correlated with ΔP, along with ΔP, are used as training targets to train the BiLSTM model. The input and output formulas of this model are as follows:

[0225]

[0226] Among them, f t i t o t The output values ​​of the forget gate, input gate, and output gate are c, respectively. t To update intermediate variables for state, W f W i W c W0 and h represent the weight values ​​of each node, respectively. t-1 For output, x t For input, b f b i b c b o These represent the bias values ​​for each node.

[0227] During the operation of the digital twin model, five relevant variables appearing in the simulation are input into the BiLSTM model. The model outputs the ΔP value at time t. This value is then superimposed on the simulated output power of the digital twin model to form the actual output power value of the twin.

[0228] Through the above training steps, the obtained deviation prediction model has the ability to predict the actual deviations in the subsequent twin modeling process, providing a basis for correction, thereby enhancing the dynamic adaptive capability and modeling accuracy of the digital twin model.

[0229] S104: Use the deviation correction value to correct the simulation results of the multiphysics mechanism model to form a data-mechanism dual-driven digital twin model.

[0230] Specifically, the server superimposes the deviation correction value onto the original simulation results of the multiphysics mechanism model, thereby dynamically correcting the simulation output of the mechanism model. By applying the deviation correction value to the simulation process of the multiphysics mechanism model, the corrected model output is obtained, thus forming a data-mechanism dual-driven digital twin model. This improves the real-time performance, consistency, and accuracy of the model output, meeting the high-precision modeling requirements of wind turbine units under multiple operating conditions.

[0231] In one embodiment, the architecture of the digital twin model of the wind turbine is as follows: Figure 20 As shown.

[0232] Figure 8 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 8 As shown, after S104, it also includes:

[0233] S801: Acquire real-time operating data of wind turbine units in the wind farm at preset time intervals;

[0234] Specifically, to maintain the effectiveness and accuracy of the digital twin model of the wind turbine control system during long-term operation, it is necessary to periodically verify and update the model based on real-time operating data of the wind turbine after the twin model is built, so as to achieve dynamic maintenance of the model. The server automatically collects real-time operating data of the wind turbine at preset time intervals. Real-time operating data includes key operating parameters such as wind speed, grid voltage, current, speed, and power, which can be collected in real time through field control systems, sensor networks, or SCADA systems to ensure the timeliness and representativeness of the data.

[0235] S802: Use the real-time running data to verify the digital twin model and determine the deviation between the simulation results of the digital twin model and the real-time running data;

[0236] Specifically, the server uses the collected real-time operational data as a reference value and compares it with the simulation output of the current digital twin model under the same input conditions. By calculating the difference between the twin model output and the real-time operational data, it determines whether the current model can still accurately reflect the operating status of the wind turbine. This difference is the real-time deviation, reflecting the degree of synchronization between the twin model and the actual system.

[0237] S803: If the deviation is greater than the preset second threshold, the digital twin model is updated using the real-time running data.

[0238] Specifically, if the comparison results show that the current deviation value exceeds the preset second threshold, it indicates that the model's simulation capability has significantly decreased and cannot meet the accuracy requirements under the current operating conditions. At this time, the system will trigger the model update process, using the newly collected real-time data to perform parameter correction, deviation compensation, or retraining operations on the digital twin model in order to maintain the high fidelity of the model to the system's operating state.

[0239] Through the aforementioned periodic verification and adaptive update mechanism, the digital twin model is able to dynamically optimize as operating conditions evolve, significantly enhancing the model's online adaptability and long-term stability, and meeting the actual needs for refined monitoring and prediction throughout the entire life cycle of wind turbine units.

[0240] Figure 9 This is a flowchart illustrating a digital twin modeling method for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 9 As shown, before S101, it also includes:

[0241] S901: Acquire the operating data of wind turbine units in the wind farm;

[0242] Specifically, the server acquires raw operating data of the wind turbines during actual operation from the wind farm monitoring system or edge acquisition devices. The operating data is multi-source and multi-dimensional structured or semi-structured data, including steady-state data (minute-level data, second-level data) and transient data (millisecond-level data), containing physical quantities related to the operating status of the wind turbines such as wind speed, pitch angle, generator speed, converter output current, grid voltage, and frequency.

[0243] S902: Preprocess the running data; the preprocessing includes cleaning, noise reduction and normalization.

[0244] Specifically, to improve the accuracy of subsequent comparative analysis and the consistency of model input data, the acquired raw data needs to be standardized and preprocessed. The cleaning process removes missing, invalid, or outlier values ​​to avoid interfering with the interpretation of simulation results; denoising eliminates high-frequency noise or measurement errors through filtering, moving averages, wavelet transforms, etc., improving data stability; normalization maps data of different dimensions to a fixed numerical range (e.g., 0–1), ensuring that input variables have a uniform weight scale in subsequent calculations or models, which is beneficial for improving the robustness of comparative analysis and the efficiency of numerical computation.

[0245] Through the above preprocessing steps, the raw operating data can be transformed into a structured and standardized data format, ensuring a good correspondence and comparability with the data output from the mechanism model simulation, and providing a reliable data foundation for subsequent comparative analysis and deviation calculation.

[0246] This application provides a digital twin modeling method for wind turbine control systems. It constructs a multiphysics mechanism model of the wind turbine control system based on acquired aerodynamic, mechanical, and electrical characteristic parameters. A comparative analysis is performed between the multiphysics mechanism model and preprocessed operating data to generate deviation data. If the deviation data exceeds a preset first threshold, a deviation prediction model is used to predict the deviation based on the preprocessed operating data, generating a deviation correction value. The deviation correction value is then used to correct the simulation results of the multiphysics mechanism model, forming a data-mechanism dual-driven digital twin model. This effectively improves the output accuracy and environmental adaptability of the wind turbine digital twin system under various operating conditions, enhancing the model's ability to fit actual operating states and its continuous stability.

[0247] This process involves constructing a multiphysics mechanism model of the wind turbine control system based on acquired aerodynamic, mechanical, and electrical parameters. This achieves high-fidelity modeling of the wind turbine's operating mechanism, providing an accurate physical basis for subsequent simulation analysis. By comparing and analyzing the multiphysics mechanism model with preprocessed operating data, deviation data is generated, enabling the identification and quantification of differences between the model's simulation output and the actual operating state. If the deviation data exceeds a preset first threshold, a deviation prediction model is used to predict the deviation based on the preprocessed operating data, generating a deviation correction value. This achieves intelligent prediction and dynamic compensation for significant deviations, improving the model's adaptability to complex operating conditions. The deviation correction value is used to correct the simulation results of the multiphysics mechanism model, forming a data-mechanism dual-driven digital twin model. This integrates the data-driven correction mechanism with mechanism modeling, improving the accuracy and real-time consistency of the digital twin model. The data-mechanism dual-driven modeling method can identify changes in wind turbine performance, allowing for timely model adjustments and ensuring the digital twin model follows these performance changes.

[0248] This application can fully combine the advantages of artificial intelligence algorithm technology and multiphysics high-precision modeling and simulation technology to improve the reliability of sample library data, improve the accuracy and computational efficiency of digital twin models, improve the generalization ability of models, reduce the problem of decreased model credibility caused by erroneous data introduced by traditional data augmentation techniques in artificial intelligence algorithms, and improve the reliability of models.

[0249] Based on the same inventive concept, this application also provides a digital twin modeling device for a wind turbine control system, which can be used to implement the method described in the above embodiments, as described in the following embodiments. Since the principle of the digital twin modeling device for a wind turbine control system is similar to that of the digital twin modeling method for a wind turbine control system, the implementation of the digital twin modeling device for a wind turbine control system can refer to the implementation of the method based on software performance benchmarks; repeated details will not be elaborated further. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the system described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0250] Figure 10 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application, as shown below. Figure 10 As shown, the device includes:

[0251] Mechanism model construction unit 1001 is used to construct a multi-physics mechanism model of the wind turbine control system based on the acquired aerodynamic characteristic parameters, mechanical characteristic parameters and electrical characteristic parameters;

[0252] Specifically, the mechanism model construction unit 1001 first acquires the aerodynamic, mechanical, and electrical characteristic parameters involved in the design or actual operation of the wind turbine. These parameters reflect the basic physical characteristics and control behavior of the wind turbine at each subsystem level. Based on these parameters, a multiphysics mechanism model encompassing aerodynamics, mechanical dynamics, and electromagnetic processes can be constructed, providing an accurate physical basis for subsequent simulation analysis.

[0253] The deviation data generation unit 1002 is used to perform comparative analysis based on the multiphysics mechanism model and the preprocessed running data to generate deviation data.

[0254] Specifically, the deviation data generation unit 1002 acquires the preprocessed operating data collected during the actual operation of the wind turbine. The preprocessed operating data is compared with the simulation results obtained based on the multiphysics mechanism model, and the differences between the two are analyzed to obtain deviation data describing the model error, thereby quantifying the modeling accuracy of the multiphysics mechanism model.

[0255] The deviation correction value generation unit 1003 is used to generate a deviation correction value based on the pre-processed running data and a pre-trained deviation prediction model if the deviation data is greater than a preset first threshold.

[0256] Specifically, after completing the deviation analysis, the deviation correction value generation unit 1003 determines whether the deviation exceeds the acceptable range, i.e., whether it is greater than a preset first threshold. If it exceeds the threshold, it indicates that there is a significant difference between the current mechanism model output and the actual operating conditions, making it difficult to meet the accuracy requirements. In this case, a data-driven mechanism needs to be introduced for correction.

[0257] If the deviation exceeds a threshold, the pre-trained deviation prediction model is invoked based on the pre-processed operational data to predict the deviation and output a correction value for compensation. This correction value is a quantitative compensation for the mechanism modeling error based on historical experience and learning results from actual operating conditions.

[0258] In one embodiment, if the deviation between the model simulation results and the actual running data does not exceed the first threshold, but there is still a certain gap, instead of using a complex learning model (deviation prediction model), some key parameters inside the multiphysics mechanism model are fine-tuned or locally corrected.

[0259] The digital twin model generation unit 1004 is used to correct the simulation results of the multiphysics mechanism model using the deviation correction value, so as to form a data-mechanism dual-driven digital twin model.

[0260] Specifically, the digital twin model generation unit 1004 superimposes the deviation correction value onto the original simulation results of the multiphysics mechanism model, thereby achieving dynamic correction of the simulation output of the mechanism model. By applying the deviation correction value to the simulation process of the multiphysics mechanism model, the corrected model output is obtained, thus forming a data-mechanism dual-driven digital twin model. This improves the real-time performance, consistency, and accuracy of the model output, meeting the high-precision modeling requirements of wind turbine units under multiple operating conditions.

[0261] Figure 11 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application. Figure 10 Based on the embodiments, further, such as Figure 11 As shown, the digital twin modeling device for wind turbine control systems provided in this application also includes:

[0262] The aerodynamic model construction module 1101 is used to construct an aerodynamic model based on the aerodynamic characteristic parameters.

[0263] Mechanical model construction module 1102 is used to construct a mechanical model based on the mechanical characteristic parameters;

[0264] Electrical model construction module 1103 is used to construct an electrical model based on the electrical characteristic parameters;

[0265] The mechanism model construction module 1104 is used to construct the multiphysics mechanism model by integrating the aerodynamic model, the mechanical model and the electrical model.

[0266] Figure 12 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application. Figure 10 Based on the embodiments, further, such as Figure 12 As shown, the digital twin modeling device for wind turbine control systems provided in this application also includes:

[0267] The operating condition parameter extraction module 1201 is used to extract operating condition parameters from the preprocessed operating data.

[0268] The simulation result generation module 1202 is used to perform simulation based on the operating parameters and the multiphysics mechanism model to generate simulation results.

[0269] The deviation data generation module 1203 is used to calculate the difference between the simulation results and the preprocessed running data to obtain the deviation data.

[0270] Figure 13 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application. Figure 10Based on the embodiments, further, such as Figure 13 As shown, the digital twin modeling device for wind turbine control systems provided in this application also includes:

[0271] Dataset acquisition unit 1301 is used to acquire a pre-generated training dataset;

[0272] The model training unit 1302 is used to input the training dataset into the bidirectional long short-term memory neural network for model training to obtain the bias prediction model.

[0273] Figure 14 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application. Figure 13 Based on the embodiments, further, such as Figure 14 As shown, the digital twin modeling device for wind turbine control systems provided in this application also includes:

[0274] Historical operation data acquisition unit 1401 is used to acquire historical operation data of wind turbine units in wind farms;

[0275] Data processing unit 1402 is used to clean, denoise and normalize the historical operation data;

[0276] The feature variable extraction unit 1403 is used to extract feature variables that are highly correlated with simulation deviations from the processed historical running data based on correlation analysis.

[0277] The historical deviation data generation unit 1404 is used to perform comparative analysis based on the multiphysics mechanism model and the processed historical operation data to generate historical deviation data.

[0278] The dataset generation unit 1405 is used to generate the training dataset using the feature variables and the corresponding historical deviation data.

[0279] Figure 15 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application. Figure 14 Based on the embodiments, further, such as Figure 15 As shown, the digital twin modeling device for wind turbine control systems provided in this application also includes:

[0280] The extreme weather data generation unit 1501 is used to perform statistical analysis on the historical operational data and generate extreme weather data.

[0281] The simulation operation data generation unit 1502 is used to simulate the operating status of the wind turbine using the extreme weather data and the acquired extreme power grid environment operating condition data, and generate simulation operation data.

[0282] The dataset expansion unit 1503 is used to expand the training dataset using the simulated running data.

[0283] Figure 16 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application. Figure 14 Based on the embodiments, further, such as Figure 16 As shown, the digital twin modeling device for wind turbine control systems provided in this application also includes:

[0284] The historical operating condition parameter extraction module 1601 is used to extract historical operating condition parameters from the processed historical operating data.

[0285] The historical simulation result generation module 1602 is used to perform simulation based on the historical operating parameters and the multiphysics mechanism model to generate historical simulation results.

[0286] The historical deviation data generation module 1603 is used to calculate the difference between the historical simulation results and the processed historical running data to obtain the historical deviation data.

[0287] Figure 17 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application. Figure 10 Based on the embodiments, further, such as Figure 17 As shown, the digital twin modeling device for wind turbine control systems provided in this application also includes:

[0288] The real-time operation data acquisition unit 1701 is used to acquire the real-time operation data of the wind turbine in the wind farm at preset time intervals.

[0289] The model verification unit 1702 is used to verify the digital twin model using the real-time running data, and to determine the deviation between the simulation results of the digital twin model and the real-time running data.

[0290] The model update unit 1703 is used to update the digital twin model using the real-time running data if the deviation is greater than a preset second threshold.

[0291] Figure 18 This is a schematic diagram of the structure of a digital twin modeling device for a wind turbine control system provided in an embodiment of this application. Figure 10 Based on the embodiments, further, such as Figure 18 As shown, the digital twin modeling device for wind turbine control systems provided in this application also includes:

[0292] The operation data acquisition unit 1801 is used to acquire the operation data of the wind turbine in the wind farm;

[0293] The data preprocessing unit 1802 is used to preprocess the running data; the preprocessing includes cleaning, noise reduction and normalization.

[0294] This application provides a digital twin modeling method and apparatus for a wind turbine control system. It constructs a multiphysics mechanism model of the wind turbine control system based on acquired aerodynamic, mechanical, and electrical characteristic parameters. A comparative analysis is performed between the multiphysics mechanism model and preprocessed operating data to generate deviation data. If the deviation data exceeds a preset first threshold, a deviation prediction model is used to predict the deviation based on the preprocessed operating data, generating a deviation correction value. The deviation correction value is then used to correct the simulation results of the multiphysics mechanism model, forming a data-mechanism dual-driven digital twin model. This effectively improves the output accuracy and environmental adaptability of the wind turbine digital twin system under various operating conditions, enhancing the model's ability to fit actual operating states and its continuous stability.

[0295] This process involves constructing a multiphysics mechanism model of the wind turbine control system based on acquired aerodynamic, mechanical, and electrical parameters. This achieves high-fidelity modeling of the wind turbine's operating mechanism, providing an accurate physical basis for subsequent simulation analysis. By comparing and analyzing the multiphysics mechanism model with preprocessed operating data, deviation data is generated, enabling the identification and quantification of differences between the model's simulation output and the actual operating state. If the deviation data exceeds a preset first threshold, a deviation prediction model is used to predict the deviation based on the preprocessed operating data, generating a deviation correction value. This achieves intelligent prediction and dynamic compensation for significant deviations, improving the model's adaptability to complex operating conditions. The deviation correction value is used to correct the simulation results of the multiphysics mechanism model, forming a data-mechanism dual-driven digital twin model. This integrates the data-driven correction mechanism with mechanism modeling, improving the accuracy and real-time consistency of the digital twin model. The data-mechanism dual-driven modeling method can identify changes in wind turbine performance, allowing for timely model adjustments and ensuring the digital twin model follows these performance changes.

[0296] This application can fully combine the advantages of artificial intelligence algorithm technology and multiphysics high-precision modeling and simulation technology to improve the reliability of sample library data, improve the accuracy and computational efficiency of digital twin models, improve the generalization ability of models, reduce the problem of decreased model credibility caused by erroneous data introduced by traditional data augmentation techniques in artificial intelligence algorithms, and improve the reliability of models.

[0297] From a hardware perspective, in order to address the problems in the prior art, this application provides an embodiment of an electronic device for implementing all or part of the digital twin modeling method for the wind turbine control system, wherein the electronic device specifically includes the following:

[0298] The system comprises a processor, a 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 digital twin modeling device of the wind turbine control system and core business systems, user terminals, and related databases and other related equipment; 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 digital twin modeling method and the embodiment of the digital twin modeling device of the wind turbine control system in the embodiments, the content of which is incorporated herein, and repeated parts will not be described again.

[0299] 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.

[0300] In practical applications, parts of the digital twin modeling method for wind turbine control systems 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.

[0301] 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. 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.

[0302] Figure 19 This is a schematic block diagram illustrating the system configuration of the electronic device 9600 according to an embodiment of this application. Figure 19As 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 19 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.

[0303] In one embodiment, the digital twin modeling method for the wind turbine control system can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following control:

[0304] S101: Construct a multiphysics mechanism model of the wind turbine control system based on the obtained aerodynamic, mechanical and electrical characteristic parameters;

[0305] S102: Based on the multiphysics mechanism model and the preprocessed running data, a comparative analysis is performed to generate deviation data;

[0306] S103: If the deviation data is greater than a preset first threshold, based on the preprocessed running data, deviation prediction is performed through a pre-trained deviation prediction model to generate a deviation correction value;

[0307] S104: Use the deviation correction value to correct the simulation results of the multiphysics mechanism model to form a data-mechanism dual-driven digital twin model.

[0308] As described above, the digital twin modeling method and apparatus for wind turbine control systems provided in this application effectively improves the output accuracy and environmental adaptability of the wind turbine digital twin system under various operating conditions, and enhances the model's ability to fit actual operating states and its continuous stability. This application fully combines the advantages of artificial intelligence algorithm technology and multiphysics high-precision modeling and simulation technology to improve the reliability of sample database data, increase the accuracy and computational efficiency of the digital twin model, improve the model's generalization ability, and reduce the problem of decreased model credibility caused by erroneous data introduced by traditional data augmentation techniques in artificial intelligence algorithms, thereby improving the model's reliability.

[0309] In another embodiment, the digital twin modeling device for the wind turbine control system can be configured separately from the central processing unit 9100. For example, the data composite transmission device for the digital twin modeling device for the wind turbine control system can be configured as a chip connected to the central processing unit 9100, and the function of the digital twin modeling method for the wind turbine control system can be realized through the control of the central processing unit.

[0310] like Figure 19As 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 19 All components shown; in addition, the electronic device 9600 may also include Figure 19 For components not shown, please refer to existing technologies.

[0311] like Figure 19 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.

[0312] 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.

[0313] 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.

[0314] 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.

[0315] 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's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0316] The communication module 9110 is a transmitter / receiver 9110 that transmits and receives signals via the antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which can be the same as in a conventional mobile communication terminal.

[0317] 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 (transmitter / receiver) 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. Additionally, 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.

[0318] Embodiments of this application also provide a computer-readable storage medium capable of implementing all steps of the digital twin modeling method for wind turbine control systems with server or client execution subjects in the above embodiments. The computer-readable storage medium stores a computer program that, when executed by a processor, implements all steps of the digital twin modeling method for wind turbine control systems with server or client execution subjects in the above embodiments. For example, when the processor executes the computer program, it implements the following steps:

[0319] S101: Construct a multiphysics mechanism model of the wind turbine control system based on the obtained aerodynamic, mechanical and electrical characteristic parameters;

[0320] S102: Based on the multiphysics mechanism model and the preprocessed running data, a comparative analysis is performed to generate deviation data;

[0321] S103: If the deviation data is greater than a preset first threshold, based on the preprocessed running data, deviation prediction is performed through a pre-trained deviation prediction model to generate a deviation correction value;

[0322] S104: Use the deviation correction value to correct the simulation results of the multiphysics mechanism model to form a data-mechanism dual-driven digital twin model.

[0323] As described above, the digital twin modeling method and apparatus for wind turbine control systems provided in this application effectively improves the output accuracy and environmental adaptability of the wind turbine digital twin system under various operating conditions, and enhances the model's ability to fit actual operating states and its continuous stability. This application fully combines the advantages of artificial intelligence algorithm technology and multiphysics high-precision modeling and simulation technology to improve the reliability of sample database data, increase the accuracy and computational efficiency of the digital twin model, improve the model's generalization ability, and reduce the problem of decreased model credibility caused by erroneous data introduced by traditional data augmentation techniques in artificial intelligence algorithms, thereby improving the model's reliability.

[0324] 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.

[0325] 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.

[0326] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0327] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0328] 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 digital twin modeling method for a wind turbine control system, characterized in that, include: A multiphysics mechanism model of the wind turbine control system is constructed based on the obtained aerodynamic, mechanical, and electrical characteristic parameters. Based on the multiphysics mechanism model and the preprocessed operational data, a comparative analysis is performed to generate deviation data. If the deviation data is greater than a preset first threshold, a deviation prediction is performed based on the pre-processed running data using a pre-trained deviation prediction model to generate a deviation correction value. The simulation results of the multiphysics mechanism model are corrected using the deviation correction value to form a data-mechanism dual-driven digital twin model.

2. The digital twin modeling method for wind turbine control systems according to claim 1, characterized in that, The construction of a multiphysics mechanism model for the wind turbine control system based on the acquired aerodynamic, mechanical, and electrical characteristic parameters includes: Construct an aerodynamic model based on the aforementioned aerodynamic characteristic parameters; Construct a mechanical model based on the aforementioned mechanical characteristic parameters; Construct an electrical model based on the electrical characteristic parameters; The multiphysics mechanism model is constructed by integrating the aerodynamic model, the mechanical model, and the electrical model.

3. The digital twin modeling method for wind turbine control systems according to claim 1, characterized in that, The comparative analysis based on the multiphysics mechanism model and the preprocessed operational data generates deviation data, including: Extract operating parameters from the preprocessed operating data; Based on the operating parameters, the multiphysics mechanism model is used to perform simulation and generate simulation results; The difference between the simulation results and the preprocessed running data is calculated to obtain the deviation data.

4. The digital twin modeling method for wind turbine control systems according to claim 1, characterized in that, The steps for pre-training the bias prediction model include: Obtain the pre-generated training dataset; The training dataset is input into a bidirectional long short-term memory neural network for model training to obtain the bias prediction model.

5. The digital twin modeling method for wind turbine control systems according to claim 4, characterized in that, The steps for pre-generating the training dataset include: Obtain historical operating data of wind turbine units in wind farms; The historical operation data is cleaned, denoised, and normalized. Based on correlation analysis, feature variables that are highly correlated with simulation deviations are extracted from the processed historical operating data. Based on the multiphysics mechanism model and the processed historical operating data, a comparative analysis is performed to generate historical deviation data. The training dataset is generated using the feature variables and the corresponding historical deviation data.

6. The digital twin modeling method for wind turbine control systems according to claim 5, characterized in that, The step of pre-generating the training dataset also includes: Statistical analysis is performed on the historical operational data to generate extreme weather data; The extreme weather data and the acquired extreme power grid environment data are used to simulate the operating status of wind turbine units and generate simulated operating data; The training dataset is expanded using the simulated running data.

7. The digital twin modeling method for wind turbine control systems according to claim 5, characterized in that, The comparison and analysis based on the multiphysics mechanism model and the processed historical operating data generates historical deviation data, including: Extract historical operating condition parameters from the processed historical operating data; Based on the historical operating parameters, the multiphysics mechanism model is used to perform simulation and generate historical simulation results. The historical deviation data is obtained by calculating the difference between the historical simulation results and the processed historical operating data.

8. The digital twin modeling method for wind turbine control systems according to claim 1, characterized in that, After correcting the simulation results of the multiphysics mechanism model using the deviation correction value to form a data-mechanism dual-driven digital twin model, the method further includes: Real-time operating data of wind turbines in the wind farm is acquired at preset time intervals; The digital twin model is verified using the real-time running data to determine the deviation between the simulation results of the digital twin model and the real-time running data. If the deviation is greater than a preset second threshold, the digital twin model is updated using the real-time running data.

9. The digital twin modeling method for wind turbine control systems according to claim 1, characterized in that, Before generating deviation data through comparative analysis based on the multiphysics mechanism model and preprocessed operational data, the method further includes: Obtain operational data of wind turbine units in wind farms; The operational data is preprocessed; the preprocessing includes cleaning, noise reduction, and normalization.

10. A digital twin modeling device for a wind turbine control system, characterized in that, include: The mechanism model construction unit is used to construct a multi-physics mechanism model of the wind turbine control system based on the acquired aerodynamic, mechanical and electrical characteristic parameters. The deviation data generation unit is used to perform comparative analysis based on the multiphysics mechanism model and the preprocessed running data to generate deviation data. The deviation correction value generation unit is used to generate a deviation correction value based on the pre-processed running data and a pre-trained deviation prediction model if the deviation data is greater than a preset first threshold. The digital twin model generation unit is used to correct the simulation results of the multiphysics mechanism model using the deviation correction value, so as to form a data-mechanism dual-driven digital twin model.

11. The digital twin modeling device for wind turbine control system according to claim 10, characterized in that, The mechanism model construction unit includes: An aerodynamic model construction module is used to construct an aerodynamic model based on the aerodynamic characteristic parameters. A mechanical model building module is used to build a mechanical model based on the mechanical characteristic parameters. An electrical model construction module is used to construct an electrical model based on the electrical characteristic parameters; The mechanism model construction module is used to construct the multiphysics mechanism model by integrating the aerodynamic model, the mechanical model, and the electrical model.

12. The digital twin modeling device for wind turbine control system according to claim 10, characterized in that, The deviation data generation unit includes: The operating condition parameter extraction module is used to extract operating condition parameters from the preprocessed operating data. The simulation result generation module is used to perform simulation based on the operating parameters and the multiphysics mechanism model to generate simulation results. The deviation data generation module is used to calculate the difference between the simulation results and the preprocessed running data to obtain the deviation data.

13. The digital twin modeling device for wind turbine control system according to claim 10, characterized in that, Also includes: The dataset acquisition unit is used to acquire the pre-generated training dataset; The model training unit is used to input the training dataset into a bidirectional long short-term memory neural network for model training to obtain the bias prediction model.

14. The digital twin modeling device for wind turbine control system according to claim 13, characterized in that, Also includes: The historical operation data acquisition unit is used to acquire historical operation data of wind turbines in wind farms; The data processing unit is used to clean, denoise, and normalize the historical operating data. The feature variable extraction unit is used to extract feature variables that are highly correlated with simulation deviations from the processed historical running data based on correlation analysis. The historical deviation data generation unit is used to perform comparative analysis based on the multiphysics mechanism model and the processed historical operating data to generate historical deviation data. The dataset generation unit is used to generate the training dataset using the feature variables and the corresponding historical deviation data.

15. The digital twin modeling device for wind turbine control system according to claim 14, characterized in that, Also includes: An extreme weather data generation unit is used to perform statistical analysis on the historical operational data to generate extreme weather data; The simulated operation data generation unit is used to simulate the operating status of the wind turbine using the extreme weather data and the acquired extreme power grid environment condition data, and generate simulated operation data. A dataset expansion unit is used to expand the training dataset using the simulated running data.

16. The digital twin modeling device for wind turbine control system according to claim 14, characterized in that, The historical deviation data generation unit includes: The historical operating condition parameter extraction module is used to extract historical operating condition parameters from the processed historical operating data. The historical simulation result generation module is used to perform simulation based on the historical operating parameters and the multiphysics mechanism model to generate historical simulation results. The historical deviation data generation module is used to calculate the difference between the historical simulation results and the processed historical running data to obtain the historical deviation data.

17. The digital twin modeling device for wind turbine control system according to claim 10, characterized in that, Also includes: The real-time operation data acquisition unit is used to acquire real-time operation data of the wind turbine in the wind farm at preset time intervals. The model verification unit is used to verify the digital twin model using the real-time running data, and to determine the deviation between the simulation results of the digital twin model and the real-time running data. The model update unit is used to update the digital twin model using the real-time running data if the deviation is greater than a preset second threshold.

18. The digital twin modeling device for wind turbine control system according to claim 10, characterized in that, Also includes: The operation data acquisition unit is used to acquire the operation data of the wind turbine in the wind farm; A data preprocessing unit is used to preprocess the running data; The preprocessing includes cleaning, noise reduction, and normalization.

19. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 9.

20. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 9.

21. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 9.

Images