A method and apparatus for simulating icing on wind turbine blades

Abstract

This invention relates to a method and apparatus for simulating icing of wind turbine blades. It establishes a state description model to describe the icing state of wind turbine blades; discretizes the state description model and solves the discretized model to obtain the solution results, with accelerated processing during both the discretization and solution processes; divides a preset icing time into multiple icing time nodes, and iteratively processes the solution results according to the order of the divided icing nodes to represent the icing process of multi-field coupled blades; this invention extends the potential flow model, enabling realistic and effective simulation of the three-dimensional unsteady blade icing process under multiple operating conditions, thus improving the accuracy of blade icing simulation.

Description

Technical Field

[0001] This invention relates to the field of wind power generation technology, and in particular to a method and apparatus for simulating icing on wind turbine blades. Background Technology

[0002] Blade icing occurs when the ambient temperature drops to or below freezing point, during the phase change of water droplets on the blade surface. It is more likely to occur in low-temperature, humid environments such as mountainous areas or near lakes. Blade icing alters the boundary layer flow state, causing boundary layer separation, increasing flow resistance and losses, and in severe cases, damaging the wind turbine blade structure. For the entire wind turbine unit, icing reduces the output of some turbines, leading to unstable actual power output, and in severe cases, causing the unit to shut down, resulting in wasted wind energy resources. Research on wind turbines suitable for cold climates, improving the reliability of wind turbine units and their ability to withstand various extreme weather conditions, has become a research direction of widespread international attention in recent years.

[0003] Wind turbine blade icing is an unsteady process, with continuous interactions between the airflow field, water droplet trajectory, and icing shape. The airflow field and water droplet trajectory determine the icing state, while changes in blade geometry during icing alter surface heat flux, water droplet collection efficiency, and shear stress. Most previous studies have used icing models applicable to two-dimensional or quasi-three-dimensional environments. Two-dimensional models explore icing based on individual airfoils, while quasi-three-dimensional models superimpose icing data from individual airfoils to represent complete blade icing. Both models reduce workload but cannot fully reflect the icing state of blades in a three-dimensional flow field. Therefore, establishing a complete three-dimensional wind turbine predictive icing model to study the icing characteristics of wind turbines under extreme conditions not only meets the needs of wind resource development in cold regions and the establishment of large-scale wind farms but also aligns with market demands and policies. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and apparatus for simulating icing of wind turbine blades, which addresses the shortcomings of the prior art.

[0005] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A method for simulating icing on wind turbine blades, comprising:

[0006] Establish a state description model to describe the icing state of wind turbine blades;

[0007] The state description model is discretized, and the discretized state description model is solved to obtain the solution result. Acceleration processing is performed during the discretization and solution processes.

[0008] The preset icing time is divided into multiple icing time nodes. The solution results are iteratively processed in the order of the divided icing nodes to describe the icing process of the multi-field coupled blades, wherein the multi-field coupled blades are wind turbine blades under different physical fields.

[0009] The beneficial effects of this invention are: establishing a state description model, performing discretization and solution processing on the state description model, and accelerating the processing; iterating the solution results sequentially according to the order of the segmented icing nodes to describe the icing process of multi-field coupled blades; this invention extends the potential flow model, which can realistically and effectively simulate the three-dimensional unsteady blade icing process under multiple working conditions, and improve the accuracy of blade icing simulation.

[0010] Based on the above technical solution, the present invention can be further improved as follows.

[0011] Furthermore, the establishment of a state description model for describing the icing state of wind turbine blades specifically includes:

[0012] A three-dimensional external flow model is established based on the water droplet continuity equation, the time-averaged Navier-Stokes equation, and the energy conservation equation.

[0013] A model of droplet motion and a model of supercooled large droplet impact in a three-dimensional droplet field are established based on the continuity equation and momentum equation of the droplet.

[0014] A continuous temperature field model is established based on the mass conservation equation and the energy conservation equation;

[0015] The state description model is obtained based on the three-dimensional external flow field model, the water droplet field model, and the temperature field continuity model.

[0016] Another technical solution of the present invention to solve the above-mentioned technical problems is as follows: A wind turbine blade icing simulation device, comprising:

[0017] The model building module is used to build a state description model to describe the icing state of wind turbine blades.

[0018] The solution module is used to discretize the state description model and solve the discretized state description model to obtain the solution result, and to accelerate the discretization and solution processes.

[0019] The description module is used to divide the preset icing time into multiple icing time nodes. The solution results are iteratively processed in the order of the divided icing nodes to describe the icing process of the multi-field coupled blades, wherein the multi-field coupled blades are wind turbine blades under different physical fields.

[0020] Another technical solution of the present invention to solve the above-mentioned technical problems is as follows: a wind turbine blade icing simulation device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a wind turbine blade icing simulation method as described above.

[0021] The advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating the wind turbine blade icing simulation method provided in this embodiment of the invention;

[0024] Figure 2 This is a flowchart illustrating the process of discretizing and solving the control equations using finite element analysis tools, and accelerating the discretization and solution of the control equations using a GPU, as provided in an embodiment of the present invention.

[0025] Figure 3 A block diagram of a wind turbine blade icing simulation device provided in an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0027] Example 1:

[0028] like Figure 1 As shown, a simulation method for icing wind turbine blades includes:

[0029] S1: Establish a state description model to describe the icing state of wind turbine blades;

[0030] S2: Discretize the state description model and solve the discretized state description model to obtain the solution result. Accelerate the discretization and solution processes.

[0031] S3: Divide the preset icing time into multiple icing time nodes. Iterate the solution results in sequence according to the order of the divided icing nodes to describe the icing process of the multi-field coupled blades. The multi-field coupled blades are wind turbine blades under different physical fields.

[0032] In the above embodiments, a state description model is established, and the state description model is discretized and solved, and accelerated. The solution results are iteratively processed according to the order of the segmented icing nodes to describe the icing process of multi-field coupled blades. This invention extends the potential flow model and can realistically and effectively simulate the three-dimensional unsteady blade icing process under multiple working conditions, thus improving the accuracy of blade icing simulation.

[0033] Specifically, S1: The establishment of a state description model for describing the icing state of wind turbine blades is as follows:

[0034] A three-dimensional external flow model is established based on the water droplet continuity equation, the time-averaged Navier-Stokes equation, and the energy conservation equation.

[0035] A model of droplet motion and a model of supercooled large droplet impact in a three-dimensional droplet field are established based on the continuity equation and momentum equation of the droplet.

[0036] A continuous temperature field model is established based on the mass conservation equation and the energy conservation equation;

[0037] The state description model is obtained based on the three-dimensional external flow field model, the water droplet field model, and the temperature field continuity model.

[0038] In this process, based on the flow control equations and the mass-energy conservation law, a three-dimensional external flow field model, a droplet field model, and a continuous temperature field model are established. Those skilled in the art can establish these three-dimensional flow field, droplet field, and continuous temperature field models using techniques known in the field (e.g., computer software such as Fluent UDF, Python, Fensap-ice, and OpenFoam) based on the flow control equations and the mass-energy conservation law.

[0039] It should be noted that most wind farms are built in high-altitude mountainous areas or near lakes with good wind energy resources. Wind turbine blades are highly susceptible to icing in low-temperature and humid environments. Icing alters the boundary layer flow state, causing boundary layer separation, increasing flow resistance and losses, and in severe cases, damaging the turbine blade structure. For the entire unit, icing reduces the output of some turbines, leading to unstable actual power output, and in severe cases, causing the unit to shut down, resulting in wasted wind energy resources. Furthermore, ice that falls or melts during wind turbine operation poses a threat to workers and nearby residents. Therefore, accurate assessment of wind turbine blade icing characteristics during the design phase provides crucial technical support for wind turbine control operation and anti-icing / de-icing structure design, effectively ensuring the reliability and service life of the wind turbine and reducing overall maintenance costs.

[0040] Therefore, in an exemplary embodiment of the present invention, the specific steps for establishing a three-dimensional predictive icing model for wind turbine blades are described using the establishment of a complete three-dimensional wind turbine blade icing model as an example. This describes the specific steps for establishing a three-dimensional flow field, water droplet field, and temperature field continuity model based on the flow control equations and the mass-energy conservation law. However, the present invention is not limited to this; icing models for other structures of wind turbines, such as towers and hubs, can also be established according to design requirements.

[0041] Specifically, the establishment of a three-dimensional external flow field model using the water droplet continuity equation, the time-averaged Navier-Stokes equation, and the energy conservation equation is as follows:

[0042] The continuity equation of the water droplet is established based on the mass conservation of the infinitesimal element flowing through the control volume in the three-dimensional flow field;

[0043] The solid wall turbulence problem is analyzed based on the one-equation turbulence model, and the time-averaged Navier-Stokes equation for low-velocity viscous flow is established based on the analysis results.

[0044] An energy conservation equation is established based on the energy conservation of the infinitesimal element flowing through the control volume in a three-dimensional flow field;

[0045] A three-dimensional external flow model is obtained based on the water droplet continuity equation, the time-averaged Navier-Stokes equation, and the energy conservation equation.

[0046] The three-dimensional external flow field consists of a two-phase flow of air and liquid water droplets.

[0047] More specifically, the steps for establishing a complete three-dimensional detailed model of a wind turbine blade include: establishing a three-dimensional wind turbine blade model consisting of a blade airfoil and a blade body, wherein the blade airfoil features include its blade chord length, leading edge radius, trailing edge angle, spars, mid-curvature, maximum thickness, maximum thickness position, maximum camber, and maximum camber position.

[0048] Since the flow characteristics of each field in the icing process and their coupling characteristics are fully considered in the process of establishing a complete three-dimensional wind turbine blade icing prediction model, the icing characteristics of the blade in three-dimensional unsteady state can be obtained more closely to reality by using the model for calculation.

[0049] Furthermore, without considering the aforementioned structural features of a complete wind turbine blade, a simplified model of the wind turbine blade can be established; that is, a cylindrical model can be established, and the characteristic chord length of the cylindrical model is equal to the chord length of the wind turbine blade. The advantage of establishing a simplified model of the wind turbine blade is that it allows for rapid 3D modeling. However, the simplified model cannot assess the icing shape and icing limit on the pressure and suction surfaces of the blade.

[0050] Specifically, a model of water droplet motion in a three-dimensional water droplet field is established based on the continuity equation and momentum equation of the water droplet, as follows:

[0051] The gas-liquid two-phase flow control equations are established based on the Euler method. These equations include a droplet continuity equation and a momentum equation. Specifically, the droplet continuity equation is established by introducing the droplet volume factor and droplet velocity distribution. The droplet continuity equation is as follows:

[0052]

[0053] Where α is the droplet volume factor. For the Laplace operator, The velocity of the water droplet;

[0054] The momentum equation is:

[0055]

[0056] Among them, C D R is the drag coefficient. e Where is the Reynolds number, and K is the inertia factor. ρ is the local airflow velocity. a ρ is the air density. b Let Fr be the density of the water droplet, and Fr be the Froude number. It is the acceleration due to gravity;

[0057] The water droplet motion model is obtained based on the continuity equation and the momentum equation.

[0058] It should be understood that the continuity equation, momentum equation, and energy equation all belong to the governing equations, meaning that the governing equations have a broader scope.

[0059] Since energy conversion is not involved in the water droplet motion process or water droplet field model, the governing equations in the model establishment process only include the water droplet continuity equation and momentum equation.

[0060] Specifically, a model of supercooled large water droplet impact in a three-dimensional water droplet field is established based on the water droplet continuity equation and momentum equation, as follows:

[0061] A splashing and bouncing model of water droplets impacting a blade surface is established using parameters defined by the Ohnesorge number, where Oh = μ. d / (dρ d σ d ) 1 / 2 ,

[0062] Using the critical Weber number defined by the Pilch number and Erdman number, a droplet splitting model is established based on the critical Weber number, which is: We critical =12(1+1.077Oh) 1.6 ),

[0063] Among them, W ecritical μ is the critical number. d Let d be the water droplet viscosity, d be the water droplet size, and σ be the viscous viscosity. d The surface tension of the water droplet;

[0064] Based on the splashing and bouncing model and the droplet splitting model, a supercooled large droplet impact model is obtained.

[0065] Specifically, the establishment of a continuous temperature field model based on the mass conservation equation and the energy conservation equation is as follows:

[0066] Selecting a water film element on the blade surface, and based on the fact that the mass of the water droplets flowing into the water film element is equal to the mass of the water droplets flowing out of the water film element, the mass conservation equation is established as follows:

[0067] The energy conservation equation is established based on the surface heat conversion process during blade icing:

[0068] in, The total amount of water droplets flowing in. The amount of water droplet impact. For the amount of ice, Evaporation amount The total amount of water droplets that flowed out. For the energy of the flowing water droplets, Energy from the impact of water droplets For friction heating, Phase transition energy, For the energy of evaporating water, The energy of the flowing water droplets, For convective heat transfer, For heat conduction.

[0069] Specifically, the flow characteristics of the backflow water on the three-dimensional blade surface are represented by the velocity of the water film on the blade surface. The velocity of the water film at point y in the infinitesimal element under the action of shear force is: Considering the thickness of the water film, the overall velocity of the water film can be expressed as the average velocity:

[0070] The surface roughness of icing surfaces increases with the increase of the heat transfer rate at the gas-liquid-solid interface. Using NASA's empirical correlation coefficient for icing, the surface roughness of icing blades is calculated as follows:

[0071] Where, μ ∞ = 1.79 × 10⁻⁵ Pa·s; T is the static temperature of the air; T ∞ =288K; C1 = 0.00216176W / (m·K3 / 2); c is the airfoil chord length; V ∞ Free flow velocity; LWC is the liquid water content; (k s / c) base =0.001177.

[0072] Specifically, S2: Discretize the state description model, solve the discretized state description model, and accelerate the discretization and solution processes, specifically as follows:

[0073] The state description model is discretized using finite element analysis tools, and the discretized state description model is solved. The control equations are accelerated during the discretization and solution processes using GPU acceleration.

[0074] The steps of discretizing and solving the control equations using finite element analysis tools and accelerating the discretization and solution process of the control equations using GPUs include: discretizing the established external flow field control equations using the finite volume method to obtain a set of algebraic equations for the variables to be solved; and using the SIMPLE algorithm to find the similarity matrix and solve the set of algebraic equations.

[0075] The first-order Euler method is used to solve the continuity equation and momentum equation of the water droplet phase to obtain the water droplet volume factor and water droplet velocity distribution at each node of the spatial grid, and then the water droplet impact area and impact amount on the blade surface are obtained.

[0076] The icing characteristics depend on the water droplet collection efficiency and water droplet impact limit on the blade surface, which satisfy the law of conservation of mass and energy. By combining the mass conservation equation and energy conservation equation of the control volume element on the blade surface, the icing thickness is calculated using the ratio of the product of the icing mass and the product of the ice density and the icing surface area. Thus, the distribution of water film thickness and icing characteristics on the blade surface are obtained.

[0077] By utilizing the general-purpose parallel computing architecture CUDA, all parameters and models are transferred to the GPU, and the model is discretized to accelerate the solution process. For each mesh node, an independent thread is created to calculate its velocity, pressure, temperature, surface heat flux, and shear force, and to update the information of the mesh node, thereby improving the efficiency of icing simulation in parallel.

[0078] It should be noted that those skilled in the art can, according to the specific calculation process, apply the complete detailed model of the three-dimensional wind turbine blade established in step S1 to the corresponding three-dimensional wind turbine blade icing prediction model, and, based on the three-dimensional wind turbine blade icing prediction model obtained in step S1, use general computational fluid dynamics and finite element analysis tools to discretize and solve it to obtain the flow characteristics of each field in the wind turbine blade icing prediction model. Those skilled in the art will understand that, in the specific calculation process, one or more conditions from the wind turbine blade icing conditions or custom special conditions can be selected for calculation according to actual needs.

[0079] Furthermore, simulations using general-purpose computational fluid dynamics and finite element analysis tools offer flexibility and freedom. Simultaneously, due to the discretization and solution of flow control equations across a large number of mesh nodes, parallel computing can be used to rapidly implement the aforementioned processing and model discretization and solution processes. Moreover, the general-purpose computational fluid dynamics and finite element analysis tools are not limited to Fensap-ice, but also include general-purpose commercial software such as Fluent UDF, Python, and OpenFoam. Additionally, the simulation method can also be implemented using computer code.

[0080] Specifically, S3: The preset icing time is divided into multiple icing time nodes. The solution results are iteratively processed according to the order of the divided icing nodes to describe the icing process of the multi-field coupled blades.

[0081] The preset icing time will be divided according to the preset time scale. The time interval between each icing step will be calculated using the empirical equation in the LOWICE code. The empirical equation in the LOWICE code is as follows:

[0082]

[0083] Where Δt2 is the time interval, ρ ice C is the density of ice, C1 is an empirical parameter, and T is the density of ice.icing To summarize the freezing time, N is the number of freezing intervals, c is the blade chord length, LWC is the liquid water content, and U... rel It is the relative velocity;

[0084] The solution results are iteratively processed sequentially according to the order of the segmented icing nodes to describe the icing process of the multi-field coupled blade. The iterative processing is to update the mesh information and boundary conditions in the state description model when the solution result of the nth step is obtained, and use the solution result of the nth step as the initial condition for the icing calculation of the (n+1)th step.

[0085] The mesh information and boundary conditions in the state description model are updated. Since the mesh information stores the calculation data and results, and the boundary conditions play a constraint role, both of them change dynamically during the freezing process. Therefore, they need to be updated to ensure the accuracy of the model.

[0086] For example: If the freezing time is 30 minutes, step 3 divides it into 3 segments, each segment being 10 minutes. After performing steps 1 and 2 on the first segment, a result is obtained. Based on this result, steps 1 and 2 are performed on the second segment to obtain another result. The same process is followed for the third segment to obtain the final result.

[0087] In this process, the icing time is segmented according to the LOWICE code to reveal the details of the multi-field coupled blade icing process. Specifically, the summarized icing time is segmented according to the time scale, and the icing step time interval is calculated using empirical equations in the LOWICE code. The calculation is performed in chronological order, and the mesh information and boundary conditions are updated at the end of the nth step calculation as the initial conditions for the (n+1)th step icing calculation.

[0088] Specifically, updating the mesh information may include mesh node locations, inflow velocity, liquid water content, median droplet diameter, ambient temperature, time, radius of gyration and rotation speed, and blade geometry information, but the updated mesh information is not limited to these.

[0089] It should be noted that boundary conditions generally refer to the initial information of the grid nodes at the boundary of the fluid domain, and are a prerequisite for solving the flow control equations. For example, they may include the incoming velocity component V at the fluid domain inlet. x V y V z The operating condition settings parameters include parameters such as azimuth angle, inflow velocity, median diameter distribution of water droplets, and freezing time under a certain operating condition.

[0090] like Figure 2As shown, the process of discretizing and solving the control equations using finite element analysis tools, and accelerating the discretization and solution of the control equations using GPU, may include: external flow field model 10, using the finite volume method to discretize the established external flow field control equations to obtain a set of algebraic equations for the variables to be solved, and using the SIMPLE algorithm to find the similarity matrix and solve the set of algebraic equations (step S4); water droplet field model 11, using the first-order Euler method to solve the continuity equation and momentum equation of the water droplet phase to obtain the water droplet volume factor and water droplet velocity distribution at each node of the spatial grid, and then obtaining the water droplet impact area and impact amount on the blade surface (step S5); temperature field model 12, the icing characteristics depend on the water droplet collection efficiency on the blade surface. The rate and water droplet impact limit satisfy the law of conservation of mass and energy. The mass conservation equation and energy conservation equation of the control volume element on the blade surface are solved simultaneously. The ice thickness is calculated by using the ratio of the product of ice mass, ice density and ice surface area. Thus, the water film thickness distribution and icing characteristics on the blade surface are obtained (step S6). GPU acceleration is performed by using the general parallel computing architecture CUDA to transfer all parameters and models to the GPU and discretize the model to accelerate the solution process (step S7). For each mesh node, an independent thread is created to calculate its velocity, pressure, temperature, surface heat flux and shear force, and the information of the mesh node is updated, thereby improving the efficiency of icing simulation in parallel.

[0091] An exemplary embodiment of the present invention provides a simulation method for icing of wind turbine blades, which can effectively evaluate and optimize the icing characteristics of wind turbines in cold climates in the early stage of design. This provides key technologies for the design and development of anti-icing and de-icing structures, improves the efficiency of wind turbines in cold climates, and can also shorten the design cycle and reduce design costs, thereby ensuring the high reliability and competitiveness of wind turbines.

[0092] Example 2:

[0093] like Figure 3 As shown, a wind turbine blade icing simulation device includes:

[0094] The model building module is used to build a state description model to describe the icing state of wind turbine blades.

[0095] The solution module is used to discretize the state description model and solve the discretized state description model to obtain the solution result, and to accelerate the discretization and solution processes.

[0096] The description module is used to divide the preset icing time into multiple icing time nodes. The solution results are iteratively processed in the order of the divided icing nodes to describe the icing process of the multi-field coupled blades, wherein the multi-field coupled blades are wind turbine blades under different physical fields.

[0097] Example 3:

[0098] A wind turbine blade icing simulation device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a wind turbine blade icing simulation method as described in any of the preceding claims.

[0099] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for simulating icing on wind turbine blades, characterized in that, include: Establish a state description model to describe the icing state of wind turbine blades; The state description model is discretized, and the discretized state description model is solved to obtain the solution result. Acceleration processing is performed during the discretization and solution processes. The preset icing time is divided into multiple icing time nodes. The solution results are iteratively processed in the order of the divided icing nodes to describe the icing process of the multi-field coupled blade. The multi-field coupled blade is a wind turbine blade under different physical fields. The establishment of a state description model for describing the icing state of wind turbine blades specifically involves: A three-dimensional external flow model is established based on the water droplet continuity equation, the time-averaged Navier-Stokes equation, and the energy conservation equation. A model of droplet motion and a model of supercooled large droplet impact in a three-dimensional droplet field are established based on the continuity equation and momentum equation of the droplet. A continuous temperature field model is established based on the mass conservation equation and the energy conservation equation; A state description model is obtained based on the three-dimensional external flow field model, the water droplet field model, and the temperature field continuity model. The model for the impact of supercooled large water droplets in a three-dimensional water droplet field, established based on the continuity equation and momentum equation, is as follows: A splashing and bouncing model of water droplets impacting a blade surface is established using parameters defined by the Ohnesorge number, where Oh = μ. d / (dρ d σ d ) 1 / 2 , Using the critical Weber number defined by the Pilch number and Erdman number, a droplet splitting model is established based on the critical Weber number, which is: We critical =12(1+1.077Oh) 1.6 ), Among them, W ecritical μ is the critical number. d Let d be the water droplet viscosity, d be the water droplet size, and σ be the viscous viscosity. d The surface tension of the water droplet; Based on the splashing and bouncing model and the droplet splitting model, a supercooled large droplet impact model is obtained.

2. The method for simulating icing of wind turbine blades according to claim 1, characterized in that, A three-dimensional external flow model is established using the water droplet continuity equation, the time-averaged Navier-Stokes equation, and the energy conservation equation. Specifically: The continuity equation of the water droplet is established based on the mass conservation of the infinitesimal element flowing through the control volume in the three-dimensional flow field; The solid wall turbulence problem is analyzed based on the one-equation turbulence model, and the time-averaged Navier-Stokes equation for low-velocity viscous flow is established based on the analysis results. An energy conservation equation is established based on the energy conservation of the infinitesimal element flowing through the control volume in a three-dimensional flow field; A three-dimensional external flow model is obtained based on the water droplet continuity equation, the time-averaged Navier-Stokes equation, and the energy conservation equation.

3. The method for simulating icing of wind turbine blades according to claim 1, characterized in that, A model of water droplet motion in a three-dimensional water droplet field is established based on the continuity equation and momentum equation of the water droplet, specifically as follows: The gas-liquid two-phase flow control equations are established based on the Euler method. These equations include a droplet continuity equation and a momentum equation. Specifically, the droplet continuity equation is established by introducing the droplet volume factor and droplet velocity distribution. The droplet continuity equation is as follows: Where α is the droplet volume factor. For the Laplace operator, The velocity of the water droplet; The momentum equation is: Among them, C D R is the drag coefficient. e Where is the Reynolds number, and K is the inertia factor. ρ is the local airflow velocity. a ρ is the air density. b Let Fr be the density of the water droplet, and Fr be the Froude number. It is the acceleration due to gravity; The water droplet motion model is obtained based on the continuity equation and the momentum equation.

4. The method for simulating icing of wind turbine blades according to claim 1, characterized in that, The establishment of a continuous temperature field model based on the mass conservation equation and the energy conservation equation is specifically as follows: Selecting a water film element on the blade surface, and based on the fact that the mass of the water droplets flowing into the water film element is equal to the mass of the water droplets flowing out of the water film element, the mass conservation equation is established as follows: The energy conservation equation is established based on the surface heat conversion process during blade icing: in, The total amount of water droplets flowing in. The amount of water droplet impact. For the amount of ice, Evaporation amount The total amount of water droplets flowing out. For the energy of the flowing water droplets, Energy generated by the impact of water droplets. For friction heating, Phase transition energy, Energy for evaporating water The energy of the flowing water droplets, For convective heat transfer, For heat conduction.

5. The method for simulating icing of wind turbine blades according to claim 1, characterized in that, The state description model is discretized, and the discretized state description model is solved. Acceleration is implemented during both the discretization and solution processes, specifically as follows: The state description model is discretized using finite element analysis tools, and the discretized state description model is solved. The control equations are accelerated during the discretization and solution processes using GPU acceleration.

6. The method for simulating icing of wind turbine blades according to claim 1, characterized in that, The preset icing time is divided into multiple icing time nodes. The solution results are iteratively processed sequentially according to the order of these icing nodes to describe the icing process of the multi-field coupled blades. Specifically: The preset icing time will be divided according to the preset time scale. The time interval between each icing step will be calculated using the empirical equation in the LOWICE code. The empirical equation in the LOWICE code is as follows: Where Δt2 is the time interval, ρ ice C is the density of ice, C1 is an empirical parameter, and T is the density of ice. icing To summarize the freezing time, N is the number of freezing intervals, c is the blade chord length, LWC is the liquid water content, and U... rel It is the relative velocity; The solution results are iteratively processed sequentially according to the order of the segmented icing nodes to describe the icing process of the multi-field coupled blade. The iterative processing is to update the mesh information and boundary conditions in the state description model when the solution result of the nth step is obtained, and use the solution result of the nth step as the initial condition for the icing calculation of the (n+1)th step.

7. A wind turbine blade icing simulation device, characterized in that, include: The model building module is used to build a state description model to describe the icing state of wind turbine blades. The solution module is used to discretize the state description model and solve the discretized state description model to obtain the solution result, and to accelerate the discretization and solution processes. The description module is used to divide the preset icing time into multiple icing time nodes, and to iteratively process the solution results according to the order of the divided icing nodes to describe the icing process of the multi-field coupled blades, wherein the multi-field coupled blades are wind turbine blades under different physical fields. The establishment of a state description model for describing the icing state of wind turbine blades specifically involves: A three-dimensional external flow model is established based on the water droplet continuity equation, the time-averaged Navier-Stokes equation, and the energy conservation equation. A model of droplet motion and a model of supercooled large droplet impact in a three-dimensional droplet field are established based on the continuity equation and momentum equation of the droplet. A continuous temperature field model is established based on the mass conservation equation and the energy conservation equation; A state description model is obtained based on the three-dimensional external flow field model, the water droplet field model, and the temperature field continuity model. The model for the impact of supercooled large water droplets in a three-dimensional water droplet field, established based on the continuity equation and momentum equation, is as follows: A splashing and bouncing model of water droplets impacting a blade surface is established using parameters defined by the Ohnesorge number, where Oh = μ. d / (dρ d σ d ) 1 / 2 , Using the critical Weber number defined by the Pilch number and Erdman number, a droplet splitting model is established based on the critical Weber number, which is: We critical =12(1+1.077Oh) 1.6 ), Among them, We critical μ is the critical number. d Let d be the water droplet viscosity, d be the water droplet size, and σ be the viscous viscosity. d The surface tension of the water droplet; Based on the splashing and bouncing model and the droplet splitting model, a supercooled large droplet impact model is obtained.

8. A wind turbine blade icing simulation device, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a wind turbine blade icing simulation method as described in any one of claims 1 to 6.

Images