Numerical analysis method for predicting speed distribution of downwash airflow of unmanned aerial vehicle in rice canopy based on fluid-structure interaction method

Abstract

The invention discloses a numerical analysis method for predicting speed distribution of downwash airflow of an unmanned aerial vehicle in a rice canopy based on a fluid-solid coupling method. The numerical analysis method comprises the following steps: measuring physical parameters of rice plants based on physiological characteristics of rice; establishing a three-dimensional rice plant model based on three-dimensional geometric modeling software; dividing a three-dimensional rice plant canopy grid based on grid division software; establishing a fluid-solid coupling model of the lower washing flow field of the unmanned aerial vehicle and the rice canopy; setting an initial boundary condition and performing spatial discretization; full-coupling finite element numerical simulation of the fluid-solid coupling model is carried out; performing post-processing on the numerical simulation result based on flow field post-processing software to obtain an evolution process and spatial distribution of the downward washing flow field of the unmanned aerial vehicle and dynamic behaviors of rice plants; and analyzing the influence of the dynamic behavior of the rice plant on the downwash airflow of the unmanned aerial vehicle, and determining the distribution rule of the downwash airflow speed in the rice canopy. The problems of high time consumption, influence of external conditions, difficulty in copying, limitation and the like are effectively solved.

Description

Technical Field

[0001] The invention relates to unmanned aerial vehicle technology, and in particular to a numerical analysis method for predicting the velocity distribution of downwash airflow of unmanned aerial vehicle in a rice canopy based on a fluid-solid coupling method. Background Art

[0002] With the improvement of agricultural modernization and technological progress, the application of drones in the agricultural field is becoming more and more extensive. Drone spraying technology can improve the efficiency and accuracy of crop application, reduce the use of pesticides, reduce agricultural production costs, and also reduce the intensity of manual labor. It is an important technical means of agricultural production. In the process of drone spraying pesticides, due to the flight characteristics of the drone itself and the complex environment of farmland crops, it is difficult to ensure the spraying effect of pesticides in the canopy. Especially in the growth environment of crops such as rice fields, due to factors such as narrow crop spacing and dense canopies, the diffusion and distribution of pesticides after spraying pesticides are difficult to accurately control, which may lead to unsatisfactory spraying effects, and even cause problems such as pesticide waste and environmental pollution. At present, the research on the airflow velocity distribution in the canopy during drone spraying pesticides mainly relies on experimental methods and numerical simulation methods. The experimental method faces the challenges of high cost, high time consumption, limited experimental conditions, inability to fully consider the influence of parameters, and difficulty in real-time adjustment. Traditional numerical simulation methods mostly use unidirectional fluid-solid coupling methods, ignoring the impact of rice plant movement on the downwash airflow distribution of the UAV, resulting in a large deviation between the simulation results and the actual situation. Summary of the invention

[0003] The first purpose of the present invention is to overcome the shortcomings of the above existing technologies and provide a numerical analysis method for predicting the velocity distribution of the downwash airflow of a drone in the rice canopy based on the fluid-solid coupling method. This numerical analysis method for predicting the velocity distribution of the downwash airflow of a drone in the rice canopy based on the fluid-solid coupling method provides a scientific basis for optimizing the effect of pesticide spraying, improving the rice growth environment, enhancing the ability to control pests and diseases, and improving agricultural production efficiency.

[0004] The object of the present invention is achieved by the following technical solution: The numerical analysis method for predicting the velocity distribution of the downwash airflow of a UAV in a rice canopy based on the fluid-solid coupling method comprises the following steps:

[0005] S1. Measuring the physical parameters of rice plants based on the physiological characteristics of rice;

[0006] S2. Establishing a three-dimensional rice plant model based on three-dimensional geometric modeling software;

[0007] S3, dividing the three-dimensional rice plant canopy grid based on grid division software;

[0008] S4. Establish a fluid-solid coupling model between the downwash flow field of the UAV and the rice canopy;

[0009] S5, setting initial boundary conditions and spatial discretization;

[0010] S6. Perform a fully coupled finite element numerical simulation of the fluid-solid coupling model;

[0011] S7. Post-process the numerical simulation results based on the flow field post-processing software to obtain the evolution process and spatial distribution of the UAV downwash flow field and the dynamic behavior of the rice plants; analyze the impact of the dynamic behavior of the rice plants on the UAV downwash airflow, and clarify the distribution law of the downwash airflow velocity in the rice canopy.

[0012] Preferably, the physical parameters in step S1 include measuring characteristic parameters of rice plants and mechanical parameters of rice plants;

[0013] The characteristic parameters of rice plants include rice plant height, stem length, stem diameter, stem wall thickness, leaf length, leaf width, leaf thickness, leaf number and leaf area; the mechanical parameters of rice plants include stem elastic modulus, stem Poisson's ratio, leaf elastic modulus and leaf Poisson's ratio.

[0014] Preferably, step S3 comprises the following steps:

[0015] S31, importing the three-dimensional rice plant model into the finite element analysis software ABAQUS, and assigning the mechanical parameters of the rice plant to the material properties of the rice plant;

[0016] S32, dividing the rice plants using a hexahedral grid in a structured grid;

[0017] S33. A rice canopy model is obtained by using the circular and linear array methods. This rice canopy model is a solid mechanics model.

[0018] Preferably, step S4 includes the following specific steps:

[0019] S41. Establish a fluid mechanics model based on the lattice Boltzmann method and large eddy simulation method:

[0020]

[0021] Among them, μ t is the turbulent eddy viscosity, K w is a constant of 0.2, Δ is the unit length of the sub-grid model, H αβ and G αβ is the analytical scale strain rate tensor, δ αβ is the Kronecker symbol, g αβ , g βα and g γγare the components of the strain rate tensor;

[0022] S42, based on step S3, obtain a solid mechanics model:

[0023]

[0024] Where M is the mass matrix of rice plant nodes, are the acceleration vectors of rice plant nodes, F int is the internal force of the rice plant, F ext is the fluid force acting on the rice plant, N t is the basis function of the rice plant finite element, T is the Cauchy stress, V (m) is the volume of the mth rice plant node, V (n) is the volume of the nth rice plant node, f a ' is the force density acting on the nodes of rice plants;

[0025] S43. Using the immersed boundary method to realize the information interaction of the coupling between the fluid mechanics model and the solid mechanics model, we get:

[0026]

[0027] in, and are Euler variables and Lagrangian variables respectively, the Cartesian coordinates of the fluid Euler point x = (x, y, z), the Cartesian coordinates of the Lagrangian point of the rice canopy structure X = (o, i, j), Ω represents the flow field; Γ represents the boundary of the rice canopy structure, γ is the proportional factor of the reciprocity between the interpolation operator Φ and the expansion operator θ, δ(xX) is the discrete Dirac interpolation function, which interpolates the external force on the Lagrangian coordinates of the rice canopy structure to the force on the flow field on the Euler coordinates of the flow field.

[0028] Preferably, step S41 includes the following specific steps:

[0029] S411. Based on the Boltzmann equation, taking the velocity distribution function f as the basic variable, we get the following equations:

[0030]

[0031] f i (x+e i ·Δt,t+Δt)-f i (x,t)=Ω i , which is formula (10),

[0032] i=1,...,Q,(Q=2,3,...), this is formula (11),

[0033] Where t is the time, Δt is the time step, x is the lattice position, and e i is the discrete velocity, Ω i is the collision operator, i represents the finite velocity component of the distribution function; the velocity distribution function f is composed of the spatial position x, the particle velocity vector e i and time t constitutes a function f(x,e i ,,t), represents the distribution function f of the ith discrete velocity vector of a discrete node i The evolution law of

[0034] S412. Select the single-relaxed BGK model as the collision term, then:

[0035]

[0036] Where τ is the relaxation time, F is the external force applied to the fluid, and f i eq is the local equilibrium distribution function;

[0037] S413. Use the BGK model to discretize equations (9) to (10), and we get:

[0038]

[0039] S414. Based on formula (13), combined with large eddy simulation, a fluid mechanics model is established:

[0040]

[0041] Preferably, step S5 includes the following specific steps:

[0042] S51. In the fluid solver, specific boundary conditions are set according to the actual UAV plant protection operation conditions;

[0043] S52, establishing a spatial wind field model, wherein the bottom surface of the spatial wind field model is set as a wall boundary that enhances the wall function, the top surface is set as a zero pressure inlet boundary, and the other sides are set as zero pressure outlet boundaries; the surface of the rice plant is the wall boundary, the surface of the drone rotor is the non-slip and adiabatic inert wall boundary, and the remaining parts of the drone are set as non-slip wall boundaries;

[0044] S53. Use a discrete method to discretize the spatial fluid in the spatial wind field model.

[0045] Preferably, the spatial wind field model in step S52 establishes a 24×24m×6m virtual wind tunnel in the Xflow software, and imports a rice canopy model and a drone model to form a spatial wind field model.

[0046] Preferably, the discretization method in step S53 adopts Extended Discrete Element Method.

[0047] Preferably, step S6 includes the following specific steps:

[0048] In the fully coupled finite element numerical solution, the fluid-solid coupling between the fluid solver and the solid solver is managed using the joint simulation engine. The XML file in the fluid solver and the inp file generated by the solid solver are added to the system save file to create a joint simulation data exchange environment. The joint simulation engine is started by entering the code in the command window.

[0049] Fluid Solver and Solid Solver During the fluid-structure interaction process, displacements and coordinates are transferred from the solid solver to the fluid solver. At the same time, loads are transferred from the fluid solver to the solid solver.

[0050] The present invention has the following advantages over the prior art:

[0051] 1. The present invention can accurately simulate the distribution of downwash airflow generated by drones spraying pesticides in the rice canopy by predicting the velocity distribution of downwash airflow in the rice canopy, including the distribution and variation of downwash airflow velocity. This provides a scientific basis for optimizing the effect of pesticide spraying, improving the rice growth environment, enhancing the ability to control pests and diseases, and improving agricultural production efficiency, thereby promoting the sustainable development of agricultural production.

[0052] 2. The present invention utilizes the fluid-solid coupling method, combined with numerical simulation technology, to more accurately predict the velocity distribution of the downwash airflow of a UAV in the rice canopy. Compared with traditional methods, the present invention can better describe the dynamic behavior of rice plants under the action of the downwash airflow, as well as the influence of the dynamic behavior of rice plants on the velocity distribution of the downwash airflow in the rice canopy, thereby being able to provide more accurate results and better explain and understand the phenomenon.

[0053] 3. The present invention adopts a numerical simulation method, which can simulate and predict the velocity distribution of the downwash airflow of the UAV in the rice canopy in a relatively short time. Compared with the experimental method, this invention not only saves a lot of time and resources, but also can realize simulation under a variety of different working conditions, thereby improving research efficiency.

[0054] 4. The numerical simulation technology based on the fluid-solid coupling method of the present invention has certain feasibility and adaptability, and can better simulate complex physical phenomena and flow behaviors. This invention provides a feasible numerical simulation method for the study of UAV downwash airflow in the rice canopy, and provides new approaches and tools for research in related fields.

[0055] 5. The numerical simulation method provided by the present invention will help to better understand the UAV downwash process, so as to optimize the UAV spraying operation parameters, reduce the drift of droplets, increase the uniform deposition of droplets in the crop canopy, improve crop yield and quality, and reduce the impact on the environment.

[0056] 6. The present invention uses the finite element analysis software Abaqus to establish a three-dimensional model of the rice canopy, which can accurately restore the actual situation in the field experiment in the numerical simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a flow chart of a numerical analysis technology for predicting the velocity distribution of UAV downwash airflow in a rice canopy based on a fluid-solid coupling method of the present invention.

[0058] Figure 2 It is a schematic diagram of measuring mechanical parameters of rice plants according to the present invention.

[0059] Figure 3 It is a schematic diagram of measuring a rice plant placed on a universal testing machine according to an embodiment of the present invention.

[0060] Figure 4 It is a schematic diagram of the structure of a single rice plant according to an embodiment of the present invention.

[0061] Figure 5 Schematic diagram of row spacing and plant spacing of rice plants according to an embodiment of the present invention.

[0062] Figure 6 Schematic diagram of initial boundary conditions of the calculation domain according to an embodiment of the present invention.

[0063] Among them, 1 is the universal testing machine, 2 is the induction pressure head, 3 is the support, 4 is the drone, 5 is the rice canopy model, and 6 is the calculation domain. DETAILED DESCRIPTION

[0064] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0065] like Figure 1 As shown in FIG. 1 , a numerical analysis method for predicting the velocity distribution of the downwash airflow of a UAV in a rice canopy based on a fluid-solid coupling method includes the following steps:

[0066] S1. Measuring physical parameters of rice plants based on the physiological characteristics of rice; specifically, the physical parameters in step S1 include measuring characteristic parameters of rice plants and mechanical parameters of rice plants;

[0067] The characteristic parameters of rice plants include rice plant height, stem length, stem diameter, stem wall thickness, leaf length, leaf width, leaf thickness, leaf number and leaf area; the mechanical parameters of rice plants include stem elastic modulus, stem Poisson's ratio, leaf elastic modulus and leaf Poisson's ratio. In this embodiment, a universal testing machine is used to measure the mechanical parameters of rice, and the process is as follows:

[0068] like Figure 2 and 3 As shown in the figure, 36 rice stems were randomly selected and evenly divided into three equal parts, and samples with a middle length A1 of 70 mm were taken from each part; 36 rice top leaves were selected and evenly divided into three equal parts, and samples with a middle length A2 of 70 mm were taken from each part; first, the induction pressure head and support were installed on the universal testing machine, and the rice stem samples or leaf samples were placed on the support of the universal testing machine; then, the pressure head was started to move downward at a constant speed of 1 mm / s; finally, the deformation and load increment of the sample were collected, and the stress-displacement curve of the sample was obtained by the universal testing machine, and then the elastic modulus of the rice stem was obtained using formulas (14) and (15), and the elastic modulus of the rice leaf was obtained using formulas (16) and (17):

[0069]

[0070] Where: μ = (D–2d) / D, D is the diameter of the rice stem sample, d is the wall thickness of the rice stem sample, and μ is the influence of the wall thickness of the sample on the overall diameter of the sample; ΔG is the load increment in N, Δδ is the center displacement increment in mm; L is the horizontal support span (L = 50 mm); ν is the Poisson’s ratio; I1 and I2 are the moments of inertia of the rice stem and leaf, respectively, in mm 4 ; b and Δh are the blade width and thickness, respectively.

[0071] E1 is the elastic modulus of rice stem, and E2 is the elastic modulus of rice leaf.

[0072] S2. Establish a three-dimensional rice plant model based on three-dimensional geometric modeling software; specifically, Figures 4 to 6 As shown, for a given rice plant, a three-dimensional rice plant canopy model is established using the three-dimensional geometric modeling software SolidWorks based on the measured physical parameters of the rice plant; the single rice plant model of this embodiment has five leaves, which are distributed at 180°, and the angle θ between the leaves and the stem is between 0° and 30°; the length Y, width W and height H of the rice plant canopy model are 12m, 12m and 1m respectively, and the row spacing N and plant spacing M of the rice plant in the three-dimensional rice plant canopy model are 300mm and 300mm respectively.

[0073] S3, dividing the three-dimensional rice plant canopy grid based on the grid division software; step S3 includes the following steps:

[0074] S31, importing the three-dimensional rice plant model into the finite element analysis software ABAQUS, and assigning the mechanical parameters of the rice plant to the material properties of the rice plant;

[0075] S32, dividing the rice plants using a hexahedral grid in a structured grid;

[0076] S33. A rice canopy model is obtained by using the circular and linear array methods. This rice canopy model is a solid mechanics model.

[0077] Among them, the grid division of the stems and leaves of the three-dimensional rice plants is carried out in the Abaqus software, using the hexahedral grid in the structured grid. The stems and leaves of the rice need to be meshed encrypted. The grid division size of the stems is 0.4mm, and the grid division size of the leaves is 0.05mm. This can capture the details in the flow field more finely, improve the accuracy of the simulation, and make the calculation results closer to the actual physical phenomena, so as to better understand the movement behavior of rice plants in the flow field.

[0078] S4. Establish a fluid-solid coupling model between the downwash flow field of the UAV and the rice canopy;

[0079] Step S4 includes the following specific steps:

[0080] S41. Establish a fluid mechanics model based on the lattice Boltzmann method and large eddy simulation method:

[0081]

[0082] Among them, μ t is the turbulent eddy viscosity, K w is a constant of 0.2, Δ is the unit length of the sub-grid model, H αβ and G αβ is the analytical scale strain rate tensor, δ αβ is the Kronecker symbol, g αβ , g βα and g γγ are the components of the strain rate tensor;

[0083] Step S41 includes the following specific steps:

[0084] S411. Based on the Boltzmann equation, taking the velocity distribution function f as the basic variable, we get the following equations:

[0085]

[0086] f i (x+e i ·Δt,t+Δt)-f i (x,t)=Ωi , which is formula (10),

[0087] i=1,...,Q,(Q=2,3,...), this is formula (11),

[0088] Where t is the time, Δt is the time step, x is the lattice position, and e i is the discrete velocity, Ω i is the collision operator, i represents the finite velocity component of the distribution function; the velocity distribution function f is composed of the spatial position x, the particle velocity vector e i and time t constitutes a function f(x,e i ,,t), represents the distribution function f of the ith discrete velocity vector of a discrete node i The evolution law of

[0089] S412. Select the single-relaxed BGK model as the collision term, then:

[0090]

[0091] Where τ is the relaxation time, F is the external force applied to the fluid, and f i eq is the local equilibrium distribution function;

[0092] S413. Use the BGK model to discretize equations (9) to (10), and we get:

[0093]

[0094] For the D3Q27 lattice model used in this embodiment, the discretized particle velocity vector is e i :

[0095]

[0096] Where c is the ratio of the lattice spacing to the time step, c = 1;

[0097] In formula (13), f i eq is the equilibrium distribution function, which is expressed as:

[0098]

[0099] Where ρ is the fluid density, kg m -3 ;c s is the constant speed of sound, ms -1 ; u is the fluid velocity, ms -1 ;w i is the weight coefficient, defined as:

[0100]

[0101] The external force term F in formula (13) affects the pressure, momentum and momentum flux of the fluid, and after discretization is:

[0102]

[0103] f a is the force density acting on the fluid, and the fluid density ρ and fluid velocity u are defined as:

[0104]

[0105] S414. Based on formula (13), combined with large eddy simulation, a fluid mechanics model is established:

[0106]

[0107]

[0108] Large eddy simulation (LES) is used to simulate turbulence phenomena in the flow field of UAV. This embodiment uses the wall adaptive local eddy viscosity (WALE) model in the subgrid model for simulation, which provides consistent local eddy viscosity and near-wall characteristics for LES.

[0109] S42, based on step S3, obtain a solid mechanics model:

[0110]

[0111] Where M is the mass matrix of rice plant nodes, are the acceleration vectors of rice plant nodes, F int is the internal force of the rice plant, F ext is the fluid force acting on the rice plant, N t is the basis function of the rice plant finite element, T is the Cauchy stress, V (m) is the volume of the mth rice plant node, V (n) is the volume of the nth rice plant node, f a ' is the force density acting on the nodes of rice plants;

[0112] S43. The data exchange between the coupling interface of the downwash flow field of the UAV and the rice canopy is processed by the immersed boundary method. The fluid area is discretized by a set of Euler points, which are fixed regular Cartesian lattice points; the structural boundary of the rice canopy is discretized by a set of Lagrange points. The data of the coupling interface between the fluid grid and the structural grid is exchanged through interpolation operators and extension operators; the information interaction of the coupling between the fluid mechanics model and the solid mechanics model is realized by the immersed boundary method, and the following is obtained:

[0113]

[0114] in, and are Euler variables and Lagrangian variables respectively, the Cartesian coordinates of the fluid Euler point x = (x, y, z), the Cartesian coordinates of the Lagrangian point of the rice canopy structure X = (o, i, j), Ω represents the flow field; Γ represents the boundary of the rice canopy structure, γ is the proportional factor of the reciprocity between the interpolation operator Φ and the expansion operator θ, δ(xX) is the discrete Dirac interpolation function, which interpolates the external force on the Lagrangian coordinates of the rice canopy structure to the force on the flow field on the Euler coordinates of the flow field.

[0115] Among them, the rebound method is used to interpolate the particle velocity u of the Euler point, that is:

[0116] u(X,t)=Φ(u(x,t)), which is equation (24).

[0117] To satisfy the no-slip boundary condition, the fluid velocity at the coupling interface must be equal to the wall velocity U of the structural boundary. B :

[0118] U B (X, t) = u(X, t), which is formula (25).

[0119] Substitute equations (24) and (25) into equation (23) and transform to Lagrangian coordinates to solve for f a 'Force density;

[0120] F and F ext is the distribution function in the opposite direction, that is:

[0121] F=-F ext , which is formula (26),

[0122] Solving the above equation, we get the solution of F, which is substituted into equation (13) to update the flow field.

[0123] S5, setting initial boundary conditions and spatial discretization; Step S5 includes the following specific steps:

[0124] S51. In the fluid solver, specific boundary conditions are set according to the actual UAV plant protection operation conditions; e.g. Figure 6As shown, the bottom surface of the calculation domain of this embodiment is set as a wall boundary to enhance the wall function, the top surface is set as a zero pressure inlet boundary, and the other sides are set as zero pressure outlet boundaries; the surface of the rice plant is a wall boundary, the surface of the drone rotor is a non-slip and adiabatic inert wall boundary, and the remaining parts of the drone are set as non-slip wall boundaries. In order to truly simulate the dynamic behavior of the drone flying over rice, the drone is set to forced motion. The flight speed of drone 4 is 2m / s, the speed of the four rotors is 1667r / min, rotors one and three rotate clockwise, rotors two and four rotate counterclockwise, and the flight altitude of drone 4 is 3-5m.

[0125] S52. Establish a spatial wind field model. The bottom surface of this spatial wind field model is set as a wall boundary for enhancing the wall function, the top surface is set as a zero pressure inlet boundary, and the other sides are set as zero pressure outlet boundaries; the surface of the rice plant is the wall boundary, the surface of the UAV rotor is a no-slip and adiabatic inert wall boundary, and the remaining parts of the UAV are set as no-slip wall boundaries; the spatial wind field model establishes a 24×24m×6m virtual wind tunnel in the Xflow software, and imports the rice canopy model and the UAV model to form a spatial wind field model.

[0126] S53. Use a discrete method to discretize the spatial fluid in the spatial wind field model. The discrete method in step S53 uses the Lattice Boltzmann Method (LBM). Unlike traditional grid methods, LBM does not require grid generation, which reduces the complexity of model construction and grid division; it can also more easily handle complex geometric shapes and dynamic boundary conditions because it does not rely on traditional grid structures; LBM calculations can be naturally parallelized, which makes it more efficient when processing large-scale calculations; it can effectively handle complex fluid behaviors such as non-uniform flow, turbulence, and multiphase flow;

[0127] The spatial fluid domain discrete resolution is set to 0.05m, the resolutions near the four rotors and the rice canopy are all set to 0.005m, the resolutions near the fuselage and the nozzle are set to 0.02m, the wake adaptive resolution is 0.00125m, and the threshold is 0.000001m;

[0128] S6, performing a fully coupled finite element numerical simulation of a fluid-solid coupling model; step S6 includes the following specific steps:

[0129] In the fully coupled finite element numerical solution, the fluid-solid coupling between the fluid solver and the solid solver is managed using the joint simulation engine. The XML file in the fluid solver and the inp file generated by the solid solver are added to the system save file to create a joint simulation data exchange environment. The joint simulation engine is started by entering the code in the command window.

[0130] During the fluid-solid coupling process between the fluid solver and the solid solver, the displacement and coordinates are transmitted from the solid solver to the fluid solver. At the same time, the load is transmitted from the fluid solver to the solid solver. That is, the transmission between the fluid solver and the solid solver occurs in a bidirectional and simultaneous manner.

[0131] S7. Post-process the numerical simulation results based on the flow field post-processing software to obtain the evolution process and spatial distribution of the UAV downwash flow field and the dynamic behavior of the rice plants; analyze the impact of the dynamic behavior of the rice plants on the UAV downwash airflow, and clarify the distribution law of the downwash airflow velocity in the rice canopy.

[0132] In summary, by analyzing the distribution law of the downwash airflow velocity of the UAV in the rice canopy, the present invention can better understand the transmission and distribution characteristics of droplets in the crop canopy, optimize the UAV spraying operation parameters, improve crop yield and quality, while reducing the impact on the environment and achieving sustainable development.

[0133] The above specific implementation modes are preferred embodiments of the present invention and cannot be used to limit the present invention. Any other changes or other equivalent replacement methods that do not deviate from the technical solution of the present invention are included in the protection scope of the present invention.

Claims

1. A numerical analysis method for predicting the velocity distribution of UAV downwash airflow in the rice canopy based on the fluid-solid coupling method, characterized in that: The following steps are involved: S1. Measuring the physical parameters of rice plants based on the physiological characteristics of rice; S2. Establishing a three-dimensional rice plant model based on three-dimensional geometric modeling software; S3, dividing the three-dimensional rice plant canopy grid based on grid division software; S4. Establish a fluid-solid coupling model between the downwash flow field of the UAV and the rice canopy; S5, setting initial boundary conditions and spatial discretization; S6. Perform bidirectional coupling finite element numerical simulation of the fluid-solid coupling model; S7. Post-process the numerical simulation results based on the flow field post-processing software to obtain the evolution process and spatial distribution of the UAV downwash flow field and the dynamic behavior of rice plants; Analyze the impact of the dynamic behavior of rice plants on the downwash airflow of UAVs, and clarify the distribution pattern of downwash airflow velocity in the rice canopy.

2. The numerical analysis method for predicting the velocity distribution of the downwash airflow of a UAV in the rice canopy based on the fluid-solid coupling method according to claim 1 is characterized in that: The physical parameters in step S1 include measuring characteristic parameters of rice plants and mechanical parameters of rice plants; The characteristic parameters of rice plants include rice plant height, stem length, stem diameter, stem wall thickness, leaf length, leaf width, leaf thickness, leaf number and leaf area; the mechanical parameters of rice plants include stem elastic modulus, stem Poisson's ratio, leaf elastic modulus and leaf Poisson's ratio.

3. The numerical analysis method for predicting the velocity distribution of the downwash airflow of a UAV in a rice canopy based on the fluid-solid coupling method according to claim 2 is characterized in that: Step S3 includes the following steps: S31, importing the three-dimensional rice plant model into the finite element analysis software ABAQUS, and assigning the mechanical parameters of the rice plant to the material properties of the rice plant; S32, dividing the rice plants using a hexahedral grid in a structured grid; S33. A rice canopy model is obtained by using the circular and linear array methods. This rice canopy model is a solid mechanics model.

4. The numerical analysis method for predicting the velocity distribution of the downwash airflow of a UAV in a rice canopy based on the fluid-solid coupling method according to claim 3 is characterized in that: Step S4 includes the following specific steps: S41. Establish a fluid mechanics model based on the lattice Boltzmann method and large eddy simulation method: This is formula (1), This is formula (2), This is formula (3), Among them, μ t is the turbulent eddy viscosity, K w is a constant of 0.2, Δ is the unit length of the sub-grid model, H αβ and G αβ is the analytical scale strain rate tensor, δ αβ is the Kronecker symbol, g αβ , g βα and g γγ are the components of the strain rate tensor; S42, based on step S3, obtain a solid mechanics model: This is formula (4), This is formula (5), This is formula (6), Where M is the mass matrix of rice plant nodes, are the acceleration vectors of rice plant nodes, F int is the internal force of the rice plant, F ext is the fluid force acting on the rice plant, N t is the basis function of the rice plant finite element, T is the Cauchy stress, V (m) is the volume of the mth rice plant node, V (n) is the volume of the nth rice plant node, f a ' is the force density acting on the nodes of rice plants; S43. Using the immersed boundary method to realize the information interaction of the coupling between the fluid mechanics model and the solid mechanics model, we get: This is formula (7), This is formula (8), in, and are Euler variables and Lagrangian variables respectively, the Cartesian coordinates of the fluid Euler point x = (x, y, z), the Cartesian coordinates of the Lagrangian point of the rice canopy structure X = (o, i, j), Ω represents the flow field; Γ represents the boundary of the rice canopy structure, γ is the proportional factor of the reciprocity between the interpolation operator Φ and the expansion operator θ, δ(xX) is the discrete Dirac interpolation function, which interpolates the external force on the Lagrangian coordinates of the rice canopy structure to the force on the flow field on the Euler coordinates of the flow field.

5. The numerical analysis method for predicting the velocity distribution of the downwash airflow of a UAV in a rice canopy based on the fluid-solid coupling method according to claim 4 is characterized in that: Step S41 includes the following specific steps: S411. Based on the Boltzmann equation, taking the velocity distribution function f as the basic variable, we get the following equations: This is formula (9), f i (x+e i ·Δt,t+Δt)-f i (x,t)=Ω i , which is formula (10), i=1,...,Q,(Q=2,3,...), this is formula (11), Where t is the time, Δt is the time step, x is the lattice position, and e i is the discrete velocity, Ω i is the collision operator, i represents the finite velocity component of the distribution function; the velocity distribution function f is composed of the spatial position x, the particle velocity vector e i and time t constitutes a function f(x,e i, ,t), represents the distribution function f of the ith discrete velocity vector of a discrete node i The evolution law of S412. Select the single-relaxed BGK model as the collision term, then: This is formula (12), Where τ is the relaxation time, F is the external force applied to the fluid, is the local equilibrium distribution function; S413. Use the BGK model to discretize equations (9) to (10), and we get: This is formula (13), S414. Based on formula (13), combined with large eddy simulation, a fluid mechanics model is established: This is formula (1), This is formula (2), This is formula (3).

6. The numerical analysis method for predicting the velocity distribution of the downwash airflow of a UAV in a rice canopy based on the fluid-solid coupling method according to claim 1 is characterized in that: Step S5 includes the following specific steps: S51. In the fluid solver, specific boundary conditions are set according to the actual UAV plant protection operation conditions; S52, establishing a spatial wind field model, wherein the bottom surface of the spatial wind field model is set as a wall boundary that enhances the wall function, the top surface is set as a zero pressure inlet boundary, and the other sides are set as zero pressure outlet boundaries; the surface of the rice plant is the wall boundary, the surface of the drone rotor is the non-slip and adiabatic inert wall boundary, and the remaining parts of the drone are set as non-slip wall boundaries; S53. Use a discrete method to discretize the spatial fluid in the spatial wind field model.

7. The numerical analysis method for predicting the velocity distribution of the downwash airflow of a UAV in a rice canopy based on the fluid-solid coupling method according to claim 6 is characterized in that: The spatial wind field model in step S52 establishes a 24×24m×6m virtual wind tunnel in the Xflow software, and imports the rice canopy model and the drone model to form a spatial wind field model.

8. The numerical analysis method for predicting the velocity distribution of the downwash airflow of a UAV in a rice canopy based on the fluid-solid coupling method according to claim 6 is characterized in that: The discretization method in step S53 adopts the Lattice Boltzmann Method.

9. The numerical analysis method for predicting the velocity distribution of downwash airflow from a UAV in a rice canopy based on a fluid-solid coupling method according to claim 1, characterized in that: Step S6 includes the following specific steps: In the fully coupled finite element numerical solution, the fluid-solid coupling between the fluid solver and the solid solver is managed using the joint simulation engine. The XML file in the fluid solver and the inp file generated by the solid solver are added to the system save file to create a joint simulation data exchange environment. The joint simulation engine is started by entering the code in the command window. Fluid Solver and Solid Solver During the fluid-structure interaction process, displacements and coordinates are transferred from the solid solver to the fluid solver. At the same time, loads are transferred from the fluid solver to the solid solver.

Images