Method for predicting droplet deposition drift of strip-shaped composite canopy flying prevention pesticide application

Abstract

The invention discloses a strip-shaped composite canopy flight control pesticide application fog drop deposition drift prediction method, and relates to the technical field of plant protection spraying simulation, and the method comprises the steps: obtaining crop canopy and unmanned aerial vehicle parameters of a field experiment as first simulation parameter information; respectively constructing a rotor sliding grid model, a dynamic parameter porous medium canopy model and a canopy fog drop probability capture model by utilizing the first simulation parameter information; constructing a computational domain three-dimensional model in combination with the rotor sliding grid model and the dynamic parameter porous medium canopy model, and performing grid division; setting second simulation parameter information of strip-shaped composite canopy crop simulation spraying; and in combination with second simulation parameter information, the computational domain three-dimensional model after grid division and the canopy fog drop probability capture model, performing simulation statistics and analyzing banded composite irregularity, downwash flow field distribution under heterogeneous canopies and a fog drop deposition rule. The influence of downwash airflow and a special canopy structure on migration and deposition of fog drops is analyzed, and a reference is provided for strategy optimization of strip-shaped composite canopy unmanned aerial vehicle pesticide application.

Description

Technical Field

[0001] The present application relates to the technical field of plant protection spray simulation, and in particular to a method for predicting the deposition and drift of droplets during aerial spraying of strip-shaped composite canopies. Background Art

[0002] The soybean-corn strip intercropping system effectively increases soybean yields while ensuring stable corn production, making it a crucial measure for ensuring national grain and oil security. Large-scale implementation of this intercropping system urgently requires support from precision pesticide application strategies. Traditional field droplet deposition experiments typically use indirect measurement methods such as water-sensitive paper and polyester film sheets to measure droplet deposition distribution within the crop canopy. However, field experiments often face numerous limitations, including high costs and complex sources of error.

[0003] In recent years, CFD (Computational Fluid Dynamics) numerical simulation has become increasingly widely used in agricultural aerial spray research. This method uses Lagrangian particle tracking to simulate droplet migration trajectories. Chinese patent application number CN202211506013.7 discloses a CFD-based method for simulating inter-row wind-delivered spraying. This method uses a porous media model to represent the crop canopy structure and successfully simulates the deposition of droplets on the porous media surface under the influence of airflow. However, this method fails to reflect the dynamic changes in canopy resistance properties under the influence of auxiliary airflow. Furthermore, this method only deposits droplets on the porous media canopy surface, failing to simulate the deposition process within the canopy.

[0004] At present, there is an urgent need for a numerical simulation method of UAV droplet drift and deposition applicable to the irregular and heterogeneous strip-shaped composite canopy of soybean and corn. Summary of the Invention

[0005] In order to solve the above technical problems, this application proposes the following technical solutions: In a first aspect, the present application provides a method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy, comprising: Obtain crop canopy and drone parameters from the field experiment as first simulation parameter information; Using the first simulation parameter information, a rotor sliding grid model, a dynamic parameter porous media canopy model, and a canopy droplet probability capture model are constructed respectively; Combining the rotor sliding grid model and the dynamic parameter porous media canopy model to construct a three-dimensional model of the computational domain and perform grid division; Setting the second simulation parameter information of the strip composite canopy crop simulation spray; Combining the parameter information, the three-dimensional model of the computational domain after grid division and the canopy droplet probability capture model, simulation statistics are performed to analyze the downwash flow field distribution and droplet deposition law under the strip-shaped composite irregular and heterogeneous canopy.

[0006] In a possible implementation, obtaining crop canopy and drone parameters from a field experiment as first simulation parameter information includes: Through the aerial spraying experiment, the wind field data under the rotor under standard load was obtained, as well as the wind speed data of the airflow into and out of the two crop canopies at different flight altitude and flight speed combinations; The three-dimensional size of the equivalent porous medium canopy model of the crop is determined by field measurements of the crop strip composite planting pattern, crop plant height, and plant leaf area parameters, and the canopy leaf area density is calculated: in: LAD is the canopy leaf area density, S leaf represents the total leaf area of the plant, S Indicates the area occupied by the plant, h Indicates plant height; Determine the spray atomization parameters based on the actual drone nozzle model and operating flow rate.

[0007] In a possible implementation, the use of the first simulation parameter information to respectively construct a rotor sliding grid model, a dynamic parameter porous media canopy model, and a canopy droplet probability capture model includes: The ensemble inverse modeling method determines the rotor standard speed by comparing the simulated flow field data with the measured data at different speeds, and constructs the rotor sliding grid model; Determine the basic equation of the airflow velocity at the canopy outlet and determine the undetermined parameter function relationship of the basic equation of the airflow velocity at the canopy outlet, and obtain the dynamic parameter porous medium canopy model in combination with the porous medium model preset in the simulation software; The droplet deposition behavior in the canopy space is divided into two processes: droplet blade contact and adhesion. The deposition probability formula of the droplet canopy in each grid unit is obtained, and a canopy droplet probability capture model coupled with dynamic parameters is constructed.

[0008] In one possible implementation, the collective inverse modeling method determines the rotor standard speed by comparing simulated flow field data with measured data at different speeds, and constructs the rotor sliding grid model, including: Use an optical scanner to collect rotor point cloud data, and import the obtained point cloud data into reverse modeling software to reconstruct the rotor three-dimensional model; The rotor three-dimensional model is used to carry out a rotor downwash flow field simulation experiment through sliding grid technology. The rotor speed is determined by comparing the measured flow field data under the same operating conditions with the simulated flow field data under different speeds, and the rotor sliding grid model is obtained.

[0009] In one possible implementation, determining the basic equation for the airflow velocity at the canopy outlet and determining the undetermined parameter function relationship of the basic equation for the airflow velocity at the canopy outlet, and obtaining the dynamic parameter porous medium canopy model in combination with a porous medium model preset in the simulation software, includes: Taking the canopy inlet airflow velocity, leaf area density, inertial drag coefficient and porosity as input features, the calculation equation for the canopy outlet airflow velocity is derived: in: V is the predicted value of the airflow velocity at the canopy outlet; is the airflow velocity at the canopy entrance; is the canopy inertial drag coefficient; is the canopy porosity; k is the number of iterations; The residuals of wind speed data of two crop canopies were collected through computational simulation and field experiments. The functional relationship between the inertial drag coefficient and porosity change was constructed using the least squares method based on the trust region, and the undetermined parameters in the canopy outlet airflow velocity equation were identified. The objective function of the least squares method is as follows: The iterative calculation of the trust region algorithm is shown as follows: in: is the parameter vector to be identified, is the optimal identification parameter vector obtained using the least squares method; is the simulated measurement value of the airflow velocity at the canopy outlet; is the parameter vector at the kth iteration when the least squares method is iteratively calculated; The equation for the airflow velocity at the canopy outlet is Jacobian matrix; m is the number of experimental data groups; is the mth error function in the parameter vector The gradient vector at ; and Respectively The first and second parameters in the vector; The functional relationship between the above-mentioned inertial resistance coefficient and the canopy porosity change is loaded into the porous medium model preset in the simulation software to obtain the dynamic parameter porous medium canopy model.

[0010] In one possible implementation, the droplet deposition behavior in the canopy interior is divided into two processes: droplet contact and adhesion. The deposition probability formula of the droplet canopy in each grid unit is obtained, and a dynamic parameter-coupled canopy droplet probability capture model is constructed, including: According to the actual working conditions in the field, the fixed parameters of blade unit diameter, leaf area density, droplet density and air dynamic viscosity are determined; Obtain the droplet size, velocity inertia, and airflow velocity at the center of each grid cell, calculate the contact probability and adhesion probability between each grid cell in the canopy and the droplet in real time, and finally calculate the capture probability: in: is the capture probability; is the contact probability between the droplet and the leaf, is the droplet velocity; C 1 is the model empirical coefficient; is the time variation; is the probability of droplets adhering to leaves; is the droplet density; d is the droplet diameter; u is the air velocity; is the dynamic viscosity of air; D leaf Indicates the width of vegetation leaves; A canopy droplet deposition model with differential distribution of internal spatial deposition probability was constructed based on the deposition probability calculation formula.

[0011] In a possible implementation, the constructing of a computational domain three-dimensional model by combining the rotor sliding grid model and the dynamic parameter porous media canopy model, and performing grid division, includes: Ignoring the fuselage of the rotor sliding mesh model, a multi-rotor equivalent representation of the plant protection UAV is adopted; Based on crop phenotypic parameters, canopy wind resistance properties and droplet capture probability, a zoned differentiated porous medium equivalent method was used to characterize the crop canopy. A multi-level grid division strategy is adopted to appropriately encrypt the grids of the rotor, sprinkler, crops and their surrounding areas.

[0012] In a possible implementation, the second simulation parameter information for setting the strip-shaped composite canopy crop simulation spray includes: The DPM under the Lagrangian framework is used to simulate the droplet motion trajectory and the spray atomization parameters are set according to the actual working conditions; The boundary pressure around the crops and near the drone is set to allow the droplets to escape freely, and the ground boundary is set as a no-slip wall that can capture the droplets.

[0013] In one possible implementation, combining the second simulation parameter information, the gridded three-dimensional computational domain model, and the canopy droplet probability capture model to perform simulation statistics and analyze the downwash flow field distribution and droplet deposition pattern under the zonal composite irregular and heterogeneous canopy, includes: Importing the second simulation parameter information, the gridded three-dimensional computational domain model, and the canopy droplet probability capture model into the post-processing software CFD-post and introducing droplet deposition amount and deposition density variables; The CFD-post simulation is used to statistically analyze the distribution characteristics of the irregular heterogeneous canopy downwash flow field and the droplet drift and deposition characteristics.

[0014] One possible implementation method also includes: through simulation experiments and analysis under a variety of typical operating conditions, the downwash flow field distribution and droplet deposition patterns under the crop strip-shaped composite irregular and heterogeneous canopy are revealed, and the drone flight control strategy is optimized.

[0015] In this application example, the dynamic canopy drag coefficient and porosity changes under the interaction of downwash airflow and the crop canopy are simulated, and the droplet and airflow kinetic energy parameters of each grid cell are obtained to simulate the droplet deposition process within the canopy space with dynamic parameter coupling. The influence of downwash airflow and special canopy structure on droplet migration and deposition is analyzed, revealing the spatial distribution patterns of downwash flow and droplet deposition, providing a reference for optimizing drone spraying strategies in zonal composite canopies. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic flow chart of a method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy provided in an embodiment of the present application; Figure 2 A schematic diagram of the process of constructing a rotor sliding grid model provided in an embodiment of the present application; Figure 3 A schematic diagram of constructing a 3D rotor model provided in an embodiment of the present application; Figure 4 A schematic diagram of the three-dimensional model of the computational domain and its grid division provided in an embodiment of the present application; Figure 5 The downwash airflow distribution diagram provided in the embodiment of the present application; Figure 6 Schematic diagram of airflow distribution at different levels of soybean and corn canopies provided in the embodiments of this application; Figure 7 A side view of the droplet deposition distribution provided in an embodiment of the present application; Figure 8 Schematic diagram of droplet deposition distribution at different levels of soybean and corn canopy provided in the examples of this application. DETAILED DESCRIPTION

[0017] The present invention will be described below with reference to the accompanying drawings and specific implementation methods.

[0018] See also Figure 1 The present embodiment provides a method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy, including: S101: Acquire crop canopy and drone parameters of a field experiment as first simulation parameter information.

[0019] In this example, corn and soybean crops were used at maturity. The experimental drone was a DJ-T30 six-rotor drone equipped with 16 fan-shaped hydraulic atomizing nozzles (model SX11001), with a maximum spray flow rate of 7.2 L / min.

[0020] In order to obtain the first simulation parameter information, an aerial defense experiment was conducted. A hot wire anemometer (405i, TESTO) was used to collect wind field data below the rotor under a standard load, as well as wind speed data of airflow into and out of the two crop canopies at different flight altitude and speed combinations.

[0021] Through actual field measurements, it was found that the planting patterns of soybean and corn crops were 30 cm × 10 cm and 40 cm × 10 cm (row spacing × plant spacing), and the plant heights were 1.0 m and 2.9 m, respectively. The leaf area density of soybean and corn canopies was calculated by the following formula: 3.80 m -1 and 6.03 m -1 .

[0022] in, LAD is the canopy leaf area density; S leaf represents the total leaf area of the plant; S Indicates the area occupied by the plant; h Indicates plant height; Based on the actual operating conditions of the drone and the nozzle model, the nozzle flow rate was determined to be 0.34 L / min. The spray droplet size range was 50-250 μm, and the dispersion coefficient under the Rosin-Rammler distribution was 2.23.

[0023] S102: Use the first simulation parameter information to construct a rotor sliding grid model, a dynamic parameter porous medium canopy model, and a canopy droplet probability capture model.

[0024] The rotor sliding mesh model is constructed by using the inverse modeling method to build a three-dimensional rotor model, and the simulated flow field data at different speeds are compared with the measured data to determine the standard rotor speed. The rotor sliding mesh model is obtained by using the sliding mesh technology. The rotor sliding mesh model construction process is as follows: Figure 2 shown.

[0025] The OKIO 5M 3D scanner was used to obtain the point cloud data of the UAV rotor. After noise reduction, packaging, and repair in Geomagic Studio 12, the rotor 3D model was obtained. Figure 3 shown.

[0026] Field experiments measured the vertical component of wind speed at a horizontal section 65 cm below the UAV rotor at flight speeds of 3 m / s and 5 m / s. Simultaneously, simulation experiments were conducted based on the aforementioned 3D rotor model, obtaining eight sets of simulated wind speed data at rotor speeds of 160 rad / s, 180 rad / s, 200 rad / s, and 240 rad / s, under the same flight parameters as the measured values. Finally, by comparing the actual wind speed data at the same flight speed during actual operations, the rotor speed of 200 rad / s was determined to best match the actual wind field distribution. Rotation and translation of the rotor model were achieved using sliding mesh technology, ultimately determining the rotor sliding mesh model.

[0027] To construct a dynamic parameter porous media canopy model, first determine the basic equation for the canopy outlet airflow velocity. Then, use the trust region-based least squares method to identify the unknown parameters in the canopy outlet airflow velocity equation. Finally, load the functional relationship into the pre-set porous media model in the simulation software to obtain the dynamic parameter porous media canopy model.

[0028] Specifically, the canopy inlet airflow velocity, leaf area density, inertial drag coefficient, and porosity are used as input features to derive the canopy outlet airflow velocity calculation equation: in: V is the predicted value of the airflow velocity at the canopy outlet; is the airflow velocity at the canopy entrance; is the canopy inertial drag coefficient; is the canopy porosity; k is the number of iterations.

[0029] The residuals of wind speed data of two crop canopies were collected through computational simulation and field experiments. The functional relationship between the inertial drag coefficient and porosity change was constructed using the least squares method based on the trust region, and the undetermined parameters in the canopy outlet airflow velocity equation were identified. The objective function of the least squares method is as follows: The iterative calculation of the trust region algorithm is shown as follows: in: is the parameter vector to be identified, is the optimal identification parameter vector obtained using the least squares method; is the simulated measurement value of the airflow velocity at the canopy outlet; is the parameter vector at the kth iteration when the least squares method is iteratively calculated; The equation for the airflow velocity at the canopy outlet is Jacobian matrix; m is the number of experimental data groups; is the mth error function in the parameter vector The gradient vector at ; and Respectively The first and second parameters in the vector; The functional relationship between the above-mentioned inertial resistance coefficient and the canopy porosity change is loaded into the porous medium model preset in the simulation software to obtain the dynamic parameter porous medium canopy model.

[0030] A canopy droplet deposition model is constructed, which divides the droplet deposition behavior in the canopy space into two processes: contact and adhesion of droplets to leaves. By obtaining the deposition probability formula in the droplet canopy of each grid unit, a dynamic parameter-coupled canopy droplet probability capture model is constructed.

[0031] Specifically, in this embodiment, fixed parameters of blade unit diameter, leaf area density, droplet density and air dynamic viscosity are determined based on actual field conditions.

[0032] Obtain the droplet size, velocity inertia, and airflow velocity at the center of each grid cell, calculate the contact probability and adhesion probability between each grid cell in the canopy and the droplet in real time, and finally calculate the capture probability: in: is the capture probability; is the contact probability between the droplet and the leaf, is the droplet velocity; C 1 is the model empirical coefficient; is the time variation; is the probability of droplets adhering to leaves; is the droplet density; d is the droplet diameter; u is the air velocity; is the dynamic viscosity of air; D leaf Indicates the width of vegetation leaves.

[0033] A canopy droplet deposition model with differential distribution of internal spatial deposition probabilities is constructed based on a deposition probability calculation formula. In this embodiment, a canopy droplet capture algorithm is developed in C or C++ based on this deposition probability calculation formula. This canopy droplet capture algorithm is used to simulate, predict, or optimize the computational model of droplet deposition behavior within the crop canopy during pesticide spraying. The droplet capture algorithm is introduced into the discrete phase and porous media modules via a user-defined function module within the simulation software, enhancing the standard functionality of the simulation software and thereby constructing a canopy droplet probability capture model with differential distribution of internal spatial deposition probabilities.

[0034] A canopy droplet deposition model with differential distribution of internal spatial deposition probability was constructed based on the deposition probability calculation formula.

[0035] S103 , constructing a three-dimensional model of the computational domain by combining the rotor sliding grid model and the dynamic parameter porous medium canopy model, and performing grid division.

[0036] Based on the rotor sliding mesh model constructed by S102, the fuselage is ignored and a 6-rotor equivalent representation of the plant protection UAV is adopted, such as Figure 4 (a). Based on crop phenotypic parameters, canopy wind resistance properties, and droplet capture probability, a partitioned, differentiated porous medium was used to equivalently characterize soybean and corn canopies. The 3D model was imported into Mesh, and a multi-level meshing strategy was used to appropriately encrypt the meshes of the rotor, nozzle, crop, and surrounding areas, such as Figure 4 (b)

[0037] S104: Setting second simulation parameter information for strip-shaped composite canopy crop simulation spraying.

[0038] The DPM setup uses DPM within the Lagrangian framework to simulate droplet motion trajectories and set spray atomization parameters based on actual operating conditions. The DPM particle injection source is positioned below the rotor model based on the actual nozzle distribution pattern, and sliding mesh technology is used to control the injection source's synchronous horizontal displacement with the rotor area.

[0039] The computational domain boundary conditions were set to define the boundaries around the crops and near the drone as pressure outlets with a gauge pressure of zero, allowing droplets to escape freely. The ground boundary was set to a no-slip wall that could capture droplets.

[0040] S105, combining the second simulation parameter information, the three-dimensional model of the computational domain after grid division, and the canopy droplet probability capture model to perform simulation statistics and analyze the downwash flow field distribution and droplet deposition law under the strip-shaped composite irregular and heterogeneous canopy.

[0041] The droplet deposition visualization is performed by importing the second simulation parameter information, the meshed three-dimensional computational domain model, and the canopy droplet probability capture model into the post-processing software CFD-post, and introducing the droplet deposition amount and deposition density variables. The post-processing software CFD-post is then used to statistically analyze the distribution characteristics of the irregular heterogeneous canopy downwash flow field and the droplet drift and deposition characteristics.

[0042] The simulated UAV downwash flow pattern distribution and canopy airflow velocity distribution are as follows: Figure 5 and Figure 6 As shown in Figure 1, the downwash airflow from the drone has a strong penetration effect on the soybean canopy and the upper part of the corn canopy (2.0-2.9 m). However, the penetration ability of the middle and lower parts of the corn canopy (0-2.0 m) is significantly reduced, which will make it difficult for a large number of droplets to reach the bottom of the canopy, which is extremely detrimental to the uniformity of droplet deposition. The simulated drone spray droplet distribution and canopy deposition are shown in Figure 1. Figure 7 and Figure 8 As shown, droplet deposition in the corn canopy decreased significantly with decreasing sampling altitude, with droplet deposition in both the soybean and lower corn canopies (0-1.0 m) remaining below 2.0 μL / cm². In summary, using this combination of flight parameters for drone spraying resulted in poor droplet deposition distribution in both crop canopies, making it difficult to achieve the desired pest and disease control results.

[0043] To optimize drone spraying strategies, this example also conducted simulation experiments using various typical operational parameter combinations, comparing and analyzing the effects of varying flight altitudes and speeds on airflow and droplet penetration across the canopies of two crops. By revealing the downwash flow distribution and droplet deposition patterns under the irregular, heterogeneous, and strip-like canopies of soybeans and corn, this study provides a reference for the precise optimization of operational parameters such as flight altitude and speed during drone spraying operations.

[0044] In the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0045] The above description is merely a specific embodiment of the present application. Any person skilled in the art may easily conceive of variations or substitutions within the technical scope disclosed in this application, and such variations or substitutions shall be within the scope of protection of this application. The scope of protection of this application shall be subject to the scope of protection of the claims.

Claims

1. A method for predicting droplet deposition and drift during aerial spraying of composite canopies, characterized in that: include: Obtain crop canopy and drone parameters from the field experiment as first simulation parameter information; Using the first simulation parameter information, a rotor sliding grid model, a dynamic parameter porous media canopy model, and a canopy droplet probability capture model are constructed respectively; Combining the rotor sliding grid model and the dynamic parameter porous media canopy model to construct a three-dimensional model of the computational domain and perform grid division; Setting the second simulation parameter information of the strip composite canopy crop simulation spray; Combined with the second simulation parameter information, the three-dimensional model of the computational domain after grid division and the canopy droplet probability capture model, simulation statistics are performed to analyze the downwash flow field distribution and droplet deposition law under the strip-shaped composite irregular and heterogeneous canopy.

2. The method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy according to claim 1, characterized in that: The step of obtaining crop canopy and drone parameters of the field experiment as first simulation parameter information includes: Through the aerial spraying experiment, the wind field data under the rotor under standard load was obtained, as well as the wind speed data of the airflow into and out of the two crop canopies at different flight altitude and flight speed combinations; The three-dimensional size of the equivalent porous medium canopy model of the crop is determined by field measurements of the crop strip composite planting pattern, crop plant height, and plant leaf area parameters, and the canopy leaf area density is calculated: in: LAD is the canopy leaf area density, S leaf represents the total leaf area of the plant, S Indicates the area occupied by the plant, h Indicates plant height; Determine the spray atomization parameters based on the actual drone nozzle model and operating flow rate.

3. The method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy according to claim 1, characterized in that: The method of using the first simulation parameter information to respectively construct a rotor sliding grid model, a dynamic parameter porous medium canopy model, and a canopy droplet probability capture model includes: The ensemble inverse modeling method determines the rotor standard speed by comparing the simulated flow field data with the measured data at different speeds, and constructs the rotor sliding grid model; Determine the basic equation of the airflow velocity at the canopy outlet and determine the undetermined parameter function relationship of the basic equation of the airflow velocity at the canopy outlet, and obtain the dynamic parameter porous medium canopy model in combination with the porous medium model preset in the simulation software; The droplet deposition behavior in the canopy space is divided into two processes: droplet blade contact and adhesion. The deposition probability formula of the droplet canopy in each grid unit is obtained, and a canopy droplet probability capture model coupled with dynamic parameters is constructed.

4. The method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy according to claim 3 is characterized in that: The collective inverse modeling method determines the standard rotor speed by comparing the simulated flow field data with the measured data at different speeds, and constructs the rotor sliding grid model, including: Use an optical scanner to collect rotor point cloud data, and import the obtained point cloud data into reverse modeling software to reconstruct the rotor three-dimensional model; The rotor three-dimensional model is used to carry out a rotor downwash flow field simulation experiment through sliding grid technology. The rotor speed is determined by comparing the measured flow field data under the same operating conditions with the simulated flow field data under different speeds, and the rotor sliding grid model is obtained.

5. The method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy according to claim 3 is characterized in that: The method of determining the basic equation of airflow velocity at the canopy outlet and determining the undetermined parameter function relationship of the basic equation of airflow velocity at the canopy outlet, and obtaining the dynamic parameter porous medium canopy model in combination with the porous medium model preset in the simulation software, includes: Taking the canopy inlet airflow velocity, leaf area density, inertial drag coefficient and porosity as input features, the calculation equation for the canopy outlet airflow velocity is derived: in: V is the predicted value of the airflow velocity at the canopy outlet; is the airflow velocity at the canopy entrance; is the canopy inertial drag coefficient; is the canopy porosity; k is the number of iterations; The residuals of wind speed data of two crop canopies were collected through computational simulation and field experiments. The functional relationship between the inertial drag coefficient and porosity change was constructed using the least squares method based on the trust region, and the undetermined parameters in the canopy outlet airflow velocity equation were identified. The objective function of the least squares method is as follows: The iterative calculation of the trust region algorithm is shown as follows: in: is the parameter vector to be identified, is the optimal identification parameter vector obtained using the least squares method; is the simulated measurement value of the airflow velocity at the canopy outlet; is the parameter vector at the kth iteration when the least squares method is iteratively calculated; The equation for the airflow velocity at the canopy outlet is Jacobian matrix; m is the number of experimental data groups; is the mth error function in the parameter vector The gradient vector at ; and Respectively The first and second parameters in the vector; The functional relationship between the above-mentioned inertial resistance coefficient and the canopy porosity change is loaded into the porous medium model preset in the simulation software to obtain the dynamic parameter porous medium canopy model.

6. The method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy according to claim 3, characterized in that: The droplet deposition behavior in the canopy interior is divided into two processes: contact and adhesion of droplets to leaves. The deposition probability formula of the droplet canopy in each grid unit is obtained to construct a canopy droplet probability capture model coupled with dynamic parameters, including: According to the actual working conditions in the field, the fixed parameters of blade unit diameter, leaf area density, droplet density and air dynamic viscosity are determined; Obtain the droplet size, velocity inertia, and airflow velocity at the center of each grid cell, calculate the contact probability and adhesion probability between each grid cell in the canopy and the droplet in real time, and finally calculate the capture probability: in: is the capture probability; is the contact probability between the droplet and the leaf, is the droplet velocity; C 1 is the model empirical coefficient; is the time variation; is the probability of droplets adhering to leaves; is the droplet density; d is the droplet diameter; u is the air velocity; is the dynamic viscosity of air; D leaf Indicates the width of vegetation leaves; A canopy droplet probability capture model with differential distribution of internal spatial deposition probability is constructed based on the deposition probability calculation formula.

7. The method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy according to claim 1, characterized in that: The step of constructing a three-dimensional computational domain model by combining the rotor sliding grid model and the dynamic parameter porous media canopy model and performing grid division includes: Ignoring the fuselage of the rotor sliding mesh model, a multi-rotor equivalent representation of the plant protection UAV is adopted; Based on crop phenotypic parameters, canopy wind resistance properties and droplet capture probability, a zoned differentiated porous medium equivalent method was used to characterize the crop canopy. A multi-level grid division strategy is adopted to appropriately encrypt the grids of the rotor, sprinkler, crops and their surrounding areas.

8. The method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy according to claim 1, characterized in that: The second simulation parameter information for setting the strip-shaped composite canopy crop simulation spray includes: The DPM under the Lagrangian framework is used to simulate the droplet motion trajectory and the spray atomization parameters are set according to the actual working conditions; The boundary pressure around the crops and near the drone is set to allow the droplets to escape freely, and the ground boundary is set as a no-slip wall that can capture the droplets.

9. The method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy according to claim 1, characterized in that: The method of combining the second simulation parameter information, the three-dimensional model of the computational domain after grid division, and the canopy droplet probability capture model to perform simulation statistics and analyze the downwash flow field distribution and droplet deposition law under the strip-shaped composite irregular and heterogeneous canopy, includes: Importing the second simulation parameter information, the three-dimensional model of the computational domain after grid division, and the canopy droplet probability capture model into the post-processing software CFD-post and introducing the droplet deposition amount and deposition density variables; The CFD-post simulation is used to statistically analyze the distribution characteristics of the irregular heterogeneous canopy downwash flow field and the droplet drift and deposition characteristics.

10. The method for predicting droplet deposition and drift during aerial pesticide application in a strip-shaped composite canopy according to claim 9, characterized in that: Also includes: Through simulation experiments and analysis under various typical operating conditions, the downwash flow field distribution and droplet deposition rules under the crop strip-shaped composite irregular and heterogeneous canopy are revealed, and the drone flight control strategy is optimized.

Images