Discrete element-based root growth force measurement method

Abstract

The invention discloses a discrete element-based root growth force measurement method, which specifically comprises the following steps of: scanning a root contour in soil to obtain three-dimensional image data of root tip tissues; three-dimensional image data obtained through scanning are imported into computer software, an original image is reconstructed, a triangular network model of the root tip tissue is constructed, and surface smoothing is conducted on the surface of the triangular network model; establishing a multi-layer particle wall model by constructing particle walls with different attributes in discrete element software; setting a motion mode of the root tip tissue, solving motion behaviors of the root tip and the particle wall model based on a discrete element method, and analyzing stress and motion conditions of the root tip tissue and the particle wall; real multilayer soil wall wrapping seeds are manufactured, an analogue simulation result is verified, and the root system groundbreaking growing power is calculated according to simulation data. The root system growing power measuring method can simulate and accurately measure the root system groundbreaking growing power of the seeds in a real soil environment.

Description

Technical Field

[0001] The present invention belongs to the field of discrete element simulation of crops, and in particular relates to a method for measuring root growth force based on discrete elements. Background Art

[0002] The core advantage of the discrete element method (DEM), a numerical simulation technique based on the discrete properties of particle systems, lies in its ability to reveal the microscopic mechanical behavior of materials by accurately calculating the contact forces between particles. This method treats each particle as an independent mechanical unit, tracking its motion using Newton's laws of motion and quantifying the interactions between particles using contact mechanics models.

[0003] Seed growth is the most important stage for plants. During the germination stage, the radicle will grow and the root system will be stationed in the soil to absorb nutrients. The deeper the root system is stationed, the wider the range of nutrient absorption. However, since the growth of the root system is affected by the soil resistance, it is of great significance to measure the growth force of the roots breaking through the soil. This is conducive to exploring soil conditions suitable for seed growth and increasing the survival rate of plants. In existing research, the commonly used methods for measuring the growth force of seeds in soil mainly include making a force measuring device for measurement and changing the annular resistance. Among them,

[0004] The method of measuring by making a force measuring device is to fix the seeds by making a device, and then measure the growth force of the seed radicle during germination through direct or indirect contact with a probe with a micro-force sensor or a similar force-sensitive instrument. This method can accurately read the specific growth force, but in the test, the growth environment of the seeds mostly does not contact the soil, which is different from the actual environment, making the final data difficult to be convincing.

[0005] The method of changing the annular resistance is to create different soil environments, such as depth, moderation, viscosity, etc., to form different soil resistances for seeds, and observe the growth of seeds under different resistances through the control variable method, so as to infer the growth ability of seeds. Although this method can measure the growth ability of seeds in a real soil environment, the limitation of this method is mainly reflected in the fact that the measured soil resistance cannot truly represent the growth ability of seed roots when breaking through the soil, and the measurement result is an interval range, not a specific value.

[0006] Therefore, the main problem of the existing method is that it is difficult to take into account both the real soil environment and the accurate growth force value at the same time; the force measuring device measurement method cannot be carried out in the real soil environment, and the annular resistance method cannot directly derive the accurate root growth force. Summary of the Invention

[0007] In order to overcome the deficiencies in the prior art, the present invention provides a method for measuring root growth force based on discrete elements, which can simulate the accurate measurement of the root growth force of seeds in a real soil environment.

[0008] The technical solution of the present invention to solve the above technical problems is:

[0009] A method for measuring root growth force based on discrete elements comprises the following steps:

[0010] Step 1: Use a computed tomography scanner to scan the root system outline in the soil using 3D X-rays to obtain 3D image data of the root tip tissue;

[0011] Step 2: Import the scanned 3D image data into computer software, reconstruct the original image through the computer software, construct a triangular network model of the root apex tissue, and smooth the surface of the triangular network model;

[0012] Step 3: Establish a multi-layer granular wall model by constructing granular walls with different properties in the discrete element software;

[0013] Step 4: Set the motion mode of the root apex tissue, solve the motion behavior of the root apex and granular wall model based on the discrete element method, and analyze the force and motion of the root apex tissue and granular wall;

[0014] Step 5: Create a real multi-layer soil wall to wrap the seeds, verify the simulation results, and calculate the root growth ability by comparing the simulation data.

[0015] Preferably, the specific steps of step S1 are:

[0016] Step 101: Based on the actual planting situation, place soil in a container and plant seeds at a depth of 3-5 cm. Place the container in a constant temperature and humidity environment for 3-5 days. After the seed roots grow, transfer the container and secure it to the sample stage of the scanner.

[0017] Step 102: Select a scanning mode, set the scanning parameters of the scanner, and set the slice interval according to the required observation accuracy. Use the scanner to scan the root shape to ensure that the root apex tissue is covered, including the top, bottom, and sides, and finally obtain all slice images of the root apex tissue.

[0018] Preferably, in step S2, all the tomographic images of the root apex tissue are preprocessed to remove interference items and irrelevant items in the soil to obtain three-dimensional image data of the root apex tissue. The specific steps are:

[0019] Step 201: Based on the density difference between the root system and the soil, a grayscale value interval is determined using a statistical method. The grayscale value interval tool and a threshold segmentation algorithm are used to binarize the tomographic image to separate the root apex tissue from the soil particles. Based on the segmented binary image, a three-dimensional volume data model containing the complete morphological structure of the root apex tissue is constructed.

[0020] Step 202: Based on the constructed three-dimensional volume data model, three-dimensional geometric analysis and data quantification technology are used to extract various physical parameters of the root apex tissue, wherein the physical parameters include spatial scale parameters, spatial positioning parameters, and material density parameters;

[0021] Step 203: The 3D volume data model is converted into a triangular mesh model using the Marching Cubes algorithm, and a 3D graphic surface of the root apex tissue is constructed using the vertex set and the triangle patch connection relationship; based on the triangular mesh model, a non-local means algorithm is used to perform noise reduction processing to achieve a smooth state.

[0022] Preferably, in step S203, the process of using the non-local means algorithm to perform noise reduction is:

[0023] Obtain the normal vector of each vertex in the triangular mesh model by local plane fitting or weighted average of neighborhood patch normal vectors;

[0024] The k-nearest neighbor search or radius neighborhood search strategy is used to determine the local neighborhood vertex set of each vertex in the triangular mesh model;

[0025] By calculating the Euclidean distance and normal vector angle between vertices, a multidimensional similarity weight matrix including geometric similarity and normal vector similarity is constructed;

[0026] The vertex coordinates are iteratively updated by weighted averaging based on the weight matrix until the three-dimensional graphics surface achieves a smooth noise reduction effect.

[0027] Preferably, in step S3, the steps of constructing the multi-layer granular wall model are:

[0028] Step 301: Create particles using discrete element software and input various construction parameters of the particles, including particle radius and contact radius;

[0029] Step 302: creating a constraint framework to ensure that the generated particles are evenly laid out into a regular wall shape;

[0030] Step 303: Create bonding bonds between particles and generate force bonds between particles;

[0031] Step 304: Generate a virtual plate and compress the particles using the virtual plate to simulate the soil compaction effect;

[0032] Step 305: Save the particle wall and repeat steps 301 to 304 in sequence to generate multiple particle walls as needed.

[0033] Preferably, in step S302, the particle wall contains at least two layers of particles.

[0034] Preferably, the specific steps of step S4 are:

[0035] Step 401: importing a triangular mesh model of the root apex tissue and setting the motion state and motion parameters of the root apex tissue;

[0036] Step 402: The motion behavior of the root apical tissue and the particle wall model is solved based on the discrete element method. After the solution is completed, a table is generated in the discrete element software post-processing system to view the force and motion of each particle wall.

[0037] Preferably, in step S402, the force and movement conditions of the particle wall include the damage condition of the force bond, the force direction, and the force magnitude.

[0038] Preferably, the specific steps of step 5 are:

[0039] Step 501: preparing a soil layer corresponding to the parameters of the simulated granular wall;

[0040] Step 502: Based on the force applied to the particle wall, the prepared soil layers are arranged from small to large, and the seeds are wrapped in a spherical shape from the inside out.

[0041] Step 503: Observe the growth of the seed roots, record the soil layer that the roots can contact, correspond to the simulated particle wall data, and measure the root growth force of the seed roots.

[0042] Preferably, in step 503, the force applied to the seed root system when the force bond of the particle wall corresponding to the outermost soil layer is broken is the root growth force of the seed root system.

[0043] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0044] The root growth force measurement method based on discrete elements of the present invention uses a particle wall simulated by discrete elements to solve the force required for the roots to break through the corresponding soil layer. At the same time, multiple layers of different soil layers are prepared to wrap the seeds so that they can grow in a normal soil environment. By observing the soil layers penetrated by the roots, the root growth force of the roots can be measured by comparing the corresponding force solved by discrete element simulation, and the energy required to break through the root soil can also be estimated. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Schematic diagram of the process of the root growth force measurement method based on discrete elements of the present invention.

[0046] Figure 2 Different soil layers for construction.

[0047] Figure 3 Form force bonds between soil layers.

[0048] Figure 4 This is a cross-sectional view of the root tip tissue penetrating the first layer of soil.

[0049] Figure 5 This is a cross-sectional view of the root tip tissue penetrating the second soil layer.

[0050] Figure 6 This is a cross-sectional view of the root tip tissue penetrating the third soil layer.

[0051] Figure 7 This is a diagram showing the relationship between the force acting on the first soil layer and time.

[0052] Figure 8 This is a diagram showing the relationship between the force acting on the second soil layer and time.

[0053] Figure 9 This is a diagram showing the relationship between the force acting on the third soil layer and time.

[0054] Figure 10 Schematic diagram of wrapping seeds with multiple soil layers. DETAILED DESCRIPTION

[0055] The present invention will be described in further detail below with reference to the embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0056] See also Figures 1-10 The root growth force measurement method based on discrete elements of the present invention comprises the following steps:

[0057] Step 1: Use a CT scanner to scan the root contours in the soil using 3D X-rays to obtain 3D image data of the root tip tissue. Specifically:

[0058] Step 101: Based on the actual planting situation, place soil in a container and plant seeds at a depth of 3-5 cm. Place the container in a constant temperature and humidity environment for 3-5 days. After the seed roots grow, transfer the container and secure it to the sample stage of the scanner.

[0059] Step 102: Select a scanning mode (including eccentric rotation and concentric rotation), set the scanning parameters of the scanner, and set the slice interval according to the required observation accuracy. Use the scanner to scan the root shape to ensure that the root apex tissue is covered, including the top, bottom and sides, and finally obtain all slice images of the root apex tissue.

[0060] Step 2: Import the scanned 3D image data into computer software (e.g., Mimics InnovationSuite), reconstruct the original image using the computer software, construct a triangular network model of the root apex tissue, and smooth the surface of the triangular network model;

[0061] In this embodiment, all tomographic images of the root apex tissue need to be preprocessed to remove interference items and irrelevant items in the soil to obtain three-dimensional image data of the root apex tissue. The specific steps are as follows:

[0062] Step 201: Based on the density difference between the root system and the soil, a statistical method (e.g., the mean method) is used to determine the grayscale value interval. The grayscale value interval tool and a threshold segmentation algorithm are used to binarize the tomographic image to separate the root apex tissue from the soil particles. Based on the segmented binary image, a three-dimensional volume data model containing the complete morphological structure of the root apex tissue is constructed.

[0063] Step 202: Based on the constructed three-dimensional volume data model, three-dimensional geometric analysis and data quantification technology are used to extract various physical parameters of the root apex tissue, wherein the physical parameters include spatial scale parameters, spatial positioning parameters, and material density parameters;

[0064] Step 203: The 3D volume data model is converted into a triangular mesh model using the Marching Cubes algorithm, and a 3D graphic surface of the root apex tissue is constructed using the vertex set and the triangle patch connection relationship; based on the triangular mesh model, a non-local means algorithm is used to perform noise reduction processing to achieve a smooth state.

[0065] In step 203, the process of using the non-local means algorithm to perform noise reduction is as follows:

[0066] Obtain the normal vector of each vertex in the triangular mesh model by local plane fitting or weighted average of neighborhood patch normal vectors;

[0067] The k-nearest neighbor search or radius neighborhood search strategy is used to determine the local neighborhood vertex set of each vertex in the triangular mesh model;

[0068] By calculating the Euclidean distance and normal vector angle between vertices, a multidimensional similarity weight matrix including geometric similarity and normal vector similarity is constructed;

[0069] The vertex coordinates are iteratively updated by weighted averaging based on the weight matrix until the three-dimensional graphics surface achieves a smooth noise reduction effect.

[0070] Step 3: In the discrete element software, a multi-layer granular wall model is established by constructing granular walls with different properties, specifically:

[0071] Step 301: Create particles using discrete element software and input various construction parameters of the particles, including particle radius and contact radius;

[0072] In this embodiment, when creating different particle walls, parameters such as the Poisson's ratio, shear modulus, density, static friction coefficient, dynamic friction coefficient, and collision restitution coefficient of the particles can be adjusted as needed. The parameters used in this embodiment are shown in the following table:

[0073]

[0074] Step 302: creating a constraint framework to force the generated particles to be evenly spread out into a regular wall shape; wherein, when generating uniform particles, to ensure the generation of longitudinal force bonds, at least two layers of particles are generated in the particle wall.

[0075] Step 303: Create bonding bonds between particles and generate force bonds between particles;

[0076] In this embodiment, a bond is a connection between particles that can withstand a certain force. When the root tip tissue penetrates the soil layer, the force acting on the bond between the particles exceeds its threshold, causing the bond to break and the particles to separate. According to the principle of action and reaction, the maximum force that causes the bond to break can be regarded as the root growth force required for the root tip tissue to break through the soil.

[0077] When creating different particle walls, parameters such as the stiffness, stress, and cohesion radius of the bonding between particles can be adjusted as needed. The parameters in this embodiment are shown in the following table:

[0078] <![CDATA[Stiffness per unit area N / M 3 > Critical stress Pa Cohesion radius mm Particle wall 1 1e+08 1e+06 10 Particle Wall 2 5e+08 5e+06 10 Particle Wall 3 1e+09 1e+07 10

[0079] Step 304: Generate a virtual plate and compress the particles using the virtual plate to simulate the soil compaction effect;

[0080] Step 305: Save the particle wall and repeat steps 301 to 304 in sequence to generate multiple particle walls as needed.

[0081] Step 4: Set the motion mode of the root apex tissue, solve the motion behavior of the root apex and granular wall model based on the discrete element method, and analyze the force and motion of the root apex tissue and granular wall; specifically:

[0082] Step 401: importing a triangular mesh model of the root apex tissue, and setting the motion state of the root apex tissue, assigning motion speed, time, etc.;

[0083] In this embodiment, the movement mode of the root apex tissue is set to uniform linear motion. To speed up the simulation and reduce the running memory, the movement speed is set to twelve times that of the measured data, within the range of 2-7 mm / s; the angle between the movement direction of the root apex tissue and the plane of the particle wall is 45°;

[0084] Step 402: The motion behavior of the root apical tissue and the particle wall model is solved based on the discrete element method. After the solution is completed, a table is generated in the discrete element software post-processing system to view the force and motion conditions of each particle wall, wherein the force and motion conditions of the particle wall include the force bond failure condition, force direction, and force magnitude.

[0085] Among them, the forces acting on the bonds between particles are divided into normal force and tangential force. According to Hooke's law, the normal force is calculated as follows

[0086] F n =-k n δ n ;

[0087] Where: k n is the normal stiffness (N / M); δ n is the normal displacement (M);

[0088] Similarly, according to Hooke's law, the calculation formula for tangential force is as follows

[0089] F t =-k t δ t ;

[0090] Where: k t is the normal stiffness (N / M); δ t is the normal displacement (M);

[0091] When the normal force or tangential force during penetration of the root apex tissue exceeds the limit of the force bond, the force bond breaks;

[0092] at this time:

[0093] |Fn|>F n，max or |F t |>F t，max ;

[0094] Where: F n,max and F t,max are the limit values ​​of normal force and tangential force that cause the force bond to break;

[0095] The combined force of the normal force and the tangential force when the root tip tissue penetrates the soil layer is the root growth breaking force required for breaking through the soil;

[0096] Step 5: Create a real multi-layer soil wall to wrap the seeds, verify the simulation results, and calculate the root growth force by comparing the simulation data. Specifically:

[0097] Step 501: preparing a soil layer corresponding to the parameters of the simulated granular wall;

[0098] Step 502: Based on the force applied to the particle wall, the prepared soil layers are arranged from small to large, and the seeds are wrapped in a spherical shape from the inside out.

[0099] Step 503: Observe the growth of the seed root system, record the soil layer that the root system can contact, correspond to the simulated particle wall data, and measure the maximum growth force of the root system;

[0100] In this embodiment, the force applied when the force bond of the particle wall corresponding to the outermost soil layer is broken is the root growth force of the seed root system.

[0101] In addition, the energy required for the roots to break through the soil is approximately calculated:

[0102]

[0103] Where: F i is the average resistance of the granular wall in layer i (N); d i is the thickness of the i-th layer (m).

[0104] The above is a preferred embodiment of the present invention, but the embodiment of the present invention is not limited to the above content. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the scope of protection of the present invention.

Claims

1. A method for measuring root growth based on discrete elements, characterized in that: The following steps are involved: Step 1: Use a computed tomography scanner to scan the root system outline in the soil using 3D X-rays to obtain 3D image data of the root tip tissue; Step 2: Import the scanned 3D image data into computer software, reconstruct the original image through the computer software, construct a triangular network model of the root apex tissue, and smooth the surface of the triangular network model; Step 3: Establish a multi-layer granular wall model by constructing granular walls with different properties in the discrete element software; Step 4: Set the motion mode of the root apex tissue, solve the motion behavior of the root apex and granular wall model based on the discrete element method, and analyze the force and motion of the root apex tissue and granular wall; Step 5: Create a real multi-layer soil wall to wrap the seeds, verify the simulation results, and calculate the root growth ability by comparing the simulation data.

2. The root growth force measurement method based on discrete element according to claim 1, characterized in that: The specific steps of step S1 are: Step 101: Based on the actual planting situation, place soil in a container and plant seeds at a depth of 3-5 cm. Place the container in a constant temperature and humidity environment for 3-5 days. After the seed roots grow, transfer the container and secure it to the sample stage of the scanner. Step 102: Select a scanning mode, set the scanning parameters of the scanner, and set the slice interval according to the required observation accuracy. Use the scanner to scan the root shape to ensure that the root apex tissue is covered, including the top, bottom, and sides, and finally obtain all slice images of the root apex tissue.

3. The root growth force measurement method based on discrete element according to claim 2, characterized in that: In step S2, all the tomographic images of the root apex tissue are preprocessed to remove interference items and irrelevant items in the soil to obtain three-dimensional image data of the root apex tissue. The specific steps are as follows: Step 201: Based on the density difference between the root system and the soil, a grayscale value interval is determined using a statistical method. The grayscale value interval tool and a threshold segmentation algorithm are used to binarize the tomographic image to separate the root apex tissue from the soil particles. Based on the segmented binary image, a three-dimensional volume data model containing the complete morphological structure of the root apex tissue is constructed. Step 202: Based on the constructed three-dimensional volume data model, three-dimensional geometric analysis and data quantification technology are used to extract various physical parameters of the root apex tissue, wherein the physical parameters include spatial scale parameters, spatial positioning parameters, and material density parameters; Step 203: The 3D volume data model is converted into a triangular mesh model using the Marching Cubes algorithm, and a 3D graphic surface of the root apex tissue is constructed using the vertex set and the triangle patch connection relationship; based on the triangular mesh model, a non-local means algorithm is used to perform noise reduction processing to achieve a smooth state.

4. The root growth force measurement method based on discrete element according to claim 3, characterized in that: In step S203, the process of using the non-local means algorithm to perform noise reduction is as follows: Obtain the normal vector of each vertex in the triangular mesh model by local plane fitting or weighted average of neighborhood patch normal vectors; The k-nearest neighbor search or radius neighborhood search strategy is used to determine the local neighborhood vertex set of each vertex in the triangular mesh model; By calculating the Euclidean distance and normal vector angle between vertices, a multidimensional similarity weight matrix including geometric similarity and normal vector similarity is constructed; The vertex coordinates are iteratively updated by weighted averaging based on the weight matrix until the three-dimensional graphics surface achieves a smooth noise reduction effect.

5. The root growth force measurement method based on discrete element according to claim 4, characterized in that: In step S3, the steps for constructing the multi-layer granular wall model are: Step 301: Create particles using discrete element software and input various construction parameters of the particles, including particle radius and contact radius; Step 302: creating a constraint framework to ensure that the generated particles are evenly laid out into a regular wall shape; Step 303: Create bonding bonds between particles and generate force bonds between particles; Step 304: Generate a virtual plate and compress the particles using the virtual plate to simulate the soil compaction effect; Step 305: Save the particle wall and repeat steps 301 to 304 in sequence to generate multiple particle walls as needed.

6. The root growth force measurement method based on discrete element according to claim 5, characterized in that: In step S302, the particle wall contains at least two layers of particles.

7. The root growth force measurement method based on discrete element according to claim 5, characterized in that: The specific steps of step S4 are: Step 401: importing a triangular mesh model of the root apex tissue and setting the motion state and motion parameters of the root apex tissue; Step 402: The motion behavior of the root apical tissue and the particle wall model is solved based on the discrete element method. After the solution is completed, a table is generated in the discrete element software post-processing system to view the force and motion of each particle wall.

8. The root growth force measurement method based on discrete element according to claim 7, characterized in that: In step S402, the force and movement conditions of the particle wall include the damage of force bonds, the direction of force, and the magnitude of force.

9. The root growth force measurement method based on discrete element analysis according to claim 8, characterized in that: The specific steps of step 5 are: Step 501: preparing a soil layer corresponding to the parameters of the simulated granular wall; Step 502: Based on the force applied to the particle wall, the prepared soil layers are arranged from small to large, and the seeds are wrapped in a spherical shape from the inside out. Step 503: Observe the growth of the seed roots, record the soil layer that the roots can contact, correspond to the simulated particle wall data, and measure the root growth force of the seed roots.

10. The root growth force measurement method based on discrete element according to claim 8, characterized in that: In step 503, the force applied to the seed root system when the force bond of the particle wall corresponding to the outermost soil layer is broken is the root growth force of the seed root system.

Images