CN104573256B - A kind of crop plant type method for designing - Google Patents

Abstract

The invention relates to a crop plant type design method, which uses a plant model to mechanically analyze crop lodging, obtains appropriate plant height, yield and other shapes through data optimization, thereby providing reference for plant type improvement and design. The present invention is different from traditional agricultural breeding and genetic breeding to improve plant type, and is accomplished by a parameter optimization method implemented on a computer. This can reduce labor and experimental costs, shorten the experimental cycle, provide theoretical basis and practical guidance for reasonable plant height design, and is also conducive to the promotion of technology in agriculture.

Description

A kind of crop plant type design method

technical field

The invention belongs to the field of computer technology and electronic information technology, and in particular relates to a crop plant type design method based on mechanical calculation and plant model.

Background technique

With the increase of population and the decrease of arable land, the improvement of grain yield per unit area has become the primary goal of agricultural scientific research. And lodging is one of the important limiting factors of high crop yield. The lodging of crops not only brings serious loss of yield, reduces grain quality, but also brings great difficulties to harvest.

The lodging resistance of plants is mainly determined by plant height. Taking rice as an example, during the green revolution in the 1950s and 1960s, China's dwarf breeding stage was the first stage of development combining practice and plant type theory, and the research center of this stage was the plant height of crops. By reducing the plant height, the fertilizer tolerance, lodging resistance and close planting ability of the variety are significantly enhanced, thereby increasing the leaf area index and biological yield, thereby increasing the yield of the crop group, and a series of short-stalked high-yielding varieties such as Bantamia Nantes have been selected. Variety.

With the in-depth study of plant type theory and the development of production practice, agronomists have a new understanding of plant height. Dwarfing mainly improves the economic coefficient, and there is no obvious change in biological production. A major breakthrough in production must be achieved in biological production. There is a substantial increase in production. It is generally believed that to achieve ultra-high yield, biological yield is a foundation; plants need a certain height for high biological yield. Appropriately increasing the plant height can reduce the leaf area density, which is beneficial to the diffusion of CO ₂ and the light receiving of the middle and lower leaves, which is beneficial to the growth amount and later grain filling. At the same time, studies have shown that plant height has a significant positive correlation with biomass, especially under high-yield conditions, and the increase in biomass is the material basis for the increase in grain number per ear and thousand-grain weight.

Rice production has returned to the era of high stalks, but the plants are prone to lodging when they are tall. The goals of high yield and high quality in modern rice cultivation are without exception restricted by the problem of lodging. It can be said that the rice lodging problem solved by the first green revolution has become the bottleneck problem of rice cultivation technology in my country after half a century of reincarnation. According to expert predictions, to achieve a large area of super hybrid rice with a yield of 1,000 kg per mu, the plant type must be improved first, so that the rice plants will change from the current semi-dwarf and half-high stalks to tall stalks and even super-tall stalks, so as to increase the yield per unit area. Biological yield; at the same time, three prerequisites must be guaranteed, that is, the number of spikes per mu should not be reduced, the harvest index should be improved, and the lodging resistance should be enhanced. It is necessary to ensure the high yield of high-stalk rice and to have the advantage of lodging resistance of short-stalk rice. The problem brought about by this contradiction is in front of us: for a certain variety of rice, how high is the plant height? Ideal plant type desired? What is the basis for the judgment?

Virtual plants are one of the research hotspots that have developed rapidly in the past two decades. It integrates the knowledge of mathematics, botany, computer graphics and other related disciplines, and has been successfully applied to agriculture and forestry, virtual crop experiments and other fields. With the help of plant modeling technology extracted based on plant growth laws, people can analyze the growth behavior of plants and the interaction between plants and the external environment with the help of the accuracy, system and intuition of computer technology. This has become another important method for the development of plant science after traditional agriculture and genetic cultivation technology. The more representative ones in plant modeling technology are GreenLab model, L system and so on. Among them, the crop functional structure model is characterized by combining the advantages of the traditional process-based model and the morphological structure model originated from computer science, and introducing the mutual feedback mechanism of plant function and structure to express the plasticity of plants in different environments. Can virtual plant technology be used to study crop lodging, and provide quantitative basis for crop plant design by optimizing plant models? Such studies are few and far between. The present invention proposes a new method for designing plant types of crops.

Contents of the invention

In order to speed up the improvement of crop plant type, the present invention proposes a crop plant type design method, which uses virtual plant technology to study crop lodging, and provides quantitative basis for crop plant type design by optimizing the plant model.

A kind of crop plant type design method that the present invention proposes, comprises the following steps:

Step 1, through the plant model, extract the state variables related to the mechanical calculation of the crop in a certain growth period;

Step 2, establishing a finite element model based on the state variables related to the mechanical calculation;

Step 3: Carry out stress analysis of crops based on the finite element model, and determine whether lodging occurs;

Step 4: If it is judged that lodging occurs, optimize the relevant parameters in the plant model, and perform steps 1 to 4 again; otherwise, output the parameters of the current plant model.

In order to further optimize this method:

The relevant parameters in the plant model described in step 4 are parameters affecting plant height and/or yield in the plant model.

The method for finite element stress analysis described in step 3 includes the following steps:

Step 31, calculate the maximum moment M _break = σI/r that the crop stalk in the plant model can bear at the moment of breaking, where σ is the bending strength of the crop stalk, and I is the moment of inertia of the cross section of the crop stalk, r is the cross-sectional radius of the crop stalk;

In step 32, the moment M _i of each point on the stalk of the crop in the plant model is compared with M _break , and if the number of times M _i > M _break is greater than or equal to 1, it can be judged that the crop is lodging.

The method for optimizing the relevant parameters in the plant model described in step 4 is the crop height optimization of lodging-resistant crops, specifically: setting the crop yield, constructing a constrained single-objective optimization formula (5), and using the particle swarm optimization algorithm Get the maximum crop height of lodging resistant crops;

where a is the allometric growth rate of the plant model that can change the relationship between the length and thickness of the stalk, H(a) is the plant height of the crop at the allometric growth rate a, M _i (a) is the allometric growth is the moment of each point on the crop stalk when the rate is a, and M _break (a) is the maximum moment that the crop stalk can bear at the moment of breaking when the allometric growth rate is a.

The method for optimizing the relevant parameters in the plant model described in step 4 is the optimization of the crop yield of lodging-resistant crops, specifically: setting the plant height of the crop, constructing a single-objective optimization formula (6) with constraints, and using particle swarm optimization The algorithm obtains the maximum crop yield of lodging-resistant crops;

Among them, P ^fru is the sink strength parameter of crop ear, P ^int is the sink strength parameter of crop stalk, W(P ^fru ,P ^int ), M _i (P ^fru ,P ^int ), M _break (P ^fru ,P ^int ) represent respectively The crop yield under the parameters P ^fru and P ^int , the moment of each point on the crop stalk, and the maximum moment that the crop stalk can bear at the moment of breaking.

In step 2, the different drag forces of the air on the crops caused by different levels of wind speed are added as input to the construction of the finite element model.

The state variables related to mechanical calculation extracted in step 1 include crop structure information, crop material attribute information, and external force information. The object structure information includes the spatial position and quantity of crop organs, and the crop material attribute information includes crop organ elastic modulus. , Poisson's coefficient, material category, and external force information including gravity.

The method of the invention utilizes a plant model to mechanically analyze crop lodging, and obtains appropriate plant height, yield and other shapes through data optimization, thereby providing reference for plant type improvement and design. The present invention is different from traditional agricultural breeding and genetic breeding to improve plant type, and is accomplished by a parameter optimization method implemented on a computer. This can reduce labor and experimental costs, shorten the experimental cycle, provide theoretical basis and practical guidance for reasonable plant height design, and is also conducive to the promotion of technology in agriculture.

Description of drawings

Fig. 1 shows the framework diagram of the system of the present invention;

Fig. 2 shows the method flow chart of the present invention;

Fig. 3 shows the spatial beam element in the local coordinate system;

Fig. 4 shows the three-dimensional display of crop lodging mechanics analysis result;

Fig. 5 shows under different wind speeds, the maximum rice stalk height value of the optimized lodging resistant rice;

Figure 6 shows the maximum ear weight values of optimized lodging-resistant rice under different wind speeds.

detailed description

Various details involved in the technical solution of the present invention will be described in detail below in conjunction with the accompanying drawings. It should be pointed out that the described embodiments are only intended to facilitate the understanding of the present invention, rather than limiting it in any way.

As shown in Figure 1, the user obtains information such as stem diameter, internode length, panicle weight, etc. by measuring the actual crop rice. The parameters of the plant model are obtained by the actual information obtained through the method of parameter inversion and input into the model. The plant model in this embodiment is the GreenLab model, which automatically generates virtual plants that conform to the physiological process of plants on the computer, and optimizes the parameters of the virtual plants, so that the design results of ideal plant types can be obtained and displayed to the public in the form of data and 3D graphics. Users, users can guide the breeding work according to the results of parameter optimization, thereby affecting the morphological structure of actual crops.

As shown in Figure 2, this embodiment mainly includes the following steps:

Step 1, through the plant model, extract the state variables related to the mechanical calculation of the crop in a certain growth period.

The plant model to be designed is established through user setting/model calibration and plant model parameter input. The user can set the appropriate parameters of the crop functional structure model through the software interface, and save it as a parameter file; at the same time, the geometric and physiological parameters of the model can be adjusted according to the actual plant growth law, such as growth rhythm, ear weight, etc. Reverse request; at the same time, decide whether to perform parameter calibration according to the demand.

Simulate the dynamic growth of crops and determine a certain growth period. With the help of the dual-scale automaton theory proposed by the plant model GreenLab, the growth of rice conforms to the laws of physiological and ecological processes, and its topological structure is displayed in a three-dimensional visualization.

Extract state variables related to mechanical calculations, including rice structure information, rice material attribute information, and external force information. The object structure information includes the spatial position and quantity of rice organs, and the rice material attribute information includes rice organ elastic modulus, Poisson Coefficients, material classes, information on external forces including gravity.

In step 2, a finite element model is established based on state variables related to mechanical calculations.

In this embodiment, based on the state variables related to the mechanical calculation extracted in step 1, a finite element model is jointly constructed to analyze the mechanical deformation of the rice structure under the action of different wind speeds and determine whether lodging occurs. In order to be closer to the actual situation, this embodiment adds wind influence factors, and takes the different drag forces of the air on the crops caused by different levels of wind speed as input, and adds them to the construction of the finite element model.

The crop in this example is rice, and its rice stalks, leaves, and ears of rice can be abstracted into beam elements with a slender structure; the finite element modeling of rice stalks is mainly based on elastic-plastic material modeling; the rest such as leaves, ears of rice The finite element modeling can be simplified to elastic material modeling.

Step 3: Carry out force analysis of crops based on the finite element model, and determine whether lodging occurs.

In the finite element model, the linear force F _i on the i-th beam element can be calculated according to formula (1):

Among them, ρ _air is the air density, C _d is the drag coefficient, A _i is the upwind section area of the i-th beam unit, u _i is the air velocity at the i-th beam unit, and l _i is the i-th beam unit length.

Combined with rice structure information, rice material attribute information, and external force information (including gravity and wind force), the finite element force analysis of rice is carried out. The principle and process of the analysis can be expressed as follows:

As shown in Figure 3, the space beam unit may bear the action of torque in addition to axial force and bending moment. Each spatial beam unit has two end nodes, each node has 6 degrees of freedom in displacement, and the unit has 12 degrees of freedom in total; the node displacement array q ^e and node force array P ^e in its local coordinate system are as follows:

q ^e ＝[u ₁ v ₁ w ₁ θ _x1 θ _y1 θ _z1 u ₂ v ₂ w ₂ θ _x2 θ _y2 θ _z2 ] ^T

P ^e ＝[P _u1 P _v1 P _w1 M _x1 M _y1 M _z1 P _u2 P _v2 P _w2 M _x2 M _y2 M _z2 ] ^T

Among them, u ₁ , v ₁ , w ₁ , u ₂ , v ₂ , w ₂ , θ _x1 , θ _y1 , θ _z1 , θ _x2 , θ _y2 , θ _z2 are the two end nodes in the x, y, and z axis directions respectively. The deflection and the corresponding rotation angle; P _u1 , P _v1 , P _w1 , M _x1 , M _y1 , M _z1 , P _u2 , P _v2 , P _w2 , M _x2 , M _y2 , M _z2 are the corresponding lateral force and bending moment;

For beam elements of elastic materials (such as leaves), write the stiffness matrix of each beam element in the local coordinate system. Each beam element may carry either linear loads (such as gravity and wind) or concentrated loads (such as ears of rice).

By assembling the obtained stiffness matrices of each element, the overall stiffness matrix can be formed, and at the same time, all node loads are also assembled. Since the force analysis of each beam unit is initially established in the local coordinate system, it is necessary to transform each beam unit into the global coordinate system, so that the units at different positions have a common coordinate reference for the integration of each unit. .

Therefore, the overall finite element analysis equation of the beam element of the elastic material, that is, the stiffness equation is:

Similarly, for beam elements of elastic-plastic materials (such as rice stalks), the overall finite element analysis equation is:

After the finite element analysis equation is established, it must be solved. The boundary conditions of the whole structure are: the root of the crop is fixed to the ground by default. The stiffness equation of an elastic material is a general linear equation, which can be solved by methods such as Gaussian elimination; while the elastic-plastic problem (elastic-plastic problem) is generally a nonlinear equation, and the key to the research lies in the processing of the physical equation. At present, the main solving methods include direct iterative method, Newton-Raphson (N-R) iterative method, improved N-R iterative method and so on.

After the solution is completed, the stress and torque calculation of each unit is carried out, that is, the force analysis is carried out. The main purpose of force analysis is to analyze whether the crop stalks are bent and broken, that is, whether the crops are lodging.

The maximum moment M _break that the rice stalk can bear at the moment of breaking can be obtained by formula (4):

M _break = σI/r (4)

Among them, σ is the flexural strength, which reflects the maximum stress that the rice stalk can resist when it is bent, indicating the ability of the rice stalk to withstand the maximum load. I is the moment of inertia of the cross section of the rice stalk. r is the cross-sectional radius of the rice stalk.

The standard for judging whether lodging occurs is as follows: if the moment _Mi at a certain point of the rice stalk exceeds the maximum moment M _break that can be tolerated, that is, _Mi >M _break , the crop is considered to be lodging.

Step 4: If it is judged that lodging occurs, optimize the relevant parameters in the plant model, and perform steps 1 to 4 again; otherwise, output the parameters of the current plant model.

Mechanical lodging analysis was performed on rice to determine whether lodging occurred. If the rice lodging occurs, it is considered that the rice does not have the ideal plant type, and this parameter and the morphological structure of the rice are discarded. Iterate continuously from step 1 to step 4 until the plant model parameters of optimal plant height, yield and other traits are found. In terms of parameter optimization, this embodiment adopts the crop height optimization method of lodging-resistant crops and the crop yield optimization method of lodging-resistant crops respectively to design the optimal crop plant type.

As shown in Figure 4, after the design of the optimal crop plant type is obtained, the final optimized results are displayed in a three-dimensional visualization manner. In Figure 4, the complex crop structure is abstracted into a beam unit and finite element modeling is carried out. Through the mechanical analysis of the finite element model, the position where the rice stem breaks (lodging) is finally obtained (the thickened part of the crop stem in the figure Indicated), it can be seen that the position where lodging occurs is below 0.3 of the relative height. This is consistent with the location where lodging occurs in actual rice.

1. Optimization method of rice stalk height for lodging resistant rice

Since plant height is considered to be positively correlated with rice yield, in order to maximize rice yield, the optimization goal here is simplified to maximize rice plant height (expressed as H). The optimization algorithm used is Particle Swarm Optimization (PSO). Rice is also limited by not lodging while maximizing plant height. Therefore, the single-objective optimization problem with constraints can be expressed as formula (5):

Different rice morphological structures (here, morphological structures with different heights and radii) are obtained by changing the plant model parameter allometric growth rate a (the relationship between the length and thickness of the stalk can be changed).

Through the optimization algorithm, the maximum rice stalk height value optimized for lodging-resistant rice under different wind speeds can be obtained, as shown in Figure 5. As the wind continues to increase, the optimized height of the rice stalks is constantly decreasing, and the rice stalks are gradually thicker to prevent the rice from lodging and breaking. This is consistent with the actual situation: to withstand a large wind force, it is necessary to reduce the height, increase the diameter, and reduce the torque effect of the wind on the rice.

2. Optimization method of lodging resistance rice yield

Consider the distribution of biomass among various organs to obtain the desired yield of rice panicle (expressed as W). The yield of rice spikes is related to the sink strength parameters of different organs, including rice spikes, rice stalks, and rice leaves. Taking the functional structure model GreenLab as an example, consider the influence of the sink strength parameter P ^fru of rice ear or the sink strength parameter P ^int of rice stalk on rice panicle yield. While considering rice panicle yield, lodging resistance constraints should also be considered. Because the change of P ^fru and P ^int can affect both the biomass and the height of rice stalk. Simplify the problem, avoid considering the influence of rice height on lodging, fix the height of rice at a certain height, and only consider the relationship between P ^fru and ^Pint and lodging. In this case, it can be known from the relationship between mass and volume that when the biomass is distributed, the change in the biomass of the rice straw only changes the diameter of the rice straw, and the unit length of the rice straw remains unchanged. Therefore, the problem can be expressed as a constrained maximization problem:

(6) Figure 6 shows the relationship between the optimized rice panicle weight and different wind speeds. It can be seen from the figure that as the wind force continues to increase, in order not to cause lodging and breaking, the optimized rice yield decreases. This is consistent with actual situation: will bear big wind-force and will reduce rice ear weight, reduce the torque action of rice ear to paddy rice stalk. When the wind speed is about 9.4m/s, there is no parameter value and its corresponding rice ear weight value. This is because when the mechanical properties of rice straw (such as Young's modulus, elastic-plastic constitutive relationship) remain unchanged, no matter how the biomass is distributed among various organs, the rice structure at the current height cannot resist wind force. The impact of rice stalks is easy to bend and break.

By the method of the present invention, the crops of a certain variety can be mechanically analyzed in different growth periods to determine whether lodging occurs; different structures can also be analyzed for crops of different varieties (for example, the crops have leaves of different numbers and positions, different strains, etc.) Higher conditions) on crop lodging resistance. For the analysis of these lodging traits, the most suitable traits such as plant height and yield can be obtained through calculation, thus providing a basis for crop plant type design.

The method of the invention can also be applied to the plant type design of various crops such as wheat, and has broad application space.

The difference between the method of the present invention and the traditional methods of agricultural breeding and genetic breeding for plant type improvement lies in that the method is completed by realizing parameter optimization of plant type design on a computer. The method can be used to analyze the lodging problem of crops under the action of gravity and wind, and has the characteristics of saving time and labor, simple operation and direct view compared with the traditional method.

The above is only a specific implementation mode in the present invention, but the scope of protection of the present invention is not limited thereto. Anyone familiar with the technology can understand the conceivable transformation or replacement within the technical scope disclosed in the present invention. All should be covered within the scope of the present invention, therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims (5)

1. a kind of crop plant type method for designing, it is characterised in that comprise the following steps：

Step 1, by plant-growth model, extracts related to the Mechanics Calculation state variable under a certain growth period of crop；

Step 2, FEM model is set up based on the state variable related to Mechanics Calculation；

Step 3：Enter the force analysis of row crop based on the FEM model, judge whether to lodge；

Step 4：If it is determined that lodging, then optimize the relevant parameter in plant model, step 1 to step 4 is performed again； Otherwise export the parameter of current plant model；

Relevant parameter described in step 4 in plant model is the parameter of influence plant height and/or yield in plant model；

The method of the force analysis of finite element described in step 3, comprises the following steps：

Step 31, the maximum moment M that crop stem can bear in the moment being broken in calculating plant model_break=σ I/ R, wherein σ are the bending strength of crop stem, and I is the moment of inertia of cross-section of crop stem, and r is the cross section half of crop stem Footpath；

Step 32, by the torque M of each point on crop stem in plant model_iRespectively with M_breakContrast, if there is M_i>M_breakTime Number is more than or equal to 1, then can determine whether to be lodged for crop.

2. according to the method described in claim 1, it is characterised in that optimize the relevant parameter in plant model described in step 4 Method for lodging resistance crop crop plant height optimize, specially：Setting crop yield, builds single objective with constraints Formula (5), and the maximum plant height of crop resistant to lodging is obtained using particle cluster algorithm；

$\left\{\begin{array}{cc}Max& H\left(a\right)\\ subjectto& M_i\left(a\right)<M_{break}\left(a\right)\end{array}\right.---\left(5\right)$

Wherein a is the allometry relative growth rate that can change the relation plant model between the length of stalk and rugosity, and H (a) is friction speed Crop plant height when growth rate is a, M_iA () is the torque of each point on crop stem when allometry relative growth rate is a, M_breakA () is different The maximum moment that crop stem can bear in the moment being broken when fast growth rate is a.

3. according to the method described in claim 1, it is characterised in that optimize the relevant parameter in plant model described in step 4 Method for lodging resistance crop crop yield optimization, specially：Setting crop plant height, the single goal for building belt restraining is excellent Change formula (6), and the maximum crop yield of crop resistant to lodging is obtained using particle cluster algorithm；

$\left\{\begin{array}{cc}Max& W(P^{fru},P^{\mathrm{int}})\\ subjectto& M_i(P^{fru},P^{\mathrm{int}})<M_{break}(P^{fru},P^{\mathrm{int}})\end{array}\right.---\left(6\right)$

Wherein P^fruThe strong parameter in storehouse of crop fruit ear, P^intThe strong parameter in storehouse of crop stem, W (P^fru,P^int)、M_i(P^fru,P^int)、 M_break(P^fru,P^int) it is illustrated respectively in P^fruAnd P^intTorque, the crop stem of each point in crop yield, crop stem under parameter In the maximum moment that the moment being broken can bear.

4. according to the method any one of claim 1-3, it is characterised in that draw different grades of wind speed in step 2 The air for rising adds the structure of FEM model to the different drag forces of crop as input.

5. according to the method described in claim 4, it is characterised in that the state related to Mechanics Calculation extracted in step 1 becomes Amount includes Crop Structure information, crop material properties information, the information of external force, and thing structural information includes the space of crop organ Position and quantity, crop material properties information include elastic modelling quantity, Poisson's coefficient, the material classification of crop organ, external force Information includes gravity.