Method for optimizing stand column position and framework section of sunlight greenhouse with single-tube arch structure

Abstract

The invention discloses a method for optimizing the stand column position and the framework section of a sunlight greenhouse of a single-tube arch structure. The method comprises the steps that a three-dimensional finite element model comprising an arch frame, stand columns and straining beams is established; determining the optimal distance from the stand column to the north wall through parametric analysis; sensitivity analysis is carried out, and key skeleton section parameters influencing the allowable snow load are identified; establishing a random forest model to establish a relationship between different skeleton section parameter combinations and allowable snow load values; and on the basis of the allowable snow load value predicted by the random forest model, the optimal skeleton section combination is determined in combination with the calculated steel consumption. The optimal stand column position can be effectively determined, the ultimate bearing capacity and the allowable snow load value of the arch structure are remarkably improved, meanwhile, the steel consumption is reduced by optimizing the cross section of the framework, the technical problem that the structural safety and economical efficiency of a large-span sunlight greenhouse under the snow load condition are difficult to balance is solved, and the construction cost is reduced. And a reliable theoretical basis and a practical optimization means are provided for the design of the sunlight greenhouse.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural facility engineering, in particular to a method for optimizing the column position and the skeleton section of a single-pipe arch structure sunlight greenhouse. BACKGROUND

[0002] Sunlight greenhouse is an important facility for winter vegetable production in northern China, and its structural safety directly affects production efficiency. In recent years, single-pipe arch structure has been widely used in sunlight greenhouses due to its convenient construction and low cost. However, as the span of the greenhouse increases, snow load becomes the main factor leading to structural failure. Current researches mainly focus on two-dimensional model analysis, which fails to fully consider the coupling effect of arch, column and tie beam, and lacks a systematic optimization method for section, resulting in high steel consumption or insufficient safety margin.

[0003] In addition, factors such as non-uniform snow load, crop load and initial geometric defects further increase the risk of structural safety. Although some individual studies have evaluated the impact of column position through finite element analysis, there is no complete optimization design system to meet the dual requirements of structural safety and steel economy.

[0004] In order to solve the above problems, an ideal technical solution is needed. SUMMARY

[0005] To solve the above problems, the present application provides an optimization method for scientifically determining the column position of a single-pipe arch structure sunlight greenhouse through three-dimensional finite element analysis, sensitivity analysis and random forest model, and accurately optimizing the area of the skeleton section, thereby significantly reducing the steel consumption while ensuring safety.

[0006] To achieve the above purpose, the present application provides an optimization method for the column position and skeleton section of a single-pipe arch structure sunlight greenhouse, comprising the following steps: step one: establishing a three-dimensional finite element model containing arch, column and tie beam; step two: determining the optimal distance of the column from the north wall through parametric analysis; step three: performing sensitivity analysis to identify key skeleton section parameters affecting the allowable snow load; step four: establishing a random forest model to establish the relationship between different skeleton section parameter combinations and allowable snow load values; step five: based on the predicted allowable snow load value of the random forest model, combining the calculated steel consumption, to determine the optimal skeleton section combination.

[0007] The application can determine the optimal position of the stand column from the north wall by establishing a three-dimensional finite element model of the single-tube arch structure sunlight greenhouse and through parameterized analysis. Through sensitivity analysis, the most critical skeleton section parameters affecting the allowable snow load can be determined. According to the results of the sensitivity analysis, through orthogonal test design, representative section parameter combinations can be obtained and simulated. Then by constructing a random forest model between the above-mentioned representative section parameters and the allowable snow load value, based on the allowable snow load value predicted by the random forest model, combined with the calculated steel consumption, the optimal skeleton section combination can be finally determined.

[0008] Further, the load types of the three-dimensional finite element model in step one include dead load, snow load and crop load, and four load working condition combinations are adopted for analysis, including: working condition 1: dead load + uniform snow load; working condition 2: dead load + non-uniform snow load; working condition 3: dead load + uniform snow load + crop load; working condition 4: dead load + non-uniform snow load + crop load.

[0009] In the scheme, the loads of the sunlight greenhouse are subdivided into three cases and combined into four load working conditions in the process of establishing the three-dimensional finite element model, which can be closer to the actual load situation, and the analyzed structure position will be more reasonable, which is more conducive to implementation.

[0010] Further, the snow load is applied to the arch structure in the form of node force, and the node force Q The calculation formula is as follows: Node 1: Q 1= (Δ x 1 + Δx 2 / 2)∙ μ 1∙ s 0 Node i: Q i = (Δ x i / 2 + Δx i+1 / 2)∙ μ i ∙ s 0 Node n-1: Q n-1 = (Δ x n + Δx n-1 / 2)∙ μ n-1 ∙ s 0 In the formula, Δ x 1、 Δx 2 is the horizontal projection distance between node 1 and the adjacent two nodes, μ1 is the snow load distribution coefficient of node 1, Δ x i , Δx i+1 is the horizontal projection distance between the two adjacent nodes of node i, μ i is the snow load distribution coefficient of node i, Δ x n 、 Δ x n-1 is the horizontal projection distance between the two adjacent nodes of node n-1, μ n-1 is the snow load distribution coefficient of node n-1, Δ s 0 is the basic snow pressure.

[0011] Further, the parameterized analysis in step two uses the arc length method to simulate the failure process under the action of snow load, calculates the ultimate bearing capacity and the allowable snow load value, and thereby determines the optimal position of the column.

[0012] In the scheme, the arc length method in the ANSYS Workbench software is used to simulate the failure process of snow load, which can make the calculation result more stable and efficient.

[0013] Further, the sensitivity analysis in step three uses the SOBOL global sensitivity analysis method to analyze the sensitivity of the skeleton section parameters of the arch, column and pull beam to the allowable snow load.

[0014] In the scheme, through the sensitivity analysis, it can be determined that the sensitivity of the outer diameter of the pull beam is the highest, which provides an optimization direction for the subsequent skeleton section design.

[0015] Further, the random forest model in step four is based on the skeleton section combination generated by the orthogonal test to predict the allowable snow load value.

[0016] In the scheme, the random forest model can quickly predict the allowable snow load value of different skeleton section combinations, making the analysis result more accurate and the analysis process more efficient.

[0017] Further, the specific process of step five is: S1: determine the snow load design value; S2: construct the skeleton section combination; S3: use the random forest model to predict the allowable snow load value, and select the section combination higher than the snow load design value, and calculate the steel consumption; S4: determine the initial optimal combination of the skeleton section according to the steel consumption; S5: verify the initial optimal combination of the skeleton section using the finite element method; S6: determine whether the predicted allowable snow load value is greater than the snow load design value, if yes, proceed to the next step, if not, adjust the parameters of the skeleton section and re-perform S5; S7: determine the skeleton section combination that meets the snow load design value and has the least steel consumption.

[0018] The scheme can quickly screen out a section combination that meets the snow load design value and has the lowest steel consumption, so as to achieve the dual requirements of structural safety and steel economy.

[0019] Further, the three-dimensional finite element model in step one introduces the influence of initial geometric defects, and the second-order mode of eigenvalue buckling analysis is used as the initial defect form.

[0020] In the scheme, the three-dimensional finite element model is more reasonable by introducing initial geometric defects, so that the optimization result is closer to the application scene in actual life, and the accuracy of the application is also embodied.

[0021] The application has outstanding substantial features and significant progress compared with the prior art, specifically: 1. The three-dimensional finite element model accurately simulates the structural response under snow load, avoiding the limitations of two-dimensional models. By using the method of the application, after optimizing the position of the column, the ultimate bearing capacity and allowable snow load value of the sunlight greenhouse can be greatly improved, effectively preventing the phenomenon of strength failure and instability damage of the sunlight greenhouse.

[0022] 2. By using SOBOL global sensitivity analysis and random forest model, the size of the skeleton section of the sunlight greenhouse can be accurately optimized, the consumption of steel is reduced under the premise of safety, and the construction cost is reduced. And the application can be used in the construction process of large-span sunlight greenhouse, which significantly reduces the steel consumption under the premise of safety.

[0023] 3. The application integrates finite element analysis and engineering practice, and provides a complete solution from theory to application. Moreover, the optimization result of the application is verified, and the reliability is high, which can provide direct guidance for the design of sunlight greenhouse.

[0024] 4. The application considers various influencing factors such as non-uniform snow load, crop load and initial geometric defects, so it can be used in the design of sunlight greenhouse under different geographical environment and load conditions. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 The step flow chart of the application; Figure 2 The structure diagram of the sunlight greenhouse with single-pipe arch structure in the application; Figure 3 The initial geometric defect diagram of the three-dimensional finite element model in the application; Figure 4 The specific parameter diagram of the 12-meter-span sunlight greenhouse in the application; Figure 5: Schematic diagram of snow load distribution coefficient of the greenhouse roof and crop suspension system of the present invention; Figure 6 : A schematic diagram of node division in this invention; Figure 7 : A schematic diagram illustrating the influence of different column positions on the ultimate bearing capacity of the arch structure in this invention; Figure 8 : A schematic diagram illustrating the influence of different column positions on the allowable snow load of the arch structure in this invention; Figure 9 : A schematic diagram of the deformation of the arch structure 6 meters from the north wall of the column when the snow load reaches the allowable snow load under load condition 2 in this invention; Figure 10 This invention presents a schematic diagram of the first-order sensitivity index and global sensitivity index of the arch frame, columns, and tie beams under allowable snow load when the column is 6m away from the north wall. Figure 11 : A schematic diagram showing the allowable snow load corresponding to different cross-sectional combinations of arch frames, columns, and tie beams in this invention; Figure 12 : A schematic diagram illustrating the correlation between the simulated values ​​from the finite element method and the calculated values ​​from the random forest method in this invention; Figure 13 : A detailed flowchart of step five in this invention; Figure 14 This is a schematic diagram comparing the maximum stress of snow load obtained by finite element simulation in this invention with the maximum stress obtained by experiments in the literature. Detailed Implementation

[0026] The technical solution of the present invention will be further described in detail below through specific embodiments.

[0027] like Figure 1 As shown, this invention provides a method for optimizing the column position and frame cross-section of a single-tube arch structure greenhouse, including the following steps: Step 1: Establish a three-dimensional finite element model including the arch frame, columns, and tie beams; Step 2: Determine the optimal distance between the columns and the north wall through parametric analysis; Step 3: Perform sensitivity analysis to identify key frame cross-section parameters affecting the allowable snow load; Step 4: Establish a random forest model to establish the relationship between different combinations of frame cross-section parameters and the allowable snow load value; Step 5: Based on the allowable snow load value predicted by the random forest model, combined with the calculated steel consumption, determine the optimal frame cross-section combination.

[0028] Specifically, step one involves establishing a three-dimensional finite element model including the arch frame, columns, and tie beams: In this embodiment, ANSYS Workbench software is used to construct the three-dimensional finite element model of the greenhouse. For example... Figure 2As shown, the model includes structural components such as a single-tube arch frame, columns, and tie beams, all made of Q235 steel (yield strength 235 MPa, Young's modulus 206 GPa). The simulation was performed using the BEAM188 element in ANSYS Workbench software, considering both geometric and material nonlinear behavior. Figure 2 As shown, the bottom of the single-tube arch frame is fixed in concrete. During the simulation, to avoid overestimating the snow load bearing capacity of the single-tube arch frame, the column bases are hinged to the ground, while the single-tube arch frame and the tie beam are rigidly constrained. In actual implementation, the rigid constraint can be in the form of clamps or bolted connections. In the actual simulation, as... Figure 3 As shown, in order to make the three-dimensional finite element model of the present invention more reasonable and the optimization results obtained more closely resemble the application scenarios in real life, the three-dimensional finite element model of the present invention introduces the influence of initial geometric defects, and adopts the second-order mode of eigenvalue buckling analysis as the initial defect morphology, and introduces it by amplifying the second-order mode by 30 times to simulate actual construction errors.

[0029] The above three-dimensional finite element model is based on a typical 12-meter span solar greenhouse in northern my country. For example... Figure 4 As shown, the specific parameters of this common 12-meter span greenhouse are: span 12.0 meters, ridge height 5.45 meters, north wall height 3.49 meters, horizontal projection distance of the rear roof 1.54 meters, and slope angles of the rear and front roofs of 57° and 28°, respectively. Furthermore, the covering materials used in actual construction include 0.2mm polyethylene film (load 7 N / m²) and foamed polyethylene insulation blanket (load 7 N / m²).

[0030] Meanwhile, according to the Chinese standard GB / T 51183-2016, the load types in the three-dimensional finite element model in step one include dead load, snow load, and crop load. The dead load is automatically calculated by ANSYS Workbench software based on the structure's self-weight, with additional fixed equipment loads (70 N / m²), membrane loads, and insulation blanket loads. Snow loads are calculated based on both uniform and non-uniform distributions, such as... Figure 5 As shown in the figure, the snow load distribution coefficient µ is determined based on the roof conditions. rb μ r μ rm These are the snow load distribution coefficients for the rear roof, the roof, and the maximum snow load distribution coefficient for the roof. The parameters 1.54m, 3.35m, 6.71m, 10.06m, 1.08m, 11.01m, and 60° in the figure can be calculated according to the Chinese standard GB / T 51183-2016. The crop load acts on the arch frame through the suspension system, calculated at 150 N / m², and decomposed into horizontal and vertical components.

[0031] Four load cases are adopted to analyze the arch structure, including: case 1: dead load + uniform snow load; case 2: dead load + non-uniform snow load; case 3: dead load + uniform snow load + crop load; case 4: dead load + non-uniform snow load + crop load. The snow load is applied to the arch structure in the form of node force, and the node division is shown in Figure 6 The calculation formula of the node force Q is as follows: Node 1: Q 1= (Δ x 1 + Δx 2 / 2)∙ μ 1∙ s 0 Node i: Q i = (Δ x i / 2 + Δx i+1 / 2)∙ μ i ∙ s 0 Node n-1: Q n-1 = (Δ x n + Δx n-1 / 2)∙ μ n-1 ∙ s 0 wherein Δ x 1、 Δx 2 is the horizontal projection distance between node 1 and the adjacent two nodes, μ 1 is the snow load distribution coefficient of node 1, Δ x i , Δx i+1 is the horizontal projection distance of node i and the adjacent two nodes, μ i is the snow load distribution coefficient of node i, Δ x n 、 Δ x n-1 is the horizontal projection distance of node n-1 and the adjacent two nodes, μ n-1 is the snow load distribution coefficient of node n-1, s 0 is the basic snow pressure.

[0032] Specifically, step two: determine the optimal distance of the column from the north wall through parametric analysis: first of all, it should be pointed out that the distance of the column from the north wall in the present application is 2.0 meters to 11.0 meters, wherein each different column position is spaced 1 meter apart. Parametric analysis is to simulate the failure process under the action of snow load by using the arc length method, and through nonlinear analysis, the ultimate bearing capacity and allowable snow load of the column under different positions are calculated. As shown in Figure 7 and Figure 8 , the influence of column position on the ultimate bearing capacity and allowable snow load shows a parabolic trend, and through comparative analysis, it can be concluded that when the column is 6 meters away from the north wall, i.e. at a position of 50% of the span, the ultimate bearing capacity and allowable snow load obtained are the largest and the performance is the best. As shown in Figure 9 , in the case of load case 2, when the snow load reaches the allowable snow load, the deformation of the arch structure at the position of 6 meters away from the north wall of the column, wherein the deformation of the arch is magnified ten times to show, the deformation characteristics can be seen more clearly. Through the deformation analysis of the arch structure, it is shown that when the snow load reaches the allowable snow load, the top of the front roof of the arch structure bends inward, the north wall inclines outward, and the maximum displacement occurs in the middle of the south half of the front roof, while the column can effectively suppress the severe deformation of the arch structure, thereby enhancing the stability of the overall arch structure.

[0033] Specifically, step three: perform sensitivity analysis to identify key skeleton section parameters affecting the allowable snow load: SOBOL global sensitivity analysis method is used to analyze the sensitivity of the skeleton section parameters of the arch, column and tie beam to the allowable snow load. As shown in Figure 10 , the sensitivity analysis focuses on the outer diameter of the arch (Da), the outer diameter of the column (Dc), the outer diameter of the tie beam (Dt), and the wall thickness of the arch (Ta), the wall thickness of the column (Tc), and the wall thickness of the tie beam (Tt), and the results show that the first-order sensitivity index S1 of the outer diameter of the tie beam (Dt) is the highest, S1 = 0.68, and the total effect index ST = 0.75, the first-order sensitivity index S1 of the outer diameter of the arch (Da) is lower than that of the outer diameter of the tie beam (Dt), and the first-order sensitivity index S1 of the outer diameter of the column (Dc), the wall thickness of the arch (Ta), the wall thickness of the column (Tc), and the wall thickness of the tie beam (Tt) is too small and can be ignored.

[0034] Specifically, step four: establish a random forest model to establish the relationship between different skeleton section parameter combinations and allowable snow load values: in the actual operation process, first generate 200 skeleton section combinations based on orthogonal test, i.e. the combination of the outer diameter and wall thickness of the arch, column and tie beam, and then calculate the allowable snow load of each section combination by finite element method. 120 groups of data are trained and 80 groups of data are verified by using the random forest model, the input parameters include the section modulus of the outer diameter and wall thickness of the arch, column and tie beam, and the output result is the predicted value of the allowable snow load, as shown in Figure 11The diagram shows the allowable snow load corresponding to different combinations of arch frames, columns, and tie beam sections. Figure 12 As shown in the figure, the allowable snow load values ​​for each section combination calculated by the finite element method are highly correlated with the predicted allowable snow load values ​​obtained using the random forest model. Each point in the figure represents a set of section combinations, the light blue area reflects the fluctuation range of the data, and the red straight line is the fitted trend line. The calculated coefficient of determination R² of the random forest model is 0.90, indicating that the model has a high degree of fit to the data and can be used to quickly predict allowable snow load values. This solves the problem of long calculation times previously encountered when using software such as ANSYS, thus improving work efficiency.

[0035] Specifically, such as Figure 13 As shown, the specific process of step five is as follows: S1: Determine the snow load design value; S2: Construct the skeleton section combination; S3: Use the random forest model to predict the allowable snow load value, filter out the section combination that is higher than the snow load design value, and calculate the steel consumption; S4: Determine the initial optimal combination of skeleton sections based on the steel consumption; S5: Use the finite element method to verify the initial optimal combination of skeleton sections; S6: Determine whether the predicted allowable snow load value is greater than the snow load design value. If yes, proceed to the next step; if no, adjust the parameters of the skeleton section and repeat S5; S7: Determine the skeleton section combination that meets the snow load design value and has the least steel consumption.

[0036] Step five of this invention optimizes the dimensions of the skeleton cross-section with the goal of minimizing the total steel consumption. First, the snow load design value is determined, and all possible skeleton cross-section combinations are constructed, i.e., combinations of the outer diameter and wall thickness of the arch frame, columns, and tie beams. The allowable snow load for each cross-section combination is predicted using a random forest model. Simultaneously, the steel consumption for each cross-section combination is calculated using a steel consumption calculation model, resulting in an initial optimal combination of skeleton cross-sections. This initial optimal combination is then verified using the finite element method to ensure that the strength failure of the arch structure corresponding to this initial optimal combination precedes instability failure. Since some cross-section combinations that do not meet the allowable snow load requirements may be retained in practical applications of the random forest model, a new round of screening is needed to ensure the accuracy of the final result. It is determined whether the allowable snow load value of the initial optimal combination is greater than the snow load design value. If the allowable snow load value is greater than the snow load design value, the initial optimal combination is determined to be the best skeleton cross-section combination; otherwise, the skeleton cross-section parameters need to be adjusted, and the finite element method is used again for verification. After a series of screenings, the final frame section combination that meets the snow load design value and requires the least amount of steel was obtained. As shown in Table 1, the steel consumption in the table represents the amount of steel used per unit ground area. The data in Table 1 shows that the steel consumption of the optimized frame section combination is reduced by 21.4%-43.7% compared to the traditional single-tube arch.

[0037] Table 1: Comparison of steel consumption of optimized skeleton cross-section combination and traditional single-pipe arch

[0038] where the snow load design value is 250 N·m -2 The steel consumption of the arch structure in the present application is reduced by 43.7% compared to that of the traditional single-pipe arch structure, and the snow load design value is 300 N·m -2 The steel consumption of the arch structure in the present application is reduced by 21.4% compared to that of the traditional single-pipe arch structure. Therefore, through comparison of the data, it can be concluded that the steel consumption calculation model of the present application is effective, and can reduce the steel consumption under the premise of safety.

[0039] Finally, as shown in Figure 14 the reference, the maximum stress measured value of the arch structure under different snow depths in the reference and the simulated stress value of the three-dimensional finite element model in the present application are compared, and the difference value is less than 4%. Therefore, the reliability of the optimization method of the present application is further verified.

[0040] The above reference is from Lee, S.; Lee, J.; Jeong, Y.; Choi, W (2020). Development of a structural analysis model for pipe structures to reflect ground conditions. Biosystems Engineering. 197, 231-244. (Author: Lee, S.; Lee, J.; Jeong, Y.; Choi, W, Publication Year: 2020, Paper Title: Development of a structural analysis model for pipe structures to reflect ground conditions, Journal Name: Biosystems Engineering, Publication Information: Journal, Vol. 197, pp. 231-244).

[0041] In summary, the present application realizes the balance between safety and economy by integrating finite element analysis and random forest model, and provides a reliable foundation for large-scale application. Moreover, the optimization method of the present application has been verified in practical engineering, and can be popularized to similar sunlight greenhouse structures in other regions of China, which is conducive to the sustainable development of agriculture. The calculation and optimization in the present application can be realized by software such as ANSYS Workbench and Python, ensuring operability and repeatability.

[0042] It should be pointed out finally that the above examples are only used to illustrate the technical solutions of the present application but not to limit it; although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the specific embodiments of the present application can be modified or some technical features can be replaced by equivalent ones; without departing from the spirit of the technical solutions of the present application, which should be covered in the technical solution range claimed by the present application.

Claims

1. A method for optimizing the column positions and frame cross-sections of a single-tube arch structure greenhouse, characterized in that: The method includes the following steps: Step 1: Establish a three-dimensional finite element model including the arch frame, columns, and tie beams; Step 2: Determine the optimal distance between the column and the north wall through parametric analysis; Step 3: Conduct sensitivity analysis to identify key skeleton section parameters that affect the allowable snow load; Step 4: Establish a random forest model to establish the relationship between different combinations of skeleton cross-section parameters and allowable snow load values; Step 5: Based on the allowable snow load value predicted by the random forest model and combined with the calculated steel usage, determine the optimal combination of skeleton cross sections.

2. The method for optimizing the column position and frame cross-section of a single-tube arch structure greenhouse according to claim 1, characterized in that: The load types of the three-dimensional finite element model described in step one include dead load, snow load, and crop load, and four load case combinations are used for analysis, including: Case 1: dead load + uniform snow load; Case 2: dead load + non-uniform snow load; Case 3: dead load + uniform snow load + crop load; Case 4: dead load + non-uniform snow load + crop load.

3. The method for optimizing the column position and frame cross-section of a single-tube arch structure greenhouse according to claim 2, characterized in that: The snow load is applied to the arch structure in the form of nodal forces. Q The calculation formula is as follows: Node 1: Q 1 = (Δ x 1 +Δx 2 / 2)∙ μ 1∙ s 0 Node i: Q i = (Δ x i / 2 +Δx i+1 / 2)∙ μ i • s 0 Node n-1: Q n-1 = (Δ x n +Δx n-1 / 2)∙ μ n-1 • s 0 In the formula Δ x 1. Δx 2 represents the horizontal projection distance between node 1 and its two adjacent nodes. μ 1 represents the snow load distribution coefficient at node 1, Δ x i Δx i+1 μ is the horizontal projection distance between two adjacent nodes of node i. i Let Δ be the snow load distribution coefficient at node i. x n , Δx n-1 μ is the horizontal projection distance between two adjacent nodes of node n-1. n-1 Let n be the snow load distribution coefficient at node n-1. s 0 represents the basic snow load.

4. The method for optimizing the column position and frame cross-section of a single-tube arch structure greenhouse according to claim 1, characterized in that: The parametric analysis described in step two uses the arc length method to simulate the failure process under snow load, calculates the ultimate bearing capacity and allowable snow load value, and thus determines the optimal position of the column.

5. The method for optimizing the column position and frame cross-section of a single-tube arch structure greenhouse according to claim 1, characterized in that: The sensitivity analysis described in step three uses the SOBOL global sensitivity analysis method to analyze the sensitivity of the skeleton section parameters of the arch frame, columns and tie beams to the allowable snow load.

6. The method for optimizing the column position and frame cross-section of a single-tube arch structure greenhouse according to claim 1, characterized in that: The random forest model described in step four predicts the allowable snow load value based on the skeleton section combination generated by orthogonal experiments.

7. The method for optimizing the column position and frame cross-section of a single-tube arch structure greenhouse according to claim 1, characterized in that: The specific process of step five is as follows: S1: Determine the snow load design value; S2: Construct the skeleton section combination; S3: Use the random forest model to predict the allowable snow load value, filter out the section combination that is higher than the snow load design value, and calculate the steel consumption; S4: Determine the initial optimal combination of skeleton sections based on the steel consumption; S5: Use the finite element method to verify the initial optimal combination of skeleton sections; S6: Determine whether the predicted allowable snow load value is greater than the snow load design value. If yes, proceed to the next step; if not, adjust the parameters of the skeleton section and repeat step S5. S7: Determine the combination of frame sections that meets the snow load design value and uses the least amount of steel.

8. The method for optimizing the column position and frame cross-section of a single-tube arch structure greenhouse according to claim 1, characterized in that: The three-dimensional finite element model described in step one incorporates the influence of initial geometric defects and uses the second-order mode of eigenvalue buckling analysis as the initial defect morphology.

Images