Method for burying heating pipes near the root zone of fruit trees

Abstract

The present invention discloses a method for burying heating pipes near the root zone of a fruit tree. The method obtains the distribution of heating temperature fields in the root zone of a fruit tree under different heating pipe burying schemes based on computational fluid dynamics simulation, and records the temperature values ​​of each monitoring point in the root zone of the fruit tree; then the firefly algorithm is applied to iteratively optimize the parameters, and the position of the firefly individual is updated during each iteration, and the position of the firefly individual represents the spacing and depth parameters of the heating pipe burial. For a new set of burial spacing and burial depth parameters, calling the KNN machine learning model will output the corresponding monitoring point temperature values, and then return to the firefly algorithm to determine whether the monitoring point temperature meets the constraint conditions, and optimize the heating pipe burial spacing and burial depth parameters with the goal of uniformity of the heating temperature field distribution in the root zone of the fruit tree, so as to provide a numerical basis for the burial of the heating pipe.

Description

Technical Field

[0001] The invention relates to the field of soil heating in the root zone of fruit trees in a greenhouse, and is suitable for the design of a heating pipeline burying scheme near the root zone of fruit trees. Background Art

[0002] Temperature is a crucial factor for the normal growth of fruit trees, especially when the outdoor temperature drops sharply at night in winter. In order to ensure that the fruit trees are not damaged by frost, the greenhouse needs to be heated at night in winter. The common heating method is to heat the air in the greenhouse. However, during the heating process, the air is gathered at the top of the greenhouse due to the buoyancy of heat, which does not have a good heating effect on the fruit trees planted in the greenhouse. Once the heating equipment stops working, the air temperature in the greenhouse will drop rapidly. The soil temperature in the root zone of the fruit tree affects the solubility of various nutrients such as gases and inorganic salts in the soil, which in turn affects the ion exchange activity and microbial activity, and ultimately affects the growth of the fruit tree root system and the fruit tree yield. Therefore, the soil near the root zone of the fruit tree can be heated at night in winter to achieve a suitable growth temperature for the fruit tree.

[0003] For heating the soil near the root zone of fruit trees, a method similar to floor heating is often used. Pipes are buried near the root zone of fruit trees, and hot water is used as a medium to circulate in the pipes to achieve the purpose of heating the nearby soil. Unlike the floor heating system of commercial housing, the root zone of fruit trees is more sensitive to temperature. However, in practical applications, the burial of pipelines mostly depends on the experience of workers, and the optimal pipeline burial spacing and burial depth are not determined in combination with the root zone structure and growth requirements of fruit trees. In addition, some researchers have used numerical simulation methods to explore the effects of different pipeline burial schemes on the temperature field of the crop root zone, and then select the optimal pipeline burial scheme. However, they also use the cross-comparison method to conduct several groups of experiments and select the best one from these groups. The amount of experimental data is too small, which makes it difficult to accurately find the optimal pipeline burial scheme. Summary of the invention

[0004] In order to overcome the problems that the existing heating pipe burying method relies on experience, lacks accurate data basis, and does not consider the root zone structure of fruit trees, the present invention provides a heating pipe burying method based on flow field analysis and parameter optimization.

[0005] The technical concept of the present invention is: based on computational fluid dynamics (CFD) simulation, the distribution of heating temperature field in the root zone of fruit trees under different heating pipe burial schemes is obtained, and the temperature values ​​of each monitoring point in the root zone of the fruit trees are recorded. When the Firefly Algorithm (FA) algorithm is applied, the position of the individual fireflies is updated during each iteration, and the position of the individual fireflies represents the spacing and depth parameters of the heating pipe burial. For a new set of burial spacing and burial depth parameters, calling the KNN machine learning model will output the corresponding monitoring point temperature value, and then return to the FA algorithm to determine whether the monitoring point temperature meets the constraints, and optimize the heating pipe burial spacing and burial depth parameters with the goal of uniformity of the heating temperature field distribution in the root zone of the fruit tree.

[0006] The technical solution adopted by the present invention to solve its technical problem is:

[0007] A method for burying a heating pipe near a fruit tree root zone, the steps of which are as follows:

[0008] S1. For the fruit trees where heating pipes are to be buried, a computational fluid dynamics simulation model is constructed to obtain the heating temperature field distribution of the fruit tree root zone under different pipe burial parameter combinations, and temperature monitoring points are distributed in the fruit tree root zone; the pipe burial parameter combination includes two pipe burial parameters, namely, the burial spacing and burial depth of the heating pipes;

[0009] S2. Determine the buried spacing range and buried depth range of the heating pipeline in advance according to the suitable growth temperature range of the fruit tree root zone, and sample a series of parameter points in each of the two ranges to form different pipeline buried parameter combinations; perform CFD simulation on each pipeline buried parameter combination in a computational fluid dynamics simulation model to obtain the temperature values ​​of monitoring points at different positions in the fruit tree root zone, thereby forming a training sample set consisting of a series of training samples;

[0010] S3. Taking the temperature of all monitoring points being within the suitable growth temperature range of the fruit tree root zone as a constraint condition, and the uniformity of the heating temperature field distribution in the fruit tree root zone as the objective function, the optimization algorithm is used to iteratively optimize the two pipeline burial parameters of the burial spacing and burial depth of the heating pipeline within the burial spacing range and burial depth range; each round of iteration calls the KNN machine learning algorithm to predict the temperature value of each monitoring point under the current pipeline burial parameters based on the training sample set, and then judge whether the temperature of each monitoring point meets the constraint conditions. If they all meet the constraint conditions, the objective function value is calculated; in the optimization process, the optimal burial spacing and optimal burial depth of the heating pipeline are finally determined by maximizing the uniformity of the heating temperature field distribution in the fruit tree root zone.

[0011] Preferably, two heating pipes need to be buried at the roots of each fruit tree, and the two heating pipes are symmetrically arranged on both sides of the main root.

[0012] Preferably, the computational fluid dynamics simulation model is constructed by fluid simulation software Fluent, and during the construction, the simulation boundary conditions, initial conditions and solution methods need to be set according to the actual environmental conditions in the greenhouse where the fruit trees are located.

[0013] Preferably, in S3, the specific method of calling the KNN machine learning algorithm to predict the temperature value of each monitoring point under the current pipeline burial parameters based on the training sample set is as follows:

[0014] S311, using the input current pipeline burial parameters as the feature vector of the test sample, using the pipeline burial parameters of the training sample as the feature vector of the training sample, calculating the Euclidean distance between the feature vector of the test sample and each training sample, and determining the nearest neighbor K adjacent training samples; wherein the calculation formula of the Euclidean distance between the test sample and the i-th training sample is:

[0015]

[0016] Where: x i (1) ,x i (2) The buried spacing and buried depth of the heating pipe corresponding to the i-th training sample, x j (1) ,x j (2) The buried spacing and buried depth of the heating pipes corresponding to the samples are tested respectively;

[0017] S312, based on the Euclidean distance between the test sample and each adjacent training sample, a weight is assigned to each adjacent training sample by using a Gaussian function; wherein the weight of the i-th adjacent training sample is:

[0018]

[0019] Where: μ and σ are the mean and standard deviation of the Euclidean distance values ​​from the test sample to the K neighboring training samples, respectively;

[0020] S313, weighted operation is performed on the monitoring point temperature values ​​of K neighboring training samples to obtain the monitoring point temperature value of the test sample, and the calculation formula is:

[0021]

[0022] Furthermore, the K is preferably 4.

[0023] Preferably, the optimization algorithm is a firefly algorithm, and in each round of iteration, the two-dimensional spatial position coordinates of the firefly individuals represent the two parameters to be optimized, namely, the buried pipe spacing and the buried pipe depth. The temperature values ​​of different monitoring points distributed within the root zone of the fruit tree are obtained by calling the KNN machine learning model, and the inverse of the temperature standard deviation of all monitoring points is used as the objective function representing the luminous brightness of the fireflies. The temperature of all monitoring points is within the suitable growth temperature range of the root zone of the fruit tree as a constraint condition, thereby controlling the firefly individuals with weak luminous brightness to move toward the firefly individuals with strong luminous brightness to achieve the position update of this round of iteration; when the iteration termination condition is reached, the optimization process is terminated, and the final optimal buried pipe spacing and optimal buried pipe depth are output.

[0024] Furthermore, the objective function formula is:

[0025]

[0026] Where f(i) represents the objective function value of the i-th firefly, n is the number of temperature monitoring points distributed in the root zone of the fruit tree, T i,k is the temperature of the kth monitoring point corresponding to the spatial position of the i-th firefly, T i is the average temperature of all n monitoring points.

[0027] Preferably, the iteration termination condition is that the number of iterations reaches a preset maximum number of iterations.

[0028] The beneficial effects of the present invention are mainly manifested in:

[0029] 1) Combined with the structural characteristics of the fruit tree root zone, the CFD simulation method was used to obtain the heating temperature field distribution of the fruit tree root zone under different pipeline burial schemes.

[0030] 2) The temperature values ​​of monitoring points under different heating pipe burial parameters are quickly obtained through machine learning methods.

[0031] 3) Based on the FA-KNN optimization algorithm, all the heating pipe burial parameters within the solution range can be traversed, and the uniformity of the temperature field distribution near the root zone of the fruit tree is used as the evaluation criterion to finally determine the optimal heating pipe burial spacing and depth, providing a numerical basis for the heating pipe burial. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a schematic diagram of the fruit tree root area and the buried heating pipes. In the figure, number 1 is heating pipe 1, number 2 is heating pipe 2, and A~G represent different temperature monitoring points.

[0033] Figure 2 This is the MAE curve for different K values.

[0034] Figure 3 This is the RMSE curve for different K values.

[0035] Figure 4 This is a flow chart for optimizing the buried parameters of heating pipes near the root zone of fruit trees based on the FA-KNN algorithm. DETAILED DESCRIPTION

[0036] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0037] In a preferred embodiment of the present invention, the Red Beauty citrus grown in a greenhouse is used as a target to optimize the heating pipe burial scheme. According to the characteristics of the Red Beauty citrus variety, its root system is generally buried at a depth of about 25 cm, including a main root of 10 cm and side roots and fibrous roots of 15 cm. In addition, the extension radius of the fibrous roots is approximately 25 cm. In order to ensure that the temperature in the root zone of the fruit tree is within the appropriate growth range of the plant, temperature monitoring points can be set according to the structural characteristics of the main root, side root, and fibrous root of the fruit tree to reflect the temperature field conditions within the entire root zone of the fruit tree. In this embodiment, if Figure 1 As shown in the figure, in order to provide uniform temperature conditions for the root zone of fruit trees at night in winter, Figure 1 The heating pipes shown in the numbers 1 and 2 are buried symmetrically with the main root of the fruit tree as the axis. Therefore, the temperature environment on both sides of the fruit tree root zone should be consistent. So only one side of the fruit tree needs to be set up as follows Figure 1 There are a total of 7 temperature monitoring points: A, B, C, D, E, F, and G.

[0038] The specific steps of the method for burying heating pipes near the root zone of fruit trees in this embodiment are as follows:

[0039] 1. CFD simulation model construction

[0040] The Red Beauty citrus in the greenhouse is planted in rows. Since the planting parameters of each row are basically the same, a simulation is performed for a single row of Red Beauty citrus in this embodiment. A computational fluid dynamics (CFD) simulation model is constructed based on the fluid simulation software Fluent to obtain the heating temperature field distribution of the root zone of the fruit tree under different pipeline burial parameter combinations. The pipeline burial parameter combination to be optimized includes two pipeline burial parameters: the burial spacing and burial depth of the heating pipeline. In this simulation model, for different burial spacing and burial depth of the heating pipeline, it is only necessary to modify the position of the heating pipeline in the CFD model to obtain the heating temperature field distribution of the root zone of the fruit tree and the temperature values ​​of each monitoring point under different burial schemes.

[0041] 2. Acquisition of training samples based on CFD simulation

[0042] In the CFD simulation model, according to Figure 1 In the distribution form of the simulation, 7 temperature monitoring points are distributed in the root area of ​​the fruit tree. At the same time, the boundary conditions, initial conditions and solution methods of the simulation are reasonably set according to the actual greenhouse conditions. After the simulation, the temperature values ​​of each monitoring point at each moment can be obtained. The greenhouse heating system is turned on at 6 pm, and the outdoor temperature drops to the lowest point of the day at 12 o'clock at night. Therefore, it is necessary to monitor whether the temperature near the root area of ​​the fruit tree meets the growth needs of the fruit tree after the heating system has been working for 6 hours. Therefore, according to the suitable temperature for the growth of the root area of ​​the fruit tree is 15℃~25℃, the parameter range of the buried spacing and buried depth of the heating pipeline can be roughly determined through CFD simulation. A series of parameter points are sampled in the buried spacing range Range1 and the buried depth range Range2 of the heating pipeline to form different pipeline buried parameter combinations. Each pipeline buried parameter combination contains a buried spacing value in the buried spacing range Range1 and a buried depth value in the buried depth range Range2. For each pipeline burial parameter combination, CFD simulation is performed in the computational fluid dynamics simulation model to obtain the temperature values ​​of monitoring points at different locations in the fruit tree root zone, which are used as training samples for the next step of machine learning. A series of training samples corresponding to all pipeline burial parameter combinations are obtained as a training sample set.

[0043] 3. Parameter Optimization Based on FA-KNN

[0044] In this embodiment, a FA-KNN optimization method is formed by combining the firefly algorithm (FA) and the K-nearest neighbor (KNN) machine learning algorithm to globally optimize the two pipeline burial parameters of burial spacing and burial depth. The basic process of parameter optimization based on FA-KNN is: the temperature of all monitoring points is within the suitable growth temperature range of the fruit tree root zone as a constraint condition, the uniformity of the heating temperature field distribution in the fruit tree root zone is used as the objective function, and the optimization algorithm is used to iteratively optimize the two pipeline burial parameters of the burial spacing and burial depth of the heating pipeline in the aforementioned burial spacing range Range1 and burial depth range Range2; each round of iteration calls the KNN machine learning algorithm to predict the temperature value of each monitoring point under the current pipeline burial parameter based on the training sample set, and then judge whether the temperature of each monitoring point meets the constraint condition, and if all meet, the objective function value is calculated. In the optimization process, by maximizing the uniformity of the heating temperature field distribution in the fruit tree root zone, the heating temperature field distribution in the fruit tree root zone is as uniform as possible, and finally the optimal burial spacing and optimal burial depth of the heating pipeline are determined.

[0045] The input variables of the KNN machine learning algorithm are the burial spacing and burial depth of the heating pipes. When a new set of heating pipe burial parameters is input, the KNN algorithm will use the Euclidean distance as the criterion to find the K nearest training samples around the new input sample, assign different influence weights to the K training samples according to the distance, and determine the output value of the new input sample, that is, the temperature value of each monitoring point, through weight distribution. In this embodiment, the specific method of calling the KNN machine learning algorithm to predict the temperature value of each monitoring point under the current pipeline burial parameters based on the training sample set is as follows:

[0046] 1) Take the current pipeline burial parameters as the feature vector x of the test sample j =(x j (1) ,x j (2) ), taking the pipeline laying parameters of the training sample as the feature vector of the training sample, calculate the feature vector of the test sample and each training sample (the feature vector of the i-th training sample is recorded as x i =(x i (1) ,x i (2) )) by calculating and comparing the distance between the test sample and the training sample, the K training samples with the smallest distance can be determined, which are recorded as the K nearest neighbor training samples. The calculation formula of the Euclidean distance between the test sample and the i-th training sample is:

[0047]

[0048] Where: x i(1) ,x i (2) The buried spacing and buried depth of the heating pipe corresponding to the i-th training sample, x j (1) ,x j (2) The buried spacing and buried depth of the heating pipes corresponding to the samples are tested respectively;

[0049] S312: Based on the Euclidean distance between the test sample and each adjacent training sample, a weight is assigned to each adjacent training sample by using a Gaussian function. The weight of the i-th adjacent training sample is:

[0050]

[0051] Where: μ and σ are the mean and standard deviation of the Euclidean distance values ​​from the test sample to the K neighboring training samples, respectively.

[0052] The output value of the test sample is determined by the output values ​​of the K nearest samples in the surrounding training set. Samples that are close are given a larger weight, and samples that are far away are given a smaller weight. Different influence weights are given to these K neighboring training samples according to the distance through the Gaussian function.

[0053] S313, multiply the monitoring point temperature values ​​of K neighboring training samples by the corresponding influence weights to perform a weighted operation, then add up the results, and then divide the accumulated sum by the sum of the weights to obtain the monitoring point temperature value y of the test sample j , and its calculation formula is:

[0054]

[0055] In the application of KNN machine learning algorithm, the most important thing is the selection of K value. Figure 2 and Figure 3 As shown, whether the mean absolute error MAE (Mean Absolute Error) or the root mean square error RMSE (Root Mean Squared Error) is used as the evaluation standard for machine learning effect, the error value is the smallest when K = 4. Therefore, K = 4 is selected in this embodiment.

[0056] The role of the above-mentioned KNN machine learning algorithm is to predict the temperature of the root zone of the fruit tree for each set of heating pipe burial parameters, and the optimization of the parameters needs to be implemented based on the firefly algorithm. In this embodiment, the specific parameter optimization process of the firefly algorithm is as follows:

[0057] In the process of optimizing the buried parameters of the heating pipe, the uniformity of the temperature field distribution in the root zone of the fruit tree is taken as the goal, and the standard deviation of the temperature of all monitoring points near the root zone of the fruit tree is calculated. The smaller the calculated value, the more uniform the temperature field distribution. The firefly algorithm is that the larger the target value, the stronger the brightness of the individual fireflies, and then find the optimal value. Therefore, the opposite number of the standard deviation of the temperature of all monitoring points can be used as the objective function of FA, and the temperature of the monitoring point must meet the temperature range suitable for the growth of fruit trees. Its objective function and constraints can be expressed as:

[0058]

[0059] 15≤T i,k ≤25,(k=1,2,3,4,5,6,7) (5)

[0060] Where n is the number of temperature monitoring points, T i,k is the temperature of the kth monitoring point corresponding to the space where firefly i is located, T i is the average temperature of all monitoring points.

[0061] like Figure 4 As shown in FIG. 1 , the FA-KNN parameter optimization process includes setting initial parameters, population size Npop=20, population individual dimension d=2, maximum number of iterations Tmax=100, attraction β=1, light intensity absorption coefficient γ=1, step size factor α=0.2, and random population initialization. The 2D spatial position of the firefly individual represents the two parameters to be optimized, namely, the buried pipe spacing and buried pipe depth. The KNN machine learning model is called to obtain the temperature values ​​of different monitoring points distributed within the root zone of the fruit tree. The firefly luminescence brightness is obtained by calculating the objective function value in the above formula (4). At the same time, the temperature of all monitoring points is within the suitable growth temperature range of the root zone of the fruit tree as a constraint condition, that is, to determine whether the temperature of the monitoring point satisfies the above formula (5). If the constraint condition is met, the firefly population is moved, and the firefly individuals with weak luminescence brightness are controlled to move to the firefly individuals with strong luminescence brightness to achieve the position update of this round of iteration. If the constraint condition is not met, the next round of iteration is directly carried out.

[0062] In this embodiment, the movement of individual fireflies is controlled according to the brightness of the light, and the position of individual fireflies i with weak light brightness moves to individual fireflies j with strong light brightness to achieve position update:

[0063] x i,k (t+1)=x i,k (t)+β(x j,k (t)-x i,k (t))+α(rand-1 / 2) (6)

[0064] In the formula, t is the number of iterations, x j,k(t) and x i,k (t) are the k-dimensional positions of fireflies j and i at the tth iteration, x i,k (t+1) is the k-dimensional position of the space where the firefly i is located after the current iteration is completed, and rand is a random number that obeys a uniform distribution on [0,1]. In this embodiment, the spatial position of the firefly has only two dimensions, so the dimension k in formula (6) is 2.

[0065] The firefly algorithm ends the optimization process after reaching the iteration termination condition. In this embodiment, the iteration termination condition is that the number of iterations reaches the maximum number of iterations Tmax. Therefore, the optimal value of the current number of iterations can be recorded to determine whether the maximum number of iterations Tmax has been reached. If the maximum number of iterations has not been reached, the above optimization process is repeated; if the maximum number of iterations has been reached, the optimization process is terminated and the final optimal buried pipe spacing and optimal buried pipe depth are output.

[0066] The above examples are only specific embodiments of the present invention. Obviously, the present invention is not limited to the above examples, and many variations are possible. All variations that can be directly derived or associated with the contents disclosed by a person skilled in the art should be considered as the protection scope of the present invention.

Claims

1. A method for burying heating pipes near the root zone of fruit trees, characterized in that: Here are the steps: S1. For the fruit trees where heating pipes are to be buried, a computational fluid dynamics simulation model is constructed to obtain the heating temperature field distribution of the fruit tree root zone under different pipe burial parameter combinations, and temperature monitoring points are distributed in the fruit tree root zone; the pipe burial parameter combination includes two pipe burial parameters, namely, the burial spacing and burial depth of the heating pipes; S2. Determine the buried spacing range and buried depth range of the heating pipeline in advance according to the suitable growth temperature range of the fruit tree root zone, and sample a series of parameter points in each of the two ranges to form different pipeline buried parameter combinations; perform CFD simulation on each pipeline buried parameter combination in a computational fluid dynamics simulation model to obtain the temperature values ​​of monitoring points at different positions in the fruit tree root zone, thereby forming a training sample set consisting of a series of training samples; S3, taking the temperature of all monitoring points being within the suitable growth temperature range of the fruit tree root zone as a constraint condition, taking the uniformity of the heating temperature field distribution in the fruit tree root zone as the objective function, and using the optimization algorithm to iteratively optimize the two pipeline burial parameters of the burial spacing and burial depth of the heating pipeline within the burial spacing range and burial depth range; each round of iteration calls the KNN machine learning algorithm to predict the temperature value of each monitoring point under the current pipeline burial parameters based on the training sample set, and then judge whether the temperature of each monitoring point meets the constraint conditions, and if they all meet the constraint conditions, calculate the objective function value; in the optimization process, by maximizing the uniformity of the heating temperature field distribution in the fruit tree root zone, the optimal burial spacing and optimal burial depth of the heating pipeline are finally determined; In S3, the specific method of calling the KNN machine learning algorithm to predict the temperature value of each monitoring point under the current pipeline burial parameters based on the training sample set is as follows: S311, using the input current pipeline burial parameters as the feature vector of the test sample, using the pipeline burial parameters of the training sample as the feature vector of the training sample, calculating the Euclidean distance between the feature vector of the test sample and each training sample, and determining the nearest neighbor K adjacent training samples; wherein the calculation formula of the Euclidean distance between the test sample and the i-th training sample is: Where: x i (1) ,x i (2) The buried spacing and buried depth of the heating pipe corresponding to the i-th training sample, x j (1) ,x j (2) The buried spacing and buried depth of the heating pipes corresponding to the samples are tested respectively; S312, assigning a weight to each adjacent training sample by using a Gaussian function based on the Euclidean distance between the test sample and each adjacent training sample; The weight of the i-th neighboring training sample is: Where: μ and σ are the mean and standard deviation of the Euclidean distance values ​​from the test sample to the K neighboring training samples, respectively; S313, weighted operation is performed on the monitoring point temperature values ​​of K neighboring training samples to obtain the monitoring point temperature value of the test sample, and the calculation formula is: The optimization algorithm is a firefly algorithm, and in each round of iteration, the two-dimensional spatial position coordinates of the firefly individuals represent the two parameters to be optimized, namely, the buried pipe spacing and the buried pipe depth. The temperature values ​​of different monitoring points distributed within the root zone of the fruit tree are obtained by calling the KNN machine learning model. The inverse of the temperature standard deviation of all monitoring points is used as the objective function representing the luminous brightness of the fireflies. The temperature of all monitoring points is within the suitable growth temperature range of the root zone of the fruit tree as a constraint condition, thereby controlling the firefly individuals with weak luminous brightness to move toward the firefly individuals with strong luminous brightness to achieve the position update of this round of iteration; when the iteration termination condition is reached, the optimization process is terminated, and the final optimal buried pipe spacing and optimal buried pipe depth are output.

2. The method for burying heating pipes near the root zone of fruit trees as claimed in claim 1, characterized in that: Two heating pipes need to be buried at the roots of each fruit tree, and the two heating pipes are symmetrically arranged on both sides of the main root.

3. The method for burying heating pipes near the root zone of fruit trees as claimed in claim 1, characterized in that: The computational fluid dynamics simulation model is constructed by using the fluid simulation software Fluent. During the construction, the boundary conditions, initial conditions and solution methods of the simulation need to be set according to the actual environmental conditions in the greenhouse where the fruit trees are located.

4. The method for burying heating pipes near the root zone of fruit trees as claimed in claim 1, characterized in that: The K is preferably 4.

5. The method for burying heating pipes near the root zone of fruit trees as claimed in claim 1, characterized in that: The objective function formula is: Where f(i) represents the objective function value of the i-th firefly, n is the number of temperature monitoring points distributed in the root zone of the fruit tree, T i,k is the temperature of the kth monitoring point corresponding to the spatial position of the i-th firefly, T i is the average temperature of all n monitoring points.

6. The method for burying heating pipes near the root zone of fruit trees as claimed in claim 1, characterized in that: The iteration termination condition is that the number of iterations reaches a preset maximum number of iterations.

Images