A plant irrigation method and computer program product based on soil sediment migration analysis

Abstract

The present invention discloses a plant irrigation method and computer program product based on soil sediment migration analysis. The method is as follows: using a soil water movement mathematical model to simulate water flow movement under different irrigation schemes, and obtaining the characteristic parameters of the corresponding migration vortex; for each irrigation scheme, according to its migration vortex characteristic parameters, using a soil solute migration mathematical model to simulate and analyze the soil sediment migration process to obtain the corresponding sediment migration width, and thereby establish a quantitative relationship between the migration vortex characteristic parameters and the sediment migration width; determining the target range of sediment migration width according to the restoration characteristics of plants planted in contaminated soil; according to the above quantitative relationship, obtaining the migration vortex characteristic parameters corresponding to the target range of sediment migration width, and thereby determining the corresponding irrigation scheme, and irrigating the plants in the contaminated soil accordingly. The method can prevent pollutants from migrating and spreading to soil areas that cannot be restored by plants due to irrigation.

Description

Technical Field

[0001] The present invention relates to the technical field of contaminated soil plant remediation, and in particular to a plant irrigation method and a computer program product based on soil sediment migration analysis. Background Art

[0002] As soil pollution becomes increasingly serious, the industry has proposed a phytoremediation method to remediate contaminated soil. The operation is to plant relevant plants in the contaminated soil. These plants will absorb and accumulate pollutants in the soil or convert pollutants in the soil into non-toxic substances during their growth. During the growth of plants, users need to irrigate the plants to ensure their normal growth. Irrigation of plants usually forms a rotating flow of water in the soil, generating a migration vortex, which will drive the migration and diffusion of pollutants deposited in the soil. The intensity and direction of the migration vortex formed under different irrigation schemes will be different. The intensity and direction of the migration vortex will affect the degree of migration and diffusion of sediments (sediments including pollutants) in the soil (the degree of migration and diffusion is characterized by the migration width). The intensity and direction of the migration vortex are affected by the degree of water flow convergence, and the degree of water flow convergence is related to factors such as irrigation water speed, irrigation water volume, and irrigation duration. Under different irrigation water speeds, irrigation water volumes, and irrigation durations (i.e., different irrigation schemes), the intensity and direction of the migration vortex may vary greatly, resulting in significant differences in the degree of sediment migration and diffusion. If the migration and diffusion of sediments is too large, pollutants will migrate and diffuse to unexpected soil areas, such as migrating and diffusing to soil areas that cannot be repaired by plants, resulting in the expansion of the scope of soil pollution and affecting the effectiveness of plant restoration. Summary of the invention

[0003] The technical problem to be solved by the present invention is how to prevent pollutants from migrating and spreading to soil areas that cannot be repaired by plants due to irrigation during the process of plant repair of contaminated soil.

[0004] In order to solve the above technical problems, the present invention provides a plant irrigation method based on soil sediment migration analysis, comprising the following steps:

[0005] S1. Establish a mathematical model of soil water movement;

[0006] S2. Use a soil water movement mathematical model to simulate water flow movement under different irrigation schemes, and obtain characteristic parameters of migration vortices formed under different irrigation schemes, wherein the characteristic parameters include intensity and direction;

[0007] S3. For each irrigation scheme, according to its migration eddy characteristic parameters, a soil solute migration mathematical model is used to simulate and analyze the soil sediment migration process, and the sediment migration width corresponding to the migration eddy characteristic parameters is obtained;

[0008] S4. establishing a quantitative relationship between the migration eddy characteristic parameters and the sediment migration width according to the sediment migration width corresponding to the migration eddy characteristic parameters under the various irrigation schemes described above;

[0009] S5. Determine the target range of the migration width of sediments in the contaminated soil based on the restoration characteristics of the plants planted in the contaminated soil;

[0010] S6. According to the quantitative relationship between the migration eddy characteristic parameters and the sediment migration width, the migration eddy characteristic parameters corresponding to the target range of sediment migration width are obtained, and the corresponding irrigation plan is determined accordingly;

[0011] S10. Irrigating the plants in the contaminated soil according to the determined irrigation plan.

[0012] Further, the following steps are included between step S6 and step S10:

[0013] S7, irrigating the plants in the contaminated soil according to the irrigation scheme determined in step S6;

[0014] S8. After a preset time has passed since the irrigation is completed, the sediment migration is detected, and the sediment migration width is calculated based on the sediment migration, and it is determined whether the sediment migration width is within the above-mentioned migration width target range;

[0015] S9. If the sediment migration width is within the above migration width target range, the current irrigation plan is used as the optimal irrigation plan; if the sediment migration width exceeds the above migration width target range, the irrigation plan is automatically adjusted according to the deviation degree of the target range until the sediment migration width is within the migration width target range, and the adjusted irrigation plan is used as the optimal irrigation plan;

[0016] Step S10 specifically involves irrigating the plants in the contaminated soil according to the optimal irrigation plan.

[0017] Further, step S7 specifically includes:

[0018] A plurality of circles of sediment sensors are arranged around the contaminated soil to detect the sediment concentration at the location of the sensors, wherein the circles of sediment sensors are spaced apart from the edge of the contaminated soil toward the outside, and each circle of sediment sensors includes a plurality of sediment sensors spaced apart in the circumferential direction;

[0019] The sediment migration situation is obtained based on the sediment concentration detected by the sediment sensors in each circle.

[0020] Furthermore, step S10 specifically includes:

[0021] Taking the values ​​of various irrigation parameters of the optimal irrigation plan as the highest threshold, under the premise of not exceeding the highest threshold of the irrigation parameters, determine whether the optimal irrigation plan needs to be adjusted according to the current plant growth stage. If necessary, automatically adjust at least one irrigation parameter of the optimal irrigation plan and irrigate the plants according to the adjusted irrigation plan. Otherwise, irrigate according to the optimal irrigation plan, where the irrigation parameters include irrigation water speed and irrigation duration.

[0022] Furthermore, step S1 specifically establishes a soil moisture movement mathematical model based on the characteristics of soil moisture movement, using the Richards equation as a basic control equation and the Van Genuchten model to describe soil hydraulic characteristics.

[0023] Furthermore, step S2 specifically comprises:

[0024] Finite difference or finite element method is used to solve the soil water movement mathematical model under different irrigation schemes to obtain the numerical solution of soil water movement, so as to obtain the water movement parameters at different time and space positions under different irrigation schemes. The water movement parameters include soil water content, water flow velocity and water flow direction.

[0025] For each irrigation scheme, the flow field visualization analysis method is used to analyze the spatial distribution characteristics of the migration vortex according to the water movement parameters at different time and spatial positions under the irrigation scheme, and the vorticity of the migration vortex is obtained. The vorticity is used to jointly characterize the two characteristic parameters of the migration vortex, namely the intensity and direction.

[0026] Furthermore, in step S4, the system uses a long short-term memory network (LSTM) to establish a quantitative relationship between the migration eddy current characteristic parameters and the sediment migration width.

[0027] Furthermore, step S5 specifically comprises:

[0028] Use the pre-trained pollution remediation area range model to obtain the corresponding pollution remediation area range according to the remediation characteristic factors of the plants planted on the contaminated soil;

[0029] The target range of sediment migration width is determined based on the obtained pollution remediation area.

[0030] Further, in step S5, the pollution remediation area range prediction model is trained and constructed according to the following steps:

[0031] Use machine learning algorithms to establish a pollution remediation area range prediction model;

[0032] The principal component analysis method was used to extract multiple key influencing factors of phytoremediation as restoration characteristic factors;

[0033] Taking the restoration characteristic factor of a plant as input and the pollution restoration area range of the plant as output, a set of training samples is constructed, and multiple sets of such training samples are constructed accordingly;

[0034] The above training samples are used to train the pollution remediation area prediction model, so that it has the ability to derive the corresponding pollution remediation area range according to the restoration characteristic factors of plants.

[0035] The present invention also provides a computer program product, comprising a computer program, which implements the plant irrigation method based on soil sediment migration analysis as described above when executed.

[0036] The present invention provides a plant irrigation method based on soil sediment migration analysis, wherein step S5 determines a target range of migration width of sediments in the soil, step S6 determines an irrigation scheme whose sediment migration width meets the target range of migration width, and step S10 irrigates plants in the contaminated soil according to the above irrigation scheme, so that the sediment migration width in the contaminated soil does not exceed the target range of migration width, thereby preventing pollutants from migrating and spreading to soil areas that cannot be repaired by plants due to irrigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic flow chart of a first embodiment of a plant irrigation method based on soil sediment migration analysis.

[0038] Figure 2 It is a schematic flow chart of a second embodiment of a plant irrigation method based on soil sediment migration analysis. DETAILED DESCRIPTION

[0039] The present invention is further described in detail below in conjunction with specific implementation methods.

[0040] First embodiment

[0041] This embodiment is implemented by programming a computer code and executing it by a processor. Figure 1 The computer program product of the plant irrigation method based on soil sediment migration analysis is shown. The implementation process of the method is described in detail below by taking a computer processor executing the computer program product as an example.

[0042] S1. Establish a mathematical model of soil moisture movement.

[0043] The technician first establishes a soil water movement mathematical model in a computer based on the characteristics of soil water movement, using the Richards equation as the basic control equation and the Van Genuchten soil water retention curve model to describe the soil hydraulic characteristics. The specific process is as follows: First, the Van Genuchten soil water retention curve model is used to describe the soil hydraulic characteristics, wherein the soil hydraulic conductivity adopts an exponential model, the parameter α is 0.02 cm-1, and the parameter n is 1.8; second, the alternating direction implicit difference format ADI is used to discretize the Richards equation, the time step is 1h, and the spatial grid size is 10cm×10cm×10cm; the Van Genuchten soil water retention curve model and the Richards equation are combined as a soil water movement mathematical model. The van Genuchten soil water retention curve model (VG model) and the Richards equation are both commonly used soil hydrological models in the industry. In this embodiment, the two are combined as a soil water movement mathematical model.

[0044] S2. Use the soil moisture movement mathematical model to simulate the water flow movement under different irrigation schemes, and obtain the characteristic parameters of the migration vortex formed under different irrigation schemes, including intensity and direction.

[0045] After the computer has established the mathematical model of soil moisture movement, it can use the mathematical model to simulate the water flow movement under different irrigation schemes and obtain the characteristic parameters of the migration vortex formed under different irrigation schemes, including intensity and direction.

[0046] This embodiment pre-designs the following 12 irrigation schemes:

[0047] Irrigation plan 1: irrigation water rate 60L / h, irrigation time 0.5h.

[0048] Irrigation plan 2: irrigation water rate 80L / h, irrigation time 0.5h.

[0049] Irrigation plan 3: irrigation water rate 100L / h, irrigation time 0.5h.

[0050] Irrigation plan 4: irrigation water rate 120L / h, irrigation time 0.5h.

[0051] Irrigation plan 5: irrigation water rate 60L / h, irrigation time 1h.

[0052] Irrigation plan 6: irrigation water rate 80L / h, irrigation time 1h.

[0053] Irrigation plan 7: irrigation water rate 100L / h, irrigation time 1h.

[0054] Irrigation plan 8: irrigation water rate 120L / h, irrigation time 1h.

[0055] Irrigation plan 9: irrigation water rate 60L / h, irrigation time 1.5h.

[0056] Irrigation plan 10: irrigation water rate 80L / h, irrigation time 1.5h.

[0057] Irrigation plan 11: irrigation water rate 100L / h, irrigation time 1.5h.

[0058] Irrigation plan 12: irrigation water rate 120L / h, irrigation time 1.5h.

[0059] The computer uses the soil water movement mathematical model to simulate the soil water flow movement for each irrigation scheme one by one, and uses the finite difference or finite element method to solve the soil water movement numerical solution of the soil water movement mathematical model under different irrigation schemes, so as to obtain the water movement parameters at different time and space positions under different irrigation schemes. The water movement parameters include soil moisture content, water flow velocity and water flow direction. For each irrigation scheme, the computer uses the streamline tracing method through the ParaView software to generate the water flow movement trajectory according to the simulation results under the irrigation scheme, that is, the water movement parameters. When analyzing the characteristics of water flow rotation movement, the computer uses the least squares method to fit the center position of the rotation movement, uses the maximum likelihood estimation to determine the rotation radius, and calculates the rotation speed through the central difference format, so that the migration vortex movement caused by irrigation can be obtained. The computer further uses the flow field visualization analysis method to analyze the spatial distribution characteristics of the migration vortex. Specifically, the ParaView software is used to perform LIC texture lighting rendering on the migration vortex flow field at a certain moment to obtain the spatial distribution image of the migration vortex at that moment, and then the average vorticity and vortex core of the migration vortex area are extracted from the image. In this embodiment, the intensity and direction of the migration vortex are jointly characterized by the characteristic parameter of vorticity. According to the above process, the computer obtains the characteristic parameter vorticity and vortex core position coordinates of the migration vortex under each irrigation scheme. Since the characteristic parameter vorticity can jointly characterize the intensity and direction, obtaining the vorticity of the migration vortex is equivalent to obtaining the intensity and direction of the migration vortex.

[0060] S3. For each irrigation scheme, according to the characteristic parameters of the migration eddy current, the soil solute migration mathematical model is used to simulate and analyze the soil sediment migration process, and the sediment migration width corresponding to the characteristic parameters of the migration eddy current is obtained.

[0061] For each irrigation scheme, the computer simulates and analyzes the soil sediment migration process based on the migration eddy characteristic parameters and uses the soil solute migration mathematical model to obtain the sediment migration width corresponding to the migration eddy characteristic parameters. There are many common soil solute migration mathematical models, such as the convection-diffusion equation (CDE), the dynamic-immobile water model (MIM), the capillary bundle model of solute migration, the parallel soil column model, the random convection migration model, etc. (reference website https: / / www.docin.com / p-1664416841.html ). This embodiment uses the convection-diffusion equation (CDE) to simulate and analyze the soil sediment migration process. Specifically, the computer uses the convection-diffusion equation (CDE) to simulate and analyze the migration process of soil sediment over time after irrigation is completed according to the irrigation scheme based on the migration vortex characteristic parameters of each irrigation scheme. In this embodiment, the sediment migration width after a preset time, for example, 2 hours after the irrigation is completed is used as the sediment migration width corresponding to the migration vortex under each irrigation scheme. Therefore, after simulating the migration process of soil sediment over time after the irrigation is completed, the computer obtains the sediment migration boundary 2 hours after the irrigation is completed, and then calculates the average distance between the sediment migration boundary at this time and the original boundary of the sediment, and uses the average distance as the soil sediment migration width.

[0062] S4. According to the sediment migration width corresponding to the migration eddy characteristic parameters under the above-mentioned various irrigation schemes, a quantitative relationship between the migration eddy characteristic parameters and the sediment migration width is established.

[0063] According to the sediment migration width corresponding to the migration eddy characteristic parameters under the above-mentioned various irrigation schemes, the computer establishes a quantitative relationship between the migration eddy characteristic parameters and the sediment migration width. For example, a long short-term memory network LSTM is used to establish a nonlinear regression model between the two. The quantitative relationship between the two is characterized by the nonlinear regression model. Specifically: an LSTM network structure including 1 input layer, 2 LSTM hidden layers (10 neurons in each layer), and 1 fully connected output layer is established as a nonlinear regression model; then, according to the sediment migration width corresponding to the migration eddy characteristic parameters under the above-mentioned various irrigation schemes, multiple groups of training samples with sediment migration width as input and migration eddy characteristic parameter vorticity as output are constructed, and the nonlinear regression model is trained using these training samples to obtain a trained nonlinear regression model. The nonlinear regression model can be used to predict the corresponding migration vorticity characteristic parameters according to the sediment migration width.

[0064] S5. Determine the target range of the migration width of sediments in the contaminated soil based on the restoration characteristics of the plants planted in the contaminated soil.

[0065] The computer determines the target range of the migration width of sediments in the contaminated soil based on the restoration characteristics of the plants planted in the contaminated soil. The specific process is as follows:

[0066] The computer extracts the key influencing factors in the plant restoration process through principal component analysis based on the restoration characteristics of different plants. The computer first determines the number of common factors based on the principle that the eigenvalue is greater than 1, and then determines the number of principal components to be extracted to be 3 based on the scree plot and the explained variance ratio. Then the maximum variance rotation method is used to rotate the factor loading matrix. SPSS software is used for factor extraction and rotation. The rotated factor loading matrix shows that the first principal component mainly reflects the nutrient absorption capacity, the second principal component mainly reflects the root distribution, and the third principal component mainly reflects the transpiration water consumption. Therefore, the three key influencing factors of plant root distribution, transpiration water consumption, and nutrient absorption capacity are extracted as restoration characteristic factors. The computer predicts the pollution remediation area range based on the corresponding relationship between the restoration characteristic factors of different plants and their pollution remediation area range. Specifically, the computer pre-uses a machine learning algorithm such as a convolutional neural network to establish a pollution remediation area range prediction model, and then uses the principal component analysis method to extract multiple key influencing factors of plant remediation as restoration characteristic factors, and uses the restoration characteristic factor of a plant as input and the pollution remediation area range of the plant as output to construct a set of training samples, and then constructs multiple sets of such training samples. The pollution remediation area range prediction model is trained using the above training samples, so that it has the ability to derive the corresponding pollution remediation area range based on the restoration characteristic factors of the plants. In this way, the computer can subsequently use the pollution remediation area range prediction model to derive the corresponding pollution remediation area range based on the restoration characteristic factors of the plants. The computer determines the three restoration characteristic factors of the plants planted in the contaminated soil, namely, root distribution, transpiration water consumption, and nutrient absorption capacity, based on the type and planting range of the plants planted in the contaminated soil. Then, the computer uses a pre-trained pollution restoration area range model to obtain the corresponding pollution restoration area range based on the restoration characteristic factors of the plants planted in the contaminated soil. Then, the target range of the sediment migration width is determined based on the obtained pollution restoration area range. In this embodiment, the pollution restoration area range is used as the target boundary range of the sediment migration, and then the target range of the sediment migration width is determined based on the original boundary of the sediment and the target boundary range of the sediment migration.

[0067] S6. According to the quantitative relationship between the migration eddy characteristic parameters and the sediment migration width, the migration eddy characteristic parameters corresponding to the target range of sediment migration width are obtained, and the corresponding irrigation plan is determined accordingly.

[0068] The computer obtains the migration vortex characteristic parameters corresponding to the target range of the migration width of the sediment according to the quantitative relationship between the migration vortex characteristic parameters and the sediment migration width, and determines the corresponding irrigation scheme accordingly. As described above, the present embodiment uses a nonlinear regression model to represent the quantitative relationship between the migration vortex characteristic parameters and the sediment migration width. Therefore, the computer specifically uses the nonlinear regression model to input the middle value of the target range of the sediment migration width (i.e., the average value of the two end point values ​​of the target range) into the nonlinear regression model. The nonlinear regression model obtains the characteristic parameter vorticity of the corresponding migration vortex according to the input middle value of the sediment migration width as the migration vortex characteristic parameter corresponding to the target range of the sediment migration width. Then, according to the corresponding relationship between each irrigation scheme and its migration vortex characteristic parameter, the irrigation scheme corresponding to the characteristic parameter vorticity of the migration vortex corresponding to the target range of the sediment migration width is reversed. For example, preferably, an existing AI general large model is used to learn the correspondence between irrigation plans and migration vortex characteristic parameters, so that it has the ability to output corresponding irrigation plans based on the migration vortex characteristic parameters. In this way, the computer can input the characteristic parameter vorticity of the migration vortex corresponding to the target range of sediment migration width into the learned AI general large model, and the AI ​​general large model outputs the corresponding irrigation plan.

[0069] S10. Irrigating the plants in the contaminated soil according to the determined irrigation plan.

[0070] The computer controls the irrigation system to irrigate the plants in the contaminated soil according to the determined irrigation plan, so that the sediment migration width in the contaminated soil does not exceed the migration width target range, and can prevent pollutants from migrating and spreading to soil areas that cannot be repaired by plants due to irrigation.

[0071] Second embodiment

[0072] This embodiment is implemented by programming a computer code and executing it by a processor. Figure 2 The computer program product of the plant irrigation method based on soil sediment migration analysis is shown. The implementation process of the method is described in detail below by taking a computer processor executing the computer program product as an example.

[0073] This embodiment is substantially the same as the first embodiment. The following only describes the differences between this embodiment and the first embodiment, and the similarities are not repeated here. Please refer to the first embodiment.

[0074] The first embodiment is to irrigate according to the irrigation scheme determined by the characteristic parameter vorticity of the migration vortex corresponding to the target range of the sediment migration width. This embodiment takes into account that the characteristic parameters of the migration vortex corresponding to the irrigation scheme obtained by simulation may have errors, which may cause the sediment migration width to be offset after irrigation according to the irrigation scheme determined by the characteristic parameter vorticity of the migration vortex corresponding to the target range of the sediment migration width, that is, it may not necessarily be within the target range of the sediment migration width. For this reason, this embodiment does not execute step S10 after step S6 of the first embodiment, but first executes steps S7 to S9. Specifically as follows:

[0075] S7. Irrigate the plants in the contaminated soil according to the irrigation scheme determined in step S6.

[0076] In this embodiment, the plants in the contaminated soil are first irrigated according to the determined irrigation scheme.

[0077] S8. After a preset time has passed since the irrigation was completed, the sediment migration situation is detected, and the sediment migration width is calculated based on the sediment migration situation to determine whether the sediment migration width is within the above-mentioned migration width target range.

[0078] After the irrigation is completed and a preset time has passed, that is, 2 hours have passed, the sediment migration is detected. In this embodiment, multiple circles of sediment sensors surrounding the contaminated soil are arranged around the contaminated soil to detect the sediment concentration at the location of the sensor. The computer obtains the sediment migration according to the sediment concentration detected by each circle of sediment sensors. The sediment sensor is a specific type of pollutant detection sensor. The corresponding sensor is selected according to the main pollutants in the contaminated soil. For example, if the contaminated soil is mainly heavy metal polluted, a heavy metal detection sensor is selected as the sediment sensor. If the contaminated soil is mainly saline-alkali polluted, a salt + PH sensor is selected. Each circle of sediment sensors is distributed from the edge of the contaminated soil to the outside at intervals, and each circle of sediment sensors includes multiple sediment sensors distributed at intervals in the circumferential direction. The periphery of the contaminated soil belongs to non-contaminated soil. Under normal circumstances, the corresponding pollutants cannot be detected in the peripheral soil or the detected pollutant content is within the normal range. If the pollutants in the contaminated soil migrate and diffuse due to irrigation, they will diffuse into the peripheral non-contaminated soil, and the content of the pollutants diffused into the peripheral non-contaminated soil will change. If a sediment sensor detects a change in the sediment concentration at its location, it means that the sediment has diffused to the location of the sediment sensor. Therefore, after the irrigation is completed and a preset time has passed, that is, 2 hours have passed, the computer obtains the sediment migration situation based on the sediment concentration detected by each circle of sediment sensors outside the contaminated soil. Specifically, the sediment diffusion point is determined based on the sediment concentration detected by each circle of sediment sensors, and then each diffusion point is connected to form a sediment migration boundary, which is used as the sediment migration situation. The computer calculates the sediment migration width based on the sediment migration situation, that is, the sediment migration boundary, specifically, the average distance between the sediment migration boundary and the original sediment boundary (with the contaminated soil boundary as the original sediment boundary) is calculated, and the average distance is used as the sediment migration width. The computer then determines whether the sediment migration width is within the above-mentioned migration width target range.

[0079] S9. If the sediment migration width is within the above-mentioned migration width target range, the current irrigation plan will be used as the optimal irrigation plan; if the sediment migration width exceeds the above-mentioned migration width target range, the irrigation plan will be automatically adjusted according to the degree of deviation from the target range until the sediment migration width is within the migration width target range, and the adjusted irrigation plan will be used as the optimal irrigation plan.

[0080] If the sediment migration width is within the above migration width target range, the current irrigation plan is used as the optimal irrigation plan; if the sediment migration width exceeds the above migration width target range, the irrigation plan is automatically adjusted according to the degree of deviation from the target range, and then steps S7 and S8 are repeated until the sediment migration width is within the migration width target range, and the adjusted irrigation plan is used as the optimal irrigation plan. This embodiment controls the irrigation system to irrigate plants in contaminated soil according to the optimal irrigation plan.

[0081] Step S10: irrigating the plants in the contaminated soil according to the optimal irrigation plan.

[0082] The computer controls the irrigation system to irrigate the plants in the contaminated soil according to the determined optimal irrigation plan, so that the sediment migration width in the contaminated soil does not exceed the migration width target range, and can prevent pollutants from migrating and spreading to soil areas that cannot be repaired by plants due to irrigation.

[0083] This embodiment takes into account that the water required by plants at different growth stages is different. Therefore, the optimal irrigation schemes at different growth stages are different. The irrigation water volume of the optimal irrigation scheme determined above may exceed the irrigation water volume of the optimal irrigation scheme at certain growth stages of the plant. Therefore, the computer uses the values ​​of the various irrigation parameters of the optimal irrigation scheme as the highest threshold. Under the premise of not exceeding the highest threshold of the irrigation parameters, the computer determines whether the optimal irrigation scheme needs to be adjusted according to the current plant growth stage. If necessary, at least one irrigation parameter of the optimal irrigation scheme is automatically adjusted, and the plants are irrigated according to the adjusted irrigation scheme. Otherwise, the irrigation is carried out according to the optimal irrigation scheme, wherein the irrigation parameters include the irrigation water speed and the irrigation duration. The irrigation water volume is determined by both the irrigation water speed and the irrigation duration. The irrigation water volume can be adjusted by adjusting the irrigation water speed and / or the irrigation duration of the optimal irrigation scheme.

[0084] The above is only an implementation method of the invention, and does not limit the scope of patent protection. Those skilled in the art can make non-substantial changes or substitutions based on the invention, which still fall within the scope of patent protection.

Claims

1. A plant irrigation method based on soil sediment migration analysis, characterized in that: The following steps are involved: S1. Establish a mathematical model of soil water movement; S2. Use a soil water movement mathematical model to simulate water flow movement under different irrigation schemes, and obtain characteristic parameters of migration vortices formed under different irrigation schemes, wherein the characteristic parameters include intensity and direction; S3. For each irrigation scheme, according to its migration eddy characteristic parameters, a soil solute migration mathematical model is used to simulate and analyze the soil sediment migration process, and the sediment migration width corresponding to the migration eddy characteristic parameters is obtained; S4. establishing a quantitative relationship between the migration eddy characteristic parameters and the sediment migration width according to the sediment migration width corresponding to the migration eddy characteristic parameters under the various irrigation schemes; S5. Determine the target range of the migration width of sediments in the contaminated soil based on the restoration characteristics of the plants planted in the contaminated soil; S6. According to the quantitative relationship between the migration eddy characteristic parameters and the sediment migration width, the migration eddy characteristic parameters corresponding to the target range of sediment migration width are obtained, and the corresponding irrigation plan is determined accordingly; S10. Irrigating the plants in the contaminated soil according to the determined irrigation plan.

2. The plant irrigation method based on soil sediment migration analysis as claimed in claim 1, characterized in that: The method includes the following steps performed between steps S6 and S10: S7, irrigating the plants in the contaminated soil according to the irrigation scheme determined in step S6; S8. After a preset time has passed since the irrigation is completed, the sediment migration is detected, and the sediment migration width is calculated based on the sediment migration, and it is determined whether the sediment migration width is within the above-mentioned migration width target range; S9. If the sediment migration width is within the above migration width target range, the current irrigation plan is used as the optimal irrigation plan; If the sediment migration width exceeds the above migration width target range, the irrigation plan is automatically adjusted according to the degree of deviation from the target range until the sediment migration width is within the migration width target range, and the adjusted irrigation plan is used as the optimal irrigation plan; Step S10 specifically involves irrigating the plants in the contaminated soil according to the optimal irrigation plan.

3. The plant irrigation method based on soil sediment migration analysis as claimed in claim 2, characterized in that: Step S7 specifically: A plurality of circles of sediment sensors are arranged around the contaminated soil to detect the sediment concentration at the location of the sensors, wherein the circles of sediment sensors are spaced apart from the edge of the contaminated soil toward the outside, and each circle of sediment sensors includes a plurality of sediment sensors spaced apart in the circumferential direction; The sediment migration situation is obtained based on the sediment concentration detected by the sediment sensors in each circle.

4. The plant irrigation method based on soil sediment migration analysis as claimed in claim 2, characterized in that: Step S10 specifically: Taking the values ​​of various irrigation parameters of the optimal irrigation plan as the highest threshold, under the premise of not exceeding the highest threshold of the irrigation parameters, determine whether the optimal irrigation plan needs to be adjusted according to the current plant growth stage. If necessary, automatically adjust at least one irrigation parameter of the optimal irrigation plan and irrigate the plants according to the adjusted irrigation plan. Otherwise, irrigate according to the optimal irrigation plan, where the irrigation parameters include irrigation water speed and irrigation duration.

5. The plant irrigation method based on soil sediment migration analysis according to claim 1 or 2, characterized in that: Specifically, step S1 establishes a soil moisture movement mathematical model based on the characteristics of soil moisture movement, using the Richards equation as the basic control equation and the Van Genuchten model to describe the soil hydraulic characteristics.

6. The plant irrigation method based on soil sediment migration analysis as claimed in claim 5, characterized in that: Step S2 specifically: Finite difference or finite element method is used to solve the soil water movement mathematical model under different irrigation schemes to obtain the numerical solution of soil water movement, so as to obtain the water movement parameters at different time and space positions under different irrigation schemes. The water movement parameters include soil water content, water flow velocity and water flow direction. For each irrigation scheme, the flow field visualization analysis method is used to analyze the spatial distribution characteristics of the migration vortex according to the water movement parameters at different time and spatial positions under the irrigation scheme, and the vorticity of the migration vortex is obtained. The vorticity is used to jointly characterize the two characteristic parameters of the migration vortex, namely the intensity and direction.

7. The plant irrigation method based on soil sediment migration analysis according to claim 1 or 2, characterized in that: In step S4, the system uses a long short-term memory network (LSTM) to establish a quantitative relationship between the characteristic parameters of the migration eddy current and the width of the sediment migration.

8. The plant irrigation method based on soil sediment migration analysis according to claim 1 or 2, characterized in that: Step S5 specifically: Use the pre-trained pollution remediation area range model to obtain the corresponding pollution remediation area range according to the remediation characteristic factors of the plants planted on the contaminated soil; The target range of sediment migration width is determined based on the obtained pollution remediation area range.

9. The plant irrigation method based on soil sediment migration analysis as claimed in claim 8, characterized in that: In step S5, the pollution remediation area range prediction model is trained and constructed according to the following steps: Use machine learning algorithms to establish a pollution remediation area range prediction model; The principal component analysis method was used to extract multiple key influencing factors of phytoremediation as restoration characteristic factors; Taking the restoration characteristic factor of a plant as input and the pollution restoration area range of the plant as output, a set of training samples is constructed, and multiple sets of such training samples are constructed accordingly; The above training samples are used to train the pollution remediation area prediction model, so that it has the ability to derive the corresponding pollution remediation area range according to the restoration characteristic factors of plants.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed, the plant irrigation method based on soil sediment migration analysis as described in any one of claims 1 to 9 is implemented.

Images