Camellia oleifera forest planting ecological digital monitoring management system based on Internet of Things

Abstract

The invention discloses a camellia oleifera forest planting ecology digital monitoring management system based on the Internet of Things, and relates to the technical field of planting ecology supervision. Dynamic ecological grid data is combined with an Internet of Things sensor and unmanned aerial vehicle patrol data, a blank area is filled through a spatial interpolation technology, and a real-time camellia oleifera forest ecological monitoring system is formed; dynamically updating a disease and pest risk distribution map through a disease and pest diffusion model and unmanned aerial vehicle patrol data, and identifying a high-risk area in combination with diffusion path simulation; the unmanned aerial vehicle is linked to perform quantitative spraying of biopesticide, prevention and control operation is performed only in a high-risk area, the pesticide usage amount is reduced, and soil microbial communities and an ecological system are protected; through nutrient diffusion and permeation simulation, the influence of a fertilization scheme on soil nutrient loss is evaluated, the fertilization amount and fertilization time are optimized, nutrient waste and water pollution risks are reduced, the irrigation amount and strategy are dynamically adjusted, the water resource utilization efficiency is optimized, and excessive irrigation or drought pressure is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of planting ecological supervision, and particularly to a digital monitoring and management system for the planting ecology of oil-tea camellia forests based on the Internet of Things. Background Art

[0002] Currently, the supervision methods for oil-tea camellia forest planting mainly stay in the basic monitoring stage, relying on sensors such as soil humidity and temperature to collect environmental parameters, and it is difficult to meet the comprehensive requirements of ecological protection and efficient planting.

[0003] In terms of pest control, traditional methods mainly include relying on manual experience for pest identification and control, or using chemical control means. However, the method relying on manual experience has a large lag and poor control effect; while chemical control is prone to overuse of pesticides, which is easy to damage the soil microbial community and weaken the self-regulation ability of the ecosystem.

[0004] In terms of ecological management, traditional supervision methods are mostly limited to the monitoring of single indicators, such as soil moisture content, nitrogen and phosphorus concentration, etc., and it is difficult to comprehensively reflect the dynamic changes of the oil-tea camellia forest ecosystem, and the data utilization rate is low; at the same time, most of these monitoring data are only simply recorded and lack in-depth analysis. Even if a large amount of data is monitored, it is difficult to predict ecological risks and adjust planting strategies in a timely manner. In the face of emergencies such as extreme weather, traditional methods are difficult to provide effective support.

[0005] To solve the deficiencies in pest control and ecological management, some traditional solutions introduce automated equipment, such as drones for automated data collection. However, these solutions still require manual image interpretation or manual adjustment of prevention and control strategies, resulting in insufficient coordination and automation level of the entire supervision system. Therefore, there is an urgent need for a digital monitoring and management system for the planting ecology of oil-tea camellia forests based on the Internet of Things to solve such problems. Summary of the Invention

[0006] In view of the existing problems above, the present invention is proposed.

[0007] The present invention provides a digital monitoring and management system for the planting ecology of oil-tea camellia forests based on the Internet of Things to solve the problems of lack of joint analysis of monitoring data and insufficient intelligence of prevention and control means in traditional solutions.

[0008] To solve the above technical problems, the present invention provides the following technical solutions:

[0009] The present invention provides a digital monitoring and management system for the planting ecology of oil-tea camellia forests based on the Internet of Things, which includes

[0010] An oil-tea camellia forest monitoring module, which monitors the ecological environment of the oil-tea camellia forest, collects real-time data and monitors meteorological conditions;

[0011] The real-time data monitoring methods include the real-time monitoring of the Internet of Things (IoT) sensor network and the regular inspection monitoring of drones.

[0012] The IoT sensor network is deployed in the oil-tea camellia forest area to collect environmental parameters, including soil humidity, nutrient concentration, air temperature and humidity, and light intensity data.

[0013] The drone uses multi-spectral imaging technology for regular inspections to collect pest and disease images and vegetation conditions.

[0014] The digital twin platform, based on dynamic modeling technology, creates a digital twin model of the oil-tea camellia forest, displaying soil nutrient distribution, pest and disease risk areas, irrigation suggestions, and predicted crop growth data.

[0015] The intelligent prevention and control module is used to link drones to predict pests and diseases, dynamically update the pest and disease risk distribution map based on the pest and disease risk areas combined with the drone inspection data, and link drones to spray biological pesticides, and carry out prevention and control operations only in high-risk areas.

[0016] The ecological management module, based on the twin model, provides ecological management suggestions, which include fertilization suggestions and irrigation suggestions.

[0017] Furthermore, the management steps of the ecological digital monitoring and management system for tea forest planting include:

[0018] Step S1, collection of ecological data of the oil-tea camellia forest

[0019] Integrate the data collected by sensors and drones into the dynamic ecological data of the oil-tea camellia forest area, including environmental parameters, pest and disease images, and vegetation conditions.

[0020] Step S2, model the dynamics of the digital twin model and simulate the ecological process of oil-tea camellia forest planting

[0021] Use dynamic modeling technology to create a digital twin model of the oil-tea camellia forest and simulate and predict the ecological process of oil-tea camellia forest planting.

[0022] Step S3, intelligent prevention and control

[0023] Based on the simulation results provided by the digital twin model and the real-time monitoring data, link the intelligent prevention and control and ecological management modules for real-time prevention and control.

[0024] Furthermore, in step S1, the integration method of the ecological data in the oil-tea camellia forest area is as follows:

[0025] Let the environmental parameters collected by the sensor in real time include soil humidity W s , nutrient concentration N t , air temperature and humidity T a ,H aand the light intensity L, filtering the noise data during the acquisition process:

[0026] Among them, P i (t) is the weighted smoothed value of sensor i at time t, S j (t) is the original sampled value of sensor j at time t, w j is the weight factor, used to adjust the influence of different sensors, allocated according to the historical accuracy of the sensors, k is the number of sensors deployed in the area, the weighted smoothing eliminates outliers, and the output set of credible values of the environmental parameters is {W s , N t , T a , H a , L};

[0027] Suppose the UAV inspection image includes the vegetation coverage V c and the pest distribution image B d (x, y), and the pest distribution is calculated using a multispectral data classification model: Among them, B d (x, y) is the pest distribution value at the position (x, y), R(x, y), G(x, y), NIR(x, y) are the reflection values of the red, green, and near-infrared bands respectively, α, β, γ are spectral coefficients, based on multispectral model correction, σ(·) is the normalization function, used to map the distribution value to the standard range [0, 1];

[0028] Integrate the point information of the sensor data and the spatial information of the UAV image into a unified grid G(x, y), and calculate the dynamic ecological parameters of each grid, Among them, D(x, y) is the dynamic ecological parameter on the grid (x, y), P i (t) is the weighted value of sensor i at time t, B d (x, y) is the pest distribution matrix value, λ i is the weight factor, used to adjust the contribution of the sensor to the grid parameters, η is the weight of the UAV data, used to balance the influence of the sensor and the image data, and m is the number of sensors within the grid range;

[0029] Construct the time series grid data into a dynamic model:

[0030] E(x, y, t) = φ · E(x, y, t - 1) + (1 - φ) · D(x, y, t), where E(x, y, t) is the comprehensive ecological parameter of the grid (x, y) at time t, E(x, y, t - 1) is the comprehensive ecological parameter of the previous moment, D(x, y, t) is the dynamic ecological parameter of the current moment, and φ is the time decay coefficient, which controls the weight allocation between historical data and current data.

[0031] Furthermore, in step S2, the simulation and prediction process includes:

[0032] Based on the dynamic ecological data in step S1, a virtual oil-tea camellia forest model is created using dynamic modeling techniques;

[0033] Simulate the ecological process in the digital twin platform, including:

[0034] Soil nutrient distribution simulation: Analyze the impact of different fertilization schemes on soil nutrient loss and surrounding water body pollution;

[0035] Pest and disease spread simulation: Combine pest and disease images and meteorological conditions (wind speed, humidity, temperature) to predict the spread path and high-risk areas of pests and diseases;

[0036] Irrigation mode simulation: Evaluate the impact of different irrigation schemes on water resource utilization efficiency and soil moisture content;

[0037] Output the simulation results of the ecological process, including soil nutrient distribution, pest and disease risk areas, irrigation suggestions, and predicted crop growth data.

[0038] Furthermore, in step S2, the method for creating the virtual oil-tea camellia forest model is:

[0039] Based on the dynamic ecological grid data E(x, y, t) in step S1, use spatial interpolation method to fill the uncovered area, where, is the ecological parameter of the grid point (x, y) at time t after interpolation, E(x i , y j , t) is the ecological parameter of the sampling point (x i , y j ) closest to (x, y), is the weight factor, based on the reciprocal of the Euclidean distance between grid points, p is the attenuation coefficient, Ω is the neighborhood set containing k sampling points around (x, y), expand the discrete sampling point information into a continuous spatial distribution, and generate complete ecological data

[0040] Based on the spatial ecological data Simulate the growth dynamics of the oil-tea camellia forest:

[0041] G(x, y, t + 1) = G(x, y, t) + f(N t (x, y, t), W s (x, y, t), L(x, y, t)) - δ(G(x, y, t)), where G(x, y, t + 1) is the vegetation growth amount of the grid point (x, y) at time t + 1, G(x, y, t) is the vegetation growth amount of the grid point (x, y) at time t, Nt (x, y, t) represents the soil nutrient concentration, W s (x, y, t) represents the soil moisture, L(x, y, t) represents the light intensity, and f(·) is the growth function, defined as f = α1·N t +α2·W s +α3·L, where α1, α2, and α3 are growth influence coefficients, and δ(G(x, y, t)) is the natural attenuation function, defined as δ = β·G(x, y, t), and β is the attenuation coefficient;

[0042] Combined with the pest and disease distribution data B d (x, y, t) and the meteorological conditions wind speed V w 、humidity H a and temperature T a , a pest and disease diffusion model is established:

[0043] where B d (x, y, t + 1) is the pest and disease distribution value at the grid point (x, y) at time t + 1, is the diffusion term, k is the diffusion coefficient, and g(V w , H a , T a ) is the meteorological driving term, defined as g = γ1·V w +γ2·H a +γ3·T a , where γ1, γ2, and γ3 are driving influence coefficients, and by coupling diffusion and meteorological driving, the spatial diffusion trend of pests and diseases is predicted;

[0044] Integrate vegetation growth and pest and disease diffusion to build a comprehensive ecological model:

[0045] M(x, y, t) = G(x, y, t) - ω·B d (x, y, t), where M(x, y, t) is the comprehensive ecological parameter, representing the ecological quality of the grid (x, y) at time t, G(x, y, t) is the vegetation growth amount, and B d (x, y, t) is the pest and disease distribution value, and ω is the pest and disease influence weight coefficient.

[0046] Furthermore, in step S2, the ecological process is simulated in the digital twin platform:

[0047] Simulate nutrient diffusion and infiltration. The spatio-temporal variation of the nutrient concentration in the soil is described by the convection-diffusion equation: where N(x, y, t) is the nutrient concentration of the (x, y) grid in the soil, D n is the nutrient diffusion coefficient, which depends on the soil type, v is the water flow velocity vector, representing the nutrient convection direction and intensity, and S f(x, y, t) is the fertilizer input amount, L w (x, y, t) is the nutrient loss amount;

[0048] After fertilization, part of the nutrients are lost and enter the water body, and the change in its concentration is:

[0049] Among them, C w (t) is the nutrient concentration in the water body, V w is the water body volume, Ω is the soil area, L w (x, y, t) is the nutrient loss amount in the soil,

[0050] Simulate the spread of pests and diseases. The distribution of pests and diseases changes with time and space as:

[0051] Among them, B(x, y, t) is the pest and disease density, D b is the diffusion coefficient, indicating the pest and disease transmission rate, G b (B, T, H) is the growth function, depending on the current density B, temperature T, and humidity H, G b = α b ·B·(1 - B / K), where α b is the growth rate, K is the environmental carrying capacity, M b (V w , T) is the meteorological-driven mortality rate of pests and diseases, restricted by wind speed V w and high temperature T;

[0052] Combined with the wind speed field and humidity field, calculate the pest and disease diffusion path:

[0053] Among them, is the speed field of pest and disease diffusion, is the wind speed field vector, is the pest and disease density gradient;

[0054] Simulate the irrigation mode. The soil water content distribution after irrigation is described by the Richards equation:

[0055] Among them, W(x, y, t) is the soil water content, K s (W) is the soil hydraulic conductivity, K s (W)=K sat ·exp(-β·(W max -W)), where K sat is the saturated hydraulic conductivity, β is the soil characteristic coefficient, R(x, y, t) is the evapotranspiration rate;

[0056] Evaluate the water resource utilization efficiency of the irrigation scheme, Among them, Ei is the irrigation efficiency of Scheme i, V i is the volume of irrigation water.

[0057] Furthermore, intelligent pest control is carried out in step S3:

[0058] Based on the pest diffusion simulation results in step S2, combined with the UAV inspection data, the pest risk distribution map is dynamically updated, and the UAV is linked to spray biological pesticides, and the prevention and control operations are only carried out in high-risk areas.

[0059] Furthermore, an ecological management strategy is implemented in step S3:

[0060] According to the soil nutrient distribution data simulated in S2, adjust the fertilization amount and time;

[0061] Based on the irrigation mode simulation data, carry out irrigation in combination with the real-time data of soil humidity and water resource utilization rate.

[0062] Furthermore, in step S3, intelligent pest control is carried out:

[0063] The pest diffusion prediction result B in step S2 m (x, y, t) and the real-time distribution data B obtained from UAV inspection u (x, y, t) are fused to dynamically update the pest risk distribution:

[0064] Among them, B r (x, y, t) is the updated pest risk distribution map, B m (x, y, t) is the risk distribution obtained from pest diffusion simulation, B u (x, y, t) is the real-time distribution data analyzed from UAV inspection images, w m is the weight of the simulation result, dynamically adjusted according to historical prediction accuracy, w u is the weight of UAV data, adjusted based on the resolution and timeliness of inspection images;

[0065] Based on the updated distribution map B r (x, y, t), set the risk threshold τ, and the area above the threshold is defined as the high-risk area:

[0066] If B r (x, y, t) ≥ τ, then R(x, y, t) = 1,

[0067] If B r (x, y, t) < τ, then R(x, y, t) = 0, where R(x, y, t) is the high-risk area identifier, 1 represents high risk, 0 represents non-high risk, and τ is the pest risk threshold, determined by historical monitoring and crop economic loss analysis;

[0068] According to the spatial distribution of the high-risk area R(x, y, t), calculate the spraying path and dosage of the drone. The spraying path optimization formula is:

[0069] Among them, L spray is the total length of the drone spraying path, d i is the distance between the centers of two adjacent high-risk areas in the path, (x i , y i ) are the coordinates of the center point of the high-risk area through which the path passes. The constraint condition is that the path must cover all high-risk areas;

[0070] The spraying dosage calculation formula is Q(x, y) = λ·B r (x, y, t), where Q(x, y) is the dosage sprayed on the (x, y) grid, λ is the dosage demand coefficient per unit pest density, determined based on pesticide characteristics and target control efficiency, and B r (x, y, t) is the risk distribution value, dynamically adjusting the spraying intensity;

[0071] The drone operates according to the optimized path L spray and the spraying dosage Q(x, y), and after spraying, feeds back the spraying coverage map C(x, y, t):

[0072] If Q(x, y)>0, then C(x, y, t) = 1,

[0073] If Q(x, y) = 0, then C(x, y, t) = 0, where C(x, y, t) is the spraying coverage status, 1 indicates covered, 0 indicates not covered, and Q(x, y) is the spraying dosage, determining the coverage status.

[0074] Furthermore, the method for adjusting the fertilization amount and time in step S3 is as follows:

[0075] According to the soil nutrient distribution data N(x, y, t) and the crop demand model in step S2, determine the fertilization amount F(x, y) and fertilization time T for each grid f ,

[0076] Fertilization amount: F(x, y) = max(0, D n (x, y) - N(x, y, t))·A(x, y), where F(x, y) is the fertilization amount of the grid (x, y), D n (x, y) is the target nutrient concentration of the grid (x, y), N(x, y, t) is the current soil nutrient concentration of the grid (x, y), and A(x, y) is the grid area

[0077] Fertilization time: Among them, Tf is the fertilization time point, is the nutrient change rate of the grid (x, y), R n is the nutrient absorption rate of the crop, and Ω is the set of grids in the fertilization area;

[0078] Based on the irrigation mode simulation data W m (x, y, t) in step S2, combined with the real-time soil humidity data W r (x, y, t) and the water resource utilization rate, calculate the irrigation amount Q of each grid i (x, t) and the irrigation efficiency,

[0079] Irrigation amount: Q i (x, y) = max(0, W m (x, y, t) - W r (x, y, t)) · A(x, y), where Q i (x, y) is the irrigation amount of the grid (x, y), W m (x, y, t) is the target soil humidity simulated by the irrigation mode, W r (x, y, t) is the real-time monitored soil humidity, and A(x, y) is the grid area;

[0080] Water resource utilization rate: where E w is the water resource utilization efficiency of the irrigation scheme, V w is the total irrigation water volume, and η is the efficiency coefficient of the irrigation facility;

[0081] Path planning of drones or automatic irrigation equipment:

[0082] where L irrigation is the total length of the irrigation path, d i is the distance between adjacent irrigation points in the path, (x i , y i ) are the coordinates of the irrigation points, satisfying the irrigation amount Q i (x, y) > 0.

[0083] The beneficial effects of the present invention are:

[0084] In the present invention, dynamic ecological grid data is combined with Internet of Things sensor and drone inspection data, and the blank areas are filled by spatial interpolation technology to achieve comprehensive coverage of monitoring data, forming a real-time ecological monitoring system for oil-tea forests. The pest risk distribution map is dynamically updated through the pest diffusion model and drone inspection data, and combined with the simulation of the diffusion path, the high-risk areas are accurately identified; the drones are linked to implement quantitative spraying of biological pesticides, and the prevention and control operations are only carried out in the high-risk areas, reducing the amount of pesticide used and protecting the soil microbial community and ecosystem.

[0085] In the present invention, through the simulation of nutrient diffusion and penetration, the impact of fertilization schemes on soil nutrient loss is evaluated, and the fertilization amount and time are optimized to reduce nutrient waste and the risk of water pollution. Based on irrigation mode simulation and real-time soil moisture data, the irrigation amount and strategy are dynamically adjusted to optimize water resource utilization efficiency and avoid over-irrigation or drought stress.

[0086] In the present invention, combined with the multi-dimensional ecological simulation of the virtual oil tea forest model, it provides early prediction of nutrient loss, pest and disease spread, and water resource changes, providing early warning and decision-making support for growers. In the face of sudden situations such as wind speed and rainfall, the fertilization, irrigation, and pest and disease prevention and control strategies are quickly adjusted to reduce losses; the digital twin platform integrates multi-dimensional data, provides quantitative prediction results and ecological management suggestions through vegetation growth simulation and pest and disease spread path calculation, supports intelligent prevention and control and ecological optimization decision-making, realizes the full-process intelligent linkage from monitoring to management, and improves planting efficiency and yield quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0088] Figure 1 It is a schematic structural diagram of the ecological digital monitoring and management system for oil tea forest planting based on the Internet of Things of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0089] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings of the specification.

[0090] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention, but the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0091] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0092] Example 1, refer to Figure 1, this embodiment provides an ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things, including:

[0093] An oil-tea camellia forest monitoring module that monitors the ecological environment of the oil-tea camellia forest, collects real-time data, and monitors meteorological conditions;

[0094] The real-time data monitoring methods include real-time monitoring by the Internet of Things sensor network and regular inspection monitoring by drones.

[0095] The Internet of Things sensor network is deployed in the oil-tea camellia forest area to collect environmental parameters, including soil humidity, nutrient concentration, air temperature and humidity, and light intensity data.

[0096] The drone uses multi-spectral imaging technology to conduct regular inspections and collect pest images and vegetation conditions;

[0097] A digital twin platform that creates a digital twin model of the oil-tea camellia forest based on dynamic modeling technology, and displays soil nutrient distribution, pest risk areas, irrigation suggestions, and predicted crop growth data;

[0098] An intelligent prevention and control module that is used to link drones to predict pests and diseases, dynamically update the pest and disease risk distribution map based on the pest and disease risk areas combined with drone inspection data, and link drones to spray biological pesticides, and only perform prevention and control operations in high-risk areas;

[0099] An ecological management module that provides ecological management suggestions based on the twin model. The ecological management suggestions include fertilization suggestions and irrigation suggestions;

[0100] The management methods of the ecological digital monitoring and management system for oil-tea camellia forest planting include:

[0101] Step S1, collecting ecological data of the oil-tea camellia forest,

[0102] Integrate the data collected by sensors and drones into the dynamic ecological data of the oil-tea camellia forest area, including environmental parameters, pest images, and vegetation conditions;

[0103] In step S1, the integration method of the ecological data in the oil-tea camellia forest area is as follows:

[0104] Suppose the environmental parameters collected by the sensor in real time include soil humidity W s , nutrient concentration N t , air temperature and humidity T a ,H a and light intensity L. During the collection process, noise data is filtered:

[0105] Among them, P i (t) is the weighted smoothed value of sensor i at time t, S j (t) is the original sampling value of sensor j at time t, wj is a weight factor used to adjust the influence of different sensors and is assigned according to the historical accuracy of the sensors. k is the number of sensors deployed in the area. Weighted smoothing eliminates outliers, and the output set of credible values of environmental parameters is {W s , N t , T a , H a , L};

[0106] Suppose the UAV inspection image includes the vegetation coverage V c and the pest distribution image B d (x, y). The pest distribution is calculated using a multispectral data classification model: where B d (x, y) is the pest distribution value at the position (x, y), R(x, y), G(x, y), and NIR(x, y) are the reflection values of the red, green, and near-infrared bands respectively, α, β, and γ are spectral coefficients, based on multispectral model calibration, and σ(·) is a normalization function used to map the distribution value to the standard range [0, 1];

[0107] Integrate the point information of the sensor data and the spatial information of the UAV image into a unified grid G(x, y), and calculate the dynamic ecological parameters of each grid. where D(x, y) is the dynamic ecological parameter on the grid (x, y), P i (t) is the weighted value of sensor i at time t, B d (x, y) is the pest distribution matrix value, λ i is a weight factor used to adjust the contribution of the sensor to the grid parameters, η is the weight of the UAV data used to balance the influence of the sensor and the image data, and m is the number of sensors within the grid range;

[0108] Construct a dynamic model from the time series grid data:

[0109] E(x, y, t) = φ·E(x, y, t - 1)+(1 - φ)·D(x, y, t), where E(x, y, t) is the comprehensive ecological parameter of the grid (x, y) at time t, E(x, y, t - 1) is the comprehensive ecological parameter at the previous time, D(x, y, t) is the dynamic ecological parameter at the current time, and φ is the time decay coefficient that controls the weight allocation between historical data and current data;

[0110] Specifically, obtain the dynamic ecological data set {E(x, y, t)} based on step S1 for twin modeling and formulating ecological management strategies.

[0111] Step S2, model the dynamics of the digital twin model and simulate the ecological process of oil tea forest planting.

[0112] Using dynamic modeling technology to create a digital twin model of the oil-tea camellia forest, and simulating and predicting the ecological process of oil-tea camellia forest planting;

[0113] In step S2, the simulation and prediction process includes:

[0114] Based on the dynamic ecological data in step S1, using dynamic modeling technology to create a virtual oil-tea camellia forest model;

[0115] Simulating the ecological process in the digital twin platform, including:

[0116] Soil nutrient distribution simulation: Analyzing the impact of different fertilization schemes on soil nutrient loss and surrounding water body pollution;

[0117] Pest and disease spread simulation: Combining pest and disease images and meteorological conditions (wind speed, humidity, temperature) to predict the spread path and high-risk areas of pests and diseases;

[0118] Irrigation mode simulation: Evaluating the impact of different irrigation schemes on water resource utilization efficiency and soil moisture content;

[0119] Outputting the simulation results of the ecological process, including soil nutrient distribution, pest and disease risk areas, irrigation suggestions, and predicted crop growth data;

[0120] In step S2, the method for creating the virtual oil-tea camellia forest model is:

[0121] Based on the dynamic ecological grid data E(x, y, t) in step S1, using spatial interpolation method to fill the uncovered area, where, is the ecological parameter of the grid point (x, y) at time t after interpolation, E(x i , y j , t) is the ecological parameter of the sampling point (x i , y j ) closest to (x, y), is the weight factor, based on the reciprocal of the Euclidean distance between grid points, p is the attenuation coefficient, Ω is the neighborhood set containing k sampling points around (x, y), expanding the discrete sampling point information into a continuous spatial distribution to generate complete ecological data

[0122] Based on the spatial ecological data Simulating the growth dynamics of the oil-tea camellia forest:

[0123] G(x, y, t + 1) = G(x, y, t) + f(N t (x, y, t), W s(x, y, t), L(x, y, t)) - δ(G(x, y, t)), where G(x, y, t + 1) is the vegetation growth at grid point (x, y) at time t + 1, G(x, y, t) is the vegetation growth at grid point (x, y) at time t, and N t (x, y, t) is the soil nutrient concentration, and W s (x, y, t) is the soil humidity, L(x, y, t) is the light intensity, and f(·) is the growth function, defined as f = α1·N t + α2·W s + α3·L, where α1, α2, and α3 are growth influence coefficients, and δ(G(x, y, t)) is the natural decay function, defined as δ = β·G(x, y, t), and β is the decay coefficient;

[0124] Combined with the pest distribution data B d (x, y, t) and the meteorological conditions of wind speed V w 、humidity H a and temperature T a , a pest diffusion model is established:

[0125] where B d (x, y, t + 1) is the pest distribution value at grid point (x, y) at time t + 1, is the diffusion term, k is the diffusion coefficient, and g(V w , H a , T a ) is the meteorological driving term, defined as g = γ1·V w + γ2·H a + γ3·T a , where γ1, γ2, and γ3 are driving influence coefficients. By coupling diffusion and meteorological driving, the spatial diffusion trend of pests is predicted;

[0126] Integrate vegetation growth and pest diffusion to construct a comprehensive ecological model:

[0127] M(x, y, t) = G(x, y, t) - ω·B d (x, y, t), where M(x, y, t) is the comprehensive ecological parameter, representing the ecological quality of grid (x, y) at time t, G(x, y, t) is the vegetation growth, and B d (x, y, t) is the pest distribution value, and ω is the pest influence weight coefficient;

[0128] Specifically, through the above steps, a dynamic virtual oil tea forest model is established, and the output ecological model data M(x, y, t) is used as the basis for subsequent simulation management.

[0129] In step S2, simulate the ecological process in the digital twin platform:

[0130] Simulate nutrient diffusion and infiltration. The spatio-temporal variation of nutrient concentration in the soil is described by the convection-diffusion equation: where N(x, y, t) is the nutrient concentration in the (x, y) grid of the soil, D n is the nutrient diffusion coefficient, which depends on the soil type, v is the water flow velocity vector, indicating the nutrient convection direction and intensity, S f (x, y, t) is the fertilization input amount, L w (x, y, t) is the nutrient loss amount;

[0131] After fertilization, part of the nutrients are lost and enter the water body, and its concentration change is:

[0132] where C w (t) is the nutrient concentration in the water body, V w is the volume of the water body, Ω is the soil area, L w (x, y, t) is the nutrient loss amount in the soil,

[0133] Simulate the spread of pests and diseases. The spatio-temporal variation of the distribution of pests and diseases is:

[0134] where B(x, y, t) is the density of pests and diseases, D b is the diffusion coefficient, indicating the pest and disease transmission rate, G b (B, T, H) is the growth function, which depends on the current density B, temperature T, and humidity H, G b = α b ·B·(1 - B / K), where α b is the growth rate, K is the environmental carrying capacity, M b (V w , T) is the meteorological-driven mortality rate of pests and diseases, which is restricted by the wind speed V w and high temperature T;

[0135] Combine the wind speed field and the humidity field to calculate the pest and disease diffusion path:

[0136] where, is the velocity field of pest and disease diffusion, is the wind speed field vector, is the pest and disease density gradient;

[0137] Simulate the irrigation mode. The soil water content distribution after irrigation is described by the Richards equation:

[0138] where W(x, y, t) is the soil water content, Ks (W) is the soil hydraulic conductivity, K s (W) = K sat ·exp(-β·(W max -W)), where K sat is the saturated hydraulic conductivity, β is the soil characteristic coefficient, and R(x, y, t) is the evapotranspiration rate;

[0139] Evaluate the water resource utilization efficiency of the irrigation scheme, where E i is the irrigation efficiency of scheme i, and V i is the volume of irrigation water;

[0140] Specifically, the soil nutrient distribution is simulated as the nutrient change N(x, y, t) in the soil and the water pollution concentration C w (t), the pest and disease spread is simulated as the pest and disease density B(x, y, t) and the spread path The irrigation mode is simulated as the soil water content distribution W(x, y, t) and the irrigation efficiency E i .

[0141] Step S3, intelligent prevention and control,

[0142] Based on the simulation results provided by the digital twin model and the real-time monitoring data, link the intelligent prevention and control and ecological management modules for real-time prevention and control;

[0143] Perform intelligent pest and disease prevention and control in step S3:

[0144] Based on the pest and disease spread simulation results in step S2, combine the drone inspection data to dynamically update the pest and disease risk distribution map, and link the drone to spray biological pesticides, and only perform prevention and control operations in high-risk areas;

[0145] Execute the ecological management strategy in step S3:

[0146] Adjust the fertilization amount and time according to the soil nutrient distribution data simulated in S2;

[0147] Irrigate based on the irrigation mode simulation data, combined with the real-time data of soil humidity and water resource utilization rate;

[0148] In step S3, perform intelligent pest and disease prevention and control:

[0149] Combine the pest and disease spread prediction results B m (x, y, t) in step S2 and the real-time distribution data B u (x, y, t) obtained from drone inspections to dynamically update the pest and disease risk distribution:

[0150] where B r(x, y, t) is the updated pest risk distribution map, B m (x, y, t) is the risk distribution obtained from the simulation of pest spread, B u (x, y, t) is the real-time distribution data analyzed from the drone inspection images, w m is the weight of the simulation result, dynamically adjusted according to the historical prediction accuracy, w u is the weight of the drone data, adjusted based on the resolution and timeliness of the inspection images;

[0151] Based on the updated distribution map B r (x, y, t), set the risk threshold τ, and the area above the threshold is defined as the high-risk area:

[0152] If B r (x, y, t) ≥ τ, then R(x, y, t) = 1,

[0153] If B r (x, y, t) < τ, then R(x, y, t) = 0, where R(x, y, t) is the high-risk area identifier, 1 represents high risk, 0 represents non-high risk, and τ is the pest risk threshold, determined by historical monitoring and crop economic loss analysis;

[0154] According to the spatial distribution of the high-risk area R(x, y, t), calculate the spraying path and dosage of the drone. The spraying path optimization formula:

[0155] Where, L spray is the total length of the drone spraying path, d i is the distance between the centers of two adjacent high-risk areas in the path, (x i , y i ) is the coordinate of the center point of the high-risk area passed by the path, and the constraint condition is that the path must cover all high-risk areas;

[0156] The spraying dosage calculation formula is Q(x, y) = λ · B r (x, y, t), where Q(x, y) is the spraying dosage on the (x, y) grid, λ is the pesticide demand coefficient per unit pest density, determined based on pesticide characteristics and target control efficiency, B r (x, y, t) is the risk distribution value, dynamically adjusting the spraying intensity;

[0157] The drone performs operations according to the optimized path L spray and the spraying dosage Q(x, y), and after spraying, feedback the spraying coverage map C(x, y, t):

[0158] If Q(x, y) > 0, then C(x, y, t) = 1,

[0159] If Q(x, y) = 0, then C(x, y, t) = 0, where C(x, y, t) is the spraying coverage status, 1 indicates covered, 0 indicates uncovered, and Q(x, y) is the spraying dosage, which determines the coverage status.

[0160] Specifically, the updated pest and disease risk distribution map B r (x, y, t) is used to determine the high-risk areas, the high-risk area identifier R(x, y, t) is used to clarify the spraying scope, the spraying path L spray and the dosage Q(x, y) are used for the UAV to perform precise prevention and control, and the spraying coverage status C(x, y, t) is used to evaluate the spraying completion degree.

[0161] The method for adjusting the fertilization amount and time in step S3 is as follows:

[0162] According to the soil nutrient distribution data N(x, y, t) and the crop demand model in step S2, determine the fertilization amount F(x, y) and fertilization time T for each grid f ,

[0163] Fertilization amount: F(x, y) = max(0, D n (x, y) - N(x, y, t)) · A(x, y), where F(x, y) is the fertilization amount of grid (x, y), D n (x, y) is the target nutrient concentration of grid (x, y), N(x, y, t) is the current soil nutrient concentration of grid (x, y), and A(x, y) is the grid area

[0164] Fertilization time: where T f is the fertilization time point, is the nutrient change rate of grid (x, y), R n is the nutrient absorption rate of the crop, and Ω is the set of grids in the fertilization area;

[0165] Based on the irrigation mode simulation data W m (x, y, t) in step S2, combined with the real-time soil moisture data W r (x, y, t) and the water resource utilization rate, calculate the irrigation amount Q i (x, y) and irrigation efficiency,

[0166] Irrigation amount: Q i (x, y) = max(0, W m (x, y, t) - W r (x, y, t)) · A(x, y), where Q i (x, y) is the irrigation amount of grid (x, y), W m(x, y, t) is the target soil moisture for irrigation pattern simulation, W r (x, y, t) is the soil moisture monitored in real time, and A(x, y) is the grid area;

[0167] Water resource utilization rate: Among them, E w is the water resource utilization efficiency of the irrigation scheme, V w is the total irrigation water volume, and η is the efficiency coefficient of the irrigation facilities;

[0168] Path planning of unmanned aerial vehicle or automatic irrigation equipment:

[0169] Among them, L irrigation is the total length of the irrigation path, d i is the distance between adjacent irrigation points in the path, (x i , y i ) are the coordinates of the irrigation points, satisfying the irrigation volume Q i (x, y)>0.

[0170] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things, characterized in that: including, An oil-tea camellia forest monitoring module that monitors the ecological environment of the oil-tea camellia forest, collects real-time data and monitors meteorological conditions; The real-time data monitoring methods include real-time monitoring by the Internet of Things sensor network and regular inspection monitoring by drones. The Internet of Things sensor network is deployed in the oil-tea camellia forest area to collect environmental parameters, including soil humidity, nutrient concentration, air temperature and humidity, and light intensity data. The drones use multi-spectral imaging technology to conduct regular inspections and collect pest and disease images and vegetation conditions. A digital twin platform that, based on dynamic modeling technology, creates a digital twin model of the oil-tea camellia forest, and displays soil nutrient distribution, pest and disease risk areas, irrigation suggestions, and predicted crop growth data. An intelligent prevention and control module that is used to link drones to predict pests and diseases, dynamically update the pest and disease risk distribution map based on the pest and disease risk areas combined with the drone inspection data, and link drones to spray biological pesticides, and only carry out prevention and control operations in high-risk areas. An ecological management module that, based on the twin model, provides ecological management suggestions, and the ecological management suggestions include fertilization suggestions and irrigation suggestions.

2. The ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things according to claim 1, wherein The management steps of the ecological digital monitoring and management system for oil-tea camellia forest planting include: Step S1, collecting ecological data of the oil-tea camellia forest. Integrate the data collected by sensors and drones into the dynamic ecological data of the oil-tea camellia forest area, including environmental parameters, pest and disease images, and vegetation conditions. Step S2, dynamically modeling the digital twin model and simulating the ecological process of oil-tea camellia forest planting. Use dynamic modeling technology to create a digital twin model of the oil-tea camellia forest and simulate and predict the ecological process of oil-tea camellia forest planting. Step S3, intelligent prevention and control. Based on the simulation results provided by the digital twin model and real-time monitoring data, link the intelligent prevention and control and ecological management modules to carry out real-time prevention and control.

3. The ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things according to claim 2, characterized in that, In step S1, the integration method of the ecological data in the oil-tea camellia forest area is: Let the environmental parameters collected by the sensor in real time include soil humidity W s , nutrient concentration N t , air temperature and humidity T a , H a and light intensity L. During the collection process, noise data is filtered: Among them, P i (t) is the weighted smoothed value of sensor i at time t, S j (t) is the original sampled value of sensor j at time t, w j is the weight factor used to adjust the influence of different sensors, which is allocated according to the historical accuracy of the sensors. k is the number of sensors deployed in the area. The weighted smoothing eliminates outliers, and the output set of credible values of the environmental parameters is {W s , N t , T a , H a , L}; Let the UAV inspection image include the vegetation coverage V c and the pest distribution image B d (x,y), the pest distribution is calculated using a multispectral data classification model: where B d (x,y) is the pest distribution value at the position (x,y), R(x,y), G(x,y), MIR(x,y) are the reflection values of the red, green, and near-infrared bands respectively, α, β, γ are spectral coefficients, based on multispectral model correction, σ(·) is a normalization function used to map the distribution value to the standard range [0,1]; Integrate the point information of sensor data and the spatial information of UAV images into a unified grid G(x,y), and calculate the dynamic ecological parameters of each grid. Among them, D(x,y) is the dynamic ecological parameter on the grid (x,y), P i (t) is the weighted value of sensor i at time t, B d (x,y) is the value of the pest and disease distribution matrix, λ i is the weight factor used to adjust the contribution of the sensor to the grid parameters, η is the weight of UAV data used to balance the influence of sensor and image data, and m is the number of sensors within the grid range. Construct a dynamic model from time series grid data: E(x,y,t) = φ·E(x,y,t - 1)+(1 - φ)·D(x,y,t), where E(x,y,t) is the comprehensive ecological parameter of grid (x,y) at time t, E(x,y,t - 1) is the comprehensive ecological parameter of the previous moment, D(x,y,t) is the dynamic ecological parameter of the current moment, and φ is the time decay coefficient, which controls the weight distribution of historical data and current data.

4. An ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things according to claim 3, characterized in that, In step S2, the simulation and prediction process includes: Based on the dynamic ecological data in step S1, use dynamic modeling technology to create a virtual oil-tea camellia forest model. Simulate the ecological process in the digital twin platform, including: Soil nutrient distribution simulation: Analyze the impact of different fertilization schemes on soil nutrient loss and surrounding water body pollution. Pest and disease spread simulation: Combine pest and disease images and meteorological conditions to predict the pest and disease spread path and high-risk areas. Irrigation mode simulation: Evaluate the impact of different irrigation schemes on water resource utilization efficiency and soil moisture content. Output the simulation results of the ecological process, including soil nutrient distribution, pest and disease risk areas, irrigation suggestions, and predicted crop growth data.

5. The ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things according to claim 4, wherein, In step S2, the method of creating a virtual oil-tea camellia forest model is: Based on the dynamic ecological grid data E(x, y, t) in step S1, the spatial interpolation method is used to fill the uncovered area. Among them, is the ecological parameter of the grid point (x, y) at time t after interpolation. E(x i , y j , t) is the ecological parameter of the sampling point (x i , y j ) closest to (x, y). is the weight factor, based on the reciprocal of the Euclidean distance between grid points. p is the attenuation coefficient. Ω is the neighborhood set containing k sampling points around (x, y). The discrete sampling point information is extended to a continuous spatial distribution to generate complete ecological data. Based on spatial ecological data Simulate the growth dynamics of oil-tea camellia forests: G(x,y,t + 1) = G(x,y,t) + f(N t (x,y,t),W s (x,y,t),L(x,y,t)) - δ(G(x,y,t)), Among them, G(x, y, t+1) is the vegetation growth amount at grid point (x, y) at time t+1, G(x, y, t) is the vegetation growth amount at grid point (x, y) at time t, N t (x, y, t) is the soil nutrient concentration, W s (x, y, t) is the soil humidity, L(x, y, t) is the light intensity, f(·) is the growth function, defined as f = α1·N t +α2·W s +α3·L, where α1, α2, α3 are growth influence coefficients, δ(G(x, y, t)) is the natural decay function, defined as δ = β·G(x, y, t), and β is the decay coefficient; Combined with the pest and disease distribution data B d (x, y, t) and the meteorological condition wind speed V w , humidity H a and temperature T a , establish a pest and disease diffusion model: Among them, B d (x, y, t + 1) is the pest and disease distribution value of the grid point (x, y) at the moment t + 1, is the diffusion term, k is the diffusion coefficient, g(V w , H a , T a ) is the meteorological driving term, defined as g = γ1·V w + γ2·H a + γ3·T a , where γ1, γ2, γ3 are driving influence coefficients; Integrate vegetation growth and pest and disease spread to construct a comprehensive ecological model. M(x, y, t) = G(x, y, t) - ω·B d (x, y, t), where M(x, y, t) is the comprehensive ecological parameter, representing the ecological quality of the grid (x, y) at time t, G(x, y, t) is the vegetation growth, and B d (x, y, t) is the pest and disease distribution value, and ω is the pest and disease impact weight coefficient.

6. The ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things according to claim 5, characterized in that, In step S2, simulate the ecological process in the digital twin platform. Simulate nutrient diffusion and infiltration. The spatio-temporal variation of nutrient concentration in the soil is described by the convection-diffusion equation: where N(x, y, t) is the nutrient concentration in the (x, y) grid of the soil, D n is the nutrient diffusion coefficient, which depends on the soil type, v is the water flow velocity vector, indicating the direction and intensity of nutrient convection, S f (x, y, t) is the fertilization input amount, L w (x, y, t) is the nutrient loss amount; After fertilization, some nutrients are lost and enter the water body, and their concentration changes are as follows: Among them, C w (t) is the nutrient concentration in the water body, V w is the volume of the water body, Ω is the soil area, L w (x, y, t) is the loss amount of nutrients in the soil Simulate the spread of pests and diseases, and the changes in the distribution of pests and diseases over time and space are as follows: Among them, B(x, y, t) is the density of pests and diseases, and D b is the diffusion coefficient, representing the spread rate of pests and diseases, and G b (B, T, H) is the growth function, which depends on the current density B, temperature T, and humidity H, and G b = α b ·B·(1 - B / K), where α b is the growth rate, K is the environmental carrying capacity, and M b (V w , T) is the meteorological-driven mortality of pests and diseases, which is restricted by the wind speed V w and high temperature T; Combine the wind speed field and humidity field to calculate the spread path of pests and diseases: Among them, is the velocity field of the spread of pests and diseases, is the wind speed field vector, is the density gradient of pests and diseases; Simulate the irrigation mode, and the soil moisture content distribution after irrigation is described by the Richards equation: Among them, W(x, y, t) is the soil moisture content, and K s (W) is the soil hydraulic conductivity, and K s (W)=K sat ·exp(-β·(W max -W)), where K sat is the saturated hydraulic conductivity, β is the soil characteristic coefficient, and R(x, y, t) is the evapotranspiration rate; Evaluate the water resource utilization efficiency of irrigation schemes, where E i is the irrigation efficiency of Scheme i, and V i is the volume of irrigation water.

7. The ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things according to claim 6, characterized in that, Perform intelligent pest and disease control in step S3: Based on the pest and disease spread simulation results in step S2, dynamically update the pest and disease risk distribution map in combination with the UAV inspection data, and link the UAV to spray biological pesticides, and only perform prevention and control operations in high-risk areas.

8. An ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things according to claim 7, characterized in that, Execute the ecological management strategy in step S3: Adjust the fertilization amount and time according to the simulated soil nutrient distribution data in S2; Based on the irrigation mode simulation data, perform irrigation in combination with the real-time data of soil moisture and water resource utilization rate.

9. The ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things according to claim 8, wherein, In step S3, perform intelligent pest and disease control: The pest and disease spread prediction result B in step S2 m (x, y, t) and the real-time distribution data B obtained by drone inspection u (x, y, t) are fused to dynamically update the pest and disease risk distribution: Among them, B r (x, y, t) is the updated pest and disease risk distribution map, B m (x, y, t) is the risk distribution obtained from the simulation of pest and disease spread, B u (x, y, t) is the real-time distribution data analyzed from the UAV inspection images, w m is the weight of the simulation result, dynamically adjusted according to the historical prediction accuracy, w u is the weight of the UAV data, adjusted based on the resolution and timeliness of the inspection images; Based on the updated distribution map B r (x, y, t), set the risk threshold τ, and the area above the threshold is defined as the high-risk area: If B r (x, y, t) ≥ τ, then R(x, y, t) = 1, If B r (x, y, t) < τ, then R(x, y, t) = 0, where R(x, y, t) is the high-risk area identifier, 1 represents high risk, 0 represents non-high risk, and τ is the pest and disease risk threshold, which is determined by historical monitoring and crop economic loss analysis; According to the spatial distribution of the high-risk area R(x, y, t), calculate the spraying path and dosage of the UAV, and the spraying path optimization formula: Among them, L spray is the total length of the drone spraying path, d i is the distance between the centers of two adjacent high-risk areas in the path, (x i , y i ) is the coordinate of the center point of the high-risk area passed by the path, and the constraint condition is that the path must cover all high-risk areas; The calculation formula for the spraying dosage of pesticides is Q(x,y) = λ·B r (x,y,t), where Q(x,y) is the dosage of pesticides sprayed on the (x,y) grid, λ is the coefficient of pesticide demand per unit pest density, determined based on pesticide characteristics and target control efficiency, and B r (x,y,t) is the risk distribution value, which dynamically adjusts the spraying intensity; The drone performs operations according to the optimized path L spray and the spraying dose Q(x, y), and after spraying is completed, it feeds back the spraying coverage map C(x, y, t): If Q(x, y)>0, then C(x, y, t) = 1, If Q(x, y) = 0, then C(x, y, t) = 0, where C(x, y, t) is the spraying coverage status, 1 means covered, 0 means not covered, and Q(x, y) is the spraying dosage, which determines the coverage status.

10. An ecological digital monitoring and management system for oil-tea camellia forest planting based on the Internet of Things according to claim 9, characterized in that, The way to adjust the fertilization amount and time in step S3 is: Based on the soil nutrient distribution data N(x, y, t) and the crop demand model in step S2, determine the fertilization amount F(x, y) and fertilization time T for each grid f , Fertilizer application rate: F(x,y) = max(0, D n (x,y) - N(x,y,t)) · A(x,y), where F(x,y) is the fertilizer application rate of grid (x,y), D n (x,y) is the target nutrient concentration of grid (x,y), N(x,y,t) is the current soil nutrient concentration of grid (x,y), and A(x,y) is the grid area Fertilization time: Among them, T f is the fertilization time point, is the nutrient change rate of the grid (x, y), R n is the nutrient absorption rate of the crop, and Ω is the grid set of the fertilization area; Based on the irrigation pattern simulation data W m (x, y, t) in step S2, combined with the real-time soil moisture data W r (x, y, t) and the water resource utilization rate, calculate the irrigation amount Q of each grid i (x, t) and the irrigation efficiency Irrigation volume: Q i (x,y) = max(0, W m (x,y,t) - W r (x,y,t)) · A(x,y), where Q i (x,y) is the irrigation volume of grid (x,y), W m (x,y,t) is the target soil moisture simulated by the irrigation pattern, W r (x,y,t) is the real-time monitored soil moisture, and A(x,y) is the grid area; Water resource utilization rate: Among them, E w is the water resource utilization efficiency of the irrigation scheme, V w is the total irrigation water volume, and η is the efficiency coefficient of the irrigation facilities; Path planning of UAVs or automatic irrigation equipment: Among them, L irrigation is the total length of the irrigation path, d i is the distance between adjacent irrigation points in the path, (x i , y i ) is the coordinate of the irrigation point, and the irrigation volume Q i (x, y) > 0 is satisfied.

Images