A method for collecting orchard ecological environment information

Abstract

The application discloses a method for collecting orchard ecological environment information, which comprises the following steps: collecting soil parameters, spatial environment parameters, pest parameters and fruit parameters of the orchard in real time through deploying various collection devices, and collecting the orchard ecological basic data set by summarizing; constructing a back-stepping model to indirectly calculate key ecological parameters, including soil nutrients, fruit quality and pest spread trend, and optimizing the results through experimental verification and model correction; comparing and analyzing the deviation source of the back-stepping results and the orchard target value, and establishing a macro cause chain through data matching and a visualization tool; determining the regulation and control direction and matching executable regulation and control measures according to the deviation analysis and cause chain results; monitoring the effect after the regulation and control measures are implemented, verifying the regulation and control effect by comparing the data before and after the regulation and control, and feeding back the data to the iteration optimization of the back-stepping model. The method is scientific and efficient, and can be used for improving the production efficiency and ecological environment quality of the orchard.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ecological environment information collection, and particularly relates to a method for collecting orchard ecological environment information. BACKGROUND

[0002] With the rapid development of modern agriculture, fine management of orchards has become the key to improving production efficiency and ecological environment quality. However, traditional orchard management methods mainly rely on experience-based decision-making, and there are limitations in the collection and analysis of key ecological information such as soil, water, pests, and fruit quality. This lack of data support in management can easily lead to resource waste, yield decline, and degradation of the ecological environment.

[0003] Currently, some technologies have been applied to orchard management, such as soil sensors for monitoring water and pH, unmanned aerial vehicles for aerial photography of orchard vegetation coverage maps, or pest monitoring based on insect lures. However, the data collected by different monitoring devices is difficult to integrate, and there is a lack of a unified analysis platform. Parameters that are difficult to measure directly cannot be efficiently calculated and predicted, and there is a lack of closed-loop feedback for environmental problems, making it difficult to optimize management measures in a timely manner. Traditional monitoring methods are difficult to fully capture the spatial distribution differences of soil, environment, and pests in the orchard, resulting in insufficient targeting of management measures.

[0004] To address the above problems, the present application provides a method for collecting orchard ecological environment information, which can combine various sensing technologies, data analysis models, and feedback optimization mechanisms to form a closed-loop management mode to achieve precise management and ecological optimization of orchards. SUMMARY

[0005] The present application provides a method for collecting orchard ecological environment information to solve the problem that existing technology cannot efficiently calculate and predict parameters that are difficult to measure directly, lacks closed-loop feedback for environmental problems, and is difficult to optimize management measures in a timely manner.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] The present application provides a method for collecting orchard ecological environment information, characterized by comprising the following steps:

[0008] Step S1, real-time collection of soil parameters, spatial environment parameters, pest parameters, and fruit parameters of the orchard by deploying various collection devices, and aggregation to form an orchard ecological basic data set;

[0009] In step S1, the following sub-steps are further included:

[0010] S1-1, according to the characteristics of orchard terrain, planting planning and irrigation zoning, the orchard is divided into several preliminary management units, and the collection content and range are set, the collection content includes: soil parameters, spatial environment parameters, pest parameters, fruit parameters;

[0011] Soil parameters: moisture, pH;

[0012] Spatial environment parameters: light intensity, temperature and humidity;

[0013] Pest parameters: pest species, density;

[0014] Fruit parameters: fruit size, color;

[0015] Collection range: the preliminary division of the management unit as the arrangement range of the collection device;

[0016] S1-2, the data collection device includes soil parameter monitoring device, pest monitoring device, fruit monitoring device;

[0017] Soil parameter monitoring device: soil moisture sensor, pH sensor, temperature and humidity sensor, light intensity sensor;

[0018] Pest monitoring device: intelligent insect trapping device;

[0019] Fruit monitoring device: unmanned aerial vehicle, high-resolution camera;

[0020] The device arrangement covers terrain types, vegetation differences, and orchard center and boundary areas; according to the area and complexity of the orchard, set the sensor density, such as arranging 1 soil sensor per 100 square meters, and 1 light or temperature and humidity sensor per 500 square meters; according to historical data or observation focus, arrange in areas with significant environmental changes or prone to problems, including low-lying areas, insufficient light areas, and high pest areas; the device is close to the main road or work channel, easy to check and maintain;

[0021] S1-3, data transmission, use LoRa wireless technology to upload the collected data to the cloud, the cloud platform automatically classifies and stores the uploaded data, and forms the basic data set of the current ecological state of the orchard;

[0022] Simple aggregation and summary of real-time data, generate preliminary state diagram of orchard ecological information, including: soil moisture distribution map, light intensity distribution map, pest density distribution map, fruit distribution and size change trend map;

[0023] S1-4, based on the data collected in S1-3, the orchard is divided into more detailed management units. According to the parameters of soil moisture, pH, conductivity, etc., the units with similar soil characteristics are divided; according to the light intensity and temperature and humidity, the units with similar environmental conditions are divided; combined with the data of pest species and density, the high-pest area is marked separately; according to the size, color and distribution trend of the fruit, the areas with similar yield characteristics are divided together; combined with the terrain (high land, low land) and historical management information;

[0024] Using the clustering algorithm K-Means, similar areas are automatically divided according to the collected ecological parameters, and the clustering results are imported into the GIS platform for visualization processing of the geographic spatial data of the orchard, as shown in formula (1):

[0025]

[0026] Wherein, J is the total error sum of squares of clustering, x is the data point belonging to the i-th cluster, μ i is the center point of the i-th cluster, C i is the set of the i-th cluster, and K is the number of clusters;

[0027] Each management unit has a clear description of ecological characteristics, including soil characteristics, pest risk, fruit distribution, and the division results are output as a management unit distribution map, a detailed table of ecological parameters of each unit and a summary through the GIS platform.

[0028] Step S2, a back-stepping model is constructed to indirectly calculate key ecological parameters including soil nutrients, fruit quality and pest spread trend using the collected data, and the results are verified and optimized through experiments and model correction;

[0029] In step S2, the following sub-steps are further included:

[0030] S2-1, according to the collected data, the key parameters difficult to measure are indirectly calculated, including: soil nutrients, fruit quality, pest spread trend;

[0031] Soil nutrients: nitrogen, phosphorus, potassium;

[0032] Fruit quality: sugar content, maturity;

[0033] Pest spread trend: whole garden distribution;

[0034] S2-2, a mathematical logical relationship between ecological collection data and key parameters difficult to measure is established, and a back-stepping model is constructed: soil nutrient model, fruit quality model, pest spread model;

[0035] Soil nutrient model, linear regression formula is used to establish the linear relationship between soil moisture, conductivity, pH and nitrogen, phosphorus, potassium nutrient concentration, as shown in formula (2)-(4):

[0036] N = a1 · (H2O) + b1 · (pH) + c1 Equation (2)

[0037] P = a2 · (H2O) + b2 · (pH) + c2 Equation (3)

[0038] K = a3 · (H2O) + b3 · (pH) + c3 Equation (4)

[0039] wherein, N, P, K represent the current concentration of nitrogen, phosphorus, potassium in the soil respectively, H2O is the water content of the soil, pH is the acid-base value of the soil, a1, a2, a3 are model coefficients related to soil moisture, b1, b2, b3 are model coefficients related to soil pH, c1, c2, c3 are the intercept terms of the model;

[0040] fruit quality model, using light intensity, temperature and humidity, fruit size change rate, to establish a prediction relationship with sugar content and maturity, specifically as equation (5) - equation (6):

[0041] T_H = 0.7 · T + 0.3 · H Equation (5)

[0042] S = d1 · (Lux) + d2 · (T_H) + d3 · (v) + d4 Equation (6)

[0043] wherein, T_H is the temperature and humidity index, T is the air temperature in the orchard, H is the air humidity; S is the sugar content of the fruit, Lux is the light intensity, v is the fruit change rate, d1, d2, d3 are model coefficients related to light, temperature and humidity index, fruit change rate respectively, d4 is the intercept term of the model;

[0044] pest spread model, using trap data, wind speed, humidity and temperature to predict the spread trend of pests, specifically as equation (7):

[0045] D(t+1) = D(t) · (1 + r - c) Equation (7)

[0046] wherein, D(t) is the pest density at the current time t, D(t+1) is the pest density at the next time t+1, r is the pest reproduction rate (affected by temperature and humidity), c is the pest capture rate, 1 + r - c is the increase or decrease factor of pest density;

[0047] S2-3, the collected real-time data are taken as input, substituted into the constructed backstepping model, and the predicted values of the key parameters are calculated; a small range area is randomly selected in the orchard, and the actual measurement or experimental detection of the backstepping parameters is performed: soil nutrients: the concentrations of nitrogen, phosphorus and potassium in the soil are detected and analyzed through the laboratory; fruit quality: the sugar content and maturity of the fruit are measured manually; pest density: the actual distribution data of pests are recorded manually;

[0048] Compare the backstepping results with the small-scale experimental data, calculate the error value, optimize the model parameters for the parts with larger errors, and retrain the model with new data.

[0049] Step S3, compare the backstepping results with the orchard target values, analyze the deviation sources, establish the macro cause chain of deviation and environmental parameters through data matching and visualization tools, and intuitively present the problem causes and their spatial distribution;

[0050] In step S3, the following sub-steps are further included:

[0051] S3-1, compare and analyze the deviation between the key parameter values calculated by the backstepping model and the target values of the orchard, the deviation is the difference between the actual calculation value and the target value, mainly including soil nutrient deviation, fruit quality deviation, and pest density deviation, and analyze the cause chain of the deviation;

[0052] Cause chain of low soil nitrogen: high soil moisture → serious nitrogen leaching → low nitrogen content;

[0053] Cause chain of low fruit sugar content: insufficient light → low photosynthesis efficiency → insufficient sugar accumulation;

[0054] Cause chain of high pest density: suitable temperature and humidity → rapid reproduction of pests → local spread of pests;

[0055] S3-2, match the analyzed deviation with the environmental parameters of the orchard to establish the macro cause chain and understand the formation reasons of the deviation, and through data analysis and visualization tools, correlate the spatial distribution of the deviation with the environmental parameters of the orchard;

[0056] Use GIS tools to draw the cause chain distribution map of the orchard, display the spatial distribution relationship between the deviation and the environmental parameters, generate regional problem charts including light distribution map and pest high-risk area map, and form a visualization report of the deviation and environmental parameters;

[0057] According to the matching and visualization results, sort out the logical relationship between the deviation and the cause chain, output the spatial distribution of the deviation data, the main macro cause chain of the deviation in each region, and the key driving factors of the deviation formation.

[0058] Step S4, determine the control direction according to the deviation analysis and cause chain results, and match the executable control measures;

[0059] In step S4, the following sub-steps are further included:

[0060] S4-1, according to the results of bias analysis and macro reason mapping, a clear control direction is formulated, the difference between the key parameters and the target value is narrowed, according to the deviation degree of the key parameters, including low soil nitrogen content, low fruit sugar content and high pest density, the problems that need to be solved are determined, and then combined with the results of macro reason analysis, the core driving factors affecting the deviation are determined, so that the control direction meets the management target of the orchard as a whole;

[0061] S4-2, according to the control direction determined in S4-1, specific executable measures are matched to optimize the ecological parameters and meet the target value requirements of the orchard, and the typical control measures are designed as follows:

[0062] Soil nutrient regulation: increase soil nitrogen content, reduce irrigation frequency or use drip irrigation technology to reduce water loss; periodically apply slow-release nitrogen fertilizer in the low area;

[0063] Fruit quality regulation: improve fruit sugar content, prune the tree crown, increase the light area of the fruit, and spray potassium and magnesium element fertilizer on the leaf surface to promote sugar accumulation;

[0064] Pest control: reduce pest density, install pest nets or spray biological pesticides in the high pest area; introduce natural enemies for biological control; adjust the humidity level of the orchard to reduce the breeding conditions of pests.

[0065] Step S5, monitor the effect after the implementation of the control measures, verify the control effect by comparing the data before and after, and use the feedback data for iterative optimization of the reverse model, and finally realize closed-loop management and automatic control.

[0066] In step S5, the following sub-steps are further included:

[0067] S5-1, monitor the control effect, continuously collect soil, fruit and pest data, dynamic changes of water and nutrients, improvement trend of fruit sugar content and reduction of pest density after the implementation of the control measures;

[0068] S5-2, compare the changes of the core data before and after the adjustment, including soil nitrogen content, fruit sugar content, pest density and target value, and evaluate the effect of the control measures;

[0069] S5-3, input the feedback data after the control into the reverse model, correct the model parameters, introduce historical data and adjustment data to enrich the training set, gradually introduce more complex nonlinear model optimization reverse logic, perfect the closed-loop control, and realize automatic decision and control.

[0070] Compared with the prior art, the beneficial effects of the present application are:

[0071] The present application realizes comprehensive collection of soil, environment, pest and fruit parameters by deploying various collection devices including soil sensors, unmanned aerial vehicles and intelligent insect trapping devices, and integrates data to the cloud using wireless transmission technology to form a unified ecological information basic data set, solving the problem of data isolation in the prior art.

[0072] The present application constructs a backstepping system based on linear regression and other mathematical models, which can accurately calculate key parameters that are difficult to measure directly, such as soil nutrients, fruit sugar content and pest spread trend, improve the completeness of ecological information and the scientificity of analysis, and make up for the limitations of traditional methods.

[0073] The present application can identify ecological deviations by comparing backstepping results with orchard target values, and establish cause chains and spatial distribution maps in combination with environmental parameters, providing scientific basis for precise regulation, while the analysis ability of traditional technology for deviation causes is limited.

[0074] The present application uses feedback data for model parameter correction and optimization to form a closed-loop automated decision-making system, greatly improving the dynamic response capability and continuous optimization capability of orchard management.

[0075] The present application visualizes the collection and backstepping results as management unit distribution maps in the orchard with the help of GIS platform, captures the spatial differences of soil, environment and pests, and provides intuitive support for zoned management and precise measures, making up for the shortcomings of the prior art in considering spatial heterogeneity. BRIEF DESCRIPTION OF DRAWINGS

[0076] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments, understand that the following drawings only show some embodiments of the present application, therefore should not be regarded as limiting the scope, for those skilled in the art, on the premise of not paying creative labor, also obtain other related drawings according to these drawings.

[0077] Figure 1 It is a method flowchart of the present application. DETAILED DESCRIPTION

[0078] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of protection of the present application. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only for selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of protection of the present application.

[0079] Please refer to Figure 1 is a method flowchart for collecting orchard ecological environment information provided by an embodiment of the present application, comprising the following steps:

[0080] Step S1, real-time collection of soil parameters, spatial environment parameters, pest parameters and fruit parameters of the orchard by deploying various collection devices, and aggregation to form an orchard ecological basic data set;

[0081] In step S1, the following sub-steps are further included:

[0082] S1-1, according to the orchard topographic features, planting plan and irrigation zoning, the orchard is divided into a plurality of preliminary management units, and the collection content and range are set, the collection content including: soil parameters, spatial environment parameters, pest parameters, fruit parameters;

[0083] Soil parameters: moisture, pH;

[0084] Spatial environment parameters: light intensity, temperature and humidity;

[0085] Pest parameters: pest species, density;

[0086] Fruit parameters: fruit size, color;

[0087] Collection range: the preliminary divided management units as the arrangement range of the collection devices;

[0088] S1-2, the data collection devices include soil parameter monitoring devices, pest monitoring devices, and fruit monitoring devices;

[0089] Soil parameter monitoring devices: soil moisture sensors, pH sensors, temperature and humidity sensors, and light intensity sensors;

[0090] Pest monitoring devices: intelligent insect trapping devices;

[0091] Fruit monitoring device: UAV, high-resolution camera;

[0092] The device arrangement covers terrain types, vegetation differences, and orchard center and boundary areas; set sensor density according to orchard area and complexity, such as 1 soil sensor per 100 square meters, 1 light or temperature and humidity sensor per 500 square meters; arrange according to historical data or observation focus in areas with significant environmental changes or prone to problems, including low-lying areas, insufficient light areas, and high pest areas; devices are close to main roads or work channels for easy inspection and maintenance;

[0093] S1-3, data transmission, use LoRa wireless technology to upload collected data to the cloud, and the cloud platform automatically classifies and stores the uploaded data to form a basic data set of the current ecological state of the orchard;

[0094] Simple aggregation and summary of real-time data to generate preliminary state graphs of orchard ecological information, including soil moisture distribution map, light intensity distribution map, pest density distribution map, and fruit distribution and size trend graph;

[0095] S1-4, based on the data collected in S1-3, divide the orchard into more detailed management units, divide units with similar soil characteristics according to soil water, pH, and conductivity; divide units with similar environmental conditions according to light intensity and temperature and humidity; mark high-pest areas separately based on pest species and density data; divide areas with similar yield characteristics together based on fruit size, color, and distribution trends; combine terrain (high ground, low-lying areas) and historical management information;

[0096] Use the clustering algorithm K-Means to automatically divide similar areas based on collected ecological parameters, and import the clustering results into the GIS platform for visualization processing of the orchard's geographic spatial data, as shown in equation (1):

[0097]

[0098] Where J is the total sum of squared errors of clustering, x is the data point belonging to the i-th cluster, μ i is the center point of the i-th cluster, C i is the set of i-th clusters, and K is the number of clusters;

[0099] Each management unit has a clear ecological characteristic description, including soil characteristics, pest risk, and fruit distribution, and the division results are output as a management unit distribution map, a detailed table of ecological parameters for each unit, and a summary through the GIS platform.

[0100] It should be noted that the "management unit" can be understood as a partition unit in the orchard, each unit can independently deploy sensors, collect data and control, the boundary can be divided according to the terrain, tree spacing or artificial area; The area is determined according to the total scale of the orchard, usually the area of each unit is several hundred to several thousand square meters, which ensures that the monitoring data is representative and easy to manage, and the soil, moisture, light and vegetation conditions of each unit are relatively consistent, facilitating modeling analysis.

[0101] The soil moisture sensor is used to monitor the soil moisture, conductivity and temperature in real time; The pH sensor is used to monitor the acidity and alkalinity of the soil; The temperature and humidity sensor is used to monitor the temperature and humidity in the air; The light intensity sensor is used to measure the light distribution under the fruit tree canopy; The intelligent insect trapping device is equipped with UV insect trapping lamp and camera, which is used to monitor the type and density of insect pests; The unmanned aerial vehicle is used to shoot the vegetation coverage map and fruit distribution map; The high-resolution camera is used to assist in obtaining fruit size and color data.

[0102] GIS is a technical tool that combines geographic spatial data and attribute data, which can collect, store, analyze, visualize and manage spatial related data, visualize ecological data such as soil parameters, light intensity, pest density and fruit distribution to specific geographic locations of the orchard, help to identify ecological differences in different areas of the orchard, combine the calculation results of clustering algorithm, generate management unit distribution map on geographic space, and spatially display the ecological characteristics of the management unit, such as labeling the distribution map of soil moisture and pest density parameters.

[0103] Step S2, construct a backstepping model, indirectly calculate key ecological parameters including soil nutrients, fruit quality and pest spread trend using collected data, and optimize the results through experimental verification and model correction;

[0104] In step S2, the following sub-steps are further included:

[0105] S2-1, indirectly calculate the key parameters that are difficult to measure according to the collected data, the parameters include: soil nutrients, fruit quality, pest spread trend;

[0106] Soil nutrients: nitrogen, phosphorus, potassium;

[0107] Fruit quality: sugar content, maturity;

[0108] Pest spread trend: whole orchard distribution;

[0109] S2-2, establish mathematical logical relationship between ecological collection data and key parameters difficult to measure, construct backstepping model: soil nutrient model, fruit quality model, pest spread model;

[0110] The soil nutrient model uses a linear regression formula to establish a linear relationship between soil moisture, conductivity, pH, and nitrogen, phosphorus, and potassium nutrient concentrations, specifically as formulas (2)-(4):

[0111] N = a1 · (H2O) + b1 · (pH) + c1 Formula (2)

[0112] P = a2 · (H2O) + b2 · (pH) + c2 Formula (3)

[0113] K = a3 · (H2O) + b3 · (pH) + c3 Formula (4)

[0114] where N, P, and K represent the current concentrations of nitrogen, phosphorus, and potassium in the soil, H2O is the water content of the soil, pH is the acid-base value of the soil, a1, a2, and a3 are model coefficients related to soil moisture, b1, b2, and b3 are model coefficients related to soil pH, and c1, c2, and c3 are intercept terms of the model;

[0115] The fruit quality model uses light intensity, temperature and humidity, and fruit size change rate to establish a prediction relationship with sugar content and maturity, specifically as formulas (5)-(6):

[0116] T_H = 0.7 · T + 0.3 · H Formula (5)

[0117] S = d1 · (Lux) + d2 · (T_H) + d3 · (v) + d4 Formula (6)

[0118] where T_H is the temperature and humidity index, T is the air temperature in the orchard, H is the air humidity; S is the sugar content of the fruit, Lux is the light intensity, v is the fruit change rate, d1, d2, and d3 are model coefficients related to light, temperature and humidity index, and fruit change rate, and d4 is the intercept term of the model;

[0119] The pest spread model uses trap data, wind speed, humidity, and temperature to predict the spread trend of pests, specifically as formula (7):

[0120] D(t+1) = D(t) · (1 + r - c) Formula (7)

[0121] where D(t) is the pest density at the current time t, D(t+1) is the pest density at the next time t+1, r is the pest reproduction rate (affected by temperature and humidity), c is the pest capture rate, and 1 + r - c is the pest density increase / decrease factor;

[0122] S2-3, input the collected real-time data as input into the inverse model constructed, calculate the predicted value of the key parameters; randomly select a small area in the orchard, measure or test the inverse parameters in practice: soil nutrients: measure the concentrations of nitrogen, phosphorus and potassium in the soil through laboratory testing; fruit quality: measure the sugar content and maturity of the fruit by artificial measurement; pest density: manually record the actual distribution data of pests;

[0123] Compare the inverse results with the small-scale experimental data, calculate the error value, and optimize the model parameters for the parts with larger errors, and retrain the model with new data.

[0124] It should be noted that the reason why the key parameters are difficult to collect directly is that the concentrations of nitrogen, phosphorus and potassium in the soil need to be detected in the laboratory, which is time-consuming and costly; the fruit quality, such as sugar content, needs to be measured destructively and cannot be measured in real time; the spread trend of pests is affected by many environmental factors and needs to be calculated through dynamic prediction.

[0125] Soil moisture, electrical conductivity and pH value are key factors affecting the content of nitrogen, phosphorus and potassium. Excessive moisture may lead to leaching of nitrogen and decrease of phosphorus and potassium availability, and pH value affects the dissolution and activity of nutrients; therefore, based on these correlations, a linear regression model can describe the mathematical relationship between moisture, electrical conductivity, pH value and nutrient concentration.

[0126] The sugar content and maturity of the fruit are affected by light intensity, temperature and humidity, and the growth rate of the fruit. Light intensity directly determines the efficiency of photosynthesis, thereby affecting sugar accumulation.

[0127] Temperature and humidity determine the suitability of the fruit growth environment and affect the maturity speed. The change rate of fruit size reflects the dynamic state of fruit development and can be used as an indirect measurement index.

[0128] Pest density is affected by many factors such as environmental temperature and humidity, wind speed and pest trapping devices. High temperature and humidity are usually conducive to the reproduction of pests, while the capture rate of pest trapping devices reduces the spread trend. Time series model can dynamically describe the change rule of pest density.

[0129] Step S3, compare the inverse results with the target values of the orchard, analyze the sources of deviation, establish the macro cause chain of deviation and environmental parameters through data matching and visualization tools, and intuitively present the causes and spatial distribution of the problem;

[0130] In step S3, the following sub-steps are further included:

[0131] S3-1, compare the key parameter values calculated by the inverse model with the target values of the orchard to analyze the deviation, which is the difference between the actual calculated value and the target value, mainly including soil nutrient deviation, fruit quality deviation and pest density deviation, and analyze the cause chain of the deviation;

[0132] Reason chain for low soil nitrogen: high soil moisture → severe nitrogen leaching → low nitrogen content;

[0133] Reason chain for low fruit sugar content: insufficient light → low photosynthesis efficiency → insufficient sugar accumulation;

[0134] Reason chain for high pest density: suitable temperature and humidity → rapid pest reproduction → local pest spread;

[0135] S3-2. Match the analyzed deviations with the environmental parameters of the orchard to establish macroscopic reason chains and understand the causes of the deviations. Through data analysis and visualization tools, correlate the deviations with the spatial distribution of environmental parameters in the orchard;

[0136] Use GIS tools to draw reason chain distribution maps of the orchard to show the spatial distribution relationship between deviations and environmental parameters, generate regional problem charts including light distribution maps and pest high-risk area maps, and form a visual report of deviations and environmental parameters;

[0137] Based on the matching and visualization results, sort out the logical relationship between deviations and reason chains, output the spatial distribution of deviation data, the main macroscopic reason chains of each region's deviations, and the key driving factors of deviation formation.

[0138] It should be noted that the complete reason chain of deviations includes the following:

[0139] Reason chain for soil nutrient deviation:

[0140] Possible reasons for low soil nitrogen content:

[0141] 1. High soil moisture → intensified nitrogen leaching → low nitrogen content; 2. Low or high pH value → inhibited microbial activity → reduced nitrogen mineralization and transformation efficiency; 3. Uneven or insufficient fertilization → unbalanced local soil nitrogen supply; 4. Low soil organic matter content → lack of stable nitrogen source → reduced nitrogen content.

[0142] Possible reasons for low soil phosphorus content:

[0143] 1. High pH value (alkaline soil) → phosphorus combines with calcium to form insoluble calcium phosphate → reduced phosphorus utilization rate; 2. Insufficient application of organic fertilizer → lack of available phosphorus source; 3. Excessive rainfall or irrigation → severe surface loss of phosphorus.

[0144] Possible reasons for low soil potassium content: 1. High-intensity absorption by crops → insufficient potassium reserves; 2. Rainwater erosion and leaching → soil potassium loss; 3. Unreasonable potassium fertilizer application method → potassium does not effectively penetrate to the crop root zone.

[0145] Reason chain for fruit quality deviation:

[0146] Possible reasons for low sugar content: 1. Insufficient light → low photosynthetic efficiency → reduced sugar accumulation; 2. Low temperature and humidity → slow fruit ripening process → limited sugar synthesis; 3. Insufficient soil potassium content → hindered sugar transport in fruit.

[0147] Possible reasons for low maturity:

[0148] 1. Unsuitable temperature and humidity → deteriorating fruit growth environment → delayed maturity; 2. High fruit density → insufficient nutrient allocation in tree → prolonged single fruit ripening time; 3. Weakened root absorption capacity → insufficient nutrient supply.

[0149] Reason chain for high pest density:

[0150] 1. Suitable temperature and humidity → accelerated pest reproduction; 2. Improper orchard management → delayed or inadequate control measures; 3. Insufficient number of natural enemies → weakened natural control; 4. Complex vegetation at orchard boundaries → providing shelter and transmission path for pests.

[0151] Step S4, determine the control direction according to the deviation analysis and reason chain results, and match the executable control measures;

[0152] In step S4, it further includes the following sub-steps:

[0153] S4-1, according to the results of deviation analysis and macroscopic reason mapping, formulate clear control direction, narrow the gap between key parameters and target value, according to the deviation degree of key parameters, including low soil nitrogen content, low sugar content of fruit, high pest density, determine the problem that needs to be solved first, and then combine the results of macroscopic reason analysis, clear the core driving factors affecting the deviation, ensure that the control direction meets the management target of the whole orchard;

[0154] S4-2, according to the control direction determined in S4-1, match specific executable measures, realize the optimization of ecological parameters, meet the requirements of orchard target value, and the typical control measures are designed as follows:

[0155] Soil nutrient regulation: increase soil nitrogen content, reduce irrigation frequency or use drip irrigation technology to reduce water loss; regularly apply slow-release nitrogen fertilizer in low area;

[0156] Fruit quality regulation: improve fruit sugar content, prune tree crown, increase fruit light area; spray potassium and magnesium element fertilizer on leaf surface to promote sugar accumulation;

[0157] Pest control: reduce pest density, install pest net or spray biological pesticide in high pest area; introduce natural enemies for biological control; adjust the humidity level of the orchard to reduce the breeding conditions of pests.

[0158] Note that the following is a supplement and refinement of soil nutrients, fruit quality and pest control measures:

[0159] Soil nutrient regulation:

[0160] For low nitrogen content: 1. Apply slow-release nitrogen fertilizer: Regularly apply slow-release fertilizer in areas with low nitrogen content to reduce nitrogen loss; 2. Optimize irrigation method: Reduce excessive irrigation, use drip irrigation or micro-sprinkling, avoid nitrogen leaching with water; 3. Plant green manure crops: Plant green manure (such as alfalfa) in idle areas of the orchard, increase soil nitrogen level through nitrogen fixation.

[0161] For low phosphorus content: 1. Use organic fertilizer: Apply phosphorus-containing organic fertilizer (such as compost, farmyard manure) to increase phosphorus content; 2. Adjust soil pH: Apply sulfur powder or ammonium sulfate to adjust soil pH to neutral in alkaline areas, improve the effectiveness of phosphorus; 3. Prevent phosphorus loss: Reduce irrigation loss on sloping land, reduce surface water flow through mulching.

[0162] For low potassium content: 1. Apply potassium sulfate fertilizer: Supplement soluble potassium source in areas with low potassium content; 2. Recycle biomass: Use pruned branches and fallen leaves of fruit trees, compost and return to soil to supplement potassium elements; 3. Plant deep-rooted plants: Use deep-rooted plants to absorb potassium in deep soil, increase surface potassium content.

[0163] Fruit quality regulation:

[0164] For low sugar content: 1. Optimize tree pruning: Increase fruit light area through pruning to improve light conditions; 2. Apply foliar fertilizer: Spray foliar fertilizer containing potassium, magnesium, boron and other elements to promote sugar accumulation; 3. Adjust fruit load: Reasonable thinning, reduce fruit load per plant, improve sugar distribution in single fruit.

[0165] For low maturity of fruit: 1. Control temperature and humidity: Avoid the adverse effects of low temperature and high humidity on maturity by adjusting irrigation frequency and time; 2. Promote nutrient supply: Combine soil nutrient regulation to enhance root absorption efficiency; 3. Artificial ripening: Spray low-concentration ethylene preparation before picking to promote fruit ripening.

[0166] For uneven fruit color: 1. Adjust fruit orientation: Rotate fruit by hand or adjust support structure to ensure that fruit is fully exposed to light; 2. Use reflective film: Lay reflective film at the base of the fruit tree to increase light levels for fruit under the tree canopy; 3. Regulate light quality: Install supplemental lighting (red and blue light combination) to improve fruit color.

[0167] Pest control:

[0168] For high pest density: 1. Release natural enemies: release ladybugs or trichogramma in areas with high pest incidence for biological control; 2. Spray biological pesticides: use microbial source pesticides (such as Bacillus thuringiensis preparation) for safe pest control; 3. Install insect trapping equipment: install more ultraviolet insect trapping lamps in management units to control pests.

[0169] For pest spread: 1. Boundary vegetation cleaning: remove weeds and host plants at the boundary of the orchard to cut off the pest spread path; 2. Humidity control: reduce orchard humidity to reduce the environment suitable for pest reproduction; 3. Isolation protection: set up physical isolation nets in areas where pests are easy to invade to prevent pest spread.

[0170] For specific pest species:

[0171] Aphids: 1. Spray vegetable oil soap solution to prevent adhesion and respiration; 2. Plant aphid-repellent plants (such as garlic) as natural protection.

[0172] Leaf roller: 1. Set up sex pheromone traps to interfere with mating; 2. Use low-toxicity biological pesticides for targeted control.

[0173] Step S5, monitor the effect after implementing the control measures, compare the data before and after to verify the control effect, and use the feedback data for iterative optimization of the reverse model, finally realize closed-loop management and automatic control.

[0174] In step S5, the following sub-steps are also included:

[0175] S5-1, monitor the control effect, after implementing the control measures, continuously collect soil, fruit and pest data, dynamic changes of water and nutrients, improvement trend of fruit sugar content, and reduction of pest density;

[0176] S5-2, compare the core data before and after adjustment, including soil nitrogen content, fruit sugar content, pest density, and changes from target value, to evaluate the effect of control measures;

[0177] S5-3, input the feedback data after control into the reverse model, correct the model parameters, introduce historical data and adjustment data to enrich the training set, gradually introduce more complex nonlinear model optimization reverse logic, perfect the closed-loop control, and realize automatic decision and control.

[0178] It should be noted that the reason for introducing complex nonlinear model optimization reverse logic is that the linear or simple regression model currently used has certain interpretability and computational efficiency, but when dealing with complex orchard ecosystems, it may have the following shortcomings:

[0179] Complex nonlinear relationship of ecological parameters, there may be nonlinear relationship between the ecological data of the orchard; for example, the influence of excessive or insufficient soil moisture on nitrogen absorption may be parabolic rather than linear; the comprehensive influence of light and temperature and humidity on fruit sugar content has a threshold effect, and a single linear model may not accurately describe it.

[0180] Data dimension and feature interaction, there may be complex interaction between multiple parameters, a single variable is difficult to explain the target parameter independently, and a simple model is difficult to capture the interaction between these high-dimensional features; in addition, with the increase of data size and the improvement of real-time requirement, a simple model may not be able to effectively process massive data or adapt to dynamic changes.

[0181] The implementation method of introducing complex nonlinear model: expanding the data set, combining the collected data with the backstepping results to generate a new training data set, containing more dimensions and feature interactions; data standardization, standardizing or normalizing all parameters to eliminate dimensional differences and avoid the excessive influence of certain features on the model; feature engineering, through feature selection or generating new features, to improve the generalization ability of the model.

[0182] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application has various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for collecting information on the ecological environment of an orchard, characterized in that, The method comprises the following steps: Step S1, real-time collection of soil parameters, spatial environment parameters, pest parameters and fruit parameters of the orchard by deploying multiple collection devices, and aggregation to form an orchard ecological basic data set; Step S2, constructing a back-stepping model, indirectly calculating key ecological parameters including soil nutrients, fruit quality and pest spread trend by using the collected data, and verifying and optimizing the results through experiments; Step S3, comparing the back-stepping results with the target values of the orchard, analyzing the deviation sources, establishing a macro cause chain of the deviation and the environmental parameters through data matching and visualization tools, and intuitively presenting the problem causes and their spatial distribution; Step S4, determining the control direction according to the deviation analysis and cause chain results, and matching executable control measures; Step S5, monitoring the effect after the implementation of the control measures, verifying the control effect by comparing the data before and after the implementation, and using the feedback data for iterative optimization of the back-stepping model, so as to finally realize closed-loop management and automatic control; In step S2, the following sub-steps are further included: S2-1, indirectly calculating the key parameters that are difficult to measure according to the collected data, the parameters including soil nutrients, fruit quality and pest spread trend; Soil nutrients: nitrogen, phosphorus and potassium; Fruit quality: sugar content and maturity; Pest spread trend: whole orchard distribution; S2-2, establishing a mathematical logical relationship between the ecological collection data and the key parameters that are difficult to measure, and constructing a back-stepping model: soil nutrient model, fruit quality model and pest spread model; The soil nutrient model adopts a linear regression formula to establish a linear relationship between soil moisture, pH and nitrogen, phosphorus and potassium nutrient concentrations, specifically as formula (2) to formula (4): N=a1·(H2O)+b1·(pH)+c1 Formula (2) P=a2·(H2O)+b2·(pH)+c2 Formula (3) K=a3·(H2O)+b3·(pH)+c3 Formula (4) Wherein, N, P and K respectively represent the current concentrations of nitrogen, phosphorus and potassium in the soil, H2O is the water content of the soil, pH is the acid-base value of the soil, a1, a2 and a3 are model coefficients related to soil moisture, b1, b2 and b3 are model coefficients related to soil pH, and c1, c2 and c3 are intercept terms of the model; The fruit quality model uses light intensity, temperature and humidity, and fruit size change rate to establish a prediction relationship with sugar content and maturity, specifically as formula (5) to formula (6): T_H=0.7·T+0.3·H Formula (5) S=d1·(Lux)+d2·(T_H)+d3·(v)+d4 Formula (6) Wherein, T_H is the temperature and humidity index, T is the air temperature of the orchard, H is the air humidity; S is the sugar content of the fruit, Lux is the light intensity, v is the fruit change rate, d1, d2 and d3 are model coefficients related to light, temperature and humidity index and fruit change rate, and d4 is the intercept term of the model; The pest spread model uses trap data, wind speed, humidity and temperature to predict the spread trend of pests, specifically as formula (7): D(t+1)=D(t)·(1+r-c) Formula (7) Wherein, D(t) is the pest density at the current time t, D(t+1) is the pest density at the next time t+1, r is the pest reproduction rate, which is affected by temperature and humidity, c is the pest capture rate, and 1+r-c is the increase or decrease factor of the pest density; S2-3, the collected real-time data are taken as inputs and substituted into the constructed inverse model to calculate the predicted values of the key parameters; a small range area is randomly selected in the orchard, and the inverse parameters are actually measured or experimentally detected: soil nutrients: the concentrations of soil nitrogen, phosphorus, and potassium are detected and analyzed in the laboratory; fruit quality: the sugar content and maturity of the fruit are measured manually; pest density: the actual distribution data of the pests are manually recorded; The inverse results are compared with the small-range experimental data to calculate the error values, the model parameters are optimized for the parts with larger errors, and the model is retrained in combination with new data.

2. The method of collecting ecological environment information of an orchard according to claim 1, characterized in that: In step S1, the following sub-steps are further included: S1-1, according to the terrain features, planting plan, and irrigation zoning of the orchard, the orchard is divided into a plurality of preliminary management units, and the collection content and range are set, the collection content including: soil parameters, spatial environment parameters, pest parameters, and fruit parameters; Soil parameters: water content, pH; Spatial environment parameters: light intensity, temperature and humidity; Pest parameters: pest species, density; Fruit parameters: fruit size, color; Collection range: the preliminary division of the management unit as the arrangement range of the collection device; S1-2, the data collection device includes soil parameter monitoring equipment, pest monitoring equipment, and fruit monitoring equipment; Soil parameter monitoring equipment: soil moisture sensor, pH sensor, temperature and humidity sensor, and light intensity sensor; Pest monitoring equipment: intelligent insect trapping device; Fruit monitoring equipment: unmanned aerial vehicle and high-resolution camera; The equipment arrangement covers terrain types, vegetation differences, and the center and boundary areas of the orchard; according to the area and complexity of the orchard, the sensor density is set, 1 soil sensor is arranged per 100 square meters, and 1 light or temperature and humidity sensor is arranged per 500 square meters; according to historical data or observation focus, the arrangement is made in the areas where the environmental changes are significant or the problems are prone to occur, including low-lying areas, insufficient light areas, and high-pest areas; the equipment is close to the main road or work channel, which is easy to check and maintain; S1-3, data transmission, the collected data are uploaded to the cloud by using LoRa wireless technology, the cloud platform automatically classifies and stores the uploaded data, and forms a basic data set of the current ecological state of the orchard; The real-time data are simply aggregated and summarized to generate a preliminary state diagram of the ecological information of the orchard, including: soil moisture distribution diagram, light intensity distribution diagram, pest density distribution diagram, and fruit distribution and size change trend diagram; S1-4, based on the data collected in S1-3, the orchard is divided into more detailed management units. According to the soil moisture, pH, and electrical conductivity parameters, units with similar soil characteristics are divided. According to the light intensity and temperature and humidity, units with similar environmental conditions are divided. Combined with pest species and density data, high-pest areas are marked separately. According to the size, color, and distribution trend of the fruit, areas with similar yield characteristics are divided together. Combined with the terrain and historical management information; Using the clustering algorithm K-Means, similar areas are automatically divided according to the collected ecological parameters. The clustering results are imported into the GIS platform for visualization processing of the geographic spatial data of the orchard, as shown in formula (1): where J is the total sum of squared errors of clustering, x is a data point belonging to the i-th cluster, μ i is the center point of the i-th cluster, C i is the set of i-th cluster, and K is the number of clusters. Each management unit has a clear description of its ecological characteristics, including soil characteristics, pest risk, and fruit distribution. The division results are output as a management unit distribution map, a detailed table of ecological parameters for each unit, and a summary through the GIS platform.

3. The method of collecting ecological environment information of an orchard according to claim 1, characterized in that: In step S3, the following sub-steps are further included: S3-1, compare the key parameter values calculated by the inverse model with the target values of the orchard to analyze the deviation, which is the difference between the actual calculated value and the target value. The main manifestations are: soil nutrient deviation, fruit quality deviation, and pest density deviation. Analyze the cause chain of the deviation; The cause chain of low soil nitrogen: high soil moisture → serious nitrogen leaching → low nitrogen content; The cause chain of low fruit sugar content: insufficient light → low photosynthesis efficiency → insufficient sugar accumulation; The cause chain of high pest density: suitable temperature and humidity → rapid reproduction of pests → local spread of pests; S3-2, match the analyzed deviation with the environmental parameters of the orchard to establish a macro cause chain and understand the formation reasons of the deviation. Through data analysis and visualization tools, associate the deviation with the spatial distribution of environmental parameters; Use GIS tools to draw the cause chain distribution map of the orchard to show the spatial distribution relationship between the deviation and the environmental parameters, generate regional problem charts including light distribution map and pest high-risk area map, and form a visualization report of the deviation and environmental parameters; According to the matching and visualization results, sort out the logical relationship between the deviation and the cause chain, output the spatial distribution of the deviation data, the main macro cause chain of each regional deviation, and the key driving factors of the deviation.

4. The method of collecting ecological environment information of an orchard according to claim 1, characterized in that: In step S4, the following sub-steps are further included: S4-1, based on the results of deviation analysis and macro cause mapping, determine the control direction to narrow the gap between the key parameters and the target values. According to the deviation degree of the key parameters, including low soil nitrogen content, low fruit sugar content, and high pest density, determine the problems that need to be solved first. Combined with the results of macro cause analysis, determine the core driving factors that affect the deviation to ensure that the control direction meets the overall management goal of the orchard; S4-2, according to the control direction determined in S4-1, match specific executable measures to optimize the ecological parameters and meet the target value requirements of the orchard. The control measures are designed as follows: Soil nutrient regulation: increase soil nitrogen content, reduce irrigation frequency or use drip irrigation technology, reduce water loss; regularly apply slow-release nitrogen fertilizer in low areas; Fruit quality regulation: improve fruit sugar content, prune the tree crown, increase the light area of the fruit; spray potassium and magnesium element fertilizer on the leaf surface to promote sugar accumulation; Pest control: reduce pest density, install insect-proof nets or spray biological pesticides in high-incidence areas; introduce natural enemies for biological control; adjust the humidity level of the orchard to reduce the breeding conditions of pests.

5. The method of claim 1, wherein in step S5, further comprising the following sub-steps: S5-1, monitoring and regulating effect, after implementing the regulation measures, continuously collecting soil, fruit and pest data, dynamic changes of water and nutrients, improvement trend of fruit sugar content, reduction of pest density; S5-2, compare the core data before and after adjustment, including soil nitrogen content, fruit sugar content, pest density, and the change of target value, evaluate the effect of regulation measures; S5-3, input the feedback data after regulation into the backstepping model, modify the model parameters, introduce historical data and adjustment data to enrich the training set, gradually introduce more complex nonlinear model optimization backstepping logic, perfect the closed-loop control, realize automatic decision and regulation. ​

Images