Farmland irrigation control method and system

Abstract

The invention relates to the technical field of agricultural irrigation, and particularly discloses a farmland irrigation control method and system, and the method comprises the steps: constructing a multi-sensor fusion monitoring network, and collecting soil, weather and plant physiological data in real time; transmitting data to an edge computing node through LoRa wireless communication; utilizing a lightweight neural network model deployed by edge computing nodes to calculate and predict crop water demand and accurate irrigation volume in combination with evapotranspiration; the irrigation strategy is executed through the intelligent irrigation execution system; and remote monitoring and abnormity alarm are realized by means of a remote monitoring and management platform based on WebGIS. Precise and intelligent management of farmland irrigation is realized, the utilization rate of water resources and fertilizers can be effectively improved, irrigation abnormity can be timely handled, the labor management cost is reduced, and the growth requirements of crops are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural irrigation, and in particular to a farmland irrigation control method and system. Background Art

[0002] In agricultural production, irrigation is a key factor influencing crop yield and quality. Traditional irrigation methods often lack precision, often based on experience or fixed irrigation schedules. Dynamic adjustments based on soil moisture, meteorological conditions, and actual crop water requirements are difficult to implement. This can lead to water waste and adversely affect crop growth due to under- or over-irrigation.

[0003] Advances in sensor, communication, and artificial intelligence technologies have made intelligent farmland irrigation possible. However, existing intelligent irrigation systems still face shortcomings in comprehensive data collection, efficient data processing, and accurate irrigation decisions. For example, some systems rely solely on a single soil moisture sensor to make irrigation decisions, ignoring the impact of meteorological conditions and crop physiological status on water demand. Other systems employ centralized data processing, resulting in delayed data transmission and inefficient processing, making it difficult to respond to irrigation needs in a timely manner.

[0004] Therefore, a farmland irrigation control method and system has become an urgent problem to be solved. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a farmland irrigation control method and system. By building a multi-sensor fusion monitoring network, using edge computing for real-time data processing and decision-making, and combining it with an intelligent irrigation execution system, precise control of farmland irrigation can be achieved, water resource utilization efficiency can be improved, and healthy crop growth can be promoted.

[0006] To solve the above technical problems, the present invention provides a technical solution: a farmland irrigation control method, comprising the following steps:

[0007] S1. Build a multi-sensor fusion monitoring network using a distributed soil moisture sensor array, meteorological stations, and plant physiological sensors using time domain reflectometry to collect soil volumetric water content, meteorological parameters, and plant physiological indicators in real time.

[0008] S2, transmits the data collected by the multi-sensor fusion monitoring network to the edge computing node through LoRa wireless communication;

[0009] S3. Deploy a lightweight neural network model on edge computing nodes to predict crop water requirements and precise irrigation amounts through evapotranspiration calculations based on historical data and real-time sensor inputs, taking into account factors such as soil type, crop variety, and growth stage.

[0010] S4, using a zone-controlled solenoid valve network, pressure-compensated drip irrigation tape, and a water-fertilizer integrated module to implement the irrigation strategy;

[0011] S5. Through the WebGIS-based remote monitoring and management platform, farmland distribution, irrigation status and sensor data are displayed to achieve remote monitoring and control, and automatically alarm in case of abnormal situations.

[0012] Furthermore, the distributed soil moisture sensor array is evenly distributed in different areas of the farmland and penetrates into the soil at multiple depths of 10 cm-30 cm. The soil moisture sensors are calibrated every month or according to the actual use environment.

[0013] Furthermore, the meteorological parameters monitored in real time by the weather station include rainfall, light intensity, temperature, humidity and wind speed.

[0014] Furthermore, the plant physiological sensor includes a leaf water potential sensor using a pressure chamber method and a stem flow rate sensor using a heat pulse method, and the plant physiological sensor is installed at a representative position of the crop plant.

[0015] Furthermore, the lightweight neural network model uses a multi-layer perceptron and is trained and optimized based on soil moisture, meteorological parameters, plant physiological indicators and corresponding crop water requirement historical data under different soil types, crop varieties and growth stages.

[0016] Furthermore, the evapotranspiration is calculated using the modified Penman-Montes formula, as follows:

[0017]

[0018] Among them, ET0 is the reference crop evapotranspiration, R n is the net radiation, G is the soil heat flux, T is the average temperature, u2 is the wind speed at 2 meters, e s is the saturated water vapor pressure, e a is the actual water vapor pressure, Δ is the slope of the saturated water vapor pressure-temperature curve, and γ is the hygrometer constant;

[0019] Determine the crop coefficient K based on crop variety and growth stage c , calculate the actual crop evapotranspiration ET c :

[0020] ET c =K c ×ET0;

[0021] Among them, K c Dynamically adjust according to crop variety and growth stage.

[0022] Furthermore, the calculation method of the precise irrigation amount is as follows:

[0023] Input the real-time sensor data into the trained lightweight neural network model to obtain the initial predicted water demand W p ;

[0024] The initial predicted water requirement is corrected based on the actual crop evapotranspiration ETc to calculate the precise irrigation amount I:

[0025]

[0026] Among them, K s is the soil moisture correction coefficient, K b is the crop growth correction coefficient, and η is the irrigation system efficiency.

[0027] Furthermore, the water-fertilizer integration module monitors the soil nutrient content and the nutrient demand of crops during their growth stages through sensors, automatically controls the fertilizer injection system, mixes the fertilizer with irrigation water in the required proportion, and delivers the mixed fertilizer to the roots of the crops.

[0028] Furthermore, the remote monitoring and management platform displays the location, boundaries, and irrigation status of farmland areas in the form of maps and updates sensor data in real time in the form of charts;

[0029] The remote monitoring and management platform sets a pipeline pressure threshold and a normal range for sensor data. When it is detected that the pipeline pressure is lower than the threshold or the sensor data exceeds the normal range, an alarm message is sent via SMS or APP push, and the alarm area and alarm type are highlighted on the management interface.

[0030] The present invention also provides a farmland irrigation control system for implementing the above-mentioned farmland irrigation control method, comprising:

[0031] A multi-sensor fusion monitoring network, including a distributed soil moisture sensor array, weather stations, and plant physiological sensors, is used to collect soil volumetric water content, meteorological parameters, and plant physiological indicators in real time;

[0032] The data transmission module uses the LoRa wireless communication module to transmit the data collected by the multi-sensor fusion monitoring network to the edge computing node;

[0033] Edge computing nodes are used to deploy lightweight neural network models that predict crop water requirements and precise irrigation amounts through evapotranspiration calculations based on historical data and real-time sensor inputs, taking into account factors such as soil type, crop variety, and growth stage.

[0034] The intelligent irrigation execution system, including a zone-controlled solenoid valve network, pressure-compensated drip irrigation tape, and a water and fertilizer integration module, executes irrigation operations based on the irrigation strategy output by the edge computing node;

[0035] The remote monitoring and management platform displays farmland distribution, irrigation status, and sensor data based on WebGIS, enabling remote monitoring and control, and automatically alarming in the event of abnormal conditions.

[0036] The advantages of the present invention compared with the prior art are:

[0037] The present invention provides more comprehensive data support for irrigation decision-making by constructing a multi-sensor fusion monitoring network and comprehensively considering soil moisture, meteorological parameters and plant physiological indicators, thereby improving the accuracy of irrigation decision-making.

[0038] The present invention utilizes edge computing nodes for data processing and decision-making, avoiding the delay caused by data transmission to the cloud, and can respond to irrigation needs in real time, thereby improving irrigation efficiency.

[0039] The present invention adopts a lightweight neural network model and evaporation and transpiration calculation method, combined with factors such as soil type, crop variety, and growth stage, to accurately predict crop water demand and irrigation amount, thus achieving precise irrigation and effectively saving water resources.

[0040] The water-fertilizer integration module of the present invention can automatically control fertilizer injection according to the soil nutrient content and the nutrient demand of the crop growth stage, realize the synchronous supply of water and fertilizer, improve the fertilizer utilization rate, and promote crop growth.

[0041] The invention is based on a remote monitoring and management platform of WebGIS, which facilitates users to understand the farmland irrigation status and sensor data in real time, realizes remote monitoring and control, and promptly issues alarms in case of abnormal situations, thereby improving the intelligent level of farmland management. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of a farmland irrigation control method of the present invention.

[0043] Figure 2 It is a system block diagram of a farmland irrigation control system of the present invention. DETAILED DESCRIPTION

[0044] Various exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention.

[0045] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0046] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0047] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0048] The farmland irrigation control method and system of the present invention will be further described in detail below with reference to the accompanying drawings.

[0049] Combined with attachment Figure 1-2 , the present invention is introduced in detail.

[0050] A farmland irrigation control method, the specific steps are as follows:

[0051] 1. Construct a multi-sensor fusion monitoring network: A multi-sensor fusion monitoring network is constructed using a distributed soil moisture sensor array using time-domain reflectometry, a weather station, and plant physiological sensors. The distributed soil moisture sensor array is evenly distributed across different areas of the farmland and penetrates the soil at multiple depths, ranging from 10cm to 30cm. The soil moisture sensors are calibrated monthly or based on actual usage conditions to collect real-time soil volumetric moisture content. The weather station monitors meteorological parameters such as rainfall, light intensity, temperature, humidity, and wind speed in real time. Plant physiological sensors, including leaf water potential sensors using the pressure chamber method and stem flow rate sensors using the heat pulse method, are installed at representative locations on crop plants to collect plant physiological indicators.

[0052] 2. Data transmission: The data collected by the multi-sensor fusion monitoring network is transmitted to the edge computing node through LoRa wireless communication.

[0053] 3. Calculate and predict water requirements and irrigation amounts: A lightweight neural network model is deployed on edge computing nodes. This model uses a multi-layer perceptron and is trained and optimized based on soil moisture, meteorological parameters, plant physiological indicators, and historical data on crop water requirements for different soil types, crop varieties, and growth stages. Based on historical data and real-time sensor input, crop water requirements and precise irrigation amounts are predicted through evaporation and transpiration calculations, taking into account factors such as soil type, crop variety, and growth stage. Evaporation and transpiration are calculated using the modified Penman-Montes formula:

[0054] The calculation formula for reference crop evapotranspiration ET0 is:

[0055]

[0056] Among them, ET0 is the reference crop evapotranspiration, R n is the net radiation, G is the soil heat flux, T is the average temperature, u2 is the wind speed at 2 meters, e s is the saturated water vapor pressure, e a is the actual water vapor pressure, Δ is the slope of the saturated water vapor pressure-temperature curve, and γ is the hygrometer constant;

[0057] Determine the crop coefficient K based on crop variety and growth stage c , calculate the actual crop evapotranspiration ET c :

[0058] ET c =K c ×ET0;

[0059] Among them, K c Dynamically adjust according to crop variety and growth stage.

[0060] Input the real-time sensor data into the trained lightweight neural network model to obtain the initial predicted water demand W p ; Combined with the actual crop evapotranspiration ETc, the initial predicted water demand is corrected to calculate the precise irrigation amount I:

[0061]

[0062] Among them, K s is the soil moisture correction coefficient, K b is the crop growth correction coefficient, and η is the irrigation system efficiency.

[0063] 4. Execute irrigation strategies: An intelligent irrigation execution system, comprised of a zone-controlled solenoid valve network, pressure-compensated drip irrigation tape, and a water-fertilizer integration module, executes irrigation strategies. The water-fertilizer integration module uses sensors to monitor soil nutrient content and crop nutrient needs during growth stages. It automatically controls the fertilizer injection system, mixing fertilizer with irrigation water in the desired ratio and delivering it to the crop roots.

[0064] 5. Remote Monitoring and Management: A WebGIS-based remote monitoring and management platform displays farmland distribution, irrigation status, and sensor data, enabling remote monitoring and control. The platform displays farmland location, boundaries, and irrigation status on a map, and updates sensor data in real-time charts. Pipeline pressure thresholds and sensor data normal ranges are set. When pipeline pressure falls below the threshold or sensor data exceeds the normal range, an alarm is sent via SMS or app push notifications, and the alarm area and type are highlighted on the management interface.

[0065] Based on the above farmland irrigation control method, the present invention further provides a farmland irrigation control system, comprising:

[0066] Multi-sensor fusion monitoring network: includes distributed soil moisture sensor arrays, weather stations, and plant physiological sensors, which are used to collect soil volumetric moisture content, meteorological parameters, and plant physiological indicators in real time.

[0067] Data transmission module: uses LoRa wireless communication module to transmit data collected by the multi-sensor fusion monitoring network to the edge computing node.

[0068] Edge computing nodes: Used to deploy lightweight neural network models. Based on historical data and real-time sensor input, they calculate evapotranspiration to predict crop water requirements and precise irrigation amounts, taking into account soil type, crop variety, and growth stage.

[0069] Intelligent irrigation execution system: includes a zone-controlled solenoid valve network, pressure-compensated drip irrigation belts, and a water-fertilizer integrated module, which performs irrigation operations based on the irrigation strategy output by the edge computing node.

[0070] Remote monitoring and management platform: Based on WebGIS, it displays farmland distribution, irrigation status and sensor data, enables remote monitoring and control, and automatically issues alarms in case of abnormal situations.

[0071] The specific implementation process of the farmland irrigation control method and system of the present invention is as follows:

[0072] Example 1: Wheat Planting Farmland Irrigation

[0073] Building a monitoring network: Distributed soil moisture sensor arrays were evenly distributed across wheat fields, buried at soil depths of 10 cm, 20 cm, and 30 cm, and calibrated monthly. A weather station was installed to monitor real-time meteorological parameters such as rainfall, light intensity, temperature, humidity, and wind speed. Leaf water potential sensors and stem flow rate sensors were installed on representative wheat plants to collect plant physiological indicators.

[0074] Data Transmission and Processing: Collected data is transmitted to edge computing nodes via LoRa wireless communication. A trained and optimized multi-layer perceptron lightweight neural network model is deployed at the edge computing node. Based on historical data and real-time sensor input, combined with factors such as wheat growth stage and soil type, a modified Penman-Montes formula is used to calculate evapotranspiration, thereby predicting wheat water requirements and precise irrigation amounts.

[0075] Irrigation execution: The intelligent irrigation execution system controls the irrigation area through a zone-controlled solenoid valve network based on the irrigation strategy output by the edge computing node, and uses pressure-compensated drip irrigation belts for precise irrigation. At the same time, the water and fertilizer integration module mixes fertilizer with irrigation water and delivers it to the roots of the wheat according to soil nutrients and wheat growth needs.

[0076] Remote Monitoring: The WebGIS-based remote monitoring and management platform displays farmland distribution, irrigation status, and sensor data in real time. When abnormal pipeline pressure or sensor data outside normal range is detected, the platform sends an alert via SMS and app, and highlights the alarm area and type on the management interface for timely response.

[0077] Example 2: Irrigation of corn-growing farmland in the north

[0078] The farmland covers an area of ​​50 mu, the soil type is loam, and the corn planted is in the jointing stage.

[0079] 1. Building a multi-sensor fusion monitoring network

[0080] In this corn field, a distributed array of soil moisture sensors was evenly distributed across different areas, with sensors buried at soil depths of 10 cm, 20 cm, and 30 cm. The soil moisture sensors were calibrated at the beginning of each month to ensure data accuracy. A weather station was installed to monitor various meteorological parameters in real time. Leaf water potential sensors and stem flow rate sensors were installed at appropriate locations on leaves and stems of representative corn plants. At a given moment, the multi-sensor fusion monitoring network collected the following data:

[0081] Soil moisture data: The volumetric water content of the soil at a depth of 10 cm is 20%, at a depth of 20 cm it is 22%, and at a depth of 30 cm it is 23%.

[0082] Meteorological parameters: rainfall is 0mm, light intensity is 800W / m 2 , average temperature T = 25℃, humidity is 60%, wind speed u2 = 2m / s at 2m height. The net radiation R is calculated. n =600W / m 2 , soil heat flux G=20W / m 2 , saturated water vapor pressure e s =31.7hPa, actual water vapor pressure e a =19hPa, and the slope of the saturated water vapor pressure-temperature curve Δ=0.4hPa / ℃ and the hygrometer constant γ=0.66hPa / ℃ are calculated by the relevant formula.

[0083] Plant physiological indicators: leaf water potential is -1.2MPa, stem flow rate is 0.8g / (cm 2 ·h).

[0084] 2. Data Transmission

[0085] Through the LoRa wireless communication module, the soil moisture data, meteorological parameters and plant physiological indicator data collected by the above multi-sensor fusion monitoring network are transmitted to the edge computing node in real time and stably.

[0086] 3. Calculate and forecast water demand and irrigation volume

[0087] Deploy the trained and optimized multi-layer perceptron lightweight neural network model on the edge computing node.

[0088] Calculate evapotranspiration:

[0089] The reference crop evapotranspiration ET0 is calculated according to the modified Penman-Montes formula:

[0090] ET0=(0.4×600-20)×(900 / (25+273)×2)×((31.7-19) / (0.4+0.66)) / (0.4+0.66),

[0091] The calculated reference crop evapotranspiration ET0 is 5.2 mm / d.

[0092] The crop coefficient K of corn at the jointing stage c After consulting relevant data and combining local actual conditions, it was determined to be 1.2, and the actual crop evapotranspiration ET was calculated. c :

[0093] ET c =K c ×ET0, K c =1.2 and ET0 = 5.2 mm / d are substituted into the formula to obtain the actual crop evapotranspiration ET. c It is 6.24mm / d.

[0094] Calculate exact irrigation amount:

[0095] Input the real-time sensor data into the trained lightweight neural network model to obtain the initial predicted water demand W p =120mm. Set the soil moisture correction coefficient K s =0.9, crop growth correction coefficient K b =1.1, irrigation system efficiency η = 0.85. Calculate the exact irrigation amount: W p =120mm, K s =0.9, K b Substituting η = 1.1 and η = 0.85 into the formula, we can calculate that the precise irrigation amount I is 139.8 mm.

[0096] 4. Implement irrigation strategies

[0097] The intelligent irrigation execution system receives the irrigation strategy output by the edge computing node, which precisely irrigates the field at a rate of 139.8 mm. A zone-controlled solenoid valve network activates irrigation channels in corresponding areas based on the field's zoning. Pressure-compensated drip irrigation tape ensures a stable flow of irrigation water to the corn roots under varying terrain and pressure conditions. The integrated water and fertilizer module automatically controls the fertilizer injection system based on sensor-monitored soil nutrient content (e.g., 80 mg / kg nitrogen, 30 mg / kg phosphorus, and 100 mg / kg potassium) and corn's nutrient requirements during the jointing phase (nitrogen requirements increase, while phosphorus and potassium requirements remain relatively stable). This mixes nitrogen, phosphorus, and potassium fertilizers with irrigation water in an appropriate ratio (e.g., nitrogen:phosphorus:potassium = 3:1:2) and precisely delivers them to the crop's roots.

[0098] 5. Remote Monitoring and Management

[0099] The WebGIS-based remote monitoring and management platform clearly displays the regional location, boundaries, and current irrigation status of the corn-growing farmland in the form of a map (showing that each sub-area is irrigating according to the calculated irrigation volume), and updates the data collected by the sensors in real time in the form of charts. The pipeline pressure threshold is set to 0.2MPa, and the sensor data is within the normal range (such as the normal range of soil moisture is 15%-30%, the normal range of temperature is 15℃-35℃, etc.). During the irrigation process, if the pipeline pressure of a certain sub-area is detected to be 0.15MPa at a certain moment, which is lower than the set threshold, the platform immediately sends an alarm message to the management personnel via SMS and APP push, and highlights the alarm area and alarm type (pipeline pressure is too low) in a striking color and flashing effect on the management interface, so that the management personnel can take timely measures to carry out maintenance and ensure the normal operation of the irrigation system.

[0100] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. A farmland irrigation control method, characterized in that: The following steps are involved: S1. Build a multi-sensor fusion monitoring network using a distributed soil moisture sensor array, meteorological stations, and plant physiological sensors using time domain reflectometry to collect soil volumetric water content, meteorological parameters, and plant physiological indicators in real time. S2, transmits the data collected by the multi-sensor fusion monitoring network to the edge computing node through LoRa wireless communication; S3. Deploy a lightweight neural network model on edge computing nodes to predict crop water requirements and precise irrigation amounts through evapotranspiration calculations based on historical data and real-time sensor inputs, taking into account factors such as soil type, crop variety, and growth stage. S4, using a zone-controlled solenoid valve network, pressure-compensated drip irrigation tape, and a water-fertilizer integrated module to implement the irrigation strategy; S5. Through the WebGIS-based remote monitoring and management platform, farmland distribution, irrigation status and sensor data are displayed to achieve remote monitoring and control, and automatically alarm in case of abnormal situations.

2. A farmland irrigation control method according to claim 1, characterized in that: The distributed soil moisture sensor array is evenly distributed in different areas of the farmland and penetrates into the soil at multiple depths of 10 cm to 30 cm. The soil moisture sensors are calibrated monthly or according to the actual use environment.

3. A farmland irrigation control method according to claim 2, characterized in that: The meteorological parameters monitored in real time by the weather station include rainfall, light intensity, temperature, humidity and wind speed.

4. A farmland irrigation control method according to claim 3, characterized in that: The plant physiological sensor comprises a leaf water potential sensor using the pressure chamber method principle and a stem flow rate sensor using the heat pulse method. The plant physiological sensor is installed at a representative position of the crop plant.

5. The farmland irrigation control method according to claim 4, characterized in that: The lightweight neural network model uses a multi-layer perceptron and is trained and optimized based on soil moisture, meteorological parameters, plant physiological indicators under different soil types, crop varieties, and growth stages, as well as corresponding historical data on crop water requirements.

6. A farmland irrigation control method according to claim 5, characterized in that: The evapotranspiration is calculated using the modified Penman-Montes formula as follows: Among them, ET0 is the reference crop evapotranspiration, R n is the net radiation, G is the soil heat flux, T is the average temperature, u2 is the wind speed at 2 meters, e s is the saturated water vapor pressure, e a is the actual water vapor pressure, Δ is the slope of the saturated water vapor pressure-temperature curve, and γ is the hygrometer constant; Determine the crop coefficient K based on crop variety and growth stage c , calculate the actual crop evapotranspiration ET c : AND c =K c ×ET0; Among them, K c Dynamically adjust according to crop variety and growth stage.

7. A farmland irrigation control method according to claim 6, characterized in that: The calculation method of the precise irrigation amount is as follows: Input the real-time sensor data into the trained lightweight neural network model to obtain the initial predicted water demand W p ; The initial predicted water requirement is corrected based on the actual crop evapotranspiration ETc to calculate the precise irrigation amount I: Among them, K s is the soil moisture correction coefficient, K b is the crop growth correction coefficient, and η is the irrigation system efficiency.

8. The farmland irrigation control method according to claim 7, characterized in that: The water-fertilizer integration module monitors the soil nutrient content and the nutrient demand of crops during their growth stages through sensors, automatically controls the fertilizer injection system, mixes fertilizer and irrigation water in the required proportion, and delivers the mixed mixture to the crop roots.

9. The farmland irrigation control method according to claim 8, characterized in that: The remote monitoring and management platform displays the location, boundaries, and irrigation status of farmland areas in the form of maps and updates sensor data in real time in the form of charts; The remote monitoring and management platform sets a pipeline pressure threshold and a normal range for sensor data. When it is detected that the pipeline pressure is lower than the threshold or the sensor data exceeds the normal range, an alarm message is sent via SMS or APP push, and the alarm area and alarm type are highlighted on the management interface.

10. A farmland irrigation control system, used to implement the farmland irrigation control method according to any one of claims 1 to 9, characterized in that: include: A multi-sensor fusion monitoring network, including a distributed soil moisture sensor array, weather stations, and plant physiological sensors, is used to collect soil volumetric water content, meteorological parameters, and plant physiological indicators in real time; The data transmission module uses the LoRa wireless communication module to transmit the data collected by the multi-sensor fusion monitoring network to the edge computing node; Edge computing nodes are used to deploy lightweight neural network models that predict crop water requirements and precise irrigation amounts through evapotranspiration calculations based on historical data and real-time sensor inputs, taking into account factors such as soil type, crop variety, and growth stage. The intelligent irrigation execution system, including a zone-controlled solenoid valve network, pressure-compensated drip irrigation tape, and a water and fertilizer integration module, executes irrigation operations based on the irrigation strategy output by the edge computing node; The remote monitoring and management platform displays farmland distribution, irrigation status, and sensor data based on WebGIS, enabling remote monitoring and control, and automatically alarming in the event of abnormal conditions.

Images