An Internet of Things-based integrated water and fertilizer control system and method

Abstract

This invention relates to the field of water and fertilizer control technology, and particularly to an integrated water and fertilizer control system and method based on the Internet of Things (IoT). The method includes the following steps: acquiring measured environmental data, historical environmental data, and regional geographic information of a farmland area; performing spatial modeling and time-series learning on the farmland area to construct a digital twin model of the farmland; using the digital twin model to simulate the state of areas without sensor deployment, obtaining virtual sensor node data; collecting multispectral image data of crops and extracting crop physiological characteristic data; and constructing a crop physiological state model using the measured environmental data, virtual sensor node data, and crop physiological characteristic data to generate crop physiological state assessment data. This invention, through multi-source data fusion and three-dimensional visualization integration, achieves data-driven precise control and intelligent management of the entire process in an integrated water and fertilizer system, comprehensively improving the intelligence level of irrigation and fertilization and resource utilization efficiency.

Description

Technical Field

[0001] This invention relates to the field of water and fertilizer control technology, and in particular to an integrated water and fertilizer control system and method based on the Internet of Things. Background Technology

[0002] Traditional water and fertilizer systems rely heavily on manual experience or timed control, making it difficult to adapt to the dynamic needs of crop growth. The introduction of IoT technology has enabled agricultural production to possess environmental sensing, remote control, and data-driven capabilities, propelling integrated water and fertilizer management towards intelligence and automation. Current advanced technologies generally integrate soil moisture, meteorological, and water quality sensors, controlling irrigation and fertilization processes through preset rules. However, these methods still suffer from insufficient data coverage, delayed response mechanisms, and coarse environmental modeling. Especially in areas with limited sensor distribution or complex farmland terrain, comprehensive environmental conditions are difficult to perceive, leading to biased control strategies and impacting water and fertilizer utilization efficiency.

[0003] Although existing integrated water and fertilizer technology has achieved a certain level of intelligence, it still has the following shortcomings: First, it lacks accurate perception of areas without sensor deployment, resulting in imprecise control of irrigation and fertilization range; second, it lacks in-depth modeling of crop physiological state, relying solely on soil parameters to judge crop needs, leading to inaccurate fertilizer allocation decisions; third, the system is mostly rule-driven, lacking dynamic response mechanisms based on machine learning and expert models; and fourth, it lacks visualization support, making it difficult for operation and maintenance personnel to grasp the overall status of farmland in real time. Summary of the Invention

[0004] Therefore, it is necessary for the present invention to provide an integrated water and fertilizer control system and control method based on the Internet of Things to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above objectives, an integrated water and fertilizer control method based on the Internet of Things includes the following steps:

[0006] Step S1: Obtain measured environmental data, historical environmental data, and regional geographic information of the farmland area, and perform spatial modeling and time series learning on the farmland area to construct a digital twin model of the farmland; use the digital twin model of the farmland to simulate the state of the area without sensor deployment to obtain virtual sensor node data;

[0007] Step S2: Collect multispectral image data of crops and extract crop physiological characteristic data; construct a crop physiological state model using environmental measured data, virtual sensor node data and crop physiological characteristic data, and generate crop physiological state assessment data;

[0008] Step S3: Collect water quality monitoring data in real time; dynamically adjust the fertilizer-liquid ratio based on crop physiological status assessment data and water quality monitoring data, and control the operation of multi-channel fertilizer mixing pumps to generate fertilizer mixing control parameters;

[0009] Step S4: Determine the control decision data for irrigation duration, flow rate, and fertilizer application based on virtual sensor node data and fertilizer application control parameters;

[0010] Step S5: Drive the solenoid valve, irrigation pump and fertilizer pump to perform irrigation and fertilization operations based on the control decision data, and generate fertilization and irrigation execution log data;

[0011] Step S6: Based on the virtual sensor node dataset, crop physiological status assessment data, water quality monitoring data, and fertilization and irrigation execution log data, construct a 3D visualized smart farm twin dashboard and generate a twin visual data view.

[0012] This invention overcomes the key problems of insufficient data coverage and lag in traditional water and fertilizer systems by introducing multi-source data fusion, digital twin reconstruction, multispectral sensing, dynamic water quality control, and three-dimensional visualization integration. It achieves continuous sensing and dynamic completion of farmland environmental conditions, accurately reconstructing regional environmental distribution even when sensors cannot be fully deployed, effectively improving the spatial precision of control. Combined with crop canopy multispectral image analysis, it can obtain intuitive parameters of crop growth status, avoiding reliance solely on soil data to determine crop nutrient requirements, thereby improving the targeting and scientific nature of fertilizer-solution ratios. By linking crop health indicators with water quality status to adjust fertilization plans, it achieves simultaneous optimization of irrigation and fertilization decisions. This system ensures precise input of water and fertilizer resources; control parameters at each stage are generated based on real-time data, avoiding rigid responses caused by manually set rules and enabling the system to self-adjust; control information throughout the process is recorded and integrated in real time, further supporting subsequent operational traceability and strategy evaluation; simultaneously, by constructing a farm visualization dashboard in a 3D scene, environmental information, crop status, operation paths, and historical operation data are integrated and displayed, allowing managers to intuitively grasp the overall operational status of the farmland and conduct interactive queries, improving system operation and maintenance efficiency and management transparency. Overall, it promotes the upgrade of the integrated water and fertilizer system from rule-triggered to data-driven, significantly improving the intelligence level and resource utilization efficiency of agricultural irrigation and fertilization.

[0013] Preferably, the present invention also provides an IoT-based integrated water and fertilizer control system for executing the above-described IoT-based integrated water and fertilizer control method, wherein the IoT-based integrated water and fertilizer control system includes:

[0014] The twin modeling module is used to acquire measured environmental data, historical environmental data, and regional geographic information of farmland areas, and to perform spatial modeling and time series learning on farmland areas to construct a digital twin model of farmland; the digital twin model of farmland is used to simulate the state of areas without sensor deployment to obtain virtual sensor node data;

[0015] The crop assessment module is used to collect multispectral image data of crops and extract crop physiological characteristic data; it uses environmental measured data, virtual sensor node data and crop physiological characteristic data to construct a crop physiological state model and generate crop physiological state assessment data.

[0016] The water and fertilizer regulation module is used to collect water quality monitoring data in real time; dynamically adjust the fertilizer-liquid ratio based on crop physiological status assessment data and water quality monitoring data, control the operation of multi-channel fertilizer mixing pumps, and generate fertilizer mixing control parameters.

[0017] The intelligent decision-making module is used to determine the control decision data of irrigation duration, flow rate and fertilizer application based on virtual sensor node data and fertilizer application control parameters;

[0018] The execution control module is used to drive the solenoid valves, irrigation pumps and fertilizer pumps to perform irrigation and fertilization operations based on control decision data, and generate fertilization and irrigation execution log data.

[0019] The visual twin module is used to build a 3D visualized smart farm twin dashboard based on virtual sensor node datasets, crop physiological status assessment data, water quality monitoring data, and fertilization and irrigation execution log data, generating a twin visual data view.

[0020] This invention, through the collaborative work of the aforementioned modules, forms a complete closed-loop intelligent water and fertilizer integrated control system. This enhances the system's adaptability and control precision to the complex dynamic environment of farmland, achieving fully automated management from data acquisition, status assessment, decision analysis to execution control and visualization. The spatial and temporal correlation analysis capabilities of the twin modeling module effectively compensate for sensor coverage blind spots, ensuring the integrity and continuity of environmental status data across the entire farmland area. The multispectral information extraction and physiological characteristic analysis of the crop assessment module enable real-time perception of dynamic changes in crop nutrient requirements, providing a scientific basis for precision fertilization. The cooperation between the water and fertilizer regulation module and the intelligent decision-making module ensures that water and fertilizer supply dynamically adapts to the real-time needs of crops, significantly improving nutrient utilization efficiency and crop growth balance. The execution control module ensures the stable and precise operation of each control unit, achieving synchronous and efficient execution of irrigation and fertilization processes. The visual twin module provides a comprehensive and intuitive display of farm status, facilitating rapid understanding of the overall production situation, tracing historical operation processes, and conducting abnormal status diagnosis. Throughout the system's operation, the data flow and logical connection between modules continuously improve the automation, intelligence, and efficient management level of the smart farm. Attached Figure Description

[0021] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0022] Figure 1This is a flowchart illustrating the steps of an integrated water and fertilizer control system and method based on the Internet of Things according to the present invention.

[0023] Figure 2 for Figure 1 A detailed flowchart of step S1;

[0024] Figure 3 for Figure 1 A detailed flowchart of step S2. Detailed Implementation

[0025] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0026] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0027] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0028] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a water and fertilizer integration control method based on the Internet of Things, the method comprising the following steps:

[0029] Step S1: Obtain measured environmental data, historical environmental data, and regional geographic information of the farmland area, and perform spatial modeling and time series learning on the farmland area to construct a digital twin model of the farmland; use the digital twin model of the farmland to simulate the state of the area without sensor deployment to obtain virtual sensor node data;

[0030] Step S2: Collect multispectral image data of crops and extract crop physiological characteristic data; construct a crop physiological state model using environmental measured data, virtual sensor node data and crop physiological characteristic data, and generate crop physiological state assessment data;

[0031] Step S3: Collect water quality monitoring data in real time; dynamically adjust the fertilizer-liquid ratio based on crop physiological status assessment data and water quality monitoring data, and control the operation of multi-channel fertilizer mixing pumps to generate fertilizer mixing control parameters;

[0032] Step S4: Determine the control decision data for irrigation duration, flow rate, and fertilizer application based on virtual sensor node data and fertilizer application control parameters;

[0033] Step S5: Drive the solenoid valve, irrigation pump and fertilizer pump to perform irrigation and fertilization operations based on the control decision data, and generate fertilization and irrigation execution log data;

[0034] Step S6: Based on the virtual sensor node dataset, crop physiological status assessment data, water quality monitoring data, and fertilization and irrigation execution log data, construct a 3D visualized smart farm twin dashboard and generate a twin visual data view.

[0035] In this embodiment of the invention, reference is made to Figure 1 The diagram shown is a flowchart illustrating the steps of an IoT-based integrated water and fertilizer control method according to the present invention. In this example, the IoT-based integrated water and fertilizer control method includes the following steps:

[0036] Step S1: Obtain measured environmental data, historical environmental data, and regional geographic information of the farmland area, and perform spatial modeling and time series learning on the farmland area to construct a digital twin model of the farmland; use the digital twin model of the farmland to simulate the state of the area without sensor deployment to obtain virtual sensor node data;

[0037] In this embodiment of the invention, firstly, soil sensors, air environment monitoring devices, and light acquisition modules deployed in farmland areas are used to acquire six types of environmental data: soil temperature, soil moisture, soil conductivity, air temperature, air humidity, and light intensity. The acquisition frequency for each type of data is fixed at once every 30 minutes, and the data format is a structured numerical record corresponding to a timestamp. Next, the same six types of environmental monitoring data from the past five years are retrieved from an established historical database of the farmland area. Data is extracted at a daily time granularity and segmented by month to form a historical environmental data set. Subsequently, a drone equipped with a multispectral camera system and an RTK high-precision positioning device is used to perform a full-coverage low-altitude flight scan of the farmland area, acquiring geographical features such as regional topography, boundaries, irrigation ditches, and changes in surface elevation. This data is then processed through rasterization and spatial resampling. Geographic information is standardized to a spatial resolution of 0.5 meters. After the above data collection is completed, a three-dimensional data interpolation method and time-smoothing weighted processing are used to superimpose and fuse the measured data and historical data according to spatial coordinates and time axis, forming a spatial-temporal joint numerical matrix at each sampling node, which constitutes the basic data of farmland spatial structure and dynamic changes. Based on the above fused data, the estimated values ​​of soil temperature, humidity, electrical conductivity and air temperature, humidity and light in areas without sensor deployment are calculated according to a fixed grid point array (e.g., one node every 10 meters). The estimation adopts a distance-weighted average method based on known nodes, where the weighting factor is set as a function inversely proportional to the square of the distance, and a threshold radius of 50 meters is set, with the weight being zero if it exceeds 50 meters. Finally, an equally spaced virtual node dataset is formed in the entire area. The data of each virtual node consists of the same six indicators as the measured nodes.

[0038] Step S2: Collect multispectral image data of crops and extract crop physiological characteristic data; construct a crop physiological state model using environmental measured data, virtual sensor node data and crop physiological characteristic data, and generate crop physiological state assessment data;

[0039] In this embodiment of the invention, firstly, a multispectral imaging device installed on an agricultural automated track vehicle is used to take parallel strip-shaped images of the crop canopy in the farmland area. The imaging device used includes the visible light band (wavelength range of 400–700 nm), the red-edge band (wavelength range of 700–740 nm), and the near-infrared band (wavelength range of 740–900 nm), with an imaging resolution of 0.1 meters. The imaging frequency is once every 7 days. All images undergo radiometric correction and geometric registration processing after acquisition. Subsequently, the acquired multispectral images are processed by band segmentation, based on red light and... The Normalized Difference Vegetation Index (NDVI) was calculated using the near-infrared band as: (Near-infrared band value - Red band value) / (Near-infrared band value + Red band value). The calculation result was then linked to geographic coordinates to generate a raster map. Simultaneously, the mean and standard deviation of RGB color values ​​were extracted from the visible light image to determine the distribution of leaf color intensity. Furthermore, the curvature variation of crop leaf edges was identified using image edge detection methods. The degree of leaf curling was graded by calculating the second derivative of the edge segments, with curling levels ranging from 0 to 3, corresponding to no curling, slight curling, moderate curling, and... Severe curling; in addition, for the distribution of reflectance of each band on each pixel, the variation range of the difference between the reflectance of the five bands is statistically analyzed to extract the reflectance response gradient, and a crop physiological characteristic dataset consisting of NDVI index value, color statistical value, curling degree level and reflectance gradient is constructed; then, this dataset is uniformly time-calibrated and spatially matched with the environmental measured data and virtual sensor node data obtained in step S1, and the data is aligned according to the plot number and collection time to form a unified data structure for each plot; subsequently, multi-temporal crop growth in each plot is analyzed. The characteristic numerical sequence is processed by moving average (window width set to 3 days) to reduce the impact of short-term fluctuations. The difference between the current NDVI value, color mean, curl level and historical mean is compared at fixed time intervals (every 7 days). If the NDVI value decreases by more than 10%, the yellowing of the color mean exceeds 15%, and the curl level increases by more than 1 level, and any two of the above conditions are met at the same time, the current crop status of the area is marked as "abnormal", otherwise it is marked as "normal". Finally, crop physiological status assessment data containing plot number, timestamp and status label fields are generated.

[0040] Step S3: Collect water quality monitoring data in real time; dynamically adjust the fertilizer-liquid ratio based on crop physiological status assessment data and water quality monitoring data, and control the operation of multi-channel fertilizer mixing pumps to generate fertilizer mixing control parameters;

[0041] In this embodiment of the invention, an integrated online water quality monitoring module, comprising an electrode-type pH probe, a conductivity sensor, and a TDS sensor, is first installed at the inlet of farmland water source. The module collects pH value, conductivity (in μS / cm), and total dissolved solids concentration (TDS) (in mg / L) every 10 minutes. The collection range is limited to pH value 5.5–8.5, conductivity 0–2000 μS / cm, and TDS... Water quality data, ranging from 0 to 1200 mg / L, is collected and transmitted to the central control unit via an RS485 bus. Next, based on the crop physiological status assessment data generated in step S2, each plot is classified as "normal" or "abnormal" according to the status field. Then, the water quality data and crop status data are associated by plot number. Subsequently, the required fertilizer concentration adjustment direction is determined based on the plot's crop status and water quality indicators. Specifically, if the plot status is "abnormal" and the pH is less than 6.0 or greater than 7.5, the phosphorus solution ratio is increased by 10%; if the conductivity is less than 400 μS / cm, the nitrogen solution ratio is increased by 15%; if the TDS is greater than 900 mg / L, the potassium solution ratio is decreased by 10%. Each adjustment is recorded as a percentage, and the adjustment results are used to generate fertilizer setting parameters for each plot. The system comprises a set of parameters, including the ratios of nitrogen, phosphorus, and potassium solutions (totaling 100%) and the target concentrations (in mg / L) for each type of fertilizer solution. The target concentration ranges are set as follows: nitrogen solution 100–300 mg / L, phosphorus solution 50–150 mg / L, and potassium solution 80–250 mg / L. Based on these parameters, the system drives the operation control unit of the electronically controlled multi-channel fertilizer mixing pump. The PWM signal controls the opening time ratio of the solenoid valves in each fertilizer solution channel, ensuring that the actual output fertilizer solution concentration matches the set value. The fertilizer liquid mixing process is precisely measured by a three-channel flow meter. The duration of a single fertilizer mixing task does not exceed 20 minutes, and the output pressure is stably controlled between 0.3–0.5 MPa. Finally, the system packages the original data of each operation, including the plot number, fertilizer solution ratio, flow rate, duration, and water quality, to generate the fertilizer mixing control parameters.

[0042] Step S4: Determine the control decision data for irrigation duration, flow rate, and fertilizer application based on virtual sensor node data and fertilizer application control parameters;

[0043] In this embodiment of the invention, firstly, the virtual sensor node data generated in step S1 is spatially interpolated at 0.5-meter intervals to obtain the current values ​​and 72-hour trends of three indicators—soil moisture (in %), soil electrical conductivity (in mS / cm), and soil temperature (in °C)—at various locations in the farmland. The soil moisture threshold is set between 30% and 70%, the electrical conductivity threshold is set between 0.2 and 2.5 mS / cm, and the temperature threshold is set between 10 and 35 °C. Secondly, combining the nitrogen, phosphorus, and potassium solution concentrations and the total fertilizer solution ratio set for each plot in the fertilizer application control parameters in step S3, the target fertilizer application amount for each plot is uniformly converted into fertilizer solution volume per square meter (in L / m²). 2 The conversion method is to divide the total fertilizer requirement (in liters) by the plot area (in m²). 2 Then, for each plot, the irrigation flow rate (in L / min) is determined proportionally based on the proportion of pixels in the virtual nodes within that plot where soil moisture is below 35% or electrical conductivity is below 0.3 mS / cm. If the proportion of low-humidity areas is greater than 50%, the irrigation flow rate is set to 120% of the standard value; if the proportion is between 30% and 50%, it is set to 110% of the standard value; otherwise, it is set to the standard value. The standard flow rate is fixed at 50 L / min based on the pipe network diameter and the irrigation pump's output capacity. Then, based on the total fertilizer application and the set irrigation flow rate, the irrigation duration for each plot is calculated, using the total fertilizer liquid volume (in L / min). The fertilizer concentration is calculated by dividing the irrigation flow rate (in L / min) by the total amount of fertilizer. If the calculated value exceeds 30 minutes, the total amount is forcibly divided into two applications with a 3-hour interval to avoid excessive water pressure on the crop roots in a short period of time. At the same time, the fertilizer concentration is adjusted according to the ratio between the pixel area with soil conductivity below 0.3 mS / cm and the fertilizer concentration. If the proportion of this area exceeds 60%, the total amount of fertilizer solution is increased by 10%. All control parameters are recorded on a plot-by-plot basis, including plot number, set irrigation duration (in minutes), set irrigation flow rate (in L / min), and total amount of fertilizer solution (in L), and finally, control decision data is generated.

[0044] Step S5: Drive the solenoid valve, irrigation pump and fertilizer pump to perform irrigation and fertilization operations based on the control decision data, and generate fertilization and irrigation execution log data;

[0045] In this embodiment of the invention, firstly, based on the control decision data generated in step S4, the irrigation control center receives three instruction parameters corresponding to each plot number: irrigation duration (in minutes), irrigation flow rate (in L / min), and total fertilizer solution volume (in L). The corresponding solenoid valve address mapping table is then retrieved, and the PLC controller issues an opening command to the solenoid valve at the specified address using the Modbus communication protocol. The solenoid valve response time is controlled within 1 second. The solenoid valve's on / off action is fed back to the controller in real time, and the irrigation pump is started after the controller confirms the status is correct. The irrigation pump is a variable frequency constant pressure pump with a rated power of 1.5kW. After startup, the pump speed is adjusted by the frequency converter according to the flow control parameters to stabilize the output pressure at 0.4MPa. A flow meter is connected to the outlet to monitor flow changes in real time and samples every 10 seconds. If the results of three consecutive samplings deviate from the target flow value by more than ±5%, the speed is automatically adjusted to compensate for the error. Poor; After the irrigation pump runs for 5 minutes, the three-channel electric fertilizer mixing pump is started through the relay control device. The opening of the flow valve of each channel is controlled according to the fertilizer-liquid ratio set in step S3. The response time of the flow valve is 500 milliseconds. After the fertilizer solution is injected into the main pipe, it is completely mixed with the irrigation water flow. The total output is detected by the mixing flow meter. If the mixing concentration is lower than the set value, the flow rate of the channel is increased by 10% by the booster pump. During the fertilization process, the data of all solenoid valves, electric pumps, flow meters and pressure sensors are recorded by the real-time monitoring module of the control center, including the on and off time, operating current, flow value, pressure curve and fertilizer ratio. After the task is completed, the fertilization and irrigation execution log data file is automatically generated. Each record includes the plot number, irrigation start time, end time, total fertilizer amount (in L), irrigation duration (in minutes), fertilizer ratio of each channel, solenoid valve opening time, electric pump operating status and flow meter reading sequence.

[0046] Step S6: Based on the virtual sensor node dataset, crop physiological status assessment data, water quality monitoring data, and fertilization and irrigation execution log data, construct a 3D visualized smart farm twin dashboard and generate a twin visual data view.

[0047] In this embodiment of the invention, firstly, the virtual sensor node dataset generated in step S1 is imported into a GIS 3D geographic rendering engine. Coordinate matching and spatial mapping are performed on the soil temperature (°C), humidity (%), electrical conductivity (mS / cm), air temperature and humidity, and light intensity of each node. A basic environmental layer is constructed using a 0.5-meter grid. Each parameter is assigned an RGB value range using color gradient encoding, with soil humidity encoded using the blue channel and light intensity using the yellow channel. All parameter values ​​are updated every 10 minutes. Next, the crop physiological state assessment data generated in step S2 is associated with the plot number and timestamp. The canopy color distribution is rendered in the crop layer using the NDVI index range (0–1), and the canopy undulation is corresponding to the leaf curling level (0–3), achieving a 3D visual expression of crop growth status. A curling level of 0 corresponds to a smooth surface, and level 3 corresponds to a maximum undulation height of 0.5 meters. Subsequently, the pH, electrical conductivity, and TDS water quality monitoring data collected in step S3 are introduced, and a system is constructed at each water inlet node. The water quality layer segments the data in a time-series format, calculating the average value hourly. Changes in water quality indicators are coded with a color scale. pH values ​​deviating from the neutral range (6.5–7.5) are marked in red, and high TDS areas are enhanced with gray shading to indicate excessively high concentrations. The fertilization and irrigation execution log data generated in step S5 is then parsed into a task trajectory layer. Water flow paths are drawn based on the solenoid valve control time periods, dynamically displaying the water and fertilizer delivery direction in a linear animation, overlaid with annotation text indicating fertilizer ratios and amounts. The trajectory animation plays frame-by-frame on a plot-by-plot basis, with a time axis granularity set to 1 minute. These four layers are uniformly overlaid into the same 3D coordinate system using geographic coordinates and timestamps. A 3D visualized smart farm scene is generated using a WebGL rendering engine. Layer selection buttons, sliding time axes, and plot filtering conditions are set in the user interface. Finally, a 3D twin dashboard integrating environmental status, crop status, water quality indicators, and fertilization execution is constructed. A twin visual data view is generated using panoramic bird's-eye views, enlarged local views, and time-series graphs for continuous monitoring and backtracking by the water and fertilizer control system.

[0048] This invention overcomes the key problems of insufficient data coverage and lag in traditional water and fertilizer systems by introducing multi-source data fusion, digital twin reconstruction, multispectral sensing, dynamic water quality control, and three-dimensional visualization integration. It achieves continuous sensing and dynamic completion of farmland environmental conditions, accurately reconstructing regional environmental distribution even when sensors cannot be fully deployed, effectively improving the spatial precision of control. Combined with crop canopy multispectral image analysis, it can obtain intuitive parameters of crop growth status, avoiding reliance solely on soil data to determine crop nutrient requirements, thereby improving the targeting and scientific nature of fertilizer-solution ratios. By linking crop health indicators with water quality status to adjust fertilization plans, it achieves simultaneous optimization of irrigation and fertilization decisions. This system ensures precise input of water and fertilizer resources; control parameters at each stage are generated based on real-time data, avoiding rigid responses caused by manually set rules and enabling the system to self-adjust; control information throughout the process is recorded and integrated in real time, further supporting subsequent operational traceability and strategy evaluation; simultaneously, by constructing a farm visualization dashboard in a 3D scene, environmental information, crop status, operation paths, and historical operation data are integrated and displayed, allowing managers to intuitively grasp the overall operational status of the farmland and conduct interactive queries, improving system operation and maintenance efficiency and management transparency. Overall, it promotes the upgrade of the integrated water and fertilizer system from rule-triggered to data-driven, significantly improving the intelligence level and resource utilization efficiency of agricultural irrigation and fertilization.

[0049] Preferably, step S1 includes the following steps:

[0050] Step S11: Collect data on soil temperature, soil moisture, soil electrical conductivity, air temperature and humidity, and light intensity in the farmland area to obtain environmental measurement data;

[0051] Step S12: Obtain historical environmental data for the farmland area;

[0052] Step S13: Collect information on the topography, boundaries, and geographic features of the farmland area, and digitize it to obtain regional geographic information;

[0053] Step S14: Construct a farmland regional spatial model based on measured environmental data, historical environmental data, and regional geographic information to obtain a farmland spatial structure model;

[0054] Step S15: Construct a farmland environmental time series based on measured environmental data and historical environmental data to obtain a farmland time dynamic model;

[0055] Step S16: Construct a digital twin model of farmland based on the farmland spatial structure model and the farmland temporal dynamic model;

[0056] Step S17: Simulate the state of areas without sensor deployment based on the farmland digital twin model to obtain virtual sensor node data.

[0057] In this embodiment of the invention, wireless soil monitoring terminals and meteorological acquisition modules are first deployed in a farmland area at 20-meter intervals. The soil monitoring terminals are used to collect soil temperature (°C, range 0–40), soil moisture (%, range 10–90), and soil electrical conductivity (mS / cm, range 0.1–2.5). The meteorological acquisition module collects air temperature (°C, range -10–45), air humidity (%, range 10–95), and light intensity (lux, range 1000–100000). The data collection period is set to 10 minutes, and the data is transmitted via L... The oRa wireless transmission gateway aggregates and stores environmental measurement data in a data server. Subsequently, it retrieves historical meteorological data and soil moisture monitoring records from the past five years stored in the agricultural data archive system, aligning the recorded times to hourly units and unifying the format with the measured data to form a historical environmental dataset. Next, a drone equipped with a multispectral photography device and lidar is used to perform a low-altitude scan of the entire area, extracting geographic features such as elevation, slope, boundaries, ditches, canals, and roads. The raw point cloud data is then converted into a DEM (Digital Elevation Map) and vector layers, with a spatial resolution limited to 0.5 meters. All geographic information... Information was unified to the UTM coordinate system through projection coordinate transformation to form regional geographic information data. After data acquisition, environmental measurement data, historical environmental data, and regional geographic information were rasterized according to spatial location using coordinate overlay. Bilinear interpolation was used to calculate the average soil and meteorological values ​​for each 0.5-meter grid node, constructing a farmland spatial structure grid composed of six environmental elements and forming a farmland spatial structure model. Subsequently, trend fitting was performed on the measured values ​​of each sensor node over 30 days. A fixed-time moving average method (window width of 6 hours) was used to extract the rate of change and periodic variation amplitude of soil and meteorological elements. Dynamic data sequences are generated based on time indexes to construct a farmland time dynamic model. Based on spatial structure grids and time series change curves, each grid node is defined as a six-dimensional vector, with the six dimensions being current soil temperature, soil moisture, electrical conductivity, air temperature, air humidity, and light intensity, respectively. In areas without sensor deployment, an inverse distance weighting method is used for data estimation, with the weighting function defined as inversely proportional to the square of the distance and the weight cutoff radius limited to 30 meters. Six-dimensional virtual data points are generated in each undeployed area. Finally, the spatial environmental values ​​and estimated values ​​of all grid nodes are integrated to form a complete data atlas, serving as virtual sensor node data.

[0058] This invention integrates measured farmland environmental data, historical meteorological and soil moisture data, and high-precision geographic information to construct a farmland structural foundation covering both spatial and temporal dimensions. This foundation accurately reflects the distribution characteristics and changing patterns of the farmland's internal microenvironment, providing high-resolution and highly continuous environmental input for subsequent irrigation and fertilization control. By establishing a coupling system between spatial structure and temporal changes, the invention effectively enhances the completeness and dynamism of farmland state expression. Even when facing areas with complex terrain or missing data, the system can still reconstruct the environmental state based on surrounding multidimensional information, enhancing the system's environmental awareness. The digitized geographic element information supports precise correspondence between environmental data and spatial location, enabling refined control granularity. By generating virtual sensor node data, dynamic compensation for blind spots in sensor deployment is achieved, overcoming the problem of incomplete data coverage due to cost or terrain limitations in actual deployment. This ensures the continuity of environmental parameters across the entire field, providing high-quality underlying data support for precise decision-making and significantly improving the spatial perception capability and scientific decision-making of the entire water and fertilizer control system.

[0059] Preferably, step S2 includes the following steps:

[0060] Step S21: Collect visible light, red edge, and near-infrared multispectral images of the crop canopy to obtain crop multispectral image data;

[0061] Step S22: Extract NDVI index, leaf color, curling degree and spectral reflectance characteristics of crop based on crop multispectral image data to obtain crop physiological characteristic data;

[0062] Step S23: Based on the measured environmental data, virtual sensor node data and crop physiological characteristic data, a fusion model of crop growth environment and crop status is performed to obtain a crop physiological status model;

[0063] Step S24: Assess the current crop health and nutritional status based on the crop physiological state model to obtain crop physiological state assessment data.

[0064] In this embodiment of the invention, firstly, during the middle and late stages of crop growth, an automatic inspection device equipped with a multispectral camera is used to collect images of the crop canopy every 10 meters along the planting rows. Each image includes visible light (400–700 nm), red-edge (700–740 nm), and near-infrared (740–900 nm) wavelengths, with a resolution set at 1024 × 1024 pixels. The image acquisition time is fixed between 10:00 and 11:00 AM daily to avoid the influence of changes in the illumination angle on the spectral reflectance. Pixel-level normalized vegetation index (NDVI) is calculated for the red and near-infrared channels of the acquired images using the formula NDVI = (NI... The NDVI index is calculated as (R-Red) / (NIR+Red), and the resulting value is mapped to a grayscale image for storage. The mean and standard deviation of the NDVI index for each image are recorded in a structured table. Then, a color channel extraction tool is used to extract the main color tone of the leaf surface from the RGB channel data of the visible light image. The color of each pixel is converted to the HSV color space, and the proportion of pixels with hue values ​​between 90° and 150° is statistically analyzed to determine the degree of green health. This proportion is set as the leaf surface color index. Next, image edge detection is performed. The Canny operator is used to extract the contour of the leaf edge curve and calculate its curvature. If the mean edge curvature is greater than 0.25, it is marked as curl level 3. If it is between 0.25 and 0.25, it is marked as curl level 3. Values ​​between 0.1 and 0.25 are marked as Level 2, otherwise Level 1 or 0. This level indicates the degree of leaf curling. In each band channel, five-dimensional spectral reflectance indicators are extracted based on pixel reflectance variation trends: red light reflectance intensity, near-infrared reflectance intensity, red edge jump amplitude, mean reflectance difference, and maximum reflectance gradient. Finally, the NDVI index, leaf color index, curling level, and the five spectral reflectance values ​​are merged into complete crop physiological characteristic data. Crop physiological characteristic data from the same time period within the same plot are selected, and corresponding environmental measured data (including soil moisture, temperature, conductivity, light intensity, and air temperature and humidity) and virtual sensor node data are synchronized. The spatial location of each plot is matched by its plot number and paired by its timestamp. Environmental values ​​of neighboring sensor nodes are matched at the center point of each image and fused into single-point growth environment feature values ​​using a distance-weighted average method (radius set to 10 meters). All image records for each plot are analyzed. If the NDVI index is below 0.6 and the proportion of green pixels is below 70%, and the curl level is above 2, or if any two of the five spectral reflectance values ​​deviate from the historical average by more than 20%, the crop status for that period is assessed as "abnormal". Otherwise, it is assessed as "normal". Finally, the status assessment results, timestamps and key feature values ​​of each plot are summarized to form crop physiological status assessment data.

[0065] This invention acquires multispectral images encompassing visible light, red-edge, and near-infrared bands, enabling comprehensive perception of crop canopy reflectivity and precise extraction of crop growth, color changes, and leaf structure, providing a rich foundation for constructing physiological characteristic data. By extracting image features, it obtains indicators such as NDVI index, leaf color, and curling degree, which directly reflect crop photosynthetic intensity, nutrient absorption, and stress response, overcoming the shortcomings of traditional water and fertilizer systems that rely solely on soil data and cannot identify the true needs of crops. Integrating crop physiological characteristics with environmental monitoring data allows for the construction of a more targeted crop growth status expression system, extending water and fertilizer regulation beyond soil-side parameters to include crop health assessment, significantly improving the scientific rigor of crop nutrient diagnosis and the accuracy of regulatory decisions. The final crop physiological status assessment data possesses both timeliness and regionality, reflecting individual crop physiological dynamics and supporting differentiated fertilization strategies at the field scale, providing crucial decision-making support for site-specific and seedling-specific policies in precision agriculture.

[0066] Preferably, step S22 includes the following steps:

[0067] Step S221: Calculate the normalization of pixel values ​​in the red band and near-infrared band based on crop multispectral image data to obtain NDVI index image data;

[0068] Step S222: Extract color features from the visible light band of the crop multispectral image data to obtain crop leaf color data;

[0069] Step S223: Analyze the edge texture and deformation curvature of crop leaf areas based on crop multispectral image data, extract leaf curling degree features, and obtain leaf curling degree data;

[0070] Step S224: Perform statistical analysis on the reflectance changes of each band based on crop multispectral image data, extract multidimensional spectral response indicators, and obtain crop spectral reflectance characteristic data;

[0071] Step S225: Based on NDVI index image data, crop leaf color data, leaf curling degree data and crop spectral reflectance characteristic data, feature fusion processing is performed to obtain crop physiological characteristic data.

[0072] In this embodiment of the invention, for the collected crop multispectral image data, pixel values ​​in the red band (center wavelength 660nm) and near-infrared band (center wavelength 850nm) are extracted. For each pixel, the Normalized Difference Vegetation Index (NDVI) is calculated point-by-point according to the formula NDVI = (NIR - Red) / (NIR + Red). The results are output in grayscale image form, with grayscale values ​​limited to 0 to 255, corresponding to NDVI values ​​between 0 and 1. All NDVI images are named with plot numbers and stored in the dataset. For the visible light bands (R, G, B) of the same image... Color channel extraction was performed, converting each pixel to the HSV color space. The mean, mode, and skewness of the H (hue) component were extracted to characterize the consistency of the dominant hue and color of crop leaves. The proportion of pixels with H values ​​in the 90–150 degree range was set as the green coverage index, accurate to two decimal places. First, the image was enhanced using an edge enhancement filter. Then, the Canny edge detection method was used to extract the leaf contours. The detected edge curves were recorded in coordinate form, and the second derivative was calculated to obtain the local curvature of each edge. If the average curvature was greater than... A value of 0.25 is defined as severe curling; values ​​between 0.15 and 0.25 are considered moderate curling; and values ​​below 0.15 are considered mild curling or a flattened state. The curling degree is graded from 0 to 3. The final generated leaf curling degree data includes the image number, region ID, and corresponding curling level. Pixel reflectance is extracted for five bands: red, green, blue, red edge, and near-infrared. For each pixel, the mean, standard deviation, and reflection difference between adjacent bands for each band are calculated. The maximum reflection difference, average reflection gradient, and red edge jump amplitude of the entire image are statistically analyzed to generate the spectrum. Five response indicators were included, with numerical precision uniformly retained to two decimal places. The above data were integrated into a crop spectral reflectance characteristic data table. The NDVI index image data, crop leaf color data, leaf curling degree data, and crop spectral reflectance characteristic data were merged using the image number as the primary key. A 9-dimensional crop physiological feature vector was uniformly constructed for each record. The nine dimensions were, in order, mean NDVI, green coverage, hue skewness, curling degree, red light reflectance, red edge reflectance, near-infrared reflectance, maximum reflectance difference, and average gradient. All vector data were exported as a structured physiological feature dataset.

[0073] This invention systematically processes pixel information from different bands in crop multispectral images, enabling comprehensive extraction of crop physiological state characteristics from multiple dimensions and improving the accuracy and dimensionality of crop growth identification. By constructing NDVI index images using red and near-infrared bands, it achieves quantitative analysis of crop greenness and photosynthetic activity. Color feature extraction reflects changes in leaf pigmentation, helping to determine nitrogen or chlorophyll deficiency. Edge texture and curvature analysis reveals leaf curling trends, effectively reflecting the crop's morphological response to stresses such as drought and high temperatures. Statistical processing of reflectance in each band generates multidimensional spectral indicators, which can identify potential diseases and nutrient imbalances. Finally, the above image-derived information is fused into a unified crop physiological feature vector, which not only enhances the ability to express crop health status but also provides a decision-making basis based on the actual crop condition for precise water and fertilizer regulation, avoiding misjudgments caused by single parameters and improving the data reliability and adaptability of the diagnostic process.

[0074] Preferably, step S3 includes the following steps:

[0075] Step S31: Collect pH value, conductivity and total dissolved solids of water sources in farmland area in real time to obtain water quality monitoring data;

[0076] Step S32: Based on crop physiological status assessment data and water quality monitoring data, dynamically adjust the concentrations of nitrogen, phosphorus, and potassium solutions in the fertilizer solution to obtain fertilizer solution ratio parameters;

[0077] Step S33: Control the operation of the multi-channel fertilizer mixing pump based on the fertilizer solution ratio parameters to obtain fertilizer mixing control parameters.

[0078] In this embodiment of the invention, an integrated online water quality monitoring module is first installed at the front end of the main water source inlet pipeline for farmland. This module includes a pH electrode sensor, a conductivity measurement probe, and a TDS infrared detection element. The sensor sampling frequency is set to once every 5 minutes, and the measurement accuracies are ±0.1 pH unit, ±2% conductivity reading (unit μS / cm), and ±5%. TDS value (unit: mg / L). All sensors are connected to the central monitoring controller via RS485 bus and upload sampling data to the water and fertilizer control server in real time as water quality monitoring data. Based on the status category marked in the crop physiological status assessment data corresponding to each farmland area, the area is divided into "normal" or "abnormal" categories. If the status is "abnormal" and the pH value in the water quality monitoring data is less than 6.0 or greater than 7.5, the target concentration of phosphorus solution is increased by 10% to 20%, with the increase value determined according to the degree of deviation. If the conductivity is less than 400 μS / cm, the target concentration of nitrogen solution is increased by 15%. If the TDS value is higher than 900 mg / L, the target concentration of potassium solution is reduced by 10%. The original target concentrations are limited to 200 mg / L for nitrogen solution, 100 mg / L for phosphorus solution, and 150 mg / L for potassium solution. The adjusted concentration variation range does not exceed ±30%. The obtained adjusted values ​​are output as fertilizer solution ratio parameter tables for each crop plot, indexed by plot number, and include the parameters for each fertilizer solution. The system includes a concentration value, adjustment range, and application cycle marker. The fertilizer mixing control unit controls the operation of a multi-channel fertilizer mixing pump based on the fertilizer solution ratio parameters. Each channel corresponds to nitrogen, phosphorus, and potassium solutions. Channel control is achieved by precisely adjusting the opening of proportional solenoid valves via PWM signals, ensuring real-time concentration accuracy within ±5% during liquid mixing. Simultaneously, three flow sensors are installed in the main mixing pipeline to detect the flow rates of the three solutions. Error correction is performed every 30 seconds between the actual and set ratios. The fertilizer mixing pumps operate sequentially by first introducing clean water for 3 minutes to establish a base flow rate, then opening each channel sequentially. The injection time for each solution is controlled according to the flow rate calculated from the concentration. The controller records the channel opening time, valve opening, and actual flow rate in real time. Fertilizer control parameters include plot number, set concentration for each channel (mg / L), flow rate (L / min), duration (minutes), and total mixed solution volume. All recorded parameters are directly output to the irrigation decision-making stage for subsequent control of the irrigation pumps and solenoid valves.

[0079] This invention introduces real-time monitoring of the pH, conductivity, and total dissolved solids content of irrigation water sources, enabling dynamic control of water acidity / alkalinity, salinity, and dissolved substance concentration. This provides precise water quality background information for fertilizer solution regulation, preventing decreased nutrient utilization or increased crop stress risks due to abnormal water quality. Combined with crop physiological state assessment data, targeted adjustments to nitrogen, phosphorus, and potassium solution concentrations are achieved, ensuring precise matching of fertilizer solution ratios to current crop nutrient needs and environmental carrying capacity, improving the timeliness and effectiveness of fertilizer input. By driving a multi-channel fertilizer pump using fertilizer solution ratio parameters, automatic control and real-time output of fertilizer solution concentration and ratio are achieved, ensuring continuous, accurate, and responsive fertilization operations. This eliminates the inadequacy of traditional fixed-ratio methods to adapt to environmental changes and crop heterogeneity, fundamentally improving the intelligence level and resource allocation efficiency of the water and fertilizer regulation process.

[0080] Preferably, step S4 includes the following steps:

[0081] Step S41: Analyze the environmental status of areas without sensor deployment within the farmland area based on virtual sensor node data to obtain virtual environmental status data;

[0082] Step S42: Calculate the water and fertilizer requirements of the farmland area based on fertilizer control parameters and virtual environment status data to obtain water and fertilizer regulation demand data;

[0083] Step S43: Based on the water and fertilizer regulation demand data, perform intelligent decision analysis on irrigation duration, irrigation flow and fertilizer application to obtain control decision data.

[0084] In this embodiment of the invention, the virtual sensor node data generated in step S17 is first called, and an environmental analysis is performed on each 0.5m × 0.5m grid cell in all areas without sensor deployment. Six types of data are extracted: soil temperature (°C), soil moisture (%), soil conductivity (mS / cm), air temperature (°C), air humidity (%), and light intensity (lux). Clustering and statistical analysis are performed according to spatial location. The average and maximum values ​​of the six indicators are calculated for each 1-acre plot to generate virtual environmental status data for that plot. Based on the nitrogen, phosphorus, and potassium solution concentrations recorded in the fertilizer application control parameters and the target fertilizer solution volume for each plot, the current nutrient input requirements of the crop are standardized and calculated. Simultaneously, soil moisture below 35% and electrical conductivity below 0.3 mS / cm are used as thresholds for water and nutrient deficiency. If any plot in the virtual environment meets either of these two conditions for more than 6 hours, it is marked as requiring supplemental irrigation or fertilization. For areas meeting these conditions, the water requirement is calculated based on the degree of water deficiency. The calculation method is: 40% of the standard target soil moisture content minus the current soil moisture, multiplied by the soil volume (based on 0.3 m³ / m²).3 The required water volume is calculated in liters. The fertilizer requirement is calculated by subtracting the current conductivity from the standard conductivity of 0.8 mS / cm, multiplied by the soil unit volume and fertilizer concentration coefficient (each 0.1 mS / cm corresponds to 15 mg / L). This yields the water and fertilizer regulation requirements per acre, including the required water volume, fertilizer volume, and target concentrations of the three fertilizer solutions. Based on these water and fertilizer regulation requirements, the corresponding irrigation duration and flow rate are determined. The irrigation flow rate is fixed at 50 L / min according to the pipe specifications. The irrigation duration = required water volume / 50. The fertilizer amount is calculated proportionally based on the required concentrations of the three liquids and injected gradually within 5 minutes before irrigation. The injection sequence for each fertilizer solution is phosphorus solution, potassium solution, and nitrogen solution, with a 1-minute interval. The controller outputs the irrigation duration (in minutes), irrigation flow rate (in L / min), and injection volume of the three fertilizer solutions (in L) for each plot as the final control decision data.

[0085] This invention achieves comprehensive perception of farmland environmental conditions by utilizing virtual sensor node data, compensating for information gaps caused by insufficient actual sensor deployment and improving the accuracy and completeness of environmental monitoring. Combined with fertilizer control parameters, it accurately calculates the water and fertilizer requirements of farmland, effectively avoiding waste of water and fertilizer resources and improving resource utilization efficiency. Through intelligent decision analysis, it enables scientific regulation of irrigation duration, flow rate, and fertilizer application, not only ensuring the optimal growth environment for crops but also enhancing the automation and intelligence of irrigation and fertilization management, thereby promoting energy conservation, emission reduction, and increased yield and efficiency in agricultural production.

[0086] Preferably, step S41 includes the following steps:

[0087] Step S411: Based on the virtual sensor node data, spatial reconstruction of soil temperature, soil moisture and soil electrical conductivity at each location is performed to obtain soil environmental state simulation data;

[0088] Step S412: Perform time-series analysis on the air temperature, air humidity and light intensity at each location in the virtual sensor node dataset to obtain meteorological environmental state simulation data;

[0089] Step S413: Based on soil environmental state simulation data and meteorological environmental state simulation data, perform soil-meteorological coupling analysis on the environmental state of the farmland area where no sensors are deployed to obtain virtual environmental state data.

[0090] In this embodiment of the invention, firstly, the virtual sensor node data generated in step S17 is called to extract soil temperature (in °C, range 0–40), soil moisture (in %, range 10–90), and soil conductivity (in mS / cm, range 0.1–2.5) from all areas without sensor deployment. All data are then used to establish a 0.5m × 0.5m regular grid based on spatial location. Using known nodes as interpolation cores, a distance-weighted average method is employed for spatial reconstruction, where the weighting coefficient is 1 / square of the distance, and the calculation cutoff radius is set to 30 meters; values ​​outside this range are not included in the calculation. Each grid cell outputs estimated values ​​for the three soil parameters and is timestamped, forming a complete soil environmental state simulation dataset. From the same virtual sensor node data, air temperature (in °C, range -10–45), air humidity (in %, range 10–95), and light intensity (in lux, range 1000–100000) are extracted and used to construct a 10-minute time interval for each... For each node's time series, a 48-hour sliding statistical analysis was performed on the data for each parameter, calculating four indicators: average, maximum, minimum, and magnitude of change. A total of 12 data points (3 types of parameters × 4 indicators) were output for each node, forming a meteorological environmental state simulation dataset. The spatially reconstructed soil environmental state data and the time-series meteorological environmental state data were matched one-to-one according to node location. Coupled analysis was performed on each grid cell using the following methods: if soil moisture is below 35% and light intensity is above 60,000 lux for 6 consecutive hours, it is marked as a high-evaporation zone; if soil conductivity is below 0.3 mS / cm and air humidity is below 40%, it is marked as a nutrient-losing zone; if soil temperature is above 35℃ and air temperature is above 40℃, it is marked as a high-temperature stress zone. Each region can be assigned multiple labels, and all coupled analysis results are recorded as attribute fields attached to the corresponding spatial cells, ultimately forming virtual environmental state data containing environmental element values ​​and environmental state labels.

[0091] This invention combines spatial reconstruction and temporal analysis to achieve multi-dimensional and dynamic simulation of soil and meteorological environments in farmland areas, thereby enhancing the accuracy of environmental perception. Through coupled analysis of soil and meteorological data, it can more comprehensively and accurately reflect environmental changes in areas without sensor deployment, filling monitoring blind spots and enhancing the integrity and reliability of environmental information. This high-precision environmental simulation provides a scientific basis for subsequent agricultural management decisions, helps to achieve precision agriculture, and improves the optimization of crop growth environment and resource utilization efficiency.

[0092] Preferably, step S5 includes the following steps:

[0093] Step S51: Based on the control decision data, control the on / off state of the solenoid valve and drive the opening and closing of the irrigation channel to obtain the solenoid valve control execution data;

[0094] Step S52: Based on the control decision data, control the start and stop of the irrigation pump, and adjust the irrigation flow and duration to obtain the irrigation pump operation data;

[0095] Step S53: Based on the control decision data, perform flow control and channel selection on the fertilizer mixing pump, and execute the fertilization operation with the set fertilizer solution ratio to obtain the fertilizer mixing pump operation data;

[0096] Step S54: Record the entire irrigation and fertilization process based on the solenoid valve control execution data, irrigation pump operation data, and fertilizer pump operation data to obtain fertilization and irrigation execution log data.

[0097] In this embodiment of the invention, after receiving the control decision data generated in step S43, the central control unit calls the corresponding solenoid valve address according to the plot number, and sends a switch signal through the RS485 bus to control the on / off state of the solenoid valve. The solenoid valve response time does not exceed 1 second. The control signal is executed by relay logic triggering. Each on / off operation records the actual action time, target state, voltage and current data, and execution feedback state to generate solenoid valve control execution data. According to the irrigation duration (in minutes) and irrigation flow rate (in L / min) corresponding to each plot, the variable frequency irrigation pump is automatically called. The starting sequence is triggered 5 seconds after the solenoid valve is opened. The rated power of the irrigation pump is 1.5kW. The frequency converter adjusts the pump operating frequency to match the set flow rate. The control method is PID closed-loop regulation. When the pump is running, the flow meter provides feedback data every 10 seconds. If the deviation between the measured value and the set value exceeds 5% for 3 consecutive times, the frequency is automatically corrected. The time, current, voltage, operating frequency, and cumulative flow rate of each start and stop are recorded. All data are recorded as irrigation pump operation data. Based on the nitrogen, phosphorus, and potassium solution ratios set in the fertilizer mixing control parameters, the control unit sends signals to control the solenoid valve opening and flow threshold of the three-channel electric fertilizer mixing pump. Each channel is equipped with an independent flow sensor. When the fertilizer mixing pump starts, the clear water channel is opened to pre-flush the pipeline for 30 seconds, and then the fertilizer solution is injected in sequence according to the ratio. The target concentration is determined by the pump flow rate and injection time. The error of each channel is controlled within ±5%. During the entire fertilization process, the operating time, flow rate, actual opening, and mixed solution output of each channel of the fertilizer mixing pump are recorded to form fertilizer mixing pump operation data. The solenoid valve control execution data, irrigation pump operation data, and fertilizer mixing pump operation data are merged and archived according to timestamps to generate a complete fertilization and irrigation execution log. The log includes the plot number, irrigation start and end time, fertilizer solution injection ratio of the three channels, total irrigation flow rate, total fertilizer application, total solenoid valve opening time, pump operation status curve, and control command feedback status at each stage. This log data is stored in the central control database in real time.

[0098] This invention enables precise control of irrigation and fertilization processes in different zones of farmland, ensuring the on-demand allocation of water and fertilizer resources, effectively avoiding water waste and nutrient loss, and improving water and fertilizer utilization efficiency. It collects and records real-time operational status data of solenoid valves, irrigation pumps, and fertilizer pumps, forming a complete and traceable record of fertilization and irrigation operations, providing detailed data support for subsequent planting management, troubleshooting, and production traceability. Based on control decision data, it links various control execution units, achieving synchronous coordination of fertilization and irrigation processes, avoiding risks of waterlogging, nutrient leaching, or fertilizer damage to crop roots due to unreasonable control timing. Precise adjustment of irrigation pump flow rate and operating time ensures irrigation uniformity and suitable moisture levels in the crop root zone, improving crop growth consistency and quality stability. The real-time multi-channel flow rate adjustment function of the fertilizer pump ensures the accuracy of fertilizer solution ratio, keeping the concentration of each nutrient component within the target range to meet the nutritional needs of crops at different growth stages. Complete fertilization and irrigation execution log data can be linked with the farmland digital twin system to continuously optimize subsequent control decisions, improving the intelligence level and adaptive regulation capabilities of the entire integrated water and fertilizer system.

[0099] Preferably, step S6 includes the following steps:

[0100] Step S61: Perform spatial mapping and parameter visualization modeling on the virtual sensing node data to obtain a three-dimensional environmental perception data layer;

[0101] Step S62: Encode crop health status and render image layers on the crop physiological status assessment data to obtain a crop status visualization layer;

[0102] Step S63: Perform time-series integration and color-coding on the water quality monitoring data to obtain a water quality visualization layer;

[0103] Step S64: Perform operation path tracking and operation record graphical processing on the fertilization and irrigation execution log data to obtain the fertilization and irrigation process layer;

[0104] Step S65: Based on the 3D environmental perception data layer, crop status visualization layer, water quality visualization layer, and fertilization and irrigation process layer, the smart farm spatial scene is 3D integrated to obtain a 3D visualized smart farm twin dashboard;

[0105] Step S66: Based on the three-dimensional visualization of the smart farm twin dashboard, the spatial distribution status, crop health status and fertilization and irrigation behavior are integrated and presented, and visual interaction is designed to obtain the twin visual data view.

[0106] In this embodiment of the invention, the virtual sensing node data generated in step S17 is spatially located at a resolution of 0.5 m × 0.5 m, and a multidimensional raster dataset containing six parameters, including soil temperature (°C), soil moisture (%), soil electrical conductivity (mS / cm), air temperature and humidity (°C, %), and light intensity (lux), is constructed. An OpenGL rendering engine is used to assign independent color channels to each of the six parameters: soil moisture is mapped to the blue channel, soil electrical conductivity to the green channel, and light intensity to the yellow channel. A three-dimensional environmental perception data layer is then generated by overlaying these parameters. The crop physiological state output in step S24 is then loaded. The assessment data was used to encode the crop health status of plots based on NDVI index, curl level, and green coverage. Areas with NDVI less than 0.5 or a curl level of 3 were marked in red, and areas with high green coverage were marked in green. These three statuses were rendered into a crop status visualization layer using a 0.5-meter grid. The pH, conductivity, and TDS values ​​from water quality monitoring data, collected every 10 minutes, were compiled into a time series. pH values ​​less than 6.0 or greater than 8.0 were marked as acidic or alkaline, respectively, with corresponding colors of red and blue. Conductivity greater than 1500 μS / cm was marked as high salinity, with a color of purple. TDS values ​​greater than 1000 μS / cm were also marked as high salinity. mg / L is marked as a gray high-solids area, and dynamic changes in water quality are rendered by scrolling along a timeline, forming a water quality visualization layer. Based on the fertilization and irrigation execution log data generated in step S54, the control time of the solenoid valve, the start-up period of the irrigation pump, and the operation record of the three-channel fertilizer mixing pump are aligned in time. Irrigation path lines are drawn according to the plot number, and the line width is adjusted according to the flow rate ratio. The fertilization process is displayed with a flowing arrow animation, and floating annotation text indicating the start and end times of fertilization, the ratio, and the total liquid volume are added to form a fertilization and irrigation process layer. Using the CesiumJS 3D geographic rendering engine, the four layers are imported into a unified coordinate system for 3D rendering. By overlaying and compositing layers, setting up layer switching functions and a timeline linkage mechanism, a 3D smart farm spatial scene integrating environmental, crop, water quality, and operational data is constructed, generating a 3D visualized smart farm twin dashboard. The twin dashboard interface is equipped with a spatial status display panel, a crop health status switching button, a water quality anomaly warning pop-up, and a fertilization animation playback control bar. Through layer mutual exclusion mechanism and layer linkage control interaction logic, when the user clicks on any plot, the interface displays the current environmental values, crop status, and the parameters of the most recent fertilization. The final output of the twin dashboard is a twin visual data view that supports point selection query, time backtracking, status highlighting, and spatial positioning.

[0107] This invention achieves a three-dimensional, visualized integration of farmland environmental conditions, crop health status, water quality changes, and fertilization and irrigation processes. This allows managers to comprehensively grasp the real-time production status and historical trends of various farmland areas on a single interface, improving the intuitiveness of planting management and the accuracy of scientific decision-making. Through mapping spatially distributed data and visual interactive design, abnormal areas can be quickly identified, and the locations of water and fertilizer imbalances, pest and disease risks, or equipment malfunctions can be promptly identified, shortening response time. The multi-layer overlay display method helps analyze the correlation between various environmental factors and crop growth status, assisting in optimizing water and fertilizer control schemes and cultivation strategies. The graphical processing of fertilization and irrigation paths and operation records can be used for operational standardization assessment and mechanical operation path optimization, reducing energy consumption and labor intensity. The time-series encoding display of water quality data provides an efficient tool for long-term water source safety monitoring and water quality trend analysis, facilitating the development of water resource protection and utilization plans. The overall visualized dashboard system, through real-time linkage with an IoT platform, achieves dynamic data updates and multi-terminal remote access, significantly improving the management efficiency and intelligence level of the smart farm system.

[0108] Preferably, the present invention also provides an IoT-based integrated water and fertilizer control system for executing the above-described IoT-based integrated water and fertilizer control method, wherein the IoT-based integrated water and fertilizer control system includes:

[0109] The twin modeling module is used to acquire measured environmental data, historical environmental data, and regional geographic information of farmland areas, and to perform spatial modeling and time series learning on farmland areas to construct a digital twin model of farmland; the digital twin model of farmland is used to simulate the state of areas without sensor deployment to obtain virtual sensor node data;

[0110] The crop assessment module is used to collect multispectral image data of crops and extract crop physiological characteristic data; it uses environmental measured data, virtual sensor node data and crop physiological characteristic data to construct a crop physiological state model and generate crop physiological state assessment data.

[0111] The water and fertilizer regulation module is used to collect water quality monitoring data in real time; dynamically adjust the fertilizer-liquid ratio based on crop physiological status assessment data and water quality monitoring data, control the operation of multi-channel fertilizer mixing pumps, and generate fertilizer mixing control parameters.

[0112] The intelligent decision-making module is used to determine the control decision data of irrigation duration, flow rate and fertilizer application based on virtual sensor node data and fertilizer application control parameters;

[0113] The execution control module is used to drive the solenoid valves, irrigation pumps and fertilizer pumps to perform irrigation and fertilization operations based on control decision data, and generate fertilization and irrigation execution log data.

[0114] The visual twin module is used to build a 3D visualized smart farm twin dashboard based on virtual sensor node datasets, crop physiological status assessment data, water quality monitoring data, and fertilization and irrigation execution log data, generating a twin visual data view.

[0115] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is not limited by the foregoing description. Thus, all changes falling within the meaning and scope of the equivalents of the application are intended to be included within the scope of the invention.

[0116] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. An Internet of Things-based water and fertilizer integration control method, characterized in that, Comprise the following steps: Step S1: Obtain the environmental measurement data, historical environmental data and regional geographic information of the farmland area, and perform spatial modeling and time series learning on the farmland area to construct a farmland digital twin model; use the farmland digital twin model to simulate the state of the area where no sensor is deployed to obtain virtual sensor node data; Step S2: Collect multispectral image data of crops and extract crop physiological feature data; Use the environmental measurement data, virtual sensor node data and crop physiological feature data to construct a crop physiological state model and generate crop physiological state evaluation data; Step S3: Real-time collection of water quality monitoring data; Based on the crop physiological state evaluation data and the water quality monitoring data, dynamically adjust the fertilizer solution ratio, and control the multi-channel fertilizer pump to operate, generate the fertilizer control parameters; Step S4: Based on the virtual sensor node data and the fertilizer control parameters, determine the control decision data of irrigation time, flow and fertilizer amount; Step S5: Based on the control decision data, drive the electromagnetic valve, irrigation pump and fertilizer pump to execute irrigation and fertilization operations, and generate fertilization and irrigation execution log data; Step S6: Based on the virtual sensor node data set, crop physiological state evaluation data, water quality monitoring data and fertilization and irrigation execution log data, construct a three-dimensional visual smart farm twin board, and generate a twin visual data view; Step S6 comprises the following steps: Step S61: Spatial mapping and parameter visualization modeling of virtual sensor node data to obtain environmental three-dimensional perception data layers; Step S62: Crop health state coding and image layer rendering of crop physiological state evaluation data to obtain crop state visualization layers; Step S63: Time series integration and color marking coding of water quality monitoring data to obtain water quality visualization layers; Step S64: Operation path tracking and operation record graphical processing of fertilization and irrigation execution log data to obtain fertilization and irrigation process layers; Step S65: Three-dimensional integration of smart farm space scene based on environmental three-dimensional perception data layers, crop state visualization layers, water quality visualization layers and fertilization and irrigation process layers to obtain a three-dimensional visual smart farm twin board; Step S66: Based on the three-dimensional visual smart farm twin board, integrate and present the spatial distribution state, crop health status and fertilization and irrigation behavior, and design visual interaction to obtain a twin visual data view. 2.The Internet of Things based water and fertilizer integration control method according to claim 1, characterized in that, Step S1 comprises the following steps: Step S11: Collect soil temperature, soil humidity, soil conductivity, air temperature and humidity, and light intensity data of the farmland area to obtain environmental measurement data; Step S12: Obtain the historical environmental data of the farmland area; Step S13: Collect topography, boundary and geographic element information of the farmland area and perform digital processing to obtain regional geographic information; Step S14: Based on the environmental measurement data, historical environmental data and regional geographic information, construct a farmland area spatial model to obtain a farmland spatial structure model; Step S15: Based on the environmental measurement data and historical environmental data, construct a farmland environmental time series to obtain a farmland time dynamic model; Step S16: Based on the farmland spatial structure model and the farmland time dynamic model, construct a farmland digital twin model; Step S17: Simulate the state of the area without sensor arrangement based on the farmland digital twin model to obtain virtual sensor node data. 3.The Internet of Things based water and fertilizer integration control method according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Collect visible light, red edge, and near-infrared multispectral images of the crop canopy to obtain crop multispectral image data; Step S22: Extract NDVI index, crop leaf color, curling degree, and spectral reflection characteristics based on the crop multispectral image data to obtain crop physiological characteristic data; Step S23: Fuse the crop growth environment and crop state based on the environmental measurement data, virtual sensor node data, and crop physiological characteristic data to obtain a crop physiological state model; Step S24: Evaluate the current crop health and nutrition status based on the crop physiological state model to obtain crop physiological state evaluation data. 4.The water and fertilizer integration control method based on the Internet of Things according to claim 3, characterized in that, Step S22 includes the following steps: Step S221: Calculate the red light band and near-infrared band pixel value normalization based on the crop multispectral image data to obtain NDVI index image data; Step S222: Extract color features from the visible light band in the crop multispectral image data to obtain crop leaf color data; Step S223: Analyze the crop leaf area edge texture and deformation curvature based on the crop multispectral image data to extract leaf curling degree characteristics and obtain leaf curling degree data; Step S224: Perform statistical analysis of the reflectivity changes of each band based on the crop multispectral image data to extract multi-dimensional spectral response indicators and obtain crop spectral reflection characteristic data; Step S225: Perform feature fusion processing based on the NDVI index image data, crop leaf color data, leaf curling degree data, and crop spectral reflection characteristic data to obtain crop physiological characteristic data. 5.The Internet of Things based water and fertilizer integration control method according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Real-time collect the pH value, conductivity, and total dissolved solids of the water source in the farmland area to obtain water quality monitoring data; Step S32: Dynamically adjust the nitrogen, phosphorus, and potassium solution concentrations in the fertilizer solution based on the crop physiological state evaluation data and water quality monitoring data to obtain fertilizer solution ratio parameters; Step S33: Control the operation of the multi-channel fertilizer mixing pump based on the fertilizer solution ratio parameters to obtain fertilizer mixing control parameters. 6.The Internet of Things based water and fertilizer integration control method according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Analyze the environmental state of the area without sensor arrangement in the farmland area based on the virtual sensor node data to obtain virtual environmental state data; Step S42: Calculate the water demand and fertilizer demand of the farmland area based on the fertilizer mixing control parameters and virtual environmental state data to obtain water and fertilizer regulation demand data; Step S43: Perform intelligent decision analysis on irrigation duration, irrigation flow rate, and fertilizer amount based on the water and fertilizer regulation demand data to obtain control decision data. 7.The Internet of Things based water and fertilizer integration control method according to claim 6, characterized in that, Step S41 includes the following steps: Step S411: Spatially reconstruct soil temperature, soil moisture, and soil conductivity at each location based on the virtual sensor node data to obtain soil environmental state simulation data; Step S412: Perform time series analysis on air temperature, air humidity, and light intensity at each location in the virtual sensor node data to obtain meteorological environmental state simulation data; Step S413: soil-meteorological coupling analysis is performed on the environment state of the farmland area where no sensor is arranged based on the soil environment state simulation data and the meteorological environment state simulation data, to obtain virtual environment state data. 8.The Internet of Things based water and fertilizer integration control method according to claim 1, characterized in that, Step S5 includes the following steps: Step S51: on-off control is performed on the electromagnetic valve based on the control decision data, and opening and closing of the irrigation channel are driven, to obtain electromagnetic valve control execution data; Step S52: start-stop control is performed on the irrigation pump based on the control decision data, and irrigation flow and duration are adjusted, to obtain irrigation pump operation data; Step S53: flow control and channel selection are performed on the fertilizer preparation pump based on the control decision data, and fertilization operation of a set fertilizer liquid ratio is performed, to obtain fertilizer preparation pump operation data; Step S54: the entire process of irrigation and fertilization is recorded based on the electromagnetic valve control execution data, the irrigation pump operation data, and the fertilizer preparation pump operation data, to obtain fertilization and irrigation execution log data.

9. An Internet of Things-based water and fertilizer integrated control system, characterized in that, The water and fertilizer integrated control system based on the Internet of Things comprises: a twin modeling module, configured to acquire environment measured data, historical environment data, and regional geographic information of the farmland area, and perform spatial modeling and time series learning on the farmland area to construct a farmland digital twin model; and simulate the state of the area where no sensor is arranged by using the farmland digital twin model, to obtain virtual sensor node data; a crop evaluation module, configured to acquire multispectral image data of crops and extract crop physiological feature data; and construct a crop physiological state model by using the environment measured data, the virtual sensor node data, and the crop physiological feature data, to generate crop physiological state evaluation data; a water and fertilizer regulation module, configured to acquire water quality monitoring data in real time; dynamically adjust a fertilizer liquid ratio based on the crop physiological state evaluation data and the water quality monitoring data, and control a multi-channel fertilizer preparation pump to operate, to generate fertilizer preparation control parameters; an intelligent decision module, configured to determine control decision data of irrigation duration, flow, and fertilization amount based on the virtual sensor node data and the fertilizer preparation control parameters; an execution control module, configured to drive the electromagnetic valve, the irrigation pump, and the fertilizer preparation pump to perform irrigation and fertilization operation based on the control decision data, to generate fertilization and irrigation execution log data; a visual twin module, configured to construct a three-dimensional visual smart farm twin board based on a virtual sensor node data set, the crop physiological state evaluation data, the water quality monitoring data, and the fertilization and irrigation execution log data, to generate a twin visual data view.

Images