Irrigation decision-making method based on agricultural system model

Abstract

The invention relates to the technical field of agricultural irrigation, in particular to an irrigation decision-making method based on an agricultural system model. The method comprises the following steps: calculating a growth environment index of a target farmland to generate a crop environment growth index; crop physiological status evaluation is carried out according to the crop environment growth index, short-term water demand prediction of the field is carried out, and field crop short-term water demand data is generated; performing irrigation scheme decision processing according to the short-term water demand data of the field crops to obtain optimal irrigation scheme data; irrigation equipment is regulated and controlled by using the optimal irrigation scheme data, and irrigation water quantity influence analysis is carried out to generate irrigation water quantity influence data; and performing adaptive irrigation control optimization according to the irrigation water quantity influence data so as to obtain intelligent irrigation control strategy data. Irrigation equipment can be regulated and controlled according to dynamic requirements of crops, intelligent and accurate irrigation decision control is achieved, excessive or insufficient irrigation is effectively avoided, and the water resource utilization rate is remarkably increased.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural irrigation, and in particular to an irrigation decision method based on an agricultural system model. Background Art

[0002] Agriculture is the foundation of the national economy, and irrigation is an important part of agricultural production. Especially in today's increasingly scarce water resources, efficient and reasonable irrigation is crucial to ensuring food security and sustainable development. At present, agricultural irrigation technology is developing in the direction of automation and intelligence, and various advanced irrigation methods and technologies have emerged, such as sprinkler irrigation, drip irrigation, micro-irrigation, etc. These technologies have improved irrigation efficiency to a certain extent, but they still fail to fundamentally solve the scientific problem of irrigation decision-making. For example, irrigation decision-making based on meteorological data uses meteorological data such as temperature, rainfall, evaporation, etc. to estimate crop transpiration and soil water evaporation, thereby determining irrigation needs. However, meteorological data is usually regional and difficult to reflect the differences in field microclimates, so the accuracy of its estimation results is limited. However, even if the traditional agricultural irrigation system uses advanced irrigation equipment, it still needs to rely on manual experience to set irrigation parameters, such as irrigation time and irrigation amount. This experience-based irrigation method is difficult to adapt to the dynamic changes in the crop growth environment and cannot achieve refined water decision management. Summary of the invention

[0003] Based on this, the present invention provides an irrigation decision method based on an agricultural system model to solve at least one of the above technical problems.

[0004] To achieve the above purpose, an irrigation decision method based on an agricultural system model includes the following steps:

[0005] Step S1: synchronously collect the farmland environment of the target farmland to generate the original farmland environment status data; calculate the growth environment index according to the original farmland environment status data to generate the crop environment growth index;

[0006] Step S2: Acquire hyperspectral image data of the target farmland through the crop environmental growth index to generate crop hyperspectral image data; Evaluate crop physiological status based on crop hyperspectral image data to generate crop physiological water requirement status data; Predict short-term water demand of the field based on crop physiological water requirement status data to generate short-term water requirement data of the field crop;

[0007] Step S3: Perform multi-index environmental water demand compensation on the short-term water demand data of the field crops through the crop environmental growth index to generate field water demand warning data; perform irrigation plan decision processing based on the field water demand warning data to obtain optimal irrigation plan data; use the optimal irrigation plan data to regulate the irrigation equipment, perform water balance anomaly analysis, and generate irrigation water balance anomaly data;

[0008] Step S4: Perform irrigation water volume impact analysis on the irrigation water balance abnormality data to generate irrigation water volume impact data; perform adaptive irrigation control optimization based on the irrigation water volume impact data to obtain intelligent irrigation control strategy data.

[0009] The present invention can more accurately reflect the actual water demand of crops by synchronously collecting farmland environmental data and evaluating the crop physiological state in combination with hyperspectral image data. The synchronously collected farmland environmental data covers multiple key indicators such as soil moisture, temperature, humidity, and light, providing comprehensive and reliable basic data for the calculation of the growth environment index. The crop environmental growth index is used to guide the acquisition of hyperspectral image data, and the hyperspectral data is combined to analyze the crop physiological state, which can more accurately evaluate the crop water demand and avoid the problem of insufficient accuracy caused by the difference between regional meteorological data and field microclimate in traditional methods. This fusion of multi-source data effectively makes up for the limitations of a single data source and significantly improves the accuracy of irrigation decisions. Intelligent decision-making of irrigation plans is achieved through short-term water demand prediction of fields and multi-indicator environmental water demand compensation. The crop environmental growth index is combined with the short-term water demand data of field crops to perform multi-indicator environmental water demand compensation, which effectively considers the impact of environmental factors on crop water demand, such as temperature, humidity, light, etc., thereby generating field water demand warning data that is more in line with the actual situation. Based on this early warning data, the system automatically generates the optimal irrigation plan and intelligently regulates the irrigation equipment without manual intervention, greatly improving the irrigation efficiency and management level. By analyzing the abnormal irrigation water balance data, the system can timely discover and correct the deviation in the irrigation process, and perform adaptive irrigation control optimization according to the irrigation water volume impact data, and finally form a set of intelligent irrigation control strategies to achieve dynamic adjustment and optimization of the irrigation process. Through accurate water demand prediction and intelligent irrigation control, the waste of water resources caused by the disconnection between the irrigation amount and the actual water demand of the crop in the traditional irrigation method is avoided. Accurate water demand prediction ensures that the irrigation amount matches the actual demand of the crop and avoids excessive or insufficient irrigation. Adaptive irrigation control optimization further improves irrigation efficiency and reduces irrigation water consumption. Through water balance abnormality analysis, the system can timely discover and correct the deviation in the irrigation process, avoid water resource waste caused by human factors or equipment failure, and finally achieve the goal of water saving and increased production. Therefore, an irrigation decision-making method based on an agricultural system model of the present invention realizes accurate prediction of crop water demand by comprehensively collecting farmland environmental data and crop physiological state information. The growth environment index is used to guide the acquisition of hyperspectral images, combined with multi-index environmental water demand compensation, to automatically generate the optimal irrigation plan and dynamically adjust the irrigation equipment. Through real-time monitoring and abnormal analysis, the efficiency and reliability of the irrigation process are ensured, thereby maximizing the utilization efficiency of water resources and promoting sustainable agricultural development.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: deploying a distributed multi-sensor network based on the Internet of Things on the target farmland to obtain the initial state data of the sensor;

[0012] Step S12: synchronously collect farmland environment data according to the initial state data of the sensor to generate original farmland environment state data;

[0013] Step S13: preprocessing the original farmland environmental status data to generate preprocessed farmland environmental status data;

[0014] Step S14: performing crop environmental factor association analysis based on the pre-processed farmland environmental status data to obtain environmental factor relationship data;

[0015] Step S15: Calculate the growth environment index using the environmental factor relationship data based on the preset crop growth model to generate the crop environment growth index.

[0016] The present invention adopts a honeycomb sensor network deployment mode, each sensor node has a certain computing power and wireless communication capability, can perform data preprocessing and local data fusion, and this distributed network can effectively avoid single point failures and improve the coverage and accuracy of data collection. A central control unit is used to coordinate all sensor nodes to collect data according to a preset schedule. After receiving the data, the central control unit will align and synchronize the data according to the timestamp to ensure the consistency of the data. A central control unit is used to coordinate all sensor nodes to collect data according to a preset schedule. Each sensor node will record an accurate timestamp while collecting data. After receiving the data, the central control unit will align and synchronize the data according to the timestamp to ensure the consistency of the data. The multivariate statistical analysis method is used to reveal the correlation pattern between these environmental factors and establish a quantitative model. According to the sensitivity of crop growth to different environmental factors, the weight coefficient is pre-set. Then, a comprehensive growth environment index is calculated based on the environmental factor relationship data and the crop growth model. The index can reflect the comprehensive impact of the farmland environment on crop growth. The higher the value, the better the growth environment.

[0017] Preferably, step S2 comprises the following steps:

[0018] Step S21: setting the response spectrum range according to the crop environmental growth index to obtain real-time spectrum range response data;

[0019] Step S22: using the real-time spectral range response data to acquire hyperspectral image data of the target farmland to generate crop hyperspectral image data;

[0020] Step S23: Calculate the farmland vegetation index based on the crop hyperspectral image data to obtain the farmland vegetation distribution index;

[0021] Step S24: Acquire farmland plant physiological data; divide the target farmland into crop plots according to the farmland vegetation distribution index, and use the farmland plant physiological data to perform farmland plot distribution labeling processing to generate farmland plot distribution data;

[0022] Step S25: evaluating crop physiological status of crop hyperspectral image data through farmland plant physiological data to generate crop physiological water requirement status data;

[0023] Step S26: Based on the farmland distribution data, the preset crop growth model is used to predict the short-term water demand of the crop plots based on the crop physiological water demand status data to generate short-term water demand data of the crop plots.

[0024] The present invention sets the real-time spectral range according to the crop environmental growth index, which can effectively improve the pertinence and efficiency of hyperspectral image data acquisition. The spectral reflectance characteristics of crops will change under different growth stages and environmental conditions. By dynamically adjusting the spectral range according to the crop environmental growth index, key crop physiological information can be more targeted, and invalid data collection can be avoided, thereby improving data acquisition efficiency and the accuracy of subsequent analysis. The farmland vegetation index is calculated using the acquired hyperspectral image data, and the field is divided and marked in combination with the physiological data of the farmland plants, so that the refined management of farmland spatial information can be achieved. This provides a basis for subsequent differentiated irrigation for different fields. By dividing the farmland into different fields and marking the vegetation distribution and physiological state of each field, the crop growth status and water demand of different fields can be more accurately understood. Based on the farmland field distribution data and the preset crop growth model, the crop physiological water demand state data is used to predict the short-term water demand of the field, and the accurate prediction of the water demand of crops in different fields can be achieved. This scheme takes into account the physiological state and growth environment of the crop itself, so the prediction result is more accurate and can better meet the actual water demand of the crop.

[0025] Preferably, step S25 comprises the following steps:

[0026] Step S251: extracting leaf spectrum characteristics from crop hyperspectral image data to obtain leaf spectrum characteristic data;

[0027] Step S252: Calculate the leaf spectral reflectance on the leaf spectral characteristic data to obtain leaf spectral reflectance data;

[0028] Step S253: performing leaf temperature distribution processing according to the crop environmental growth index to generate leaf temperature distribution data;

[0029] Step S254: performing leaf stomatal conductance estimation according to the leaf temperature distribution data and the leaf spectral reflectance data to obtain leaf stomatal conductance estimation data;

[0030] Step S255: performing stomatal conductance change pattern analysis on the leaf stomatal conductance estimation data to generate stomatal conductance change pattern data;

[0031] Step S256: using the physiological data of farmland plants to invert leaf water potential, and performing crop transpiration rate analysis based on stomatal conductance change pattern data to generate plant transpiration rate data;

[0032] Step S257: Evaluate the crop physiological state based on the plant transpiration rate data to obtain crop physiological water requirement state data.

[0033] The leaf spectral reflectance of the present invention is an important indicator of the physiological state of crops, and can reflect information such as the pigment content and water content of crop leaves. By analyzing the change of leaf spectral reflectance, it can be judged whether the crop is in a state of water stress. The leaf temperature distribution is processed using the crop environmental growth index. Leaf temperature is another important indicator of the water state of crops. When crops are short of water, the leaf temperature will rise. By analyzing the leaf temperature distribution, the water state of crops can be more intuitively understood. The leaf stomatal conductance is estimated by combining the leaf temperature distribution data and the leaf spectral reflectance data, and the change pattern of the stomatal conductance is analyzed. Stomatal conductance is an indicator of the degree of stomatal opening of plant leaves, which directly affects the transpiration and photosynthesis of plants. By analyzing the change pattern of stomatal conductance, the response mechanism of crops to water stress can be more deeply understood. The leaf water potential is inverted using the physiological data of farmland plants, and the crop transpiration rate is analyzed in combination with the stomatal conductance change pattern data, and finally the crop physiological water requirement state data is obtained. Leaf water potential is an important indicator reflecting the water state of plants, and the transpiration rate directly reflects the demand of crops for water. By comprehensively analyzing these indicators, the physiological water requirement of crops can be assessed more accurately.

[0034] Preferably, step S3 comprises the following steps:

[0035] Step S31: Acquire historical crop growth data;

[0036] Step S32: performing correlation analysis on water demand physiological indicators based on crop environmental growth index and historical crop growth data to obtain key environmental water demand indicator data;

[0037] Step S33: Perform multi-indicator environmental water demand compensation on the short-term water demand data of the field crops through the key environmental water demand index data, and process the field water demand warning value to generate field water demand warning data;

[0038] Step S34: performing irrigation plan decision processing according to the field water demand warning data to obtain the optimal irrigation plan data;

[0039] Step S35: Use the optimal irrigation plan data to control the irrigation equipment to obtain the partition irrigation execution data; perform water balance anomaly analysis based on the partition irrigation execution data to generate irrigation water balance anomaly data.

[0040] The present invention uses historical crop growth data and crop environmental growth index to perform correlation analysis of water demand physiological indicators and extract key environmental water demand indicators. This enables the scheme to make full use of historical experience and, combined with the current crop growth environment, more accurately identify the key factors affecting crop water demand. The multi-index compensation mechanism can more comprehensively consider the impact of various environmental factors on crop water demand, thereby more accurately predicting crop water demand and issuing early warnings in a timely manner to avoid crop yield reduction due to water shortage. Irrigation plan decision-making is carried out based on field water demand early warning data, and irrigation equipment is regulated using the optimal irrigation plan data to achieve zoned irrigation. This avoids the drawbacks of the "one-size-fits-all" traditional irrigation method, and can perform differentiated irrigation according to the actual water demand of different fields, thereby significantly improving water resource utilization efficiency and reducing irrigation costs.

[0041] Preferably, step S34 includes the following steps:

[0042] Step S341: Calculate the irrigation water consumption according to the field water demand warning data to obtain the irrigation water demand data;

[0043] Step S342: Allocate irrigation water sources according to the irrigation water demand data to obtain irrigation water source condition data;

[0044] Step S343: performing irrigation cost processing on the irrigation water demand data through the irrigation water source condition data to generate water irrigation cost data;

[0045] Step S344: performing multi-objective irrigation water optimization based on the irrigation water demand data, the irrigation water source condition data, and the water irrigation cost data, and generating multiple irrigation scheme data;

[0046] Step S345: constructing a three-dimensional simulation model of the agricultural irrigation system to obtain a three-dimensional agricultural irrigation simulation model;

[0047] Step S346: Based on the field water demand warning data, a three-dimensional agricultural irrigation simulation model is used to simulate water transport of multiple irrigation scheme data to obtain water simulation irrigation data;

[0048] Step S347: Evaluate irrigation efficiency based on the water simulated irrigation data to generate simulated irrigation efficiency data;

[0049] Step S348: Screening the optimal irrigation scheme from multiple irrigation scheme data by simulating irrigation efficiency data to generate optimal irrigation scheme data.

[0050] The present invention calculates the irrigation water consumption based on the field water demand warning data, and performs irrigation water source allocation and irrigation cost processing. This enables the scheme to fully consider the availability of water resources and irrigation costs while meeting the crop water demand. Multi-objective irrigation water optimization is performed based on the irrigation water demand, water source conditions and irrigation costs to generate multiple alternative irrigation schemes. Multi-objective optimization can comprehensively consider multiple factors, such as water resource utilization efficiency, irrigation cost, crop water demand, etc., to avoid the situation of losing one thing while focusing on another caused by single-objective optimization. A three-dimensional simulation model of the agricultural irrigation system is constructed, and the model is used to simulate the water transport of multiple alternative irrigation schemes to obtain the simulated irrigation efficiency data of each scheme. The three-dimensional simulation technology can more realistically simulate the water transport and distribution during the irrigation process, so as to more accurately evaluate the irrigation efficiency of different schemes. The scheme screening method based on the simulation results can more effectively avoid the problems of inefficiency and water resource waste in actual irrigation, and ensure that the irrigation scheme finally selected can maximize the water resource utilization efficiency and reduce the irrigation cost while meeting the crop water demand.

[0051] Preferably, step S35 includes the following steps:

[0052] Step S351: generating irrigation system instructions for the agricultural irrigation system using the optimal irrigation plan data to generate farmland irrigation instruction data;

[0053] Step S352: accurately controlling the variable frequency pump according to the optimized farmland irrigation instruction data, and performing farmland zoning irrigation on the target farmland to obtain zoning irrigation execution data;

[0054] Step S353: Perform stratified soil moisture monitoring according to the zoned irrigation execution data to generate real-time soil moisture data;

[0055] Step S354: constructing a soil moisture spatiotemporal migration model based on real-time soil moisture data; using the soil moisture spatiotemporal migration model to evaluate the irrigation effect of each zone, and obtaining irrigation effect evaluation data;

[0056] Step S355: Perform water balance anomaly analysis based on the irrigation effect evaluation data to generate irrigation water balance anomaly data.

[0057] The present invention generates farmland irrigation instructions based on the optimal irrigation scheme data, and realizes farmland zoning irrigation through precise control of variable frequency pumps. Precise irrigation control can accurately control the irrigation time and irrigation water volume according to the actual water demand of different plots, avoiding the problem of inaccurate water volume control in traditional irrigation methods, thereby saving water resources. Layered soil moisture monitoring is performed on the execution of zoning irrigation to obtain real-time soil moisture data. Layered monitoring can more accurately reflect the moisture conditions of different soil layers, avoiding the errors caused by relying solely on surface soil moisture data for judgment. A soil moisture spatiotemporal migration model is constructed based on real-time soil moisture data, and the model is used to evaluate the effect of zoning irrigation. The soil moisture spatiotemporal migration model can simulate the movement and distribution of water in the soil during irrigation, thereby evaluating the irrigation effect more scientifically and accurately.

[0058] Preferably, step S354 includes the following steps:

[0059] Based on the real-time soil moisture data and the preset time series model, the target farmland is subjected to spatiotemporal interpolation processing, and the soil moisture spatiotemporal migration model is constructed through the soil moisture migration model;

[0060] Use the soil moisture spatiotemporal transport model to simulate soil moisture infiltration and conduct moisture transport analysis to obtain moisture transport data;

[0061] Calculate water surplus and deficit based on water migration data to generate water surplus and deficit data;

[0062] Carry out water balance analysis based on water surplus and deficit data to obtain irrigation water balance data;

[0063] The irrigation water balance data is evaluated on a regional basis to obtain irrigation effect evaluation data.

[0064] The present invention performs spatiotemporal interpolation processing based on real-time soil moisture data and a preset time series model, which makes up for the lack of spatial information caused by limited monitoring points, thereby more completely restoring the spatiotemporal distribution of farmland soil moisture. The soil moisture spatiotemporal migration model is constructed using a soil moisture migration model, which can more accurately simulate the movement and distribution of irrigation water in the soil, so as to have a deeper understanding of the impact of irrigation on soil moisture. Water profit and loss calculations can clearly reflect the supply and demand relationship of irrigation water, and water balance analysis can more comprehensively evaluate the impact of irrigation on soil moisture, such as whether the soil moisture has reached saturation, whether there is a water deficit, etc. Zoning irrigation effect evaluation is performed based on irrigation water balance data. Compared with the traditional method of evaluation based only on soil moisture data, the movement and distribution of water in the soil are taken into account, so the evaluation results are more accurate and reliable.

[0065] Preferably, step S4 comprises the following steps:

[0066] Step S41: classifying the irrigation water balance abnormality data into abnormality types to obtain irrigation abnormality type data;

[0067] Step S42: quantifying the degree of abnormality according to the irrigation abnormality type data to obtain irrigation abnormality degree data;

[0068] Step S43: performing irrigation water volume impact analysis on the irrigation water balance abnormality data through the irrigation abnormality degree data to generate irrigation water volume impact data;

[0069] Step S44: Identify the irrigation abnormality area according to the irrigation water volume impact data, and determine the irrigation control parameters to obtain the irrigation control parameter data;

[0070] Step S45: performing adaptive irrigation control optimization on the irrigation control parameter data to obtain intelligent irrigation control strategy data.

[0071] The present invention classifies the abnormality type and quantifies the abnormality degree of irrigation water balance abnormal data. This helps to more accurately identify various abnormal situations that occur during the irrigation process, such as leakage, blockage, uneven soil moisture distribution, etc., and evaluate their severity. The irrigation water volume impact analysis is performed on the irrigation water balance abnormal data according to the irrigation abnormality degree data. This can more accurately evaluate the impact of abnormal situations on the irrigation water volume, for example, leakage will lead to a reduction in irrigation water volume, and blockage will lead to excessive irrigation water volume in a local area. Identify the irrigation abnormal area according to the irrigation water volume impact data, and determine the corresponding irrigation control parameters. This realizes the precise positioning and targeted processing of abnormal situations. For example, if a certain area is found to have a leakage, the system will automatically reduce the irrigation water volume in the area, or issue an alarm to remind the staff to carry out maintenance. The irrigation control parameters are adaptively optimized for irrigation control, and intelligent irrigation control strategy data is generated. The adaptive control mechanism can dynamically adjust the irrigation control parameters according to the real-time situation and abnormal situation during the irrigation process, thereby ensuring that the irrigation system is always in the best operating state and avoiding water resource waste to the greatest extent.

[0072] Preferably, step S45 includes the following steps:

[0073] Step S451: extracting historical operation logs of the agricultural irrigation system to obtain historical operation data of the system;

[0074] Step S452: preprocessing the system historical operation data to obtain valid historical operation data;

[0075] Step S453: performing control parameter sensitivity analysis on the irrigation control parameter data through effective historical operation data to generate parameter sensitivity analysis data;

[0076] Step S454: performing system response characteristic evaluation according to the parameter sensitivity analysis data to obtain system response characteristic data;

[0077] Step S455: adjusting the control weight based on the system response characteristic data, and adaptively optimizing the control algorithm parameters of the irrigation control parameter data to generate adaptive irrigation control parameters;

[0078] Step S456: Perform intelligent irrigation decision control processing on the agricultural irrigation system through adaptive irrigation control parameters to generate an intelligent irrigation decision control strategy.

[0079] The present invention extracts the historical operation log of the agricultural irrigation system, and pre-processes the data to obtain effective historical operation data. The effective historical operation data is used to perform sensitivity analysis on the irrigation control parameters. Sensitivity analysis can help identify key parameters that have a greater impact on the performance of the irrigation system and provide direction for parameter optimization. The system response characteristics are evaluated based on the parameter sensitivity analysis data. This can provide a deeper understanding of the response law of the irrigation system to different control parameters, for example, how the system's water output, soil moisture and other indicators will change after adjusting a certain parameter. The control weight is adjusted based on the system response characteristic data, and the control algorithm parameters of the irrigation control parameters are adaptively optimized. This means that the system can automatically adjust the weight of the control parameters and optimize the control algorithm according to its own response characteristics, thereby more accurately controlling the irrigation process. Intelligent irrigation decision control processing is performed through the adaptively optimized irrigation control parameters to generate an intelligent irrigation decision control strategy. This enables the irrigation system to automatically adjust the irrigation strategy according to real-time conditions and historical experience, so as to better adapt to different environmental conditions and crop water demand laws. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 It is a schematic diagram of the steps of the irrigation decision-making method based on the agricultural system model of the present invention;

[0081] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0082] Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0083] Figure 4 This is a schematic diagram of the decision-making process of intelligent irrigation in agricultural systems;

[0084] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0085] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0086] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0087] 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 only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0088] To achieve this, please refer to Figures 1 to 4 The present invention provides an irrigation decision method based on an agricultural system model, comprising the following steps:

[0089] Step S1: synchronously collect the farmland environment of the target farmland to generate the original farmland environment status data; calculate the growth environment index according to the original farmland environment status data to generate the crop environment growth index;

[0090] Step S2: Acquire hyperspectral image data of the target farmland through the crop environmental growth index to generate crop hyperspectral image data; Evaluate crop physiological status based on crop hyperspectral image data to generate crop physiological water requirement status data; Predict short-term water demand of the field based on crop physiological water requirement status data to generate short-term water requirement data of the field crop;

[0091] Step S3: Perform multi-index environmental water demand compensation on the short-term water demand data of the field crops through the crop environmental growth index to generate field water demand warning data; perform irrigation plan decision processing based on the field water demand warning data to obtain optimal irrigation plan data; use the optimal irrigation plan data to regulate the irrigation equipment, perform water balance anomaly analysis, and generate irrigation water balance anomaly data;

[0092] Step S4: Perform irrigation water volume impact analysis on the irrigation water balance abnormality data to generate irrigation water volume impact data; perform adaptive irrigation control optimization based on the irrigation water volume impact data to obtain intelligent irrigation control strategy data.

[0093] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of the irrigation decision-making method based on the agricultural system model of the present invention. In this embodiment, the irrigation decision-making method based on the agricultural system model includes the following steps:

[0094] Step S1: synchronously collect the farmland environment of the target farmland to generate the original farmland environment status data; calculate the growth environment index according to the original farmland environment status data to generate the crop environment growth index;

[0095] In an embodiment of the present invention, for example, in a tomato planting greenhouse, a distributed wireless sensor network is used to collect farmland environmental data in real time. Sensor nodes are deployed around tomato plants, and collect data such as air temperature, air humidity, light intensity, soil temperature, soil moisture, soil EC value and CO2 concentration every 15 minutes, and transmit the data to a central server via a wireless network. After receiving the data, the server generates the original environmental status data of the farmland. Then, a crop growth model is used, such as a temperature, light, water and nutrient requirement model based on tomato growth and development, combined with the collected original environmental status data of the farmland, to calculate the crop environmental growth index. For example, a comprehensive environmental index can be calculated, which comprehensively considers the effects of factors such as temperature, humidity, light and soil moisture on tomato growth. The higher the index, the more conducive the environment is to tomato growth.

[0096] Step S2: Acquire hyperspectral image data of the target farmland through the crop environmental growth index to generate crop hyperspectral image data; Evaluate crop physiological status based on crop hyperspectral image data to generate crop physiological water requirement status data; Predict short-term water demand of the field based on crop physiological water requirement status data to generate short-term water requirement data of the field crop;

[0097] In an embodiment of the present invention, the timing and parameters for acquiring hyperspectral image data are determined according to the tomato environmental growth index calculated in step S1. For example, when the environmental growth index is lower than a certain threshold, it indicates that there is stress in the tomato growth environment, and it is necessary to obtain hyperspectral image data for further analysis. A hyperspectral imager is mounted on a drone installed on the top of a greenhouse to collect hyperspectral image data of tomato plants. After preprocessing (e.g., radiation correction, atmospheric correction, and geometric correction) of the acquired hyperspectral image data, crop hyperspectral image data is generated. Using the established relationship model between tomato spectral characteristics and physiological indicators, such as vegetation indices calculated based on reflectance of specific bands (such as normalized vegetation index NDVI, photochemical vegetation index PRI, etc.), the physiological state of tomato leaves is evaluated, such as chlorophyll content, leaf moisture content, etc., to generate crop physiological water requirement state data. Combined with the growth stage and physiological water requirement state data of tomatoes, crop water requirement models, such as the Penman-Monteith model, are used to predict the short-term water requirement data of tomato fields in the next 24 hours, that is, the water requirement of each tomato plant.

[0098] Step S3: Perform multi-index environmental water demand compensation on the short-term water demand data of the field crops through the crop environmental growth index to generate field water demand warning data; perform irrigation plan decision processing based on the field water demand warning data to obtain optimal irrigation plan data; use the optimal irrigation plan data to regulate the irrigation equipment, perform water balance anomaly analysis, and generate irrigation water balance anomaly data;

[0099] In an embodiment of the present invention, multi-index environmental water demand compensation is performed based on the tomato environmental growth index calculated in step S1 and the field short-term water demand data predicted in step S2. For example, when the environmental growth index is high, the irrigation water volume is appropriately reduced, otherwise the irrigation water volume is increased to generate field water demand warning data. According to the field water demand warning data, combined with the type of irrigation system in the greenhouse (such as a drip irrigation system), the optimal irrigation plan data is formulated, including irrigation time, irrigation amount, and irrigation area. For example, according to the water demand of each tomato plant, the water output and irrigation time of each dripper in the drip irrigation system are controlled. After the irrigation is completed, the change of soil moisture is monitored by a sensor, and the actual irrigation water volume is compared with the planned irrigation water volume, and the abnormal situation of water balance is analyzed, such as leakage, blockage, etc., to generate irrigation water balance abnormality data.

[0100] Step S4: Perform irrigation water volume impact analysis on the irrigation water balance abnormality data to generate irrigation water volume impact data; perform adaptive irrigation control optimization based on the irrigation water volume impact data to obtain intelligent irrigation control strategy data.

[0101] In an embodiment of the present invention, the irrigation water balance abnormality data generated in step S3 is analyzed, for example, by analyzing the soil moisture sensor data, it is determined whether the irrigation system has problems such as leakage or blockage, and the impact of these problems on the irrigation water volume, and the irrigation water volume impact data is generated. For example, if it is found that the soil moisture in a certain area is continuously lower than the expected value, there is a leakage or dripper blockage problem, resulting in water shortage of tomatoes in the area. According to the irrigation water volume impact data, the irrigation control strategy is adaptively optimized. For example, if the dripper blockage is detected, the water outlet pressure of the dripper is automatically increased or the irrigation time is extended; if a leak is detected, the irrigation water volume in the area is automatically reduced or the irrigation is stopped, and an alarm is issued to notify the maintenance personnel. Through continuous learning and optimization, intelligent irrigation control strategy data is finally obtained, for example, according to the environmental growth index, the crop physiological water demand state and the water balance abnormality data, the irrigation plan is dynamically adjusted to achieve precise irrigation, improve water use efficiency, and reduce manual intervention.

[0102] Preferably, step S1 comprises the following steps:

[0103] Step S11: deploying a distributed multi-sensor network based on the Internet of Things on the target farmland to obtain the initial state data of the sensor;

[0104] Step S12: synchronously collect farmland environment data according to the initial state data of the sensor to generate original farmland environment state data;

[0105] Step S13: preprocessing the original farmland environmental status data to generate preprocessed farmland environmental status data;

[0106] Step S14: performing crop environmental factor association analysis based on the pre-processed farmland environmental status data to obtain environmental factor relationship data;

[0107] Step S15: Calculate the growth environment index using the environmental factor relationship data based on the preset crop growth model to generate the crop environment growth index.

[0108] In an embodiment of the present invention, for example, in a tomato planting greenhouse, a distributed multi-sensor network based on the Internet of Things is deployed. Various sensor nodes, such as air temperature and humidity sensors, light sensors, soil temperature and humidity sensors, soil EC sensors, CO2 sensors, and leaf humidity sensors, are installed at different positions and heights of tomato plants, as well as different areas of the greenhouse. Each sensor node is assigned a unique IP address and can communicate with a central server through a wireless network (such as WiFi or Zigbee). Based on the sensor network deployed in step S11, the farmland environment synchronization acquisition program is started. According to the sensor type, measurement range, and communication parameters recorded in the sensor initial state data, the sampling frequency and data transmission protocol of each sensor are set. For example, the air temperature and humidity sensor collects data once a minute, and the soil temperature and humidity sensor collects data once every 15 minutes. The data collected by all sensors are transmitted to the central server in real time through the wireless network. The server uses message queue protocols such as MQTT to receive and parse the data, sorts and stores the received data according to timestamps, and generates farmland original environmental state data. Clean the data and use algorithms such as Kalman filtering to remove outliers and noise. For missing values, select a suitable filling method according to the missing situation. For example, for a small number of missing values, use linear interpolation or mean filling method to fill; for a large number of continuous missing values, consider using a model-based interpolation method. Based on the pre-processed farmland environmental status data generated in step S13, perform crop environmental factor association analysis. For example, to analyze the correlation between air temperature, soil moisture, light intensity and CO2 concentration, the correlation coefficients between different environmental factors can be calculated using methods such as Pearson correlation coefficient and Spearman rank correlation coefficient. A mathematical model between environmental factors can also be established using a multiple regression analysis method, such as a multiple linear regression model between tomato growth and air temperature, soil moisture and light intensity. Select a suitable tomato growth model, such as a temperature, light, water, and nutrient requirement model based on tomato growth and development, or a growth curve model combined with expert experience. Utilize the environmental factor relationship data obtained in step S14, combined with a preset crop growth model, to calculate the tomato environmental growth index. For example, the comprehensive environmental growth index of tomatoes can be weighted and calculated based on environmental factors such as air temperature, soil moisture, light intensity, CO2 concentration, and leaf surface moisture, and the weight coefficient can be adjusted according to different growth stages of tomatoes. The higher the index, the more favorable the environment is for tomato growth.

[0109] Preferably, step S2 comprises the following steps:

[0110] Step S21: setting the response spectrum range according to the crop environmental growth index to obtain real-time spectrum range response data;

[0111] Step S22: using the real-time spectral range response data to acquire hyperspectral image data of the target farmland to generate crop hyperspectral image data;

[0112] Step S23: Calculate the farmland vegetation index based on the crop hyperspectral image data to obtain the farmland vegetation distribution index;

[0113] Step S24: Acquire farmland plant physiological data; divide the target farmland into crop plots according to the farmland vegetation distribution index, and use the farmland plant physiological data to perform farmland plot distribution labeling processing to generate farmland plot distribution data;

[0114] Step S25: evaluating crop physiological status of crop hyperspectral image data through farmland plant physiological data to generate crop physiological water requirement status data;

[0115] Step S26: Based on the farmland distribution data, the preset crop growth model is used to predict the short-term water demand of the crop plots based on the crop physiological water demand status data to generate short-term water demand data of the crop plots.

[0116] As an example of the present invention, refer to Figure 2 As shown, Figure 1 Detailed implementation steps of step S2 in the embodiment are shown in the flowchart. In this embodiment, step S2 includes:

[0117] Step S21: setting the response spectrum range according to the crop environmental growth index to obtain real-time spectrum range response data;

[0118] In an embodiment of the present invention, the response spectral range of the hyperspectral imager is set according to the previously calculated environmental growth index of tomatoes. For example, if the environmental growth index indicates that tomatoes are in a state of water stress, the spectral range is set to a band that is sensitive to water, such as a near-infrared band (700-1300nm) and a short-wave infrared band (1300-2500nm). If the environmental growth index indicates that tomatoes are growing well, a wider spectral range can be selected to obtain more vegetation information. The set spectral range parameters are transmitted to the hyperspectral imager in real time to generate real-time spectral range response data.

[0119] Step S22: using the real-time spectral range response data to acquire hyperspectral image data of the target farmland to generate crop hyperspectral image data;

[0120] In an embodiment of the present invention, based on the real-time spectral range set in step S21, a track-type hyperspectral imaging system installed on the top of the greenhouse or a hyperspectral imager carried by a drone is used to collect hyperspectral image data of tomato plants. The imaging system or drone is controlled to scan according to a preset path and height to ensure that the hyperspectral image of the tomato plants in the entire greenhouse is obtained. The acquired raw hyperspectral data contains the spectral reflectance information of each pixel. The collected data is preliminarily processed, such as removing noise and bad pixels, to generate crop hyperspectral image data.

[0121] Step S23: Calculate the farmland vegetation index based on the crop hyperspectral image data to obtain the farmland vegetation distribution index;

[0122] In an embodiment of the present invention, the farmland vegetation index is calculated based on the crop hyperspectral image data obtained in step S22. For example, the normalized difference vegetation index (NDVI), enhanced vegetation index (EVI) or photochemical vegetation index (PRI) is calculated. The selection of a suitable vegetation index depends on the specific application scenario and research purpose. For example, NDVI can reflect the growth status and coverage of vegetation, and PRI can reflect the photosynthetic efficiency of vegetation. The calculated vegetation index value is assigned to each pixel of the image to generate a farmland vegetation distribution index map.

[0123] Step S24: Acquire farmland plant physiological data; divide the target farmland into crop plots according to the farmland vegetation distribution index, and use the farmland plant physiological data to perform farmland plot distribution labeling processing to generate farmland plot distribution data;

[0124] In an embodiment of the present invention, physiological data of tomato plants, such as leaf area index (LAI), chlorophyll content, stem thickness, etc., are obtained by manual measurement or sensor collection. The farmland vegetation distribution index map obtained in step S23 is combined with an image segmentation algorithm, such as K-means clustering or threshold segmentation, to divide the tomato planting area into different plots. The plots are labeled according to the physiological data of the plants in each plot, such as the average leaf area index or the average chlorophyll content. For example, a plot with a higher LAI is labeled as "vigorous growth" and a plot with a lower LAI is labeled as "slow growth".

[0125] Step S25: evaluating crop physiological status of crop hyperspectral image data through farmland plant physiological data to generate crop physiological water requirement status data;

[0126] In the embodiment of the present invention, the physiological data of the farmland plants obtained in step S24 and the crop hyperspectral image data obtained in step S22 are combined, and the established relationship model between the spectral characteristics of tomatoes and physiological indicators, such as partial least squares regression (PLSR) or support vector machine (SVM) model, is used to evaluate the physiological state of the tomato plants, such as leaf moisture content, leaf temperature, etc. According to the evaluation results, the crop physiological water requirement state data, such as the water requirement of each field or each plant, is generated.

[0127] Step S26: Based on the farmland distribution data, the preset crop growth model is used to predict the short-term water demand of the crop plots based on the crop physiological water demand status data to generate short-term water demand data of the crop plots.

[0128] In an embodiment of the present invention, based on the farmland plot distribution data generated in step S24 and the crop physiological water requirement status data generated in step S25, combined with a preset tomato growth model, such as the Penman-Monteith model or the crop coefficient method, the short-term water demand of each plot is predicted, such as the water demand in the next 24 hours or the next week. Factors such as the growth conditions, environmental conditions, and soil characteristics of different plots need to be considered during the prediction. The prediction results generate short-term water requirement data for plot crops, providing a basis for subsequent irrigation decisions. For example, a corresponding irrigation plan can be formulated based on the predicted water demand of each plot.

[0129] Preferably, step S25 comprises the following steps:

[0130] Step S251: extracting leaf spectrum characteristics from crop hyperspectral image data to obtain leaf spectrum characteristic data;

[0131] Step S252: Calculate the leaf spectral reflectance on the leaf spectral characteristic data to obtain leaf spectral reflectance data;

[0132] Step S253: performing leaf temperature distribution processing according to the crop environmental growth index to generate leaf temperature distribution data;

[0133] Step S254: performing leaf stomatal conductance estimation according to the leaf temperature distribution data and the leaf spectral reflectance data to obtain leaf stomatal conductance estimation data;

[0134] Step S255: performing stomatal conductance change pattern analysis on the leaf stomatal conductance estimation data to generate stomatal conductance change pattern data;

[0135] Step S256: using the physiological data of farmland plants to invert leaf water potential, and performing crop transpiration rate analysis based on stomatal conductance change pattern data to generate plant transpiration rate data;

[0136] Step S257: Evaluate the crop physiological state based on the plant transpiration rate data to obtain crop physiological water requirement state data.

[0137] In an embodiment of the present invention, in a tomato planting greenhouse, the acquired tomato hyperspectral image data is used to extract the spectral features of the leaves. First, the tomato leaves are separated from the background (such as soil, greenhouse structure, etc.) by using image segmentation technology, such as threshold segmentation based on vegetation index or semantic segmentation based on machine learning. Then, the spectral features are extracted for the segmented leaf area. The spectral features that can be extracted include: reflectance value of a specific band, spectral absorption features, spectral reflection peak position, spectral curve slope, etc. The extracted feature data is stored as leaf spectral feature data. Based on the leaf spectral feature data extracted in step S251, the leaf spectral reflectance is calculated. If the extracted feature itself is the reflectance value of a specific band, it can be used directly. If other spectral features are extracted, such as spectral absorption features, it is necessary to calculate the reflectance of the leaves in different bands based on the relationship between the spectral absorption features and the reflectance. The calculated leaf spectral reflectance data is stored in a matrix or table form, in which each row represents a leaf sample and each column represents the reflectance value of a band. A thermal infrared imager is used to obtain a thermal infrared image of a tomato plant. According to the crop environmental growth index calculated previously, the thermal infrared image is temperature corrected and spatially registered. For example, the leaf temperatures in different regions can be weighted averaged according to the environmental growth index to eliminate the influence of environmental factors on the leaf temperature. Based on the leaf spectral reflectance data obtained in step S252 and the leaf temperature distribution data obtained in step S253, the leaf stomatal conductance is estimated using an established model or empirical formula. For example, the leaf stomatal conductance can be estimated using a relationship model between leaf temperature and reflectance in a specific band, or using an empirical formula established using parameters such as leaf temperature, air temperature, and relative humidity. The leaf stomatal conductance estimation data obtained in step S254 is subjected to a time series analysis, such as analyzing the variation pattern of stomatal conductance within a day, or analyzing the variation pattern of stomatal conductance with environmental factors (such as light, temperature, humidity, etc.). Fourier transform, wavelet transform, and other methods can be used to analyze the periodic changes in stomatal conductance, or a regression analysis method can be used to establish a relationship model between stomatal conductance and environmental factors. The water potential of tomato leaves is measured using instruments such as a pressure chamber, or the established model is used to invert the leaf water potential based on other physiological indicators (such as relative leaf water content). Combined with the stomatal conductance change pattern data and leaf water potential data obtained in step S255, the crop transpiration rate is calculated using methods such as the Penman-Monteith formula. Based on the plant transpiration rate data calculated in step S256, the physiological water requirement of the crop is evaluated. For example, it can be determined whether the crop is in a state of water stress based on the size of the transpiration rate. If the transpiration rate is lower than the normal level, it indicates that the crop is short of water; if the transpiration rate is too high, it indicates that the crop is in a state of water oversaturation. A comprehensive evaluation can also be performed in combination with other physiological indicators, such as leaf water potential, relative leaf water content, etc.The evaluation results are stored in the form of crop physiological water requirement data, for example, expressing the water requirement of the crop in numerical form, or classifying the crops into different water requirement levels in a classified form.

[0138] Preferably, step S3 comprises the following steps:

[0139] Step S31: Acquire historical crop growth data;

[0140] Step S32: performing correlation analysis on water demand physiological indicators based on crop environmental growth index and historical crop growth data to obtain key environmental water demand indicator data;

[0141] Step S33: Perform multi-indicator environmental water demand compensation on the short-term water demand data of the field crops through the key environmental water demand index data, and process the field water demand warning value to generate field water demand warning data;

[0142] Step S34: performing irrigation plan decision processing according to the field water demand warning data to obtain the optimal irrigation plan data;

[0143] Step S35: Use the optimal irrigation plan data to control the irrigation equipment to obtain the partition irrigation execution data; perform water balance anomaly analysis based on the partition irrigation execution data to generate irrigation water balance anomaly data.

[0144] In an embodiment of the present invention, historical tomato growth data are collected. These data can come from the greenhouse management system database, including tomato plant height, stem thickness, leaf area, yield, irrigation amount, fertilizer amount and environmental data (temperature, humidity, light, etc.) of the corresponding period at different growth stages. Ensure the integrity and accuracy of the data, and clean and pre-process the data. Based on the current crop environmental growth index and the historical crop growth data obtained in step S31, a water demand physiological index association analysis is performed. For example, the change law of physiological indicators such as water demand, transpiration rate, and leaf water potential of tomatoes under different environmental growth indexes is analyzed. Statistical methods such as correlation analysis and regression analysis can be used to establish a relationship model between environmental growth index and water demand physiological indicators. Through analysis, key environmental indicators that affect tomato water demand, such as temperature, light, humidity, etc., are determined, and the thresholds or weights of these indicators are determined to generate key environmental water demand indicator data. Based on the key environmental water demand indicator data obtained in step S32, multi-indicator environmental water demand compensation is performed on the previously calculated field crop short-term water demand data. For example, if the current ambient temperature is higher than the set threshold, the water demand of the field is increased; if the light intensity is low, the water demand of the field is reduced. The compensation range can be adjusted according to the weight set in the key environmental water demand indicator data. After compensation, the field water demand warning value is set according to the critical water demand value of tomatoes and the current soil moisture status. For example, when the predicted soil moisture is lower than a certain threshold, the warning is triggered. For example, according to the warning status and water demand of each field, determine whether irrigation is required, as well as the irrigation amount, irrigation time and irrigation method of each field. Considering the capacity limitation of the irrigation system and the availability of water resources, optimize the irrigation plan, such as allocating irrigation tasks to different time periods to avoid insufficient water pressure caused by simultaneous irrigation. The irrigation plan finally generated includes specific irrigation parameters for each field, forming the optimal irrigation plan data. Based on the optimal irrigation plan data obtained in step S34, the irrigation equipment in the greenhouse, such as solenoid valves, drip irrigation systems, etc., is controlled. According to the irrigation parameters of each field, the irrigation equipment in the corresponding area is controlled to achieve zoned irrigation. The actual irrigation water volume and irrigation time of each area are recorded to generate zoned irrigation execution data. After irrigation is completed, the soil moisture changes are monitored through the soil moisture sensor, and the actual irrigation water volume is compared with the planned irrigation water volume to analyze abnormal conditions in the water balance, such as leakage, blockage, sensor failure, etc.

[0145] Preferably, step S34 includes the following steps:

[0146] Step S341: Calculate the irrigation water consumption according to the field water demand warning data to obtain the irrigation water demand data;

[0147] Step S342: Allocate irrigation water sources according to the irrigation water demand data to obtain irrigation water source condition data;

[0148] Step S343: performing irrigation cost processing on the irrigation water demand data through the irrigation water source condition data to generate water irrigation cost data;

[0149] Step S344: performing multi-objective irrigation water optimization based on the irrigation water demand data, the irrigation water source condition data, and the water irrigation cost data, and generating multiple irrigation scheme data;

[0150] Step S345: constructing a three-dimensional simulation model of the agricultural irrigation system to obtain a three-dimensional agricultural irrigation simulation model;

[0151] Step S346: Based on the field water demand warning data, a three-dimensional agricultural irrigation simulation model is used to simulate water transport of multiple irrigation scheme data to obtain water simulation irrigation data;

[0152] Step S347: Evaluate irrigation efficiency based on the water simulated irrigation data to generate simulated irrigation efficiency data;

[0153] Step S348: Screening the optimal irrigation scheme from multiple irrigation scheme data by simulating irrigation efficiency data to generate optimal irrigation scheme data.

[0154] In an embodiment of the present invention, the irrigation water consumption is calculated based on the water demand warning data of each field, such as the warning level and the soil moisture deficit. Considering factors such as crop type, growth stage, soil type, etc., for example, the irrigation water consumption for replenishing soil moisture to field water holding capacity can be calculated according to the growth stage of tomatoes and the soil moisture characteristic curve. According to the calculated irrigation water demand data, the available irrigation water sources, such as reservoirs, groundwater, rainwater collection systems, etc., are considered to allocate irrigation water sources. If multiple water sources are available, it is necessary to determine the irrigation water source of each field according to factors such as the water quality, water quantity, distance and cost of the water source, and calculate the water supply and water supply time of each water source. Based on the irrigation water source condition data and the irrigation water demand data, the water irrigation cost is calculated. Consider factors such as the water extraction cost, water transmission cost, energy consumption and maintenance cost of the irrigation system of different water sources. For example, the water extraction cost of groundwater is higher than that of the rainwater collection system, and the cost of long-distance water transmission is higher than that of short-distance water transmission. The irrigation cost data of each field are summarized to generate water irrigation cost data. Based on the irrigation water demand data, the irrigation water source condition data and the water irrigation cost data, a multi-objective optimization model is constructed. The objective function may include minimizing the irrigation cost, maximizing the irrigation efficiency, minimizing the water resource consumption, etc. The optimization model is solved using a multi-objective optimization algorithm, such as a genetic algorithm, a particle swarm algorithm, etc., to obtain a set of Pareto optimal solutions representing different irrigation schemes. A three-dimensional simulation model of a greenhouse tomato irrigation system is constructed using professional modeling software, such as SolidWorks, AutoCAD, etc. The model should include information such as the geometric structure, topography, soil type, irrigation system layout (such as the distribution of drip irrigation pipes), water source location, and sensor location of the greenhouse. The accuracy and complexity of the model should be adjusted according to actual needs. Based on the field water demand warning data and the three-dimensional agricultural irrigation simulation model constructed in step S345, the water transport simulation is performed on the multi-irrigation scheme data generated in step S344. The migration and distribution of water in the soil and the absorption of water by crops under different irrigation schemes are simulated. Based on the water simulation irrigation data obtained in step S346, the efficiency of each irrigation scheme is evaluated. Irrigation efficiency can be expressed as the ratio of the actual water absorption of the crop to the amount of water used for irrigation. The uniform distribution of water in the soil and deep leakage and other factors can also be considered. The efficiency value of each irrigation scheme is stored as simulated irrigation efficiency data. Based on the simulated irrigation efficiency data generated in step S347, combined with other objective function values, such as irrigation cost and water resource consumption, the optimal irrigation scheme is screened out from the multiple irrigation scheme data generated in step S344. Multi-objective decision-making methods, such as TOPSIS method, hierarchical analysis method, etc., can be used to sort and select schemes. The finally selected irrigation scheme and its corresponding parameters are stored as the optimal irrigation scheme data.

[0155] Preferably, step S35 includes the following steps:

[0156] Step S351: generating irrigation system instructions for the agricultural irrigation system using the optimal irrigation plan data to generate farmland irrigation instruction data;

[0157] Step S352: accurately controlling the variable frequency pump according to the optimized farmland irrigation instruction data, and performing farmland zoning irrigation on the target farmland to obtain zoning irrigation execution data;

[0158] Step S353: Perform stratified soil moisture monitoring according to the zoned irrigation execution data to generate real-time soil moisture data;

[0159] Step S354: constructing a soil moisture spatiotemporal migration model based on real-time soil moisture data; using the soil moisture spatiotemporal migration model to evaluate the irrigation effect of each zone, and obtaining irrigation effect evaluation data;

[0160] Step S355: Perform water balance anomaly analysis based on the irrigation effect evaluation data to generate irrigation water balance anomaly data.

[0161] In an embodiment of the present invention, an irrigation system instruction is generated based on the optimal irrigation scheme data. The optimal irrigation scheme data includes information such as the irrigation time, irrigation duration, and irrigation water volume of each field. Based on this information, an instruction to control the irrigation equipment is generated, such as controlling the switch of the solenoid valve, adjusting the speed of the variable frequency pump, setting the flow of the drip irrigation system, etc. The irrigation equipment in the greenhouse is controlled according to the farmland irrigation instruction data generated in step S351. For example, by adjusting the speed of the variable frequency pump, the irrigation water pressure and flow are accurately controlled; by controlling the switch of the solenoid valve, the timed and quantitative irrigation of different fields is achieved. During the irrigation process, the actual irrigation water volume and irrigation time of each field are monitored in real time. In different fields where tomatoes are planted, layered soil moisture sensors are used, such as burying multiple sensors at different depths, to monitor the soil moisture content in real time. The number and arrangement depth of the sensors should be adjusted according to the root distribution of tomatoes and soil characteristics. The data collected by the sensors are transmitted to the central server via a wireless network, and the server synchronizes the data in time and interpolates the data in space to generate real-time soil moisture data covering the entire greenhouse. Based on the real-time soil moisture data obtained in step S353, and soil characteristic parameters (such as soil porosity, saturated hydraulic conductivity, etc.), a soil moisture spatiotemporal migration model is constructed. Hydrus-1D or other similar software can be used to simulate the movement and distribution of water in the soil. Using the constructed model, the soil moisture changes under different irrigation schemes are simulated, and the simulation results are compared with the actual monitored soil moisture data to evaluate the effect of zoned irrigation. For example, the utilization efficiency of irrigation water, the uniform distribution of water in the soil, and whether the water requirements of tomatoes are met can be evaluated. Based on the irrigation effect evaluation data obtained in step S354, the abnormal situation of water balance is analyzed. For example, if the irrigation efficiency is lower than the expected value, there are problems such as leakage, dripper blockage or soil compaction; if the water distribution is uneven, the layout of the irrigation system or the irrigation parameters need to be adjusted. By analyzing the irrigation effect evaluation data and the zoned irrigation execution data, the cause of the abnormal water balance is found, and the analysis results are recorded as irrigation water balance abnormality data, which is used to guide the maintenance and optimization of the irrigation system, such as timely replacement of damaged drippers, unblocking of blocked pipes, and improvement of irrigation strategies.

[0162] Preferably, step S354 includes the following steps:

[0163] Based on the real-time soil moisture data and the preset time series model, the target farmland is subjected to spatiotemporal interpolation processing, and the soil moisture spatiotemporal migration model is constructed through the soil moisture migration model;

[0164] Use the soil moisture spatiotemporal transport model to simulate soil moisture infiltration and conduct moisture transport analysis to obtain moisture transport data;

[0165] Calculate water surplus and deficit based on water migration data to generate water surplus and deficit data;

[0166] Carry out water balance analysis based on water surplus and deficit data to obtain irrigation water balance data;

[0167] The irrigation water balance data is evaluated on a regional basis to obtain irrigation effect evaluation data.

[0168] In an embodiment of the present invention, a network of arranged soil moisture sensors is used to obtain real-time soil moisture data. Due to the limited number of sensors and their discrete distribution, spatiotemporal interpolation is required to obtain soil moisture distribution information for the entire field. First, a preset time series model, such as an ARIMA model or an LSTM neural network, is used to predict the time series data collected by each sensor to fill in the missing values ​​and improve the temporal resolution. Then, in combination with spatial interpolation methods, such as Kriging interpolation or inverse distance weighted interpolation, the soil moisture data of the sensor point is interpolated to the entire field to generate a high-resolution soil moisture distribution map. Finally, based on the physical property parameters of the soil (such as soil porosity, saturated hydraulic conductivity, soil texture, etc.) and the known soil moisture migration law (such as the Richards equation), a soil moisture spatiotemporal migration model, such as Hydrus-2D / 3D or SWMS_2D, is constructed. The constructed soil moisture spatiotemporal migration model is used to simulate the infiltration process of irrigation water in the soil. Input parameters such as irrigation water volume, irrigation method (e.g., drip irrigation, sprinkler irrigation), and initial soil moisture content to simulate the vertical and horizontal movement of water in the soil profile, as well as the process of water absorption by tomato roots. The simulation results include the change of soil moisture content at different depths and locations over time, as well as the cumulative infiltration, evaporation, and root water absorption. Based on the simulated water movement data, the soil moisture surplus and deficit are calculated. Water surplus and deficit can be defined as the difference between soil moisture input (e.g., irrigation, rainfall) and output (e.g., evaporation, transpiration). In the tomato greenhouse, the irrigation water volume, evaporation, and tomato transpiration are mainly considered. According to the water input and output at each time step, the cumulative soil moisture surplus and deficit are calculated, and water surplus and deficit data are generated. Based on the water surplus and deficit data, the soil moisture balance of the tomato greenhouse is analyzed. Water balance refers to the relationship between the input and output of soil moisture. Ideally, the irrigation water volume should be able to compensate for the soil moisture loss caused by evaporation and transpiration, so that the soil moisture remains within a range suitable for tomato growth. By analyzing the water surplus and deficit data, it is possible to determine whether irrigation is sufficient, whether there is water deficit or over-irrigation. The results of the water balance analysis, such as the mean value, standard deviation, and range of soil moisture, as well as the proportional relationship between water input and output, are saved as irrigation water balance data. Based on the irrigation water balance data, the effect of zoning irrigation is evaluated. Evaluation indicators may include water use efficiency, water uniformity, and the degree of crop water stress. Water use efficiency can be expressed as the ratio of tomato yield or biomass to irrigation water consumption. Water uniformity can be measured by the coefficient of variation of soil moisture content. The degree of crop water stress can be evaluated based on physiological indicators such as leaf water potential and stomatal conductance. The irrigation effect evaluation indicators of different zones are summarized and compared to generate irrigation effect evaluation data.

[0169] Preferably, step S4 comprises the following steps:

[0170] Step S41: classifying the irrigation water balance abnormality data into abnormality types to obtain irrigation abnormality type data;

[0171] Step S42: quantifying the degree of abnormality according to the irrigation abnormality type data to obtain irrigation abnormality degree data;

[0172] Step S43: performing irrigation water volume impact analysis on the irrigation water balance abnormality data through the irrigation abnormality degree data to generate irrigation water volume impact data;

[0173] Step S44: Identify the irrigation abnormality area according to the irrigation water volume impact data, and determine the irrigation control parameters to obtain the irrigation control parameter data;

[0174] Step S45: performing adaptive irrigation control optimization on the irrigation control parameter data to obtain intelligent irrigation control strategy data.

[0175] As an example of the present invention, refer to Figure 3 As shown, Figure 1 Detailed implementation steps of step S4 in the embodiment are shown in the flowchart. In this embodiment, step S4 includes:

[0176] Step S41: classifying the irrigation water balance abnormality data into abnormality types to obtain irrigation abnormality type data;

[0177] In the embodiment of the present invention, the irrigation water balance abnormal data is analyzed and the abnormal types are classified. Common abnormal types include: water leakage (such as drip irrigation pipe rupture, dripper detachment), blockage (such as dripper blockage, pipe blockage), sensor failure (such as abnormal soil moisture sensor reading), insufficient irrigation water, excessive irrigation water, etc. Classification can be performed according to the characteristics exhibited by the abnormal data. For example, if the soil moisture in a certain area is continuously lower than the expected value, and the actual irrigation water volume is significantly different from the planned irrigation water volume, there is a water leakage problem; if the soil moisture in a certain area is continuously higher than the expected value, and the actual irrigation water volume is not much different from the planned irrigation water volume, there is a sensor failure or an excessive irrigation water problem. Each abnormal data point is marked as a corresponding abnormal type to generate irrigation abnormal type data.

[0178] Step S42: quantifying the degree of abnormality according to the irrigation abnormality type data to obtain irrigation abnormality degree data;

[0179] In the embodiment of the present invention, the degree of abnormality is quantified according to the irrigation abnormality type data obtained in step S41. Different types of abnormalities have different quantification methods. For example, for leakage, it can be quantified by leakage amount or leakage rate; for blockage, it can be quantified by blockage area or impact range; for insufficient or excessive irrigation water, it can be quantified by the deviation value between the actual irrigation water amount and the planned irrigation water amount; for sensor failure, it can be quantified by the deviation between the sensor reading and the actual value or the volatility of the sensor reading. The quantified abnormality degree value is associated with the corresponding abnormality type data to generate irrigation abnormality degree data.

[0180] Step S43: performing irrigation water volume impact analysis on the irrigation water balance abnormality data through the irrigation abnormality degree data to generate irrigation water volume impact data;

[0181] In the embodiment of the present invention, based on the irrigation abnormality degree data, the influence of different types of abnormalities on the irrigation water volume is analyzed. For example, water leakage may lead to the loss of irrigation water volume, blockage may lead to insufficient irrigation water volume in some areas, and sensor failure may lead to inaccurate control of irrigation water volume. A model of irrigation water volume loss or deviation may be established according to the abnormality type and abnormality degree, such as a linear model or a nonlinear model. The model is used to calculate the influence degree of each abnormal data point on the irrigation water volume, and generate irrigation water volume influence data.

[0182] Step S44: Identify the irrigation abnormality area according to the irrigation water volume impact data, and determine the irrigation control parameters to obtain the irrigation control parameter data;

[0183] In an embodiment of the present invention, an irrigation abnormality area is identified based on the irrigation water volume impact data. For example, a threshold value can be set according to the size of the irrigation water volume loss or deviation, and the area exceeding the threshold value is identified as an irrigation abnormality area. For the identified irrigation abnormality area, it is necessary to determine the corresponding irrigation control parameters, such as adjusting the irrigation time, irrigation water volume, irrigation frequency, etc. The adjustment range of the control parameters should be determined based on factors such as the abnormality type, the degree of abnormality, and the crop growth conditions. For example, for areas with severe water leakage, irrigation should be stopped immediately and repairs should be carried out; for areas with clogged drippers, you can try to increase the irrigation water pressure or extend the irrigation time; for areas with insufficient irrigation water, you can increase the irrigation water volume or increase the irrigation frequency. The control parameter adjustment scheme for each abnormal area is saved as irrigation control parameter data.

[0184] Step S45: performing adaptive irrigation control optimization on the irrigation control parameter data to obtain intelligent irrigation control strategy data.

[0185] In an embodiment of the present invention, the irrigation control strategy is adaptively optimized based on the irrigation control parameter data. Adaptive optimization refers to dynamically adjusting the irrigation control parameters according to the real-time status and environmental conditions of the irrigation system to achieve the best irrigation effect. For example, the soil moisture changes and crop water requirements in the future can be predicted based on the real-time data of the soil moisture sensor, weather forecast data, and crop growth model, and the irrigation control parameters can be adjusted accordingly. The irrigation control strategy can be optimized using a machine learning algorithm, such as reinforcement learning or model predictive control. The optimized irrigation control strategy, such as the control parameter adjustment rules under different states, is saved as intelligent irrigation control strategy data, and applied to the actual operation of the irrigation system to achieve intelligent irrigation management.

[0186] Preferably, step S45 includes the following steps:

[0187] Step S451: extracting historical operation logs of the agricultural irrigation system to obtain historical operation data of the system;

[0188] Step S452: preprocessing the system historical operation data to obtain valid historical operation data;

[0189] Step S453: performing control parameter sensitivity analysis on the irrigation control parameter data through effective historical operation data to generate parameter sensitivity analysis data;

[0190] Step S454: performing system response characteristic evaluation according to the parameter sensitivity analysis data to obtain system response characteristic data;

[0191] Step S455: adjusting the control weight based on the system response characteristic data, and adaptively optimizing the control algorithm parameters of the irrigation control parameter data to generate adaptive irrigation control parameters;

[0192] Step S456: Perform intelligent irrigation decision control processing on the agricultural irrigation system through adaptive irrigation control parameters to generate an intelligent irrigation decision control strategy.

[0193] In an embodiment of the present invention, historical operation log data is extracted from the irrigation control system. These data include irrigation time, irrigation duration, irrigation water volume, water pump operation status, valve switch status, soil moisture sensor data, environmental sensor data (such as temperature, humidity, light), etc. The extracted data is stored in a database, a text file, or a file in other formats. It is necessary to select appropriate tools and methods for data extraction according to actual conditions, such as using database query statements, file reading programs, etc. The system historical operation data extracted in step S451 is preprocessed to remove noise, fill missing values, and handle abnormal values. Data cleaning techniques can be used, such as removing duplicate data, handling abnormal values ​​(such as data beyond the normal range), and filling missing data (such as using interpolation methods or average filling). Using valid historical operation data, sensitivity analysis is performed on irrigation control parameter data. Irrigation control parameters include irrigation thresholds, irrigation water volume, irrigation time, etc. The purpose of sensitivity analysis is to evaluate the degree of influence of changes in control parameters on irrigation effects. For example, the influence of changes in irrigation thresholds on soil moisture content, crop growth, and water use efficiency can be analyzed. Sensitivity analysis can be performed using methods such as variance analysis and regression analysis. Based on the parameter sensitivity analysis data, the response characteristics of the irrigation system are evaluated. The system response characteristics refer to the response speed and response amplitude of the irrigation system to the change of the control parameters. For example, the response speed of the irrigation system to the adjustment of the irrigation water volume and the time required for the soil moisture content to reach the target value can be evaluated. The system response characteristics can be evaluated using methods such as step response analysis and frequency response analysis. Based on the system response characteristic data, the parameters of the irrigation control algorithm are adjusted. For example, if the response speed of the system is slow, the gain of the control algorithm can be increased; if the stability of the system is poor, the integral time constant of the control algorithm can be reduced. The weights of different control parameters in the control algorithm can also be adjusted according to their sensitivity. For example, a larger weight is given to a parameter with higher sensitivity, and a smaller weight is given to a parameter with lower sensitivity. By adjusting the parameters and weights of the control algorithm, the control accuracy and stability of the irrigation system can be improved. The optimized control parameters are saved as adaptive irrigation control parameters. Based on the adaptive irrigation control parameters, intelligent irrigation decision control is performed on the tomato planting greenhouse. Intelligent irrigation decision control refers to dynamically adjusting the irrigation control parameters according to the real-time monitored soil moisture data, environmental data and crop growth model to achieve precise irrigation. For example, the need for irrigation can be determined based on the data from the soil moisture sensor, and the irrigation time, irrigation water volume, and irrigation method can be determined based on the adaptive irrigation control parameters. The rules and parameters of intelligent irrigation decision control are saved as intelligent irrigation decision control strategies, and applied to the actual operation of the irrigation system to achieve automated and intelligent irrigation management.

[0194] The present application is to realize accurate monitoring of farmland environment and irrigation decision-making based on crop growth model to meet the needs of different crops and environmental conditions in agricultural production. A distributed sensor network is used to collect farmland environmental data in real time, such as soil moisture, temperature, light intensity and meteorological data, and crop physiological state information is obtained through hyperspectral image data. These data are preprocessed and indexed to generate a comprehensive crop environmental growth index. Compared with traditional methods, physiological state assessment based on growth model improves the accuracy of data, can more accurately reflect the growth status and water demand status of crops, and provide a reliable basis for irrigation decision-making. Through multi-layer data fusion and analysis, the problem of single information in traditional irrigation decision-making is effectively overcome. Through multi-dimensional farmland data and model calculation, crop short-term water demand data is generated, irrigation early warning and accurate decision-making are realized, and irrigation equipment is automatically adjusted to execute the optimal irrigation plan, avoiding the problem of irrigation inaccuracy caused by regional meteorological data and field microclimate differences. This process can dynamically respond to changes in crop water demand, avoid excessive or insufficient irrigation, reduce human intervention, significantly improve the utilization efficiency of irrigation water, and promote water saving and increased production. The safety and reliability of the irrigation system are further ensured through abnormal analysis and adaptive optimization control of irrigation water balance. The system can monitor the changes in soil moisture after irrigation in real time, automatically identify abnormal conditions during irrigation such as leakage and blockage, and optimize irrigation strategies by adjusting irrigation parameters to achieve intelligent control.

[0195] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0196] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be 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 present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An irrigation decision-making method based on an agricultural system model, characterized in that: The following steps are involved: Step S1: synchronously collect the farmland environment of the target farmland to generate the original farmland environment status data; calculate the growth environment index according to the original farmland environment status data to generate the crop environment growth index; Step S2: Acquire hyperspectral image data of the target farmland through the crop environmental growth index to generate crop hyperspectral image data; Evaluate crop physiological status based on crop hyperspectral image data to generate crop physiological water requirement status data; Predict short-term water demand of the field based on crop physiological water requirement status data to generate short-term water requirement data of the field crop; Step S3: Perform multi-index environmental water demand compensation on the short-term water demand data of the field crops through the crop environmental growth index to generate field water demand warning data; perform irrigation plan decision processing based on the field water demand warning data to obtain optimal irrigation plan data; use the optimal irrigation plan data to regulate the irrigation equipment, perform water balance anomaly analysis, and generate irrigation water balance anomaly data; Step S4: Perform irrigation water volume impact analysis on the irrigation water balance abnormality data to generate irrigation water volume impact data; perform adaptive irrigation control optimization based on the irrigation water volume impact data to obtain intelligent irrigation control strategy data.

2. The irrigation decision-making method based on the agricultural system model according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: deploying a distributed multi-sensor network based on the Internet of Things on the target farmland to obtain the initial state data of the sensor; Step S12: synchronously collect farmland environment data according to the initial state data of the sensor to generate original farmland environment state data; Step S13: preprocessing the original farmland environmental status data to generate preprocessed farmland environmental status data; Step S14: performing crop environmental factor association analysis based on the pre-processed farmland environmental status data to obtain environmental factor relationship data; Step S15: Calculate the growth environment index using the environmental factor relationship data based on the preset crop growth model to generate the crop environment growth index.

3. The irrigation decision-making method based on the agricultural system model according to claim 2, characterized in that: Step S2 includes the following steps: Step S21: setting the response spectrum range according to the crop environmental growth index to obtain real-time spectrum range response data; Step S22: using the real-time spectral range response data to acquire hyperspectral image data of the target farmland to generate crop hyperspectral image data; Step S23: Calculate the farmland vegetation index based on the crop hyperspectral image data to obtain the farmland vegetation distribution index; Step S24: Acquire farmland plant physiological data; divide the target farmland into crop plots according to the farmland vegetation distribution index, and use the farmland plant physiological data to perform farmland plot distribution labeling processing to generate farmland plot distribution data; Step S25: evaluating crop physiological status of crop hyperspectral image data through farmland plant physiological data to generate crop physiological water requirement status data; Step S26: Based on the farmland distribution data, the preset crop growth model is used to predict the short-term water demand of the crop plots based on the crop physiological water demand status data to generate short-term water demand data of the crop plots.

4. The irrigation decision-making method based on the agricultural system model according to claim 3, characterized in that: Step S25 includes the following steps: Step S251: extracting leaf spectrum characteristics from crop hyperspectral image data to obtain leaf spectrum characteristic data; Step S252: Calculate the leaf spectral reflectance on the leaf spectral characteristic data to obtain leaf spectral reflectance data; Step S253: performing leaf temperature distribution processing according to the crop environmental growth index to generate leaf temperature distribution data; Step S254: performing leaf stomatal conductance estimation according to the leaf temperature distribution data and the leaf spectral reflectance data to obtain leaf stomatal conductance estimation data; Step S255: performing stomatal conductance change pattern analysis on the leaf stomatal conductance estimation data to generate stomatal conductance change pattern data; Step S256: using the physiological data of farmland plants to invert leaf water potential, and performing crop transpiration rate analysis based on stomatal conductance change pattern data to generate plant transpiration rate data; Step S257: Evaluate the crop physiological state based on the plant transpiration rate data to obtain crop physiological water requirement state data.

5. The irrigation decision-making method based on the agricultural system model according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: Acquire historical crop growth data; Step S32: performing correlation analysis on water demand physiological indicators based on crop environmental growth index and historical crop growth data to obtain key environmental water demand indicator data; Step S33: Perform multi-indicator environmental water demand compensation on the short-term water demand data of the field crops through the key environmental water demand index data, and process the field water demand warning value to generate field water demand warning data; Step S34: performing irrigation plan decision processing according to the field water demand warning data to obtain the optimal irrigation plan data; Step S35: Use the optimal irrigation plan data to control the irrigation equipment to obtain the partition irrigation execution data; perform water balance anomaly analysis based on the partition irrigation execution data to generate irrigation water balance anomaly data.

6. The irrigation decision-making method based on the agricultural system model according to claim 1, characterized in that: Step S34 includes the following steps: Step S341: Calculate the irrigation water consumption according to the field water demand warning data to obtain the irrigation water demand data; Step S342: Allocate irrigation water sources according to the irrigation water demand data to obtain irrigation water source condition data; Step S343: performing irrigation cost processing on the irrigation water demand data through the irrigation water source condition data to generate water irrigation cost data; Step S344: performing multi-objective irrigation water optimization based on the irrigation water demand data, the irrigation water source condition data, and the water irrigation cost data, and generating multiple irrigation scheme data; Step S345: constructing a three-dimensional simulation model of the agricultural irrigation system to obtain a three-dimensional agricultural irrigation simulation model; Step S346: Based on the field water demand warning data, a three-dimensional agricultural irrigation simulation model is used to simulate water transport of multiple irrigation scheme data to obtain water simulation irrigation data; Step S347: Evaluate irrigation efficiency based on the water simulated irrigation data to generate simulated irrigation efficiency data; Step S348: Screening the optimal irrigation scheme from multiple irrigation scheme data by simulating irrigation efficiency data to generate optimal irrigation scheme data.

7. The irrigation decision-making method based on the agricultural system model according to claim 5, characterized in that: Step S35 includes the following steps: Step S351: generating irrigation system instructions for the agricultural irrigation system using the optimal irrigation plan data to generate farmland irrigation instruction data; Step S352: accurately controlling the variable frequency pump according to the optimized farmland irrigation instruction data, and performing farmland zoning irrigation on the target farmland to obtain zoning irrigation execution data; Step S353: Perform stratified soil moisture monitoring according to the zoned irrigation execution data to generate real-time soil moisture data; Step S354: constructing a soil moisture spatiotemporal migration model based on real-time soil moisture data; using the soil moisture spatiotemporal migration model to evaluate the irrigation effect of each zone, and obtaining irrigation effect evaluation data; Step S355: Perform water balance anomaly analysis based on the irrigation effect evaluation data to generate irrigation water balance anomaly data.

8. The irrigation decision-making method based on the agricultural system model according to claim 7, characterized in that: Step S354 includes the following steps: Based on the real-time soil moisture data and the preset time series model, the target farmland is subjected to spatiotemporal interpolation processing, and the soil moisture spatiotemporal migration model is constructed through the soil moisture migration model; Use the soil moisture spatiotemporal transport model to simulate soil moisture infiltration and conduct moisture transport analysis to obtain moisture transport data; Calculate water surplus and deficit based on water migration data to generate water surplus and deficit data; Carry out water balance analysis based on water surplus and deficit data to obtain irrigation water balance data; The irrigation water balance data is evaluated on a regional basis to obtain irrigation effect evaluation data.

9. The irrigation decision-making method based on the agricultural system model according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: classifying the irrigation water balance abnormality data into abnormality types to obtain irrigation abnormality type data; Step S42: quantifying the degree of abnormality according to the irrigation abnormality type data to obtain irrigation abnormality degree data; Step S43: performing irrigation water volume impact analysis on the irrigation water balance abnormality data through the irrigation abnormality degree data to generate irrigation water volume impact data; Step S44: Identify the irrigation abnormality area according to the irrigation water volume impact data, and determine the irrigation control parameters to obtain the irrigation control parameter data; Step S45: performing adaptive irrigation control optimization on the irrigation control parameter data to obtain intelligent irrigation control strategy data.

10. The irrigation decision-making method based on the agricultural system model according to claim 1, characterized in that: Step S45 includes the following steps: Step S451: extracting historical operation logs of the agricultural irrigation system to obtain historical operation data of the system; Step S452: preprocessing the system historical operation data to obtain valid historical operation data; Step S453: performing control parameter sensitivity analysis on the irrigation control parameter data through effective historical operation data to generate parameter sensitivity analysis data; Step S454: performing system response characteristic evaluation according to the parameter sensitivity analysis data to obtain system response characteristic data; Step S455: adjusting the control weight based on the system response characteristic data, and adaptively optimizing the control algorithm parameters of the irrigation control parameter data to generate adaptive irrigation control parameters; Step S456: Perform intelligent irrigation decision control processing on the agricultural irrigation system through adaptive irrigation control parameters to generate an intelligent irrigation decision control strategy.

Images