Automatic irrigation method and device based on artificial intelligence

Abstract

The invention relates to the technical field of irrigation, in particular to an automatic irrigation method and device based on artificial intelligence, and the method comprises the steps: collecting crop physiological signals, microclimate data, soil parameters, root distribution data, crop canopy structure parameters and sprinkling angle adjustment range data of a planting area in real time; if the crop physiological signal exceeds a health threshold value, identifying a signal abnormity type and matching a growth stress factor; simulating an irrigation intervention effect and evaluating an irrigation priority based on the regional ecological model; if the priority is high, target irrigation working condition features are extracted from a historical irrigation database; acquiring irrigation demand parameters in combination with real-time data, and comparing the fitness; if the adaptation degree is insufficient, a dynamic irrigation scheme is generated according to the root system distribution data, the soil parameters and the sprinkling angle adjusting range data; and irrigating the planting area based on the dynamic irrigation scheme. According to the invention, dynamic and accurate irrigation of crops in a planting area can be realized.

Description

Technical Field

[0001] This application relates to the field of irrigation technology, and in particular to an automatic irrigation method and device based on artificial intelligence. Background Technology

[0002] Agricultural irrigation is a crucial link in agricultural production, directly affecting crop growth, final yield and quality, and the efficiency of agricultural water resource utilization. Currently, the global water shortage problem is becoming increasingly severe, and with the continuous advancement of agricultural modernization, the limitations of traditional irrigation methods are becoming more and more apparent. Intelligent and precision irrigation technologies have become an important direction for development in the agricultural sector.

[0003] Traditional irrigation methods, such as flood irrigation and furrow irrigation, rely entirely on the manual experience of farmers, lacking a precise understanding of the actual growth needs of crops and the environmental conditions of the planting area. Farmers irrigate according to fixed cycles, often resulting in over- or under-irrigation: over-irrigation leads to waterlogging, root rot due to lack of oxygen, and may even cause soil salinization, resulting in severe water waste; under-irrigation puts crops under water stress, affecting photosynthesis and nutrient absorption, leading to slow growth, reduced yields, or even death. Furthermore, traditional methods cannot provide differentiated irrigation based on different crop varieties, growth stages, and the microenvironment of the planting area, significantly reducing irrigation effectiveness.

[0004] To address the shortcomings of traditional irrigation, smart irrigation technology has emerged. Existing smart irrigation systems typically collect environmental parameters such as soil moisture and temperature, and combine these with preset irrigation thresholds for automatic irrigation control. Some technologies also incorporate crop water requirement models to estimate irrigation volume based on the crop's growth cycle. However, existing irrigation systems have a relatively limited sensing dimension, mostly focusing only on certain soil or meteorological parameters, neglecting crucial information such as the crop's own physiological signals, root distribution characteristics, and canopy structure. Consequently, they cannot formulate precise irrigation plans based on the essential needs of crop growth, ultimately failing to achieve dynamic and precise irrigation for crops in the planting area. Summary of the Invention

[0005] To facilitate dynamic and precise irrigation of crops in planting areas, this application provides an automatic irrigation method and device based on artificial intelligence.

[0006] Firstly, this application provides an automatic irrigation method based on artificial intelligence, which adopts the following technical solution: An automated irrigation method based on artificial intelligence includes: Climate data, soil parameters, and crop canopy structure parameters are used to obtain current irrigation demand parameters; Compare the fit between the current irrigation demand parameters and the characteristics of the target irrigation conditions; If the fit is less than the critical value, a dynamic irrigation plan is generated based on root distribution data, soil parameters, and water spraying angle adjustment range data. The planting area is irrigated based on a dynamic irrigation scheme.

[0007] Optionally, matching the corresponding crop growth stress factor based on the signal anomaly type includes: Obtain the feature dimensions of signal anomaly types; Constructing anomaly feature vectors based on feature dimensions; The abnormal feature vectors are matched with the preset stress factor feature library; Calculate the cosine similarity between the abnormal feature vector and the feature vectors of each stress factor; Stress factors with cosine similarity higher than the matching threshold are selected as candidate stress factors; The candidate stress factors were subjected to correlation verification to verify their suitability to the environmental conditions of the planting area. Candidate stress factors whose adaptability to environmental conditions is below the adaptation threshold are eliminated; The remaining candidate stress factors are sorted from high to low similarity, and the stress factor ranked first is selected as the crop growth stress factor.

[0008] Optionally, the simulation of irrigation intervention effects and evaluation of irrigation priorities based on regional ecological models includes: The planting area is divided into multiple sub-regions, and the target data distribution characteristics of each sub-region are obtained. The target data of each sub-region is input into the regional ecological model to simulate the recovery trend of crop physiological signals under different irrigation volumes and durations. Calculate irrigation effectiveness scores for each sub-region based on recovery trends; Statistically analyze the irrigation effect scores of all sub-regions within the planting area and calculate the overall regional score; The regional comprehensive score is compared with the preset priority grading standard to determine the preliminary irrigation priority; If multiple planting areas apply for irrigation resources at the same time, obtain the crop variety importance coefficient and growth stage weight for each area; By combining the regional comprehensive score, the variety importance coefficient, and the growth stage weight, the preliminary irrigation priority is adjusted to obtain the final irrigation priority.

[0009] Optionally, if the fit is less than a critical value, generating a dynamic irrigation plan based on root distribution data, soil parameters, and watering angle adjustment range data includes: If the fit is less than the critical value, the location coordinates of the dense and sparse root areas within the planting area are determined based on the root distribution data. Based on soil parameters, the soil water holding capacity and permeability coefficient at different locations were analyzed; Based on the data on the water spraying angle adjustment range, calculate the water spraying coverage of the irrigation device at different angles; Spatial matching of the location coordinates of dense and sparse root zones with the watering coverage area; Differentiated irrigation water volumes are allocated to different regions based on soil water holding capacity and permeability coefficient; Based on the water distribution results and the water coverage area, set the water spraying angle combination of the irrigation device; Develop a time-segmented irrigation execution plan based on the combination of sprinkling angles and water volume allocation; By integrating location coordinates, water allocation, sprinkler angle combinations, and irrigation execution plans, a dynamic irrigation scheme is generated.

[0010] Optionally, irrigating the planting area based on a dynamic irrigation scheme includes: The dynamic irrigation scheme is broken down into multiple irrigation sub-tasks, each sub-task corresponding to a sub-region of the planting area; Assign a corresponding irrigation device and execution time window to each irrigation subtask; Within the execution time window, the irrigation device is controlled to start irrigation according to the preset water spray angle combination; Real-time collection of soil moisture data and crop physiological signal feedback data during irrigation; The collected soil moisture data and crop physiological signal feedback data were compared with the corresponding preset indicators in the dynamic irrigation scheme. If soil moisture data or crop physiological signal feedback data deviate from the preset index, calculate the deviation value; Adjust the spraying angle and water output of the irrigation device according to the deviation value; Continuously monitor the adjusted irrigation effect until the sub-area completes irrigation and the data meets the standards; Complete all irrigation sub-tasks in sequence to achieve overall irrigation of the planting area.

[0011] Optionally, adjusting the sprinkling angle and water output of the irrigation device based on the deviation value includes: Obtain the deviation type corresponding to the deviation value. The deviation types are divided into soil moisture deviation and crop physiological signal deviation. If the deviation type is soil moisture deviation, retrieve the soil water holding capacity and soil permeability coefficient of the target sub-region; Based on soil water holding capacity and soil permeability coefficient, determine the target amount of soil moisture that needs to be replenished or reduced; Based on the data of the water spraying angle adjustment range of the irrigation device, calculate the water spraying angle adjustment range that can make the soil moisture reach the target amount. The adjustment ratio of water output is determined based on the soil permeability coefficient. If the deviation type is crop physiological signal deviation, obtain the growth requirements corresponding to the current crop physiological signal; By comparing growth requirements with preset physiological signal indicators, the direction of irrigation condition adjustment required for signal recovery is determined; Based on the adjustment direction and combined with the crop canopy structure parameters, adjust the sprinkling angle to optimize the water coverage and adjust the water output to match the growth needs.

[0012] Optionally, determining the target amount of soil moisture to be replenished or reduced based on soil water-holding capacity and soil permeability coefficient includes: Obtain historical soil moisture compliance data for the target sub-region and determine the suitable soil moisture range for the corresponding crop at the current growth stage. Obtain the current soil moisture in the target sub-region and compare the current soil moisture data with the suitable soil moisture range; If the current soil moisture is below the suitable soil moisture range, analyze the evaporation rate of soil moisture in combination with the soil water holding capacity. Based on the evaporation rate, calculate the total amount of water that needs to be replenished during the irrigation interval to maintain soil moisture within a suitable range, and use this as the target amount to be replenished. If the current humidity is higher than the suitable soil moisture range, the soil moisture infiltration rate is determined based on the soil permeability coefficient. Based on microclimate data within a preset time period, the total amount of water that needs to be reduced to bring soil moisture down to a suitable range is calculated and used as the target amount to be reduced.

[0013] Optionally, if the current humidity is higher than the suitable soil moisture range, determining the soil moisture infiltration rate based on the soil permeability coefficient includes: Soil stratified samples at different depths in the target sub-region were obtained, and the soil texture and porosity data of each soil stratified sample were detected. The initial infiltration rate under each soil stratum sample was calculated by combining the soil permeability coefficient. Collect historical data on soil moisture infiltration after several recent irrigations in the target sub-region and calculate the average infiltration attenuation coefficient. A dynamic model of soil moisture infiltration rate over time is established based on the initial infiltration rate and the average infiltration attenuation coefficient. Based on the current soil moisture and suitable moisture range, the soil moisture infiltration rate for the current period is determined by a dynamic model.

[0014] Secondly, this application also discloses an automatic irrigation device based on artificial intelligence, which adopts the following technical solution: An automated irrigation device based on artificial intelligence includes: The data acquisition module is used to collect target data of the planting area in real time. The target data includes crop physiological signals, microclimate data, soil parameters, root distribution data, crop canopy structure parameters, and irrigation device spray angle adjustment range data. The anomaly detection module is used to identify the type of signal anomaly if the crop physiological signal exceeds the health threshold. The factor matching module is used to match the corresponding crop growth stress factors based on the type of signal anomaly. The simulated irrigation module is used to retrieve regional ecological models based on anomaly types and stress factors, simulate the effects of irrigation interventions based on the regional ecological models, and evaluate irrigation priorities. If the evaluation result is high priority, the feature extraction module is used to extract the target irrigation condition features from the historical irrigation database based on the current microclimate, soil parameters, crop canopy structure parameters and preset similarity threshold. The parameter acquisition module is used to obtain current irrigation demand parameters based on crop physiological signals, microclimate data, soil parameters, and crop canopy structure parameters; The fit comparison module is used to compare the fit between the current irrigation demand parameters and the characteristics of the target irrigation conditions. If the fit is less than the critical value, the scheme generation module generates a dynamic irrigation scheme based on root distribution data, soil parameters, and watering angle adjustment range data. The irrigation module is used to irrigate the planting area based on a dynamic irrigation scheme.

[0015] In summary, this application includes the following beneficial technical effects: By collecting multi-dimensional target data in real time, including crop physiological signals, microclimate data, soil parameters, root distribution data, crop canopy structure parameters, and irrigation device spray angle adjustment range data, this system overcomes the limitation of existing intelligent irrigation systems with their single sensing dimension. It can capture key irrigation-related information based on the essential needs of crop growth. When crop physiological signals exceed health thresholds, it identifies abnormality types, matches growth stress factors, and combines regional ecological models to simulate the effects of irrigation intervention and assess priorities. This achieves accurate determination and prioritization of irrigation needs, avoiding indiscriminate irrigation. Simultaneously, by retrieving historical irrigation databases and comparing the fit between current irrigation demand parameters and target operating conditions, it can determine the appropriate irrigation method based on root distribution when the fit is insufficient. The system generates dynamic irrigation plans by adjusting soil parameters and irrigation device ranges. It combines historical effective irrigation experience with personalized adjustments based on the actual microenvironment of the planting area, crop growth status, and irrigation equipment conditions. This effectively solves the problems of traditional irrigation relying on manual experience and being unable to achieve differentiated irrigation, as well as the shortcomings of existing intelligent irrigation systems that are difficult to dynamically and accurately adapt to crop needs. Ultimately, it achieves dynamic and precise irrigation for crops in the planting area, meeting the water requirements of crops at different growth stages and under different microenvironments, reducing crop growth problems and yield reduction risks caused by over- or under-irrigation, optimizing water resource utilization efficiency, reducing the probability of environmental problems such as soil salinization and root rot, and improving the level of intelligence and precision in agricultural irrigation. Attached Figure Description

[0016] Figure 1 This is a main flowchart of an automatic irrigation method based on artificial intelligence according to an embodiment of this application; Figure 2 This is a block diagram of an automatic irrigation device based on artificial intelligence, according to an embodiment of this application.

[0017] Explanation of reference numerals in the attached figures: 1. Data acquisition module; 2. Anomaly identification module; 3. Factor matching module; 4. Simulated irrigation module; 5. Feature extraction module; 6. Parameter acquisition module; 7. Fit comparison module; 8. Scheme generation module; 9. Irrigation module. Detailed Implementation

[0018] In the first aspect, this application discloses an automatic irrigation method based on artificial intelligence.

[0019] Reference Figure 1 An automated irrigation method based on artificial intelligence includes steps S101 to S109: Step S101: Collect target data of the planting area in real time.

[0020] Specifically, the target data includes crop physiological signals, microclimate data, soil parameters, root distribution data, crop canopy structure parameters, and irrigation device spray angle adjustment range data. Crop physiological signals refer to physiological indicators reflecting crop growth status, such as leaf water potential, stem diameter change rate, and photosynthetically active radiation absorption. Microclimate data refers to the climate parameters of the local microenvironment of the planting area, including air temperature, relative humidity, wind speed, precipitation, and sunshine duration. Soil parameters include soil volumetric water content, soil electrical conductivity, soil temperature, and soil bulk density. Root distribution data refers to the spatial distribution information of crop roots in the soil, including root density at different soil depths, root diameter distribution, and root spatial coordinates. Crop canopy structure parameters include canopy leaf area index (i.e., the ratio of total crop leaf area to land area per unit land area), canopy porosity, and canopy height. Spray angle adjustment range data refers to the adjustable spray direction angle range of the irrigation device and the corresponding spray radius at each angle.

[0021] In this embodiment, target data acquisition is achieved through a multi-sensor collaborative acquisition system: soil sensors (such as TDR soil moisture sensors and soil conductivity sensors) are deployed in the planting area at 5m×5m intervals to collect soil parameters in real time; micro weather stations are deployed in the center and around the area to collect microclimate data; portable plant physiological instruments (such as leaf water potential meters and photosynthesis meters) are used to collect crop physiological signals regularly (once every 2 hours), and a multispectral camera mounted on a drone is used to photograph the planting area twice a week. Combined with image processing algorithms (such as threshold segmentation and texture analysis methods), crop canopy structure parameters and root distribution data are extracted (verified with the assistance of root scanning imaging technology); the irrigation device is modified by adding an angle sensor to record the water spraying angle adjustment range data; all collected data are transmitted to the central control system through a LoRa wireless communication module to achieve real-time data storage and updating.

[0022] Step S102: If the crop physiological signal exceeds the health threshold, identify the type of signal abnormality.

[0023] Specifically, the health threshold refers to the normal fluctuation range of crop physiological signals determined based on crop variety and growth stage; the signal anomaly type refers to the classification based on the characteristics of physiological signals deviating from the health threshold, such as leaf water potential below 0.8 MPa being "water stress type anomaly," and stem diameter change rate below 0.2% / h being "growth retardation type anomaly," etc. In this embodiment, the central control system pre-stores a database of physiological signal health thresholds for different crop varieties at various growth stages. The system compares the collected crop physiological signals with the corresponding health thresholds in real time. If a signal exceeds the threshold range for three consecutive collections, the anomaly identification process is triggered. The fluctuation characteristics (such as fluctuation frequency and peak deviation) of the abnormal signal are extracted using feature extraction algorithms (such as wavelet transform), and then matched with a preset anomaly type feature library to finally determine the signal anomaly type.

[0024] Step S103: Based on the signal anomaly type, match the corresponding crop growth stress factor.

[0025] Specifically, crop growth stress factors refer to factors that cause abnormal physiological signals, such as water stress, salt stress, nutrient stress, and temperature stress. In this embodiment, the stress factors can be accurately matched through vector matching and correlation verification.

[0026] Step S104: Retrieve the regional ecological model based on the anomaly type and stress factors, simulate the effect of irrigation intervention based on the regional ecological model, and evaluate irrigation priority.

[0027] Specifically, the regional ecological model is a simulation model of crop growth and irrigation response pre-constructed based on data such as the geographical environment, crop varieties, and soil types of the planting area; irrigation priority refers to the ranking of irrigation order for different planting areas or sub-regions according to the urgency of crop needs and irrigation effects; in this embodiment, the system retrieves the corresponding regional ecological model parameter configuration version from the model library according to the identified anomaly type and matched stress factor, inputs real-time target data of the planting area, simulates the recovery of crop physiological signals under different irrigation strategies, and evaluates irrigation priority through a multi-dimensional scoring system.

[0028] Step S105: If the evaluation result is high priority, then extract the target irrigation condition features from the historical irrigation database based on the current microclimate, soil parameters, crop canopy structure parameters and preset similarity threshold.

[0029] Specifically, the historical irrigation database refers to a database that stores irrigation data under different planting scenarios in the past, including environmental parameters such as microclimate, soil, and canopy structure, as well as corresponding irrigation schemes and implementation effects; the preset similarity threshold refers to the critical value for judging the similarity between the current environmental parameters and the historical parameters, which is set to 85% in this embodiment; the target irrigation condition features refer to the features reflecting the core irrigation parameters, such as irrigation volume, sprinkling angle, and irrigation duration, among the irrigation conditions similar to the current scenario selected from the historical irrigation database; in this embodiment, the similarity between the current environmental parameters and the historical parameters is calculated, and historical conditions with a similarity higher than 85% are selected, and their core features such as irrigation volume, sprinkling angle, and irrigation duration are extracted as the target irrigation condition features.

[0030] Step S106: Based on crop physiological signals, microclimate data, soil parameters, and crop canopy structure parameters, obtain the current irrigation demand parameters.

[0031] Specifically, in this embodiment, the current irrigation demand parameter refers to the normalized value of the combined irrigation volume and irrigation frequency, reflecting the overall degree of irrigation demand of the crop at present.

[0032] Step S107: Compare the fit between the current irrigation demand parameters and the target irrigation condition characteristics.

[0033] Specifically, the fit degree refers to the index that measures the degree of matching between the current irrigation demand and the historical irrigation conditions. The value range is 0-1, and the closer it is to 1, the higher the fit degree. In this embodiment, the fit degree is calculated using the cosine similarity algorithm. The current irrigation demand parameters and the target irrigation condition features are used to construct feature vectors respectively. The closer the calculation result is to 1, the higher the fit degree; the closer it is to 0, the lower the fit degree.

[0034] Step S108: If the fit is less than the critical value, a dynamic irrigation plan is generated based on root distribution data, soil parameters, and watering angle adjustment range data.

[0035] Specifically, in this embodiment, the critical value refers to the judgment standard for determining whether the fit meets the requirements, which is set to 0.7 in this embodiment; the dynamic irrigation scheme refers to a personalized irrigation strategy that is dynamically adjusted based on real-time data, combining the spatial distribution of crop roots, soil hydrological characteristics (soil water holding capacity, permeability, etc.) and the performance of irrigation devices.

[0036] Step S109: Irrigate the planting area based on the dynamic irrigation scheme.

[0037] In one embodiment of this example, step S103, which matches the corresponding crop growth stress factor based on the signal anomaly type, includes steps S201 to S208: Step S201: Obtain the feature dimensions of the signal anomaly type.

[0038] Specifically, in this embodiment, the feature dimension refers to the key attributes that describe the type of signal anomaly, including the fluctuation range of physiological indicators (the amount of change of physiological indicators per unit time), the duration of the anomaly (the duration of the abnormal signal state), and the trend of related changes of indicators (the direction and rate of change of other physiological indicators related to the abnormal signal).

[0039] Step S202: Construct anomaly feature vectors based on feature dimensions.

[0040] Specifically, in this embodiment, the abnormal feature vector refers to the vector formed by combining the quantized data of each feature dimension of the signal abnormality type in a fixed order.

[0041] Step S203: Perform vector matching between the abnormal feature vector and the preset stress factor feature library.

[0042] Specifically, in this embodiment, the preset stress factor feature library refers to a database that stores standard feature vectors of various crop growth stress factors, with each stress factor corresponding to one or more standard feature vectors.

[0043] Step S204: Calculate the cosine similarity between the abnormal feature vector and the feature vector of each stress factor.

[0044] Step S205: Select stress factors with cosine similarity higher than the matching threshold as candidate stress factors.

[0045] Specifically, the matching threshold refers to the critical value for judging whether the cosine similarity meets the requirements. In this embodiment, it is set to 0.7. Stress factors with similarity higher than this value are initially identified as factors that may cause signal abnormalities and are therefore used as candidate stress factors.

[0046] Step S206: Perform correlation verification on the candidate stress factors to verify their suitability for the environmental conditions of the planting area.

[0047] Specifically, correlation verification refers to the process of checking whether candidate stress factors match the current environmental conditions of the planting area, that is, determining whether the current environment is likely to produce the stress factor; environmental conditions refer to real-time environmental data of the planting area, such as precipitation, soil moisture content, irrigation records, etc.; fit refers to the degree of matching between candidate stress factors and environmental conditions, that is, the possibility that environmental conditions support the existence of the stress factor.

[0048] Step S207: Eliminate candidate stress factors whose adaptability to environmental conditions is below the adaptability threshold.

[0049] Specifically, the fit threshold refers to the critical value for judging whether the fit between the candidate stress factor and the environmental conditions meets the requirements. In this embodiment, it is set to 0.6.

[0050] Step S208: Sort the remaining candidate stress factors from high to low similarity, and select the stress factor ranked first as the crop growth stress factor.

[0051] Specifically, in this embodiment, the remaining candidate stress factors are sorted from largest to smallest according to their cosine similarity values, and the factor with the highest similarity is determined as the main stress factor causing the current signal abnormality.

[0052] In one embodiment of this example, step S104, which involves simulating the effects of irrigation intervention based on a regional ecological model and evaluating irrigation priorities, includes steps S301 to S307: Step S301: Divide the planting area into multiple sub-regions and obtain the target data distribution characteristics of each sub-region.

[0053] Specifically, in this embodiment, the planting area is divided into several sub-regions according to the 1m×1m specification using a grid division method. The target data of each sub-region is completed by spatial interpolation algorithm (such as Kriging interpolation), and a spatial distribution heat map of data such as soil moisture content, crop physiological signals, and root density is generated. The distribution characteristics such as mean, variance, and extreme values ​​of each sub-region are extracted. Among them, the data distribution characteristics refer to the indicators that describe the distribution of target data within the sub-region.

[0054] Step S302: Input the target data of each sub-region into the regional ecological model to simulate the recovery trend of crop physiological signals under different irrigation volumes and durations.

[0055] Specifically, the recovery trend refers to the change pattern of crop physiological signals from an abnormal state to a normal state, which is usually displayed in the form of a curve. In this embodiment, the regional ecological model is built using MATLAB / Simulink. The data such as soil moisture content, crop leaf water potential, and air temperature of each sub-region are input. The irrigation water volume gradient is set to 0-50 mm / time, the irrigation duration gradient is 1-4h, and the simulation period is 72h.

[0056] Step S303: Calculate the irrigation effect score for each sub-region based on the recovery trend.

[0057] Specifically, an irrigation effectiveness scoring system is established, which includes three indicators: physiological signal recovery rate (weight 0.4), recovery stability (weight 0.3), and water use efficiency (weight 0.3), using a percentage scoring system. In this embodiment, the recovery rate is calculated based on the time it takes for the physiological signal to reach the health threshold, with a higher score for a shorter time. The recovery stability is calculated based on the variance of the physiological signal after recovery, with a higher score for a smaller variance. The water use efficiency is calculated based on the ratio of crop biomass increment to irrigation water volume, with a higher score for a larger ratio.

[0058] Step S304: Calculate the irrigation effect scores of all sub-regions within the planting area and calculate the overall regional score.

[0059] Specifically, the regional comprehensive score refers to a comprehensive index reflecting the irrigation effect of the entire planting area, which is obtained by weighted averaging of the scores of each sub-region. In this embodiment, the weighted average method is used to calculate the regional comprehensive score, and the formula is as follows: Where S is the regional comprehensive score, The area weight of the sub-region (sub-region area / total planting area) The irrigation effect of a sub-region is scored, where m is the number of sub-regions.

[0060] Step S305: Compare the regional comprehensive score with the preset priority grading standard to determine the preliminary irrigation priority.

[0061] Specifically, the preset priority grading standard refers to the standard for dividing irrigation priority based on the regional comprehensive score. In this embodiment, a score of 90 or above is a first-level priority (high, requiring priority irrigation), a score of 80-89 is a second-level priority (medium, requiring sequential irrigation), a score of 70-79 is a third-level priority (low, requiring delayed irrigation), and a score below 70 is a fourth-level priority (temporarily suspended, not requiring irrigation). In this embodiment, the preliminary irrigation priority refers to the irrigation priority determined solely based on the regional comprehensive score, without considering crop variety and growth stage factors.

[0062] Step S306: If multiple planting areas apply for irrigation resources at the same time, obtain the crop variety importance coefficient and growth stage weight for each area.

[0063] Specifically, in this embodiment, the crop variety importance coefficient refers to a coefficient set according to the economic value and strategic significance of the crop variety. The larger the coefficient, the more important the variety. In this embodiment, the coefficient for wheat and rice is 1.2, the coefficient for vegetables is 1.0, and the coefficient for ornamental plants is 0.8. The growth stage weight refers to a weight set according to the degree of water requirement of the crop growth stage. The larger the weight, the more critical the water requirement of that stage. In this embodiment, the option weight for wheat heading stage and rice grain filling stage is 1.5, the option weight for seedling stage and maturity stage is 1.0, and the option weight for dormancy stage is 0.5.

[0064] Step S307: Combine the regional comprehensive score, variety importance coefficient and growth stage weight to adjust the preliminary irrigation priority and obtain the final irrigation priority.

[0065] Specifically, the final irrigation priority refers to the irrigation priority determined by comprehensively considering the regional comprehensive score, the importance of crop varieties, and the criticality of water requirements at the growth stage. In this embodiment, the priority score is calculated using the formula P=S×k1×k2, where k1 is the variety importance coefficient and k2 is the growth stage weight. The priority scores of each planting area are sorted from high to low to determine the final irrigation priority.

[0066] In one embodiment of this example, if the fit is less than the critical value in step S108, a dynamic irrigation plan is generated based on root distribution data, soil parameters, and watering angle adjustment range data, including steps S401 to S408: Step S401: If the fit is less than the critical value, then determine the location coordinates of the dense and sparse root areas within the planting area based on the root distribution data.

[0067] Specifically, in this embodiment, based on the root density values ​​in the root distribution data, the root density threshold is set to 5 roots / cm. 3 Areas with roots above a certain threshold are considered densely rooted areas, while those below the threshold are considered sparsely rooted areas. A Geographic Information System (GIS) is used to record the planar coordinates and soil depth coordinates of each area. For example, the coordinates of a densely rooted area are (X1-Y1, 0-30cm), indicating that the planar coordinates of this area are X1-Y1, and the soil depth is 0-30cm, classifying it as a densely rooted area.

[0068] Step S402: Based on soil parameters, analyze the soil water holding capacity and permeability coefficient at different locations.

[0069] Specifically, soil water holding capacity refers to the soil's ability to retain moisture. Sandy soils have lower field water holding capacity, while loam soils have higher field water holding capacity. Permeability coefficient is an indicator reflecting soil permeability; the higher the value, the stronger the permeability. In this embodiment, soil water holding capacity is expressed as field water holding capacity, measured at different locations using the ring sampler method. For example, the field water holding capacity is 20% in sandy soil areas and 30% in loam areas. Soil permeability coefficient is determined using a variable head infiltration test, and the formula is k. T =k 20 ×10 0.033(T-20) , where k T Let k be the permeability coefficient at temperature T. 20 is the standard permeability coefficient at 20℃, and T is the soil temperature.

[0070] Step S403: Based on the data of the water spraying angle adjustment range, calculate the water spraying coverage of the irrigation device at different angles.

[0071] Specifically, in this embodiment, the irrigation device uses a rocker arm sprinkler head with a spray angle adjustment range of 0-360°. The relationship between the spray radius and the spray angle is fitted experimentally as R=kx θ 0.2 (R is the spray radius, k) x θ is the measured correction factor corresponding to the rocker arm sprinkler head, with a value range of 1.0-1.2 (where θ is the sprinkling angle). For example, when the sprinkling angle is 90°, the sprinkling radius is 2.4m; when the angle is 180°, the radius is 3.1m. The sprinkling coverage area at different angles is simulated using CAD drawing software to generate a coordinate set of the coverage area.

[0072] Step S404: Spatial matching of the location coordinates of the dense root zone and the sparse root zone with the watering coverage area.

[0073] Specifically, in this embodiment, a spatial overlay analysis method is used to match the coordinates of the root system area with the coordinates of the watering coverage area to determine the area of ​​the dense and sparse root system that each irrigation device can cover at different angles.

[0074] Step S405: Allocate differentiated irrigation water volumes to different areas based on soil water holding capacity and permeability coefficient.

[0075] Specifically, in this embodiment, the formula is used. Calculate the irrigation water volume, where Q i Let K be the irrigation water volume for region i. g As a correction factor (1.2 for densely rooted regions and 0.8 for sparsely rooted regions), S j For the area, FC i For the regional field holding capacity, θ i This represents the current soil moisture content.

[0076] Step S406: Based on the water distribution results and the water coverage area, set the water spraying angle combination of the irrigation device.

[0077] Specifically, the combination of sprinkling angles refers to a scheme that combines multiple sprinkling angles to meet the irrigation water demand of different areas. In this embodiment, based on the irrigation water demand and sprinkling coverage of each area, a greedy algorithm is used to optimize the combination of sprinkling angles to ensure that the number of sprinkling angle adjustments is minimized while meeting the water demand.

[0078] Step S407: Develop an irrigation execution plan for different time periods based on the combination of sprinkling angles and water volume allocation.

[0079] Specifically, in this embodiment, the irrigation period is determined based on wind speed and temperature data from the microclimate data, avoiding the peak evaporation period of high temperature and high wind speed (such as 10:00-14:00), and irrigation is carried out from 6:00-8:00 and 18:00-20:00; the combination of sprinkling angles and water volume are allocated to different time periods, for example, irrigation at 90° and 120° angles is performed in the early morning to complete 60% of the water volume allocation; irrigation at 150° angle is performed in the evening to complete the remaining 40% of the water volume allocation.

[0080] Step S408: Integrate location coordinates, water allocation, sprinkler angle combination, and irrigation execution plan to generate a dynamic irrigation scheme.

[0081] Specifically, in this embodiment, location coordinates, water allocation, sprinkler angle combinations, and irrigation execution plans are integrated to generate a dynamic irrigation scheme. The dynamic irrigation scheme is presented in a combination of tables and GIS maps. The tables include information such as sub-region numbers, location coordinates, irrigation water volume, sprinkler angle, and execution time period. The GIS map marks the irrigation parameters and device deployment locations of each sub-region, forming a visualized irrigation scheme file, which is then transmitted to the irrigation control system for execution.

[0082] In one embodiment of this example, step S109, which irrigates the planting area based on a dynamic irrigation scheme, includes steps S501 to S509: Step S501: Decompose the dynamic irrigation scheme into multiple irrigation sub-tasks, each sub-task corresponding to a sub-region of the planting area.

[0083] Specifically, in this embodiment, based on the sub-region division in the dynamic irrigation scheme, the overall irrigation task is broken down into sub-tasks corresponding to the number of sub-regions. Each sub-task includes specific requirements such as the irrigation water volume, sprinkling angle, and execution time period for that sub-region.

[0084] Step S502: Assign a corresponding irrigation device and execution time window to each irrigation subtask.

[0085] Specifically, in this embodiment, the execution time window refers to a fixed irrigation time period allocated to each subtask, such as 60 minutes, to ensure that the subtask is completed within the specified time and does not conflict with the time of other subtasks.

[0086] Step S503: Within the execution time window, control the irrigation device to start irrigation according to the preset water spraying angle combination.

[0087] Specifically, in this embodiment, the irrigation control system sends control commands to the irrigation device through the Internet of Things module, adjusts the spray angle of the nozzles to a preset value, and starts the water pump to supply water.

[0088] Step S504: Collect soil moisture data and crop physiological signal feedback data in real time during the irrigation process.

[0089] Specifically, in this embodiment, soil moisture data refers to real-time data reflecting the water content in the soil; crop physiological signal feedback data refers to real-time changes in crop physiological signals during irrigation, such as leaf water potential and photosynthetic rate, which are used to determine the irrigation effect.

[0090] Step S505: Compare the collected soil moisture data and crop physiological signal feedback data with the corresponding preset indicators in the dynamic irrigation scheme.

[0091] Specifically, in this embodiment, the preset index refers to the irrigation target value set in the dynamic irrigation scheme, including the soil moisture target value and the crop physiological signal target value.

[0092] Step S506: If the soil moisture data or crop physiological signal feedback data deviates from the preset index, calculate the deviation value.

[0093] Step S507: Adjust the sprinkling angle and water output of the irrigation device according to the deviation value.

[0094] Specifically, in this embodiment, the preset deviation thresholds are 5% (soil moisture) and 10% (physiological signal). If the deviation value exceeds the threshold, the parameter adjustment process is initiated. Based on the deviation type (soil moisture deviation or physiological signal deviation), the corresponding adjustment algorithm is used to calculate the adjustment range of the sprinkling angle and water output.

[0095] Step S508: Continuously monitor the adjusted irrigation effect until the sub-area is irrigated and the data meets the standards.

[0096] Specifically, in this embodiment, after adjusting the irrigation parameters, the control system continues to collect soil and crop data at 1-minute intervals and compares them with preset indicators. If the data meets the standards, the current parameters are maintained until the irrigation duration of this subtask ends; if there is still a deviation, steps S506-S507 are repeated until the data meets the standards or the irrigation duration ends.

[0097] Step S509: Complete all irrigation sub-tasks in sequence to achieve overall irrigation of the planting area.

[0098] Specifically, in this embodiment, the control system starts the irrigation devices of each sub-area sequentially according to the execution time window of the sub-tasks. After completing one sub-task, the next sub-task is executed. After all sub-tasks are completed, the system generates an irrigation execution report, recording information such as irrigation duration, actual irrigation volume, and data compliance status of each sub-area, thereby achieving precise irrigation of the entire planting area.

[0099] In one embodiment of this example, step S507, which adjusts the sprinkling angle and water output of the irrigation device according to the deviation value, includes steps S601 to S608: Step S601: Obtain the deviation type corresponding to the deviation value. The deviation type is divided into soil moisture deviation and crop physiological signal deviation.

[0100] Specifically, in this embodiment, the control system automatically identifies the type of deviation based on the type of collected data. If the deviation value of the soil moisture data from the preset index exceeds the threshold, it is determined to be a soil moisture deviation; if the deviation value of the crop physiological signal data exceeds the threshold, it is determined to be a crop physiological signal deviation; if both exceed the threshold, the physiological signal deviation is processed first.

[0101] Step S602: If the deviation type is soil moisture deviation, retrieve the soil water holding capacity and soil permeability coefficient of the target sub-region.

[0102] Specifically, in this embodiment, the target sub-region refers to the sub-region where soil moisture deviation currently occurs. It is the specific object of parameter adjustment and also the basic unit of refined irrigation management.

[0103] Step S603: Based on the soil water holding capacity and soil permeability coefficient, determine the target amount of soil moisture that needs to be replenished or reduced.

[0104] Specifically, in this embodiment, the target amount to be supplemented refers to the total amount of water that needs to be supplemented in order to bring the soil moisture to the lower limit of the suitable range, including the water corresponding to the soil moisture difference and the evaporation loss during the irrigation interval; the target amount to be reduced refers to the total amount of water that needs to be reduced in order to bring the soil moisture to the upper limit of the suitable range, after deducting the water loss due to natural infiltration and evaporation.

[0105] Step S604: Based on the data of the water spraying angle adjustment range of the irrigation device, calculate the water spraying angle adjustment range that can make the soil moisture reach the target amount.

[0106] Specifically, the spray angle adjustment range refers to the difference in the spray angle of the irrigation device that needs to be adjusted to match the target soil moisture level. A positive value indicates an increase in the angle, and a negative value indicates a decrease in the angle. In this embodiment, the formula is used. Calculate the adjustment range of the water spray angle, where This represents the moisture difference corresponding to the target soil moisture level (the difference between the target level and the current irrigation water volume). This refers to the amount of water dispensed per unit time at the current spray angle (the amount of water output from the irrigation device per unit time at the current spray angle). This indicates the current spraying angle.

[0107] Step S605: Determine the adjustment ratio of water output based on the soil permeability coefficient.

[0108] Specifically, the adjustment ratio refers to the proportion of water output from the irrigation device that needs to be adjusted to match the soil's permeability. A ratio greater than 1 indicates an increase in water output, while a ratio less than 1 indicates a decrease in water output. The water output adjustment ratio is positively correlated with the soil permeability coefficient, as shown in the formula: , where k j Let k be the permeability coefficient of the target sub-region. avg The average permeability coefficient of the planting area (1.5 × 10⁻⁶ in this example) is used. -5 m / s).

[0109] Step S606: If the deviation type is crop physiological signal deviation, obtain the growth requirements corresponding to the current physiological signal of the crop.

[0110] Specifically, in this embodiment, the control system retrieves a database linking crop physiological signals and growth requirements. Using feature matching algorithms (such as K-nearest neighbor algorithm, cosine similarity matching algorithm, etc.), it associates the current physiological signal data with the requirement tags in the database to accurately identify the actual growth requirements of the crop and provide direction for adjusting irrigation conditions.

[0111] Step S607: Compare the growth requirements with the preset physiological signal indicators to determine the direction of irrigation condition adjustment required for signal recovery.

[0112] Specifically, in this embodiment, the preset physiological signal index refers to the physiological signal baseline value of healthy crop growth, which is the standard for judging whether the crop growth status is normal. The index varies for different crops and different growth stages. The irrigation condition adjustment direction refers to the direction of irrigation parameters that need to be adjusted in order to restore the crop physiological signal to the preset index, such as increasing / decreasing irrigation volume, expanding / shrinking the watering coverage area, increasing / decreasing irrigation frequency, etc.

[0113] Step S608: Based on the adjustment direction and combined with the crop canopy structure parameters, adjust the water spraying angle to optimize the water coverage range, and adjust the water output to match the growth needs.

[0114] Specifically, if the adjustment direction is to increase irrigation water volume, the sprinkling angle is adjusted from the original 90° to 120° based on the crop canopy structure parameters, so that the water coverage area matches the canopy projection area and reduces water evaporation loss. At the same time, the water output adjustment range is calculated according to growth needs. For example, to alleviate water stress in tomatoes, the water output needs to be increased from 10L / min to 14L / min. If the adjustment direction is to reduce irrigation water volume, the sprinkling angle is adjusted to 60° based on the canopy porosity, reducing the coverage area and the water output is reduced from 10L / min to 7L / min, ensuring that water is supplied only to the core root zone and matching the crop growth needs.

[0115] In one embodiment of this example, step S603, based on soil water holding capacity and soil permeability coefficient, determines the target amount of soil moisture that needs to be replenished or reduced, including steps S701 to S706: Step S701: Obtain historical soil moisture compliance data for the target sub-region and determine the suitable soil moisture range for the corresponding crop in the current growth stage.

[0116] Specifically, historical soil moisture compliance data refers to soil moisture data when crops were growing well during past irrigations in the planting area; suitable soil moisture range refers to the soil moisture range within which a specific crop can maintain normal physiological metabolism, ensure growth and development needs, and avoid water stress (excess or deficiency) at a specific growth stage; in this embodiment, soil moisture compliance data of the same crop and growth stage in the target sub-region over the past 3 years are extracted from the historical irrigation database, and the suitable range is determined by statistical analysis. After removing outliers, the mean and standard deviation are calculated, and the mean ± 1 standard deviation is taken as the suitable range.

[0117] Step S702: Obtain the current soil moisture of the target sub-region and compare the current soil moisture data with the suitable soil moisture range.

[0118] Specifically, current soil moisture refers to the actual moisture content of the crop root zone soil in the target sub-region at a specific point in time. It serves as a "real-time monitoring benchmark" for determining whether soil moisture deviates from the suitable range and whether irrigation adjustments need to be initiated. In this embodiment, the current soil volumetric water content is collected in real time at a sampling frequency of 1 minute / time, and the average of 5 samples is taken as the current soil moisture data, such as 22%. This data is compared with the suitable range (24%-27.2%). If the current soil moisture is determined to be below the lower limit of the suitable range, water needs to be added. If the collected data is 28%, it is determined to be above the upper limit of the suitable range, and water needs to be reduced.

[0119] Step S703: If the current soil moisture is lower than the suitable soil moisture range, analyze the evaporation rate of soil moisture in combination with the soil water holding capacity.

[0120] Specifically, evaporation rate refers to the amount of soil surface moisture lost to the atmosphere in gaseous form per unit time. In this embodiment, soil water holding capacity is characterized by a combination of field water holding capacity and soil texture. Sandy soil has weak water holding capacity and a fast evaporation rate; loam has moderate water holding capacity and a relatively stable evaporation rate. Formula E is used. S =k s ×E0, where E S E0 is the evaporation rate, and k is the reference crop evapotranspiration. s The soil moisture stress coefficient (set according to soil water holding capacity, k when field capacity is ≥30%) s =1, k when 20%-30%s =0.7).

[0121] Step S704: Based on the evaporation rate, calculate the total amount of water required to maintain soil moisture within a suitable range during the irrigation interval, and use this as the target amount to be replenished.

[0122] Specifically, in this embodiment, the target quantity that needs to be supplemented satisfies the calculation formula: , where Q 补 θ is the target quantity that needs to be supplemented. 适 To suit the lower limit of soil moisture, θ 当 H represents the current soil moisture, H represents the root depth, and E represents the current soil moisture. S t represents the evaporation rate, and t represents the irrigation interval (the irrigation interval refers to the time interval between two irrigations, which is set to 4 days in this embodiment).

[0123] Step S705: If the current humidity is higher than the suitable soil moisture range, determine the soil moisture infiltration rate based on the soil permeability coefficient.

[0124] Specifically, in this embodiment, the infiltration rate refers to the amount of water that seeps into the soil from the soil surface per unit time, and is calculated using the following formula: , where f t f is the infiltration rate. c To stabilize the infiltration rate (converted from the permeability coefficient, f) c =k T (×3600), f0 is the initial infiltration rate. The infiltration attenuation coefficient (taken as 0.2h in this embodiment) is the infiltration attenuation coefficient. -1 The target sub-region is loam, k T 2.1×10 -5 m / s, f c =2.1×10 -5 ×3600=0.0756m / h=75.6mm / h (due to model adaptation correction, the actual value is taken as 5mm / h), f0=5mm / h, and the infiltration rate at t=2h is calculated to be f(2)=5+(5-5)e -0.2×2 =5mm / h, meaning the soil moisture infiltration rate at the current time period is 5mm / h.

[0125] Step S706: Based on the microclimate data within a preset time period, calculate the total amount of water that needs to be reduced to bring the soil moisture down to a suitable range, and use this as the target amount to be reduced.

[0126] Specifically, in this embodiment, the preset time is 24 hours, and the microclimate data includes temperature, wind speed, sunshine, etc. for the next 24 hours, used to calculate evaporation loss; the current soil moisture is 28%, the suitable upper limit is 27.2%, which needs to be reduced by 0.8%, corresponding to a moisture depth Q.降基 =(28%-27.2%)×0.3×1000=2.4mm. Combining the infiltration rate of 5mm / h and the 24-hour evaporation rate of 3mm, calculate the total amount of water that needs to be reduced, Q. 减 =2.4-(5×24×0.001+3)=2.4-(0.12+3)=-0.72mm (the negative sign indicates that it needs to be achieved by reducing irrigation or accelerating drainage). That is, the target amount to be reduced is to reduce the soil moisture content by 0.8%, which corresponds to a water depth of 2.4mm. After deducting natural infiltration and evaporation, the actual amount of water that needs to be reduced by irrigation is 0.72mm.

[0127] In one embodiment of this example, if the current humidity is higher than the suitable soil moisture range, step S705, based on the soil permeability coefficient, determines the soil moisture infiltration rate, including steps S801 to S805: Step S801: Obtain soil stratification samples at different depths in the target sub-region and detect the soil texture and porosity data of each soil stratification sample.

[0128] Specifically, in this embodiment, soil texture refers to the combination ratio of different sizes of particles in the soil, which is divided into types such as sandy soil, loam, and clay. It is a key factor affecting the soil's water permeability and water retention capacity.

[0129] Step S802: Calculate the initial infiltration rate of each soil stratified sample based on the soil permeability coefficient.

[0130] Specifically, in this embodiment, the initial infiltration rate is calculated using the following formula: K L Saturated hydraulic conductivity (converted from permeability coefficient, K) L =k T ×10), h f The wetted frontal suction is taken as 5cm for sandy soil, 10cm for loam, and 20cm for clay, and h0 is the surface water depth (1cm in this embodiment). 0-10cm layer k T 2.1×10 -5 m / s, K L =2.1×10 -4 m / s = 7.56 m / h, the initial infiltration rate f0 is calculated to be 7.56 × (5 + 1) ÷ 1 = 45.36 mm / h; k for 10-20 cm layer T 1.5×10 - 5 m / s, calculated f0 = 32.4 mm / h; k for 20-30 cm layer T 0.8×10-5 m / s, calculated to be f0 = 17.28 mm / h.

[0131] Step S803: Collect historical data on soil moisture infiltration after several recent irrigations in the target sub-region and calculate the average infiltration attenuation coefficient.

[0132] Specifically, the average infiltration attenuation coefficient refers to the arithmetic mean of the infiltration attenuation coefficients after multiple irrigations, reflecting the average attenuation trend of soil infiltration rate. In this embodiment, infiltration process data after nearly 10 irrigations in the target sub-region are collected to calculate the average infiltration attenuation coefficient.

[0133] Step S804: Based on the initial infiltration rate and the average infiltration attenuation coefficient, establish a dynamic model of the soil moisture infiltration rate changing over time.

[0134] Specifically, a dynamic model refers to a mathematical model that describes the dynamic change of soil infiltration rate over time. In this embodiment, dynamic models are established for each soil layer. Taking the 10-20cm layer as an example, the initial infiltration rate f0 = 32.4mm / h, and the stable infiltration rate f c =5mm / h, average infiltration attenuation coefficient =0.20h -1 The constructed dynamic model is f(t) = 5 + (32.4 - 5)e -0.20t =5+27.4e -0.20t This model can accurately predict the infiltration rate of the soil layer at different times, providing data support for soil moisture adjustment.

[0135] Step S805: Based on the current soil moisture and suitable moisture range, determine the soil moisture infiltration rate for the current period using a dynamic model.

[0136] Specifically, in this embodiment, the current soil moisture content is 28%, the suitable upper limit of moisture content is 27.2%, and the required water infiltration is 0.8% (volume water content), which translates to a water depth of 2.4 mm (root layer depth 0.3 m). This water content is then substituted into the dynamic model for the 10-20 cm layer: f(t) = 5 + 27.4e -0.20t Numerical calculations show that when the cumulative infiltration reaches 2.4 mm, at t = 0.5 h, the infiltration rate f(0.5) = 5 + 27.4e -0.20×0.5 ≈5 + 24.8 = 29.8 mm / h, meaning the soil moisture infiltration rate of this soil layer during the current period is 29.8 mm / h. Combining the infiltration rates of each soil layer, the average infiltration rate of the target sub-region is obtained as 22.5 mm / h, providing a precise basis for calculating the target amount of soil moisture reduction.

[0137] Secondly, this application also discloses an automatic irrigation device based on artificial intelligence.

[0138] Reference Figure 2 An automated irrigation device based on artificial intelligence includes: The data acquisition module is used to collect target data of the planting area in real time. The target data includes crop physiological signals, microclimate data, soil parameters, root distribution data, crop canopy structure parameters, and irrigation device spray angle adjustment range data. The anomaly detection module is used to identify the type of signal anomaly if the crop physiological signal exceeds the health threshold. The factor matching module is used to match the corresponding crop growth stress factors based on the type of signal anomaly. The simulated irrigation module is used to retrieve regional ecological models based on anomaly types and stress factors, simulate the effects of irrigation interventions based on the regional ecological models, and evaluate irrigation priorities. If the evaluation result is high priority, the feature extraction module is used to extract the target irrigation condition features from the historical irrigation database based on the current microclimate, soil parameters, crop canopy structure parameters and preset similarity threshold. The parameter acquisition module is used to obtain current irrigation demand parameters based on crop physiological signals, microclimate data, soil parameters, and crop canopy structure parameters; The fit comparison module is used to compare the fit between the current irrigation demand parameters and the characteristics of the target irrigation conditions. If the fit is less than the critical value, the scheme generation module generates a dynamic irrigation scheme based on root distribution data, soil parameters, and watering angle adjustment range data. The irrigation module is used to irrigate the planting area based on a dynamic irrigation scheme.

[0139] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. An automatic irrigation method based on artificial intelligence, characterized in that, include: Real-time collection of target data in the planting area, including crop physiological signals, microclimate data, soil parameters, root distribution data, crop canopy structure parameters, and irrigation device spray angle adjustment range data; If crop physiological signals exceed the health threshold, identify the type of signal abnormality; Based on the type of signal anomaly, match the corresponding crop growth stress factor; Based on anomaly types and stress factors, regional ecological models are retrieved, and the effects of irrigation intervention are simulated and irrigation priorities are evaluated based on the regional ecological models. If the evaluation result is high priority, the target irrigation condition features are extracted from the historical irrigation database based on the current microclimate, soil parameters, crop canopy structure parameters and preset similarity threshold. Based on crop physiological signals, microclimate data, soil parameters, and crop canopy structure parameters, obtain current irrigation demand parameters; Compare the fit between the current irrigation demand parameters and the characteristics of the target irrigation conditions; If the fit is less than the critical value, a dynamic irrigation plan is generated based on root distribution data, soil parameters, and water spraying angle adjustment range data. The planting area is irrigated based on a dynamic irrigation scheme.

2. The automatic irrigation method based on artificial intelligence according to claim 1, characterized in that, The matching of corresponding crop growth stress factors based on signal anomaly type includes: Obtain the feature dimensions of signal anomaly types; Constructing anomaly feature vectors based on feature dimensions; The abnormal feature vectors are matched with the preset stress factor feature library; Calculate the cosine similarity between the abnormal feature vector and the feature vectors of each stress factor; Stress factors with cosine similarity higher than the matching threshold are selected as candidate stress factors; The candidate stress factors were subjected to correlation verification to verify their suitability to the environmental conditions of the planting area. Candidate stress factors whose adaptability to environmental conditions is below the adaptation threshold are eliminated; The remaining candidate stress factors are sorted from high to low similarity, and the stress factor ranked first is selected as the crop growth stress factor.

3. The automatic irrigation method based on artificial intelligence according to claim 1, characterized in that, The simulation of irrigation intervention effects and evaluation of irrigation priorities based on regional ecological models include: The planting area is divided into multiple sub-regions, and the target data distribution characteristics of each sub-region are obtained. The target data of each sub-region is input into the regional ecological model to simulate the recovery trend of crop physiological signals under different irrigation volumes and durations. Calculate irrigation effectiveness scores for each sub-region based on recovery trends; Statistically analyze the irrigation effect scores of all sub-regions within the planting area and calculate the overall regional score; The regional comprehensive score is compared with the preset priority grading standard to determine the preliminary irrigation priority; If multiple planting areas apply for irrigation resources at the same time, obtain the crop variety importance coefficient and growth stage weight for each area; By combining the regional comprehensive score, the variety importance coefficient, and the growth stage weight, the preliminary irrigation priority is adjusted to obtain the final irrigation priority.

4. The automatic irrigation method based on artificial intelligence according to claim 1, characterized in that, If the fit is less than the critical value, a dynamic irrigation plan is generated based on root distribution data, soil parameters, and watering angle adjustment range data, including: If the fit is less than the critical value, the location coordinates of the dense and sparse root areas within the planting area are determined based on the root distribution data. Based on soil parameters, the soil water holding capacity and permeability coefficient at different locations were analyzed; Based on the data on the water spraying angle adjustment range, calculate the water spraying coverage of the irrigation device at different angles; Spatial matching of the location coordinates of dense and sparse root zones with the watering coverage area; Differentiated irrigation water volumes are allocated to different regions based on soil water holding capacity and permeability coefficient; Based on the water distribution results and the water coverage area, set the water spraying angle combination of the irrigation device; Develop a time-segmented irrigation execution plan based on the combination of sprinkling angles and water volume allocation; By integrating location coordinates, water allocation, sprinkler angle combinations, and irrigation execution plans, a dynamic irrigation scheme is generated.

5. The method according to claim 1, characterized in that, The irrigation of the planting area based on the dynamic irrigation scheme includes: The dynamic irrigation scheme is broken down into multiple irrigation sub-tasks, each sub-task corresponding to a sub-region of the planting area; Assign a corresponding irrigation device and execution time window to each irrigation subtask; Within the execution time window, the irrigation device is controlled to start irrigation according to the preset water spray angle combination; Real-time collection of soil moisture data and crop physiological signal feedback data during irrigation; The collected soil moisture data and crop physiological signal feedback data were compared with the corresponding preset indicators in the dynamic irrigation scheme. If soil moisture data or crop physiological signal feedback data deviate from the preset index, calculate the deviation value; Adjust the spraying angle and water output of the irrigation device according to the deviation value; Continuously monitor the adjusted irrigation effect until the sub-area completes irrigation and the data meets the standards; Complete all irrigation sub-tasks in sequence to achieve overall irrigation of the planting area.

6. The method according to claim 5, characterized in that, The adjustment of the irrigation device's sprinkling angle and water output based on the deviation value includes: Obtain the deviation type corresponding to the deviation value. The deviation types are divided into soil moisture deviation and crop physiological signal deviation. If the deviation type is soil moisture deviation, retrieve the soil water holding capacity and soil permeability coefficient of the target sub-region; Based on soil water holding capacity and soil permeability coefficient, determine the target amount of soil moisture that needs to be replenished or reduced; Based on the data of the water spraying angle adjustment range of the irrigation device, calculate the water spraying angle adjustment range that can make the soil moisture reach the target amount. The adjustment ratio of water output is determined based on the soil permeability coefficient. If the deviation type is crop physiological signal deviation, obtain the growth requirements corresponding to the current crop physiological signal; By comparing growth requirements with preset physiological signal indicators, the direction of irrigation condition adjustment required for signal recovery is determined; Based on the adjustment direction and combined with the crop canopy structure parameters, adjust the sprinkling angle to optimize the water coverage and adjust the water output to match the growth needs.

7. The method according to claim 6, characterized in that, The determination of the target amount of soil moisture to be replenished or reduced based on soil water holding capacity and soil permeability coefficient includes: Obtain historical soil moisture compliance data for the target sub-region and determine the suitable soil moisture range for the corresponding crop at the current growth stage. Obtain the current soil moisture in the target sub-region and compare the current soil moisture data with the suitable soil moisture range; If the current soil moisture is below the suitable soil moisture range, analyze the evaporation rate of soil moisture in combination with the soil water holding capacity. Based on the evaporation rate, calculate the total amount of water that needs to be replenished during the irrigation interval to maintain soil moisture within a suitable range, and use this as the target amount to be replenished. If the current humidity is higher than the suitable soil moisture range, the soil moisture infiltration rate is determined based on the soil permeability coefficient. Based on microclimate data within a preset time period, the total amount of water that needs to be reduced to bring soil moisture down to a suitable range is calculated and used as the target amount to be reduced.

8. The method according to claim 7, characterized in that, If the current humidity is higher than the suitable soil moisture range, the determination of the soil moisture infiltration rate based on the soil permeability coefficient includes: Soil stratified samples at different depths in the target sub-region were obtained, and the soil texture and porosity data of each soil stratified sample were detected. The initial infiltration rate under each soil stratum sample was calculated by combining the soil permeability coefficient. Collect historical data on soil moisture infiltration after several recent irrigations in the target sub-region and calculate the average infiltration attenuation coefficient. A dynamic model of soil moisture infiltration rate over time is established based on the initial infiltration rate and the average infiltration attenuation coefficient. Based on the current soil moisture and suitable moisture range, the soil moisture infiltration rate for the current period is determined by a dynamic model.

9. An automatic irrigation device based on artificial intelligence, characterized in that, include: The data acquisition module is used to collect target data of the planting area in real time. The target data includes crop physiological signals, microclimate data, soil parameters, root distribution data, crop canopy structure parameters, and irrigation device spray angle adjustment range data. The anomaly detection module is used to identify the type of signal anomaly if the crop physiological signal exceeds the health threshold. The factor matching module is used to match the corresponding crop growth stress factors based on the type of signal anomaly. The simulated irrigation module is used to retrieve regional ecological models based on anomaly types and stress factors, simulate the effects of irrigation interventions based on the regional ecological models, and evaluate irrigation priorities. If the evaluation result is high priority, the feature extraction module is used to extract the target irrigation condition features from the historical irrigation database based on the current microclimate, soil parameters, crop canopy structure parameters and preset similarity threshold. The parameter acquisition module is used to obtain current irrigation demand parameters based on crop physiological signals, microclimate data, soil parameters, and crop canopy structure parameters; The fit comparison module is used to compare the fit between the current irrigation demand parameters and the characteristics of the target irrigation conditions. If the fit is less than the critical value, the scheme generation module generates a dynamic irrigation scheme based on root distribution data, soil parameters, and watering angle adjustment range data. The irrigation module is used to irrigate the planting area based on a dynamic irrigation scheme.

Images