Agricultural water-saving irrigation intelligent control system

Abstract

The present invention relates to the field of intelligent irrigation technology, specifically to an intelligent control system for agricultural water-saving irrigation. The system includes a growth stage identification module, a moisture monitoring module, a meteorological data synchronization module, an irrigation strategy scheduling module, and a feedback correction module. In the present invention, through time series analysis of the increase in key rice growth parameters, high-precision identification of crop stage status is achieved, the accuracy of stage determination is improved, key growth period blocks are screened, control targets are focused, and resource utilization efficiency is improved. Humidity data is collected and compared in layers, a deep and shallow dynamic response structure is constructed, and the resolution of moisture changes is enhanced. Humidity rhythms are synchronized with meteorological forecasts to improve control foresight, eliminate precipitation interference windows, and combine water demand gears to achieve block-based precision irrigation. The control results are fed back via dual indicators of humidity and leaf color difference, and a setting mapping is constructed to form a closed-loop control mechanism, enhancing adaptability and execution accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent irrigation, and in particular to an intelligent control system for agricultural water-saving irrigation. Background Art

[0002] The field of intelligent irrigation technology includes relevant technologies for precise management and allocation of water resources in agricultural production. The core content of this technology field is to achieve efficient use of water resources through real-time monitoring and data analysis of environmental factors, crop water requirements and the operating status of the irrigation system. Intelligent irrigation systems generally combine meteorological monitoring, soil moisture detection, water volume control, remote communication and other means to build a control system that can automatically make decisions and execute irrigation tasks. The entire technology field covers the collaborative work of the perception layer, control layer and execution layer, involving data collection of environmental perception equipment, data interaction of information transmission technology, and regulation of irrigation execution devices. It has the characteristics of data-driven, dynamic regulation and system integration.

[0003] Among them, the agricultural water-saving irrigation intelligent control system refers to an irrigation control system constructed through environmental data collection, moisture assessment, crop demand analysis and water supply scheduling to address the problems of water resource waste and inaccurate irrigation timing in agricultural water management. The system mainly covers the real-time collection of surface and root layer moisture data through soil moisture sensor elements, water demand calculation based on crop type and growth period preset model, and dynamic correction based on meteorological parameters such as rainfall forecast and evaporation intensity. The actuator then controls the opening and closing of the valve to allocate water sources. By integrating geographic location annotation function and original data backtracking algorithm, the moisture distribution trend in specific areas is analyzed and judged, thereby realizing intelligent irrigation decision-making. The entire process relies on sampling analysis, parameter judgment and threshold comparison to complete various control tasks.

[0004] Existing technologies generally rely on static modeling based on crop type and growth period to assess water requirements, lacking dynamic analysis of real-time growth status. This results in an inability to accurately reflect the current crop stage, given the significant regional variations in actual conditions. This can lead to mistimed or off-frequency irrigation. Moisture monitoring is generally focused on the soil surface, lacking a multi-layered response analysis mechanism. Consequently, control actions are based on a single data dimension, failing to reflect the true water infiltration state and increasing the risk of overwatering or irrigation delays. While meteorological factors are incorporated into control, they are primarily treated as independent variables, failing to establish a predictive model that integrates them with soil moisture. This makes it difficult to adjust the irrigation schedule in response to drastic changes in precipitation or wind speed. Implementation strategies often employ regional averages or fixed irrigation cycles, lacking block-based intervention and dynamic adjustment strategies, resulting in inaccurate water resource allocation. Feedback mechanisms often rely solely on post-implementation moisture change checks, lacking a mapping chain between control settings and feedback status. This makes it impossible to classify, trace, and optimize abnormal areas, limiting the adaptive optimization capabilities and closed-loop control efficiency. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an intelligent control system for agricultural water-saving irrigation.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: The agricultural water-saving irrigation intelligent control system includes:

[0007] The growth stage identification module obtains agricultural irrigation control block data, including rice leaf width, plant internode length, and tillering density. It compares the data with the crop growth stage benchmark sequence to determine the stage interval to which the current sample belongs. It then selects blocks distributed between the tillering and heading periods to generate a stage block identification map.

[0008] The moisture monitoring module collects moisture readings of the topsoil and bottom layers of the corresponding blocks based on the stage block identification map, compares the change direction and amplitude difference of the two layers of data according to the sampling time series, identifies the dynamic change characteristics of humidity in the stage block, and generates a layered moisture distribution state;

[0009] The meteorological data synchronization module identifies the recent precipitation forecast sequence and wind speed increase sequence in the region based on the stratified humidity distribution state, synchronizes the humidity change rhythm of the current block with the predicted meteorological elements, calculates the overlap ratio of the overlapping fluctuation intervals, and generates the water vapor interaction change trend;

[0010] The irrigation strategy scheduling module calls the water-gas interaction change trend, classifies the stage block map, eliminates the area overlapping with the precipitation forecast window, compares the block humidity reading with the stage water demand level, marks the time period and area that need to be regulated, and generates an irrigation intervention interval configuration table.

[0011] As a further solution of the present invention, the stage block identification map includes growth stage classification labels, block time series indexes, and tillering and heading block identification information; the stratified humidity distribution status includes the amplitude of the humidity change in the plow layer, the bottom layer humidity response delay, and the inter-layer humidity difference characteristics; the water-vapor interaction change trend includes the humidity fluctuation overlap rate, the precipitation prediction matching interval, and the wind speed increasing interference factor; the irrigation intervention interval configuration table includes the control block number, the target humidity gear, and the intervention time window.

[0012] As a further solution of the present invention, the growth stage identification module includes:

[0013] The widening parameter extraction submodule obtains agricultural irrigation control block data, identifies the widening change sequence through the same block time series, calculates the increase rate and compares it with the set benchmark interval, extracts the irrigation control blocks that meet the widening dynamic characteristics, and generates a widening characteristic response block number set;

[0014] The internode segment determination submodule calls the widening feature response block number set, extracts internode length and tiller density data, identifies internode growth rate and tiller density increment value, selects block numbers and time periods that are simultaneously within the dual intervals of internode lengthening and density increase, and generates a water-saving regulation sensitive section identification set;

[0015] The growth block determination submodule compares the tillering density and the increase trend of the time nodes in the block according to the water-saving regulation sensitive section identification set, matches the stage benchmark increase range of the tillering period and the heading period, extracts the effective blocks and spatial positions within the stage range, and generates a stage block identification map.

[0016] As a further solution of the present invention, the moisture monitoring module includes:

[0017] The topsoil humidity collection submodule collects topsoil and bottomsoil humidity readings based on the block identification map of the stage, unifies the sampling time series, removes anomalies and repairs missing segments of block data, extracts valid humidity samples, and generates a basic set of block humidity monitoring;

[0018] The humidity change comparison submodule calls the block humidity monitoring basic set, extracts the humidity change values ​​of the plow layer and the bottom layer in adjacent time periods, compares the humidity change direction and increase and decrease amplitude of the two layers, identifies the coupling characteristics, and generates a coupling change coefficient set;

[0019] The stratified humidity modeling submodule calls the coupling variation coefficient set, selects the trend stable block, extracts the increase and decrease amplitude of the plow layer humidity and the bottom layer change rate, combines the sampling time interval, sequence position, and number of sample points, calculates the stratified humidity control index value, combines the spatial range of the irrigation operation block and the time node for mapping, and generates the stratified humidity distribution state.

[0020] As a further solution of the present invention, the meteorological data synchronization module includes:

[0021] The humidity identification submodule extracts humidity data for each layer based on the stratified humidity distribution state, combines the precipitation forecast and wind speed sequence, analyzes the humidity change rhythm and the fluctuation amplitude of meteorological elements, determines whether the difference between the two is within the humidity matching threshold range, selects the number sequence that meets the conditions, and generates an irrigation intervention warning number group;

[0022] The overlapping section determination submodule selects the wind speed increase and precipitation fluctuation information of the corresponding time period based on the irrigation intervention warning number group, identifies the overlapping time period of the two types of meteorological data, calculates the wind-precipitation coupling offset, and filters the time period numbers that exceed the interactive response limit to obtain the response section number set that needs to be regulated;

[0023] The control trend derivation submodule calls the set of response section numbers that need to be controlled, compares the humidity and wind drop interaction data of the benchmark sections, extracts the rhythm overlapping positions, calculates the humidity adjustment differences of each type of interaction combination in groups, and classifies them into intervention control levels, summarizes the level change trajectories, and generates water vapor interaction change trends.

[0024] As a further solution of the present invention, the irrigation strategy scheduling module includes:

[0025] The trend classification submodule calls the water-gas interaction change trend, extracts the meteorological change direction and amplitude value of the region in the stage block map, divides the trend type according to the continuity and gradient change judgment, eliminates the block numbers that overlap with the precipitation forecast window in the time period, and generates a controllable trend block set;

[0026] The humidity comparison submodule calls the adjustable trend block set, collects the current soil moisture reading of the block and the standard range of the water demand level in the corresponding stage, compares the difference between the humidity reading and the water demand lower limit by block, calculates the irrigation intervention priority value, sorts the block numbers, and obtains the control priority block sequence;

[0027] The intervention interval identification submodule extracts the corresponding control period interval according to the block number in the control priority block sequence, sets the intervention level and irrigation period configuration according to the continuous humidity deviation, and generates an irrigation intervention interval configuration table.

[0028] As a further solution of the present invention, the system further includes a feedback correction module:

[0029] The feedback correction module detects the change in shallow humidity readings and leaf color difference indicators after the end of irrigation according to the irrigation intervention interval configuration table, screens and groups the humidity recovery amplitudes of blocks that deviate from the expected control, marks the corresponding control settings of the deviated blocks, and generates stage control offset labels;

[0030] The stage control offset label includes the humidity recovery deviation level, the leaf color difference change amplitude, and the control setting offset mark.

[0031] As a further solution of the present invention, the feedback correction module includes:

[0032] The humidity fluctuation screening submodule extracts the block irrigation end time according to the irrigation intervention interval configuration table, identifies the shallow humidity data after the corresponding time, analyzes the difference between the humidity recovery amplitude and the control target value, screens the blocks that exceed the recovery deviation threshold, and generates a humidity control deviation block set;

[0033] The leaf color difference measurement submodule extracts the color difference readings of the leaves in the corresponding blocks before and after irrigation based on the humidity control deviation block set, analyzes the color difference change amplitude, and classifies them according to the ratio of the color difference change value to the humidity recovery deviation value, filters out the blocks with inconsistent responses, and generates a water-photosynthesis imbalance distribution group;

[0034] The control offset identification submodule extracts the corresponding control parameters according to the water-photosynthesis imbalance distribution group and compares them with the control targets in the configuration table, marks the blocks and set parameters for setting the offset, and generates a stage control offset label.

[0035] Compared with the prior art, the advantages and positive effects of the present invention are:

[0036] In the present invention, by performing timeline sequence analysis on the growth trend of key rice growth parameters, high-precision identification of crop stage status can be achieved, breaking through the problem of stage judgment errors caused by relying on static models or coarse-grained time division in the past. Blocks concentrated on key growth nodes are screened to effectively focus on the target range that needs to be controlled, reducing resource waste and redundant judgment interference. Vertical layered collection and trend comparison of humidity data establishes a dynamic response structure based on deep and shallow layer differences, enhances the ability to analyze moisture changes, avoids misjudgments caused by single-layer indicators, and based on the synchronous fitting mechanism of humidity data rhythm and meteorological forecast elements, strengthens the system's response and prediction capabilities for future short-term fluctuations, and improves the foresight and stability of humidity control. By eliminating the overlapping areas of precipitation interference windows, the independence and accuracy of water resource input are guaranteed. Further combined with the stage water demand gear, the spatiotemporal accuracy of the control behavior is limited, and the block-based and high-frequency precise matching of irrigation tasks is achieved. The control results are evaluated by feedback of the dual indicators of humidity and leaf color difference, and traceable setting mapping is marked for deviations, enhancing the closed-loop controllability and optimization path identification capabilities of the control behavior. The overall process starts with stage identification, and runs through layered perception, dynamic prediction, strategy allocation and result feedback, forming a full-cycle closed-loop control mechanism with the coordinated characteristics of forward judgment and backward verification, enhancing the comprehensive performance of irrigation execution in terms of accuracy, timeliness and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a system flow chart of the present invention;

[0038] Figure 2 This is a flow chart of the growth stage identification module in the present invention;

[0039] Figure 3 This is a flow chart of the moisture monitoring module in the present invention;

[0040] Figure 4 This is a flow chart of the meteorological data synchronization module in the present invention;

[0041] Figure 5 This is a flow chart of the irrigation strategy scheduling module in the present invention;

[0042] Figure 6 This is a flow chart of the feedback correction module in the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0044] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0045] See also Figure 1 , the agricultural water-saving irrigation intelligent control system includes:

[0046] The growth stage identification module obtains agricultural irrigation control block data, including rice leaf width, plant internode length, and tillering density. Based on the increase relationship of each data item within the time axis, it compares it with the crop growth stage benchmark sequence, determines the stage interval to which the current sample belongs, selects blocks distributed between the tillering period and the heading period, and generates a stage block identification map;

[0047] The moisture monitoring module collects moisture readings from the topsoil and bottom layers of the corresponding blocks based on the phase block identification map. It compares the change direction and amplitude differences of the two layers of data according to the sampling time series, identifies the dynamic change characteristics of humidity within the phase block, and generates a stratified moisture distribution state.

[0048] The meteorological data synchronization module identifies the recent precipitation forecast sequence and wind speed increase sequence in the region based on the stratified humidity distribution state, synchronizes the humidity change rhythm of the current block with the predicted meteorological elements, calculates the coincidence ratio of overlapping fluctuation intervals, and generates the water vapor interaction change trend;

[0049] The irrigation strategy scheduling module uses the water-gas interaction change trend to classify the stage block map, eliminate the areas overlapping with the precipitation forecast window, compare the block humidity readings with the stage water demand level, mark the time periods and areas that need to be regulated, and generate the irrigation intervention interval configuration table;

[0050] The feedback correction module detects the change in shallow humidity readings and leaf color difference indicators after the end of irrigation based on the irrigation intervention interval configuration table, screens and groups the humidity recovery amplitude of the blocks that deviate from the expected control, marks the corresponding control settings of the deviated blocks, and generates stage control offset labels.

[0051] The stage block identification map includes the growth stage classification label, block time series index, tillering and heading block identification information, the stratified humidity distribution status includes the humidity change amplitude of the plow layer, the bottom layer humidity response delay, and the humidity difference characteristics between layers, the water-vapor interaction change trend includes the humidity fluctuation overlap rate, the precipitation prediction matching interval, and the wind speed increase interference factor, the irrigation intervention interval configuration table includes the control block number, the target humidity gear, and the intervention time window, and the stage control offset label includes the humidity recovery deviation level, the leaf color difference change amplitude, and the control setting offset mark.

[0052] See also Figure 2 , the growth stage recognition module includes:

[0053] The widening parameter extraction submodule obtains agricultural irrigation control block data, identifies the widening change sequence through the same block time series, calculates the increase rate and compares it with the set benchmark interval, extracts the irrigation control blocks that meet the widening dynamic characteristics, and generates a widening characteristic response block number set;

[0054] Analyze the time series of agricultural irrigation blocks to identify widening change sequences. Widening change refers to changes in irrigation conditions or environmental factors within an irrigation block over a certain period of time, which affect the growth environment of the irrigated area and cause changes in control parameters or the size of the area. In specific implementation, the original irrigation data of each block is first collected, including influencing factors such as soil moisture, temperature, and precipitation. Data analysis is performed on the irrigation block at each time node. By comparing the irrigation area and related environmental data in different time periods, the widening change pattern of the block is identified. For example, if soil moisture drops sharply during a certain period of time, the irrigated area will expand. Therefore, it is necessary to calculate the relative growth rate to determine whether the block meets the widening dynamic characteristics. This increase rate can be calculated by comparing the changes in the irrigation area of ​​the same block in different time periods. For example, if the irrigation area of ​​a block increases from 10 hectares to 15 hectares, the increase rate is 50%. Based on the preset benchmark interval and the increase rate, the blocks that meet the widening dynamic characteristics are screened out. Assuming the preset benchmark interval is 20%, the blocks with an increase rate exceeding 20% ​​will be selected and included in the widening characteristic response block number set.

[0055] The internode segment determination submodule calls the widening feature response block number set, extracts internode length and tiller density data, identifies internode growth rate and tiller density increment values, selects block numbers and time periods that are simultaneously within the dual intervals of internode lengthening and density increase, and generates a water-saving regulation sensitive segment identification set;

[0056] Internode length and tiller density data were extracted for each plot. Internode length is related to internode extension during the plant's growth cycle, while tiller density directly affects water absorption and distribution. By collecting data from each plot at different time points, the internode growth rate and tiller density increment values ​​can be calculated. The internode growth rate is calculated as the difference between the current internode length and the internode length at the previous time point, divided by the internode length at the previous time point. For example, if the internode length of a block increases from 5 cm to 6 cm between two measurements, the growth rate is (6-5) / 5 = 20%. The tiller density increase is calculated as the difference between the tiller density at the current time node and the density at the previous time node divided by the density at the previous time node. Assuming that the tiller density of a block increases from 100 plants / m2 to 120 plants / m2 between the two measurements, the increase is (120-100) / 100 = 20%. The block numbers and corresponding time periods that fall within both the internode lengthening and tiller density increasing intervals are selected. A reasonable interval value is set. For example, blocks with an internode growth rate greater than 10% and a tiller density increase greater than 15% are identified as water-sensitive regulation sections, ultimately forming a water-sensitive regulation section identification set.

[0057] The growth block determination submodule compares the tillering density and the growth trend of the time series nodes within the block based on the water-saving regulation sensitive section identification set, matches the stage benchmark growth range of the tillering period and the heading period, extracts the effective blocks and spatial positions within the stage range, and generates a stage block identification map;

[0058] The growth status of each block is judged by combining the growth trend of tiller density at different stages. According to the water-saving regulation sensitive section identification set, the tiller density data of each block in different time periods are identified and compared according to the time series nodes. The time series node is a specific stage (such as tillering period, heading period, etc.) and the growth trend of the target growth process. By comparing the growth trend of tiller density with the time series node, it is judged whether the block has entered the tillering period or heading period. For example, if the tillering density of a block increases by more than a set benchmark increase (e.g., 10%) within a specific time period and meets the growth characteristics of that stage, the block is identified as a growth stage block. A clear benchmark increase range must be defined, such as setting the benchmark increase for the tillering stage at 8%-12% and the benchmark increase for the heading stage at 5%-7%. The original data is then compared to ensure that the data for each block meets the stage characteristics. In this way, blocks that meet the growth characteristics can be effectively screened out, and a stage block identification map can be generated to provide a precise reference for irrigation regulation and control, ensuring water conservation and yield increases in agricultural production.

[0059] Table 1: Example data for internode growth rate and tiller density increase

[0060] ;

[0061] As shown in Table 1, the table lists the data of internode length and tiller density of certain blocks, as well as their corresponding growth rate and increase value. Through the data, we can further screen out blocks that meet the sensitive sections of water-saving regulation, helping to accurately identify irrigation regulation strategies.

[0062] See also Figure 3 , the moisture monitoring module includes:

[0063] The topsoil moisture collection submodule collects topsoil and bottomsoil moisture readings based on the phase block identification map, unifies the sampling time series, removes anomalies and repairs missing segments of block data, extracts valid moisture samples, and generates a basic set of block moisture monitoring.

[0064] The collection of the humidity of the plow layer and the bottom layer is partitioned. The boundary coordinate values ​​of each agricultural farming block identified in the atlas are called one by one. In combination with the embedded soil moisture sensor, four plow layer monitoring points are arranged in the range of 0-20cm and four bottom layer monitoring points are arranged in the range of 20-50cm in each block. The sampling cycle is uniformly set to 5 minutes. The data is identified and synchronized based on the collection timestamp and block code. The raw humidity data in each time period is filtered to eliminate abnormal values. When the readings of two sampling points at the same level in the same block exceed the set range [the set humidity range of the plow layer is 5%-40%, and the set humidity range of the bottom layer is 10%-45%] The abnormal segments are marked as abnormal, and the missing values ​​are repaired by linear interpolation of the front and back points. The remaining valid data after elimination are archived and organized in the order of sampling time. The archived data are grouped and organized by block ID and level to generate the distribution rate. The humidity distribution rate curve is obtained by counting the ratio of the sampling frequency to the total number of times in each humidity value interval. For example, in a certain sampling period, the humidity of the plow layer is 15%, 16%, 15%, and 14%, and that of the bottom layer is 28%, 27%, 29%, and 27%. Combined with the total sample size and the distribution interval, it is divided into 5%–10%, 10%–15%, 15%–20%, etc. The corresponding humidity distribution rate is obtained by counting the frequency, and finally the basic set of block humidity monitoring is obtained.

[0065] The humidity change comparison submodule calls the block humidity monitoring basic set, extracts the humidity change values ​​of the plow layer and the bottom layer in adjacent time periods, compares the change direction and increase and decrease amplitude of the humidity of the two layers, identifies the coupling characteristics, and generates a coupling change coefficient set;

[0066] Extract the humidity readings of the plow layer and the bottom layer in the same block between two consecutive cycles, and use the increase and decrease method to calculate the humidity change value of the plow layer as the value obtained by subtracting the previous cycle from the current cycle. The same is true for the bottom layer. Record the increase and decrease amplitudes of the plow layer and the bottom layer respectively, and judge whether their change directions are consistent. Consistent direction means that both are positive or negative, and inconsistent direction means one is positive and the other is negative. Based on this setting, construct a direction consistency index, which is recorded as 0 or 1. If the direction is consistent, it is 1, otherwise it is 0. Subtract the change values ​​of the plow layer and the bottom layer and take the absolute value as the difference in humidity change between the two layers. The coupling threshold is set to 5%, that is, when the change difference is less than 5% and the direction is consistent, it is judged to be a strong coupling state, otherwise it is a weak coupling state. For example, in a block, the humidity of the tillage layer changes from 20% to 23% in the two cycles t1 and t2, and the bottom layer changes from 30% to 31%. The direction consistency is 1, and the difference is 3%-1%=2%, which is less than 5%. It is recorded as strong coupling. Then, the strong coupling ratio of the block is calculated for all time. If the ratio of the number of strong coupling state cycles to the total number of cycles is higher than 0.7, it is marked as a trend stable block, and finally a set of coupling change coefficients is generated.

[0067] The layered humidity modeling submodule calls the coupling variation coefficient set, selects the trend stable block, extracts the increase and decrease amplitude of the arable layer humidity and the change rate of the bottom layer, and combines the sampling time interval, sequence position, and number of sample points to use the formula:

[0068] ;

[0069] Calculate the stratified humidity control index value, combine the spatial range of the irrigation operation block with the time node for mapping, and generate the stratified humidity distribution state;

[0070] in, Represents the index value of stratified humidity control, Representative Block The variation range of the plough layer humidity during the period, Indicates the Block The rate of change of bottom humidity during the period, Indicates the The sum of squares of the sampling time intervals of the block, Indicates the The average position of the sampling sequence of the block, Indicates the The number of valid sample points in the block, Indicates the total number of time periods;

[0071] The stratified moisture control index is a numerical indicator that measures the degree of coordination between moisture changes in the agricultural plough layer and the underlying soil within a time series. This value is obtained by weighting the amplitude of moisture changes in the plough layer and the rate of change of moisture in the underlying soil, and is normalized by combining structural parameters such as sampling time interval, sequence position, and sample number. It reflects the consistency of soil moisture conduction and response synchronization at the vertical level. The higher the value, the more consistent the humidity fluctuations in the plough layer and the underlying soil within the same period, and the more coupled the dynamic trends. This can be used as one of the bases for zoning priority water supply in intelligent irrigation. If the value is too low, it indicates that the moisture responses of the upper and lower layers are not synchronized, and water allocation and irrigation intensity need to be adjusted to achieve precise regulation under the agricultural water-saving irrigation strategy.

[0072] Extract the change range of the tillage layer humidity and the rate of change of the bottom layer humidity in each cycle humidity sample, the change range of the tillage layer humidity The method of obtaining is: subtract the humidity value of the previous cycle from the topsoil humidity value of the current sampling cycle, take the absolute value and retain two decimal places. The humidity unit is percentage. For example, the topsoil humidity readings of a block from t1 to t5 are 21.0%, 23.1%, 24.8%, 26.7%, and 29.0%, respectively. The change ranges are 2.1%, 1.7%, 1.9%, and 2.3%, respectively.

[0073] Bottom humidity change rate The difference between the bottom humidity of the current cycle and the previous cycle divided by the time interval , the unit is % / h, assuming that the bottom humidity is 30.0%, 30.8%, 32.2%, 34.1%, and 36.0% in sequence, and the sampling time is once every 2 hours;

[0074] The rates are: ;

[0075] The obtained rate sequence is 0.4, 0.7, 0.95, and 0.95% / h. To ensure the uniformity of the dimensions of each participating item and the equivalence of its influence, it is necessary to normalize each parameter. That is, the variation amplitude of the tillage layer and the bottom layer rate are respectively subtracted from the minimum value and then divided by the range, so that the normalized data are distributed in the interval [0, 1]. For example, the range of the tillage layer sequence is 2.3-1.7=0.6, and the minimum value is 1.7.

[0076] After normalization, they are: ;

[0077] The bottom rate range is 0.95-0.4=0.55, and the minimum value is 0.4. After normalization: ;

[0078] Finally substitute into the formula:

[0079] ;

[0080] in, is the normalized layered humidity control index value, which represents the dynamic coordination score of the topsoil-bottom layer humidity. The value range is [0, 1]. The higher the value, the stronger the matching of the humidity changes between the upper and lower layers, and the more suitable the water allocation area in the intelligent irrigation quantitative strategy. If the score is lower than 0.3, it is marked as a poorly coordinated area and requires local coupling optimization. Finally, all blocks are The index values ​​correspond to their spatial locations one by one, and a water-saving irrigation control distribution map is constructed;

[0081] : Layered humidity control index value (dimensionless, normalized);

[0082] : Change range of plough layer humidity, unit %, normalized;

[0083] : bottom humidity change rate, unit % / h, normalized;

[0084] : the sum of squares of the sampling time intervals of the block, unit h²;

[0085] : The mean of the sampling sequence position (i.e. the average of the sampling numbers);

[0086] : number of valid sampling points (unitless);

[0087] By coupling the synchronous dynamics of the topsoil and bottom layers and considering time and sample size factors, a reasonable block priority map can be generated to guide agricultural water-saving irrigation regulation. The results show that the current block coordination is relatively low and it is advisable to adjust or re-arrange the sampling points.

[0088] See also Figure 4 , the meteorological data synchronization module includes:

[0089] The humidity identification submodule extracts humidity data for each layer based on the stratified humidity distribution state. Combining it with precipitation forecasts and wind speed sequences, it analyzes the humidity variation rhythm and the fluctuation amplitude of meteorological elements, determines whether the difference between the two is within the humidity matching threshold range, selects matching number sequences, and generates irrigation intervention warning number groups.

[0090] Measured moisture data is collected from different soil layers, which can be set to three depths: 0-20cm, 20-40cm, and 40-60cm. Capacitive soil moisture sensors are deployed to regularly record the area's daily average humidity. Hourly precipitation forecast data for the next 72 hours, as well as wind speed change sequences, are retrieved from the meteorological department. During execution, the fluctuation trends of each layer's moisture data are compared with the precipitation forecast data on a daily basis to determine whether the humidity change rhythm is linked to precipitation or wind speed trends. The absolute value of the difference is calculated, and the humidity matching threshold is set to ±2%. When the difference in the change amplitude per unit time between the two sequences is less than 2%, the rhythm is considered consistent. Time points that meet the criteria are marked with numbers. For example, the humidity in a particular area varied by 32%, 34%, and 37% from May 10th to 12th, while the predicted precipitation was 0mm, 2mm, and 4mm, and the wind speed was 3m / s, 3.5m / s, and 4.2m / s, showing the same trend. The calculated differences were 2%, 2%, and 3%, respectively. The difference for the first two days was within the set threshold, so the number was assigned to the humidity rhythm matching number for the corresponding time period of the area. Similarly, all segment numbers that met the rhythm difference judgment criteria were aggregated and labeled. The ±2% threshold was constructed based on the difference in average response to evaporation loss between soil layers and wind speed disturbances. Based on years of field experimental records in the Hebei Plain, humidity changes of less than 2% do not cause crop water stress. Therefore, it has high adaptability as a synchronization judgment standard, ultimately resulting in an irrigation intervention warning number group.

[0091] The overlap section determination submodule selects the wind speed increase and precipitation fluctuation information of the corresponding time period according to the irrigation intervention warning number group, identifies the overlapping time section of the two types of meteorological data, and uses the formula:

[0092] ;

[0093] Calculate the wind drop coupling offset, filter the time period numbers that exceed the interactive response limit, and obtain the response section number set that needs to be regulated;

[0094] in, represents the windfall coupling offset, It represents the total amplitude of the wind speed sequence within the segment. Indicates the precipitation variation range value, is the fluctuation frequency of humidity rhythm, is the crossover value difference between wind speed and precipitation, is the humidity synchronization benchmark difference, is the segment superposition offset intensity factor;

[0095] The wind-precipitation coupling deviation is a comprehensive indicator that measures the intensity of the combined impact of wind speed and precipitation on soil moisture changes over a specific time period. Its core is to quantify and integrate the amplitude of wind speed fluctuations, precipitation changes, humidity fluctuation frequency, and their differences from the original humidity baseline to form a unified metric that reflects the necessity of meteorological disturbances for agricultural irrigation intervention during that period. A higher value indicates a stronger disturbance to humidity caused by the synergistic effect of wind and precipitation during that period, and a higher irrigation response requirement. This indicator ensures dimensional consistency during the calculation process by normalizing different physical quantities and superimposing them into a unified calculation model. This facilitates comprehensive comparisons across meteorological factors and provides a highly recognizable reference for subsequent irrigation decisions.

[0096] Select the meteorological data for the time period corresponding to the marked number and extract the fluctuation data of the wind speed series and precipitation series. Set the observation period corresponding to each group of numbers to 24 hours, obtain the maximum and minimum values ​​of wind speed and precipitation in each period, and calculate their fluctuation ranges, which serve as the input value source for parameters a and b. For example, in the time period numbered 20240511_03, the wind speed increased from 2.8m / s to 5.2m / s, and the precipitation increased from 0mm to 4.5mm, corresponding to the fluctuation ranges a=2.4m / s and b=4.5mm. The variance c of the humidity rhythm fluctuation frequency is obtained by extracting the hourly soil moisture change data in the numbered segment and calculating the variance value of the change frequency. Assuming that the humidity data series recorded in this period are: 31%, 33%, 35%, 36%, 38%, 37%, 36%, 34%, with a total of 7 change points and a frequency of 7 times, the variance is processed by the standard calculation formula to be c=2.1;

[0097] Parameter d is the sequence difference at the intersection of wind speed and precipitation, and needs to be dimensionally unified before processing. Since wind speed is m / s and precipitation is mm / hour, normalization is required for cross-comparison. A normalization standard is set, with wind speed normalized to a maximum of 5 m / s and precipitation normalized to a maximum of 10 mm / h. Thus, a wind speed of 2.8 m / s and precipitation of 2.4 mm / h are normalized to 0.56 and 0.24, respectively, corresponding to an intersection difference d = 0.32. The humidity synchronization benchmark difference e is set to 0.15 with reference to the original sample. This value is derived from a comparative analysis of humidity fluctuations and meteorological parameters in regional spring wheat plots over a 10-year period. The median range is 0.12 to 0.18, and the mean is taken to determine the value.

[0098] The superposition offset intensity factor f is not directly assigned as a constant, but is determined by the daily change rate of wind and precipitation fluctuations. If the change rate of wind speed or precipitation exceeds 50% within 6 hours, it is defined as a "strong disturbance" and its corresponding factor is set to 0.8. The medium disturbance (30%-50%) is 0.6, and the weak disturbance (<30%) is 0.4. In this example, the wind speed changes from 2.8 to 5.2, with a change rate of 85.7% within 6 hours, which is a strong disturbance, so f=0.8;

[0099] Substituting into the formula: ;

[0100] The wind-drop coupling offset was 59.62, which is much higher than the interactive response limit of 6.0. This result indicates that the soil moisture cooperative fluctuation caused by wind-drop changes in the time period corresponding to this number is strong, and it needs to be marked as an irrigation response area in the regulation. By introducing the sum of wind speed and precipitation amplitude, combined with the frequency change structure of the humidity rhythm, the nonlinear interaction relationship between different source terms is dimensionally normalized and integrated, so that complex meteorological factors form a unified response evaluation scale in soil moisture regulation. The result value far exceeds the response threshold, indicating that active irrigation operation should be triggered in this area, thereby obtaining the set of response section numbers that need to be regulated.

[0101] The control trend derivation submodule calls the number set of the response section that needs to be controlled, compares the humidity and wind drop interaction data of the target section, extracts the rhythm overlap position, calculates the humidity adjustment difference of each type of interaction combination by group, and classifies it into intervention control levels. The level change trajectory is summarized to generate the water vapor interaction change trend;

[0102] During the time period covered by the number, the corresponding humidity node value was retrieved, and the numerical relationship between it and the corresponding time position of the wind-drop interaction index was counted. The overlapping rhythm points were further divided and summarized according to different types of combination segments. Taking numbers 101, 103, and 106 as examples, their humidity node values ​​were 35%, 38%, and 42%, respectively. The corresponding wind speed changes were 4.0m / s, 4.6m / s, and 5.1m / s, and the precipitation values ​​were 3mm, 4mm, and 5mm. These were classified as a humidity-increased wind speed-enhanced precipitation combination. The frequency of recurrence of this combination was counted under multiple numbers. If it recurred more than 10 times within 30 days, it could be preliminarily delineated as a significant interaction trend segment. Then calculate the humidity difference value under the combination respectively, and take the difference between the maximum and minimum humidity values ​​in the group as the reference basis for the control level. For example, 42%-35%=7% belongs to the moderate control level. The reference setting of the level division interval is: 0-3% is weak, 3-7% is medium, and above 7% is strong. Arrange all level combinations in chronological order to generate a humidity control trend line, and identify the key change interval according to the fluctuation segment on the trend line, and finally generate the water vapor interaction change trend.

[0103] See also Figure 5 , the irrigation strategy scheduling module includes:

[0104] The trend classification submodule calls the water-gas interaction change trend, extracts the meteorological change direction and amplitude value of the region in the stage block map, divides the trend type according to the continuity and gradient change judgment, eliminates the block numbers that overlap with the precipitation forecast window in the time period, and generates a controllable trend block set;

[0105] The temperature, humidity and wind speed sequence data of different time periods are extracted from the block map, and a time series curve is formed for each block. The trend of the median value change of the data is used to determine whether the block has continuous fluctuations. For example, if the humidity of a block in the map shows an increasing trend within 48 hours and the amplitude value reaches 12%, it is recorded as a trend rising segment. Then, it is screened according to the set change amplitude threshold. The threshold is set to 10%, that is, if the difference between the maximum and minimum values ​​of a block in a continuous time period exceeds this value, it is demarcated as a fluctuation segment. If it is less than the threshold, it is a stable segment. The change amplitude threshold is determined by the original precipitation response data. It is set to 10%. Taking the experimental results of a certain farmland as an example, blocks with humidity fluctuations above 10% are more susceptible to water vapor interference and produce irrigation responses, so this benchmark is set; after completing the preliminary trend judgment, based on the short-term precipitation forecast of the National Meteorological Center, the precipitation forecast time window of the next 24-72 hours for each block is extracted to determine whether it overlaps with the block map time period. If there is a time overlap, the corresponding block number is removed from the trend classification set, and finally a complete trend classification table is formed. After statistics, 18 controllable area numbers that do not overlap with the precipitation time window are screened out, and the controllable trend block set is obtained.

[0106] The humidity comparison submodule calls the adjustable trend block set, collects the current soil moisture reading of the block and the standard range of the water demand level in the corresponding stage, compares the difference between the humidity reading and the water demand lower limit by block, and uses the formula:

[0107] ;

[0108] Calculate the irrigation intervention priority value, sort the block numbers, and obtain the regulation priority block sequence;

[0109] in, represents the humidity reading of the jth block, Represents the humidity reference value of the water demand level corresponding to the jth block, Represents the time span value in the j-th block graph, Represents the volatility value in the i-th block stage, is the number of blocks included in the comparison, is the irrigation intervention priority value;

[0110] The irrigation intervention priority value is a quantitative scoring indicator calculated based on the difference between the current soil moisture and crop water requirements of each agricultural block, the normalized trend span, and the humidity fluctuation amplitude. It is used to measure whether a block is currently in urgent need of irrigation intervention. The larger the value, the more urgency of intervention is higher when the humidity is far below the water requirement limit and the fluctuation is severe. The priority value is calculated through steps such as unit unification, trend normalization, and volatility weighting. It is comparable and ranked. Ultimately, it is used in irrigation regulation to automatically identify and prioritize areas requiring water-saving irrigation control.

[0111] The current soil moisture sensor readings of the included areas are collected in sequence. Such readings are uploaded regularly by sensors deployed 10 cm underground in the center of the field. The data unit is percentage (%). For example, the measured value of a certain block is 23.5%. Before data processing, it is necessary to unify the units and dimensions. Because some records are expressed in volumetric water content (unit: cm³ / cm³), they are first converted into mass water content percentage (%). After the unified format, it can be used as Participate in operations;

[0112] The water requirement level standard value of the corresponding crop in the current growth period is called from the planting management platform. For example, the lower limit of water requirement of corn in the jointing stage is set to 30%. This value is used as Items involved in comparison;

[0113] Each block in the atlas is bound to the time span information, which is recorded as item L, representing the humidity change trend sampling interval in hours. The current sampling intervals are set to 48 hours, 72 hours, and 60 hours respectively.

[0114] Volatility The volatility is calculated by combining the maximum and minimum difference ratios of the 48-hour humidity curve of each block with its standard deviation. For example, if the maximum humidity is 28%, the minimum is 20%, and the standard deviation is 3%, then the volatility is (28-20) / 3=0.22, also expressed in percentage.

[0115] After all items are sorted, they are substituted into the formula for normalization. The square root of the denominator is used to avoid the offset enhancement effect caused by different time scales. The calculation is as follows:

[0116] ;

[0117] The units of each parameter are described as follows:

[0118] : No. The measured humidity reading of the block (unit: %) is uploaded in real time through the sensor;

[0119] : No. The lower limit standard value of humidity in the current crop water demand stage of the block (unit: %), which is retrieved from the planting stage configuration database;

[0120] : No. The block's graph trend sampling period (unit: hour), used for normalization processing;

[0121] : No. The fluctuation rate of humidity change in the block (unit: %), obtained by the difference / standard deviation ratio;

[0122] : The calculated irrigation intervention priority value, which is a unitless scoring indicator used for ranking;

[0123] : The total number of blocks involved in the comparison, in this case 3;

[0124] The final result is 0.5342, which is used as the intervention priority score value to be input into the subsequent modules for sorting. This value ranks second among all current blocks and enters the first-level regulation queue, thus obtaining the regulation priority block sequence. This formula introduces the degree of water demand deviation while standardizing the time impact and considering the humidity fluctuation factor, providing a quantifiable reference for water-saving irrigation strategies.

[0125] The intervention interval identification submodule extracts the corresponding control period interval according to the block number in the control priority block sequence, sets the intervention level and irrigation period configuration according to the continuous humidity deviation, and generates an irrigation intervention interval configuration table;

[0126] According to the top-ranked areas in the regulation priority block sequence, the regulation time period identifier of the block in the atlas is extracted, and the humidity change curve is called back for judgment. If the humidity deviation exceeds 10% in two consecutive time periods, it is judged as a continuous interference area. Based on this, the intervention time period is delineated and the irrigation level is assigned. The level is set according to the irrigation intervention priority value. If the priority is greater than 0.5, it is judged as a first-level intervention, otherwise it is a second-level intervention. Taking block 3 as an example, the volatility is 0.22% and the humidity deviation value is -10.6%. It is judged as a continuous deviation area with a level one level. Combined with the current irrigation calendar, the intervention time period is configured to be once every two hours, for a total of three times. The intervention block number, irrigation period and level are integrated to generate an irrigation intervention interval configuration table.

[0127] See also Figure 6 , the feedback correction module includes:

[0128] The humidity fluctuation screening submodule extracts the block irrigation end time according to the irrigation intervention interval configuration table, identifies the shallow humidity data after the corresponding time, analyzes the difference between the humidity recovery amplitude and the control target value, screens the blocks that exceed the recovery deviation threshold, and generates a humidity control deviation block set;

[0129] The irrigation end time for each block is obtained based on the irrigation intervention interval configuration table. The irrigation end time is extracted by searching the block's corresponding time stamp and matching it with the humidity change record in the data log to determine the block's irrigation end time. For example, if irrigation ends at 10:00 AM for a block, the shallow humidity data corresponding to the irrigation end time is extracted. Humidity data is recorded continuously. The magnitude of humidity recovery is calculated by comparing the humidity value after irrigation with the predetermined target value. For example, if the target humidity is 30% and the actual humidity is 25%, the humidity recovery magnitude is a 5% difference. This recovery magnitude is then compared with the set control target value. If the recovery magnitude exceeds a set threshold (for example, 3%), the block is identified as exceeding the recovery deviation. All blocks exceeding the threshold are aggregated to form a humidity control deviation block set, which is then used for subsequent humidity control analysis. The final result is a series of blocks with excessive humidity recovery deviations.

[0130] The leaf color difference measurement submodule extracts the color difference readings of the leaves in the corresponding blocks before and after irrigation based on the humidity control deviation block set, analyzes the color difference change amplitude, and classifies them according to the ratio of the color difference change value to the humidity recovery deviation value. It then filters out the blocks with inconsistent responses and generates a water-photosynthesis imbalance distribution group.

[0131] The corresponding leaf color difference data is extracted from each block in the humidity control deviation block set. The color difference is obtained by analyzing the images of the leaves before and after irrigation. The color difference of the leaves before and after irrigation can be measured by image processing technology (such as RGB color space difference or LAB color difference model). Assuming that the leaf color before irrigation is RGB (45, 65, 32) and after irrigation is RGB (50, 70, 35), the color difference is calculated as the change in RGB space. The amplitude of the color difference change is calculated. The amplitude reflects the response of the leaves to the change in water. Assuming that the color The color difference change amplitude is 5 units (RGB difference). The ratio analysis of the color difference change amplitude and the humidity recovery deviation value of the corresponding block is performed, that is, the ratio of the color difference change amplitude to the humidity recovery deviation value. For example, assuming the color difference change amplitude is 5 and the humidity recovery deviation is 3, then the ratio is 5 / 3=1.67. The blocks are classified according to this ratio. If the ratio is large, it means that the color difference change response is more obvious, but the ratio is inconsistent with the humidity recovery amplitude. Such blocks will be screened as inconsistent response blocks and finally classified to generate water-photosynthesis imbalance distribution groups.

[0132] The control offset identification submodule extracts the corresponding control parameters based on the water-photosynthesis imbalance distribution group and compares them with the control targets in the configuration table, marks the blocks and set parameters for setting the offset, and generates the stage control offset label;

[0133] Extract the control parameters of the block. The control parameters generally include water replenishment, light intensity, temperature and other factors. The control parameters need to be compared with the control targets in the configuration table. The configuration table contains preset standard values ​​or target ranges. For example, the target water replenishment is 50L and the target light intensity is 500lux. Compare the control parameters of each block with the target values ​​in the configuration table. If the water replenishment of the block is 60L and the light intensity is 450lux, which exceeds the set water target or the light intensity does not meet the standard, it is a control offset. By marking the block with the set offset and the corresponding parameters, the label can add offset information to each block. For example, the water offset is marked as +10L and the light offset is marked as -50lux. The final label can help the subsequent decision-making analysis stage to control the offset situation. The final result is the generation of the corresponding stage control offset label.

[0134] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. The intelligent control system for agricultural water-saving irrigation is characterized by: The system comprises: The growth stage identification module obtains agricultural irrigation control block data, including rice leaf width, plant internode length, and tillering density. It compares the data with the crop growth stage benchmark sequence to determine the stage interval to which the current sample belongs. It then selects blocks distributed between the tillering and heading periods to generate a stage block identification map. The growth stage identification module includes: The widening parameter extraction submodule obtains agricultural irrigation control block data, identifies the widening change sequence through the same block time series, calculates the increase rate and compares it with the set benchmark interval, extracts the irrigation control blocks that meet the widening dynamic characteristics, and generates a widening characteristic response block number set; The internode segment determination submodule calls the widening feature response block number set, extracts internode length and tiller density data, identifies internode growth rate and tiller density increment value, selects block numbers and time periods that are simultaneously within the dual intervals of internode lengthening and density increase, and generates a water-saving regulation sensitive section identification set; The growth block determination submodule compares the tillering density and the increase trend of the time series node in the block according to the water-saving regulation sensitive section identification set, matches the stage benchmark increase range of the tillering period and the heading period, extracts the effective blocks and spatial positions within the stage range, and generates a stage block identification map; The moisture monitoring module collects moisture readings of the topsoil and bottom layers of the corresponding blocks based on the stage block identification map, compares the change direction and amplitude difference of the two layers of data according to the sampling time series, identifies the dynamic change characteristics of humidity in the stage block, and generates a layered moisture distribution state; The meteorological data synchronization module identifies the recent precipitation forecast sequence and wind speed increase sequence in the region based on the stratified humidity distribution state, synchronizes the humidity change rhythm of the current block with the predicted meteorological elements, calculates the overlap ratio of the overlapping fluctuation intervals, and generates the water vapor interaction change trend; The irrigation strategy scheduling module calls the water-gas interaction change trend, classifies the stage block map, eliminates the area overlapping with the precipitation forecast window, compares the block humidity reading with the stage water demand level, marks the time period and area that need to be regulated, and generates an irrigation intervention interval configuration table.

2. The intelligent control system for agricultural water-saving irrigation according to claim 1, characterized in that: The stage block identification map includes growth stage classification labels, block time series indexes, and tillering and heading block identification information. The stratified humidity distribution status includes the humidity change amplitude of the plow layer, the bottom layer humidity response delay, and the inter-layer humidity difference characteristics. The water-gas interaction change trend includes the humidity fluctuation overlap rate, the precipitation prediction matching interval, and the wind speed increasing interference factor. The irrigation intervention interval configuration table includes the control block number, the target humidity gear, and the intervention time window.

3. The intelligent control system for agricultural water-saving irrigation according to claim 1, characterized in that: The moisture monitoring module includes: The topsoil humidity collection submodule collects topsoil and bottomsoil humidity readings based on the block identification map of the stage, unifies the sampling time series, removes anomalies and repairs missing segments of block data, extracts valid humidity samples, and generates a basic set of block humidity monitoring; The humidity change comparison submodule calls the block humidity monitoring basic set, extracts the humidity change values ​​of the plow layer and the bottom layer in adjacent time periods, compares the humidity change direction and increase and decrease amplitude of the two layers, identifies the coupling characteristics, and generates a coupling change coefficient set; The stratified humidity modeling submodule calls the coupling variation coefficient set, selects the trend stable block, extracts the increase and decrease amplitude of the plow layer humidity and the bottom layer change rate, combines the sampling time interval, sequence position, and number of sample points, calculates the stratified humidity control index value, combines the spatial range of the irrigation operation block and the time node for mapping, and generates the stratified humidity distribution state.

4. The intelligent control system for agricultural water-saving irrigation according to claim 3, characterized in that: The meteorological data synchronization module includes: The humidity identification submodule extracts humidity data for each layer based on the stratified humidity distribution state, combines the precipitation forecast and wind speed sequence, analyzes the humidity change rhythm and the fluctuation amplitude of meteorological elements, determines whether the difference between the two is within the humidity matching threshold range, selects the number sequence that meets the conditions, and generates an irrigation intervention warning number group; The overlapping section determination submodule selects the wind speed increase and precipitation fluctuation information of the corresponding time period based on the irrigation intervention warning number group, identifies the overlapping time period of the two types of meteorological data, calculates the wind-precipitation coupling offset, and filters the time period numbers that exceed the interactive response limit to obtain the response section number set that needs to be regulated; The control trend derivation submodule calls the set of response section numbers that need to be controlled, compares the humidity and wind drop interaction data of the benchmark sections, extracts the rhythm overlapping positions, calculates the humidity adjustment differences of each type of interaction combination in groups, and classifies them into intervention control levels, summarizes the level change trajectories, and generates water vapor interaction change trends.

5. The intelligent control system for agricultural water-saving irrigation according to claim 4, characterized in that: The irrigation strategy scheduling module includes: The trend classification submodule calls the water-gas interaction change trend, extracts the meteorological change direction and amplitude value of the region in the stage block map, divides the trend type according to the continuity and gradient change judgment, eliminates the block numbers that overlap with the precipitation forecast window in the time period, and generates a controllable trend block set; The humidity comparison submodule calls the adjustable trend block set, collects the current soil moisture reading of the block and the standard range of the water demand level in the corresponding stage, compares the difference between the humidity reading and the water demand lower limit by block, calculates the irrigation intervention priority value, sorts the block numbers, and obtains the control priority block sequence; The intervention interval identification submodule extracts the corresponding control period interval according to the block number in the control priority block sequence, sets the intervention level and irrigation period configuration according to the continuous humidity deviation, and generates an irrigation intervention interval configuration table.

6. The intelligent control system for agricultural water-saving irrigation according to claim 1, characterized in that: The system also includes a feedback correction module: The feedback correction module detects the change in shallow humidity readings and leaf color difference indicators after the end of irrigation according to the irrigation intervention interval configuration table, screens and groups the humidity recovery amplitudes of blocks that deviate from the expected control, marks the corresponding control settings of the deviated blocks, and generates stage control offset labels; The stage control offset label includes the humidity recovery deviation level, the leaf color difference change amplitude, and the control setting offset mark.

7. The intelligent control system for agricultural water-saving irrigation according to claim 6, characterized in that: The feedback correction module includes: The humidity fluctuation screening submodule extracts the block irrigation end time according to the irrigation intervention interval configuration table, identifies the shallow humidity data after the corresponding time, analyzes the difference between the humidity recovery amplitude and the control target value, screens the blocks that exceed the recovery deviation threshold, and generates a humidity control deviation block set; The leaf color difference measurement submodule extracts the color difference readings of the leaves in the corresponding blocks before and after irrigation based on the humidity control deviation block set, analyzes the color difference change amplitude, and classifies them according to the ratio of the color difference change value to the humidity recovery deviation value, filters out the blocks with inconsistent responses, and generates a water-photosynthesis imbalance distribution group; The control offset identification submodule extracts the corresponding control parameters according to the water-photosynthesis imbalance distribution group and compares them with the control targets in the configuration table, marks the blocks and set parameters for setting the offset, and generates a stage control offset label.

Images