Accurate irrigation control system fusing crop water demand model and multi-source data

Abstract

The invention relates to the technical field of irrigation control, in particular to an accurate irrigation control system fusing a crop water demand model and multi-source data, and the system comprises a humidity response recognition module, a root zone division module, a communication path construction module, a time sequence regulation and control module and an irrigation control execution module. According to the method, the adaptive precision of space control and the reasonability of water supply coverage are improved by identifying the deep humidity horizon with the continuous disturbance trend and delimiting the differential root zone response section, the stable communication path is dynamically screened in combination with the signal fluctuation characteristics, the reliability of instruction transmission and the anti-interference capability of the system are enhanced, and the reliability of the system is improved. The irrigation starting sequence is adjusted at intervals by utilizing starting points of multiple types of environmental factors, the crop response rhythm is dynamically matched, the continuity of time sequence actions and the coordination of regulation rhythm are improved, a control execution sequence is constructed in combination with action numbers and standby channels, synchronous instruction execution and multi-path collaborative response are ensured, and the control efficiency is improved. And the stability, continuity and scheduling accuracy of the irrigation process in a complex environment are enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of irrigation control, in particular to a kind of precision irrigation control system fusing crop water requirement model and multi-source data. BACKGROUND

[0002] The technical field of irrigation control includes monitoring of irrigation system, calculation and control of irrigation water quantity, automatic operation and remote management of irrigation equipment and other core matters, and its overall technical system covers sensor data acquisition (such as soil moisture content, meteorological factors, crop growth state), establishment of irrigation demand model, control of execution device (such as water pump, valve, solenoid valve) and real-time adjustment of control strategy. Among them, the traditional precision irrigation control system fusing crop water requirement model and multi-source data refers to the input of crop water requirement model and multi-source sensing data fusion, and the determination of irrigation time and water quantity according to soil moisture, weather, crop growth stage and other data, and the execution of irrigation operation according to the preset irrigation plan or dynamic adjustment through the control of water pump or valve.

[0003] The prior art mainly drives irrigation control through shallow sensing information, which has the problem of insufficient precision in distinguishing the spatial difference of crop root zone. The collected data is limited to the surface soil state, it is difficult to distinguish the disturbance characteristics of underground horizon, the irrigation partition is easy to mismatch, which leads to uneven coverage of root zone water supply, poor regional response consistency, reduced crop water absorption efficiency, the control instruction mainly relies on static model deduction, lacks continuous monitoring mechanism of multi-cycle disturbance trend, time sequence response is difficult to correct in time, irrigation rhythm and actual water requirement process of crops are easy to deviate, communication mechanism is fixed and depends on node connection topology, signal strength fluctuation is not dynamically identified and path replacement strategy is not constructed, when signal attenuation or node anomaly occurs, there is a risk of control command interruption or response delay, and the system stability is limited. SUMMARY

[0004] In order to solve the technical problems existing in the prior art, the present application provides a precision irrigation control system fusing crop water requirement model and multi-source data. On the one hand, a precision irrigation control system fusing crop water requirement model and multi-source data is provided, which comprises:

[0005] The humidity response identification module collects soil humidity data of different depths in the agricultural area, extracts the water content difference as the disturbance amplitude, judges whether the increase and decrease change of adjacent sampling points changes direction, screens the horizon of disturbance trend, and generates the root layer distribution position;

[0006] The root zone division module matches the change direction of adjacent horizons based on the root layer distribution position, extracts the disturbance horizon marker, marks the same section if the markers are consistent, and marks independent response section if the markers are inconsistent, and generates a control area distribution map;

[0007] The communication path construction module extracts a control number based on the control area distribution map, calls corresponding node signal records, calculates a signal strength difference value, judges transmission abnormalities by the number of times of decrease, and generates an irrigation communication path list;

[0008] The timing regulation module judges whether the interval change direction needs to adjust the starting time point by comparing the interval between the starting time of irrigation and the change starting point of the leaf temperature, humidity and greenness of each number based on the irrigation communication path list, and generates an irrigation timing plan.

[0009] The irrigation control execution module calls the water pump and valve action of each number and issues instructions according to the irrigation timing plan, synchronizes to the control terminal and the backup path, and generates a crop irrigation control scheme.

[0010] As a further scheme of the present application, the root layer distribution position includes a disturbance amplitude change value, a disturbance direction switching point and a continuous disturbance layer position, the control area distribution map includes a continuous response area number, a section boundary marker and a layer position attribution label, the irrigation communication path list includes a signal strength fluctuation value, a transmission abnormality judgment number and a backup node path, the irrigation timing plan includes an irrigation starting interval difference value, an action sequence arrangement number and a periodic change direction identifier, and the crop irrigation control scheme includes a control instruction type, a water pump and valve corresponding number and a control response record item.

[0011] As a further scheme of the present application, the layer position of the disturbance trend refers to the soil layer position where the increase / decrease change direction is converted during the soil humidity disturbance process.

[0012] As a further scheme of the present application, the transmission abnormality refers to an abnormal state that the node signal strength in the irrigation communication path continuously decreases by more than a set number of times.

[0013] As a further scheme of the present application, the humidity response recognition module includes:

[0014] The humidity data acquisition submodule acquires soil humidity sensing data deployed at different depths in the agricultural planting area, acquires the water content change value of each depth point in a preset time period, organizes the data in time sequence and depth label, and generates a soil humidity timing data set;

[0015] The disturbance amplitude calculation submodule calculates the water content difference value of the same depth point in the continuous time sequence as the disturbance amplitude based on the soil humidity timing data set, records the disturbance direction flag, and sorts out the disturbance amplitude and direction sequence data;

[0016] The root layer position recognition submodule calls the disturbance amplitude and direction sequence data, judges whether the adjacent disturbance direction is converted, filters the depth layer position with a continuous disturbance amplitude exceeding a preset disturbance threshold, counts the frequency and continuity, and generates the root layer distribution position.

[0017] As a further scheme of the present application, the root zone division module comprises:

[0018] The horizon direction matching sub-module calls the identified disturbance horizon and the corresponding change direction mark based on the root layer distribution position, extracts adjacent horizon pairs in horizon order, compares the change direction marks pair by pair, records the comparison results, and generates a horizon direction consistency identification matrix;

[0019] The response section classification sub-module determines whether the horizon pairs with continuous consistent marks meet the merging condition according to the horizon direction consistency identification matrix, merges them into a group if they meet the condition, classifies them as independent groups if they do not meet the condition, sequentially constructs a response section set, and generates a response section structure list;

[0020] The area number generation sub-module calls the response section structure list, numbers each response section according to the horizon arrangement order, constructs a number index mapping relationship, draws a control boundary in combination with the horizon range corresponding to the number, and obtains a control area distribution map.

[0021] As a further scheme of the present application, the communication path construction module comprises:

[0022] The number signal extraction sub-module extracts control numbers according to the control area distribution map, obtains signal record data of the field control nodes corresponding to the numbers, extracts signal strength values of each node in a continuous time period, calculates the strength difference values of adjacent time points, and generates a node signal strength difference sequence;

[0023] The node anomaly discrimination sub-module calls the node signal strength difference sequence, counts the number of continuous decreases of the nodes, judges whether it exceeds the set decrease threshold, marks the node numbers with anomalies, records the normal node numbers, and generates an anomaly node screening identification set;

[0024] The path configuration generation sub-module reads the connection records between normal nodes according to the anomaly node screening identification set, screens adjacent node pairs with stable connection relationships, configures them as backup transmission paths, and arranges the node numbers and connection relationships of the paths, to generate an irrigation communication path list.

[0025] As a further scheme of the present application, the timing control module comprises:

[0026] The response starting point extraction sub-module calls the leaf temperature change starting point, the soil moisture content change starting point, and the greenness change starting point corresponding to the control numbers for the irrigation communication path list, extracts the first change time record corresponding to the indicators, and pairs with the irrigation start time to calculate the time difference value, to generate a control indicator starting point interval matrix;

[0027] The interval change judgment submodule judges whether the change direction is consistent based on the control index starting point interval matrix, marks the control number needing adjustment, and generates a starting point adjustment identification sequence.

[0028] The action timing adjustment submodule calls the starting point adjustment identification sequence, reorders all control numbers as needed, arranges the numbers needing adjustment in priority, keeps the original time interval structure unchanged, fills in the numbers without adjustment demand in the original order, and generates an irrigation timing plan.

[0029] As a further scheme of the present application, the irrigation control execution module comprises:

[0030] The instruction configuration setting submodule calls the water pump on-off action and the valve opening and closing action corresponding to the control number based on the irrigation timing plan, arranges each group of control action content in order, sets it as an independent execution instruction sequence, and generates a numbered control execution instruction set.

[0031] The control path synchronization submodule calls the numbered control execution instruction set and the irrigation communication path list, extracts the standby path information corresponding to each control number, synchronizes the execution instruction to the control terminal device in the path, records the synchronization transmission state, and generates a standby path execution state table.

[0032] The response state recording submodule detects the response signal of each control number according to the standby path execution state table, extracts the feedback action state of the number, records the response completion condition and timestamp data in order, and obtains a crop irrigation control scheme.

[0033] As a further scheme of the present application, the falling threshold refers to the number of times for judging whether the continuous falling of node signal strength is abnormal.

[0034] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects:

[0035] By identifying the deep humidity horizon with continuous disturbance trend and demarcating the differentiated root zone response section, the adaptation accuracy of spatial control and the rationality of water supply coverage are improved, the stable communication path is dynamically screened combined with the signal fluctuation characteristics, the reliability of instruction transmission and the system anti-interference ability are enhanced, the irrigation starting order is adjusted by using multiple environmental factors, the crop response rhythm is dynamically matched, the continuity of timing action and the coordination of regulation rhythm are improved, the control execution sequence is constructed combined with the action number and the standby path, the instruction synchronization execution and multi-path collaborative response are ensured, and the stability, continuity and scheduling accuracy of the irrigation process in complex environment are enhanced. BRIEF DESCRIPTION OF DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiments description. Obviously, the drawings in the following description only show some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained based on these drawings without any creative effort.

[0037] Figure 1 is a schematic diagram of the system of the present application;

[0038] Figure 2 is a schematic diagram of the system framework of the present application;

[0039] Figure 3 is a flow chart of the humidity response identification module in the present application;

[0040] Figure 4 is a flow chart of the root area division module in the present application;

[0041] Figure 5 is a flow chart of the communication path construction module in the present application;

[0042] Figure 6 is a flow chart of the timing control module in the present application;

[0043] Figure 7 is a flow chart of the irrigation control execution module in the present application. DETAILED DESCRIPTION

[0044] The technical solutions in the present application will be described below with reference to the drawings.

[0045] In the embodiments of the present application, the words such as "example", "for example" are used to represent as an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. In fact, the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two options.

[0046] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. "Of", "corresponding" and "relevant" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0047] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1. When the distinction is not emphasized, the meanings expressed are consistent.

[0048] In order to make the technical problems, technical solutions and advantages to be solved by the present application clearer, the following will be described in detail in combination with the drawings and specific embodiments.

[0049] The embodiment of the present application provides a kind of precision irrigation control system of crop water demand model and multi-source data, as shown in Figures 1-2 Precision irrigation control system schematic diagram of crop water demand model and multi-source data, the system includes:

[0050] Humidity response identification module obtains soil humidity sensing data deployed in the differentiating depth of agricultural planting area, extracts the difference of depth point as disturbance amplitude based on the collected moisture content variation sequence, judges whether the change direction is converted according to the increasing and decreasing relationship of moisture content of adjacent sampling points, combines the continuity of disturbance amplitude and direction change, filters the horizon with disturbance trend, and generates root layer distribution position;

[0051] Root area division module matches the change direction between adjacent horizons based on the root layer distribution position, calls the identified disturbance horizon and extracts the change direction mark of horizon, judges whether it belongs to the same section by comparing whether adjacent marks are consistent, if consistent, it is merged into continuous response area, if inconsistent, it is divided into independent response section, and according to horizon sequence, it is given a unique number, and a control area distribution map is generated;

[0052] Communication path construction module extracts control number according to control area distribution map, calls signal record of number corresponding field control node and extracts signal intensity difference value in continuous time period, judges whether node exists transmission abnormality by comparing the number of adjacent records, then according to the connection record between nodes, adjacent nodes that can maintain stable transmission are screened and configured as standby path, and irrigation communication path list is generated;

[0053] Timing control module calls leaf temperature change starting point, soil moisture content change starting point and greenness change starting point of control number for irrigation communication path list, interval compares the above time records with corresponding irrigation start time, determines whether the start time point needs to be adjusted by judging the change direction of continuous period interval, and rearranges the action sequence according to the number sequence, generates irrigation timing plan;

[0054] Irrigation control execution module calls water pump on-off action and valve opening and closing action corresponding to number based on irrigation timing plan and sets execution instruction according to plan sequence, synchronizes instruction content to control terminal, then calls standby path corresponding to number in irrigation communication path list to execute synchronous instruction, and records control response of number, generates crop irrigation control scheme.

[0055] The root layer distribution position includes a disturbance amplitude change value, a disturbance direction switching point, and a continuous disturbance layer position. The control area distribution map includes a continuous response area number, a section boundary marker, and a layer position attribution label. The irrigation communication channel list includes a signal strength fluctuation value, a transmission anomaly judgment number, and a backup node channel. The irrigation timing plan includes an irrigation start interval difference value, an action sequence arrangement number, and a period change direction identifier. The crop irrigation control scheme includes a control instruction type, a water pump and valve corresponding number, and a control response record item.

[0056] Specifically, as shown in Figure 2 、 3 The humidity response identification module includes:

[0057] The humidity data acquisition submodule acquires soil humidity sensing data deployed at different depths in the agricultural planting area, acquires the water content change value of each depth point within a preset time period, organizes the data in time sequence and depth label, and generates a soil humidity time series data set.

[0058] First, a representative test area is selected according to the type of crop and the root growth characteristics of the planting area. For example, in an area where corn is planted, 1 mu can be selected as a sampling unit. In this area, multiple soil humidity collection points are set at different depths from the ground to the ground, usually at intervals of 20 cm, at depths of 0 cm, 20 cm, 40 cm, 60 cm, 80 cm, 100 cm, and one sensor supporting high-precision soil humidity collection is deployed at each depth. The sensor needs to be pre-set to sample every 30 minutes, with a duration of not less than 30 days. Each collected data needs to have a complete time stamp, such as data record at 08:00 on November 1, 2025. The collected humidity values need to be uniformly transmitted to the local data receiving terminal or remote server through the communication interface. After data transmission is completed, each piece of data needs to be labeled with its corresponding depth label and collection time point information, and sorted in time dimension in turn, so that the data structure meets the structure format with depth as the primary index and time as the secondary index. During the collection process, if there is missing data or abnormal jump data, for example, if the sampling value at a certain time point during continuous sampling deviates from the average of the previous and next two time points by more than 10%, it is considered to be an abnormal value, which needs to be filled in by interpolation. The interpolation method can use the arithmetic mean of the previous and next valid data, for example, the humidity values of the two time points before and after the missing point are 20.0% and 22.0% respectively, then the missing point is filled in as 21.0%. All data are uniformly saved as a structured data table after filling in, each record includes depth, time, humidity value, and is output in CSV format file. After the above process, the data organized in time sequence and depth label is obtained, forming a soil humidity time series data set.

[0059] The disturbance amplitude calculation submodule is based on the soil moisture time series dataset. It calculates the difference in water content of continuous time series at the same depth point as the disturbance amplitude, records the disturbance direction indicator, and sorts out the disturbance amplitude and direction sequence data.

[0060] The humidity changes at each depth point within the complete time series are analyzed line by line. During execution, the corresponding time series data must be extracted according to the depth dimension, and the difference between humidity values ​​at any two adjacent time points is calculated sequentially. For example, between two data points from 08:00 to 08:30, if the humidity value at a certain depth point changes from 21.0% to 19.8%, the disturbance amplitude is 1.2%, and the disturbance direction is downward. During the calculation, the sign of the difference must be clearly recorded to indicate the direction of the disturbance. If the value at the later time point is greater than the value at the earlier time point, it is an upward disturbance, marked with a direction of 1; if it is less than the value at the earlier time point, it is a downward disturbance, marked with a direction of -1; if they are equal, it is a stationary disturbance, marked with a direction of 0. Each disturbance data point includes depth and initial value. Factors such as time, perturbation amplitude, and perturbation direction are considered. To ensure the accuracy of the analysis, before each perturbation amplitude calculation, it is necessary to verify whether the humidity data involved in the calculation has missing or abnormal markers. If so, the calculation should be skipped or the repaired data should be used instead. After the perturbation amplitude sequence is sorted out, items with significant perturbation amplitude need to be further screened out. The screening criterion is that the perturbation amplitude value is not less than a preset threshold. This threshold needs to be determined according to the local climate conditions and soil type, and is usually set at around 1.0%. For example, if the normal fluctuation range of daily moisture content in historical sampling data is ±3%, then a significant preset perturbation threshold can be set to 1.0%. During the execution process, only perturbation data that meets this condition is retained, and its perturbation direction marker is also retained, thereby constructing perturbation amplitude and direction sequence data.

[0061] The root layer location identification submodule calls the disturbance amplitude and direction sequence data, determines whether the direction of adjacent disturbances has changed, filters the depth layers where the continuous disturbance amplitude exceeds the preset disturbance threshold, counts the frequency and continuity, and generates the root layer distribution location.

[0062] The change of disturbance direction of each depth layer needs to be analyzed item by item. In the execution process, first, in the disturbance direction sequence of each depth point, the disturbance direction marks of any two adjacent time points are read one by one, and it is judged whether the signs change, that is, it is judged whether the change from rising to falling or from falling to rising occurs. If the multiplication result of the two is less than 0, it is considered that the disturbance direction has changed at this time point, and the conversion event needs to be marked in the data set. At the same time, in cooperation with the disturbance amplitude data, all records with a disturbance amplitude greater than a significant preset disturbance threshold are screened, only the disturbance records with a significant disturbance amplitude and a change in direction are retained, and then the disturbance frequency of each depth layer within the entire sampling period that meets the above conditions is counted. For example, if the total number of records of the 60cm depth layer that meet the disturbance amplitude and direction change conditions within 30 days is 18 times, the disturbance frequency is recorded as 18, and the longest length of continuous disturbance, that is, the number of continuous records that meet the conditions of disturbance direction and amplitude, is also counted. If there are effective disturbance records every day from the 10th day to the 13th day and the direction continuously changes, the maximum continuous length is 4. Finally, two parameter values of the disturbance frequency and the maximum continuous disturbance length of each depth layer are formed. According to the preset frequency threshold and the continuous length threshold, the frequency threshold can be set to 12 times to represent at least one significant disturbance every 2.5 days, and the continuous length threshold is set to 3 times to represent that there is an effective disturbance in at least three consecutive time points. According to this standard, the set of depth layers that meet the conditions is screened, and the set is the main distribution area where the root layer is located.

[0063] Specifically, as shown in Figure 2 、 4 The root area division module includes:

[0064] The layer direction matching sub-module calls the identified disturbance layer and the corresponding change direction mark based on the root layer distribution position, extracts adjacent layer pairs in order according to the layer position, compares the change direction marks pair by pair, records the comparison results, and generates a layer direction consistency identification matrix.

[0065] First, the disturbance horizon set identified in the previous module and their respective disturbance direction markers are extracted, and it is necessary to ensure that the disturbance direction marker has been defined at each horizon. For the horizon with missing direction markers, the disturbance direction trend of adjacent horizons should be used to fill in the missing data, for example, if the upper layer is 1 and the lower layer is also 1, then the missing horizon is filled in as 1. After ensuring that all horizon data is complete, arrange the disturbance horizons in order from shallow to deep, and form horizon pairs by grouping adjacent two horizons, for example, from the horizon set {20 cm, 40 cm, 60 cm, 80 cm}, horizon pairs (20-40), (40-60), (60-80) are extracted in turn. Then compare the disturbance directions of each horizon pair, read the direction marker of the upper layer and the direction marker of the lower layer, and perform equality judgment. If the two direction marker values are the same, it is marked as consistent, and marked as "1"; if they are not the same, it is marked as "0". After executing this judgment logic for each horizon pair, all results are summarized into a sequential matrix. The matrix structure is indexed by horizon pairs, and the identification result is the corresponding value. For example, the direction of horizon pair (20-40) is 1 and 1, marked as 1; the direction of horizon pair (40-60) is 1 and -1, marked as 0; the direction of horizon pair (60-80) is -1 and -1, marked as 1. The matrix format can be table structure or array structure, and each horizon pair and its direction consistency marker are recorded. This matrix is used as the basis for response section division in the subsequent module, i.e., the horizon direction consistency identification matrix is completed.

[0066] The response section classification submodule determines whether the horizon pairs with consistent markers meet the merging conditions according to the horizon direction consistency identification matrix. If they meet the merging conditions, they are merged into a group; if they do not meet the merging conditions, they are classified as independent groups. The response section set is constructed in turn, and the response section structure list is generated.

[0067] First, read the matrix in the sequence of the layer pair and its direction consistency mark, starting from the first set of layer pairs, in turn to determine whether the current layer pair has the same direction consistency mark as the previous layer pair, the judgment method is to compare whether the current and the last identification value are both 1, if so, merge the two layer pairs into the same response section, if the mark is 0 or inconsistent with the last mark, take the current layer pair as the starting new group, and restart the new response section classification operation, this process can be merged by setting the group number, whenever there is an inconsistent direction mark, a new number is added, the starting and ending layer of the section is determined by all the layer pairs involved in the merging, for example, when the direction consistency is [1, 1, 0, 1, 1], the corresponding layer pairs are (20-40), (40-60), (60-80), (80-100), (100-120), then (20-60) is a group, (60-80) is a group because of inconsistency, (80-120) is another group, in the classification process, the minimum merging condition threshold of continuous consistency needs to be set, for example, the minimum continuous consistent group number is 1, that is, as long as two groups are continuously consistent, they are merged into a response section, if the threshold is set to 2, at least three groups of continuously consistent directions are needed to be merged, otherwise they will be classified as a separate section, after traversing the entire direction consistency matrix through this logic, finally a plurality of response sections are formed, each section takes its starting layer and ending layer as the boundary, and a structure list containing all the response sections is sorted out.

[0068] The region number generation submodule calls the response section structure list, numbers each response section according to the layer arrangement order, constructs the number index mapping relationship, draws the control boundary combined with the layer range corresponding to the number, and obtains the control region distribution map;

[0069] According to the vertical layer arrangement order of the response section, the numbering is carried out in sequence from top to bottom as the basis of the sequence, and the numbering format can be set as R1, R2, R3, etc., where R is the region identifier prefix, and the numerical part is the incremental number. First, extract the shallowest layer of each response section as the starting point of the section, and then extract the deepest layer as the end point to form a layer interval. For example, response section one is (20-60), and the number is R1; response section two is (60-80) and the number is R2, and so on. In the numbering process, it is necessary to avoid skipping and repeating the number to maintain continuity. After the numbering is completed, an index mapping table needs to be constructed according to the layer range corresponding to the number, establishing a one-to-one correspondence between each number and its layer interval. Then, in the graphical drawing, each numbered section is marked with different colors or styles. In the coordinate graph with depth on the vertical axis and numbered sections on the horizontal axis, the boundary position of the control region is drawn. Each numbered block in the graph represents a response region, which can be directly labeled with its corresponding number and layer range, such as R1: 20-60 cm, R2: 60-80 cm. Finally, a complete control region distribution map is formed.

[0070] Specifically, as shown in Figure 2 , 5 The communication path construction module comprises:

[0071] The number signal extraction submodule extracts the control number according to the control area distribution map, obtains the signal record data of the corresponding field control node, extracts the signal strength value of each node in a continuous time period, calculates the strength difference value of adjacent time points, and generates a node signal strength difference value sequence.

[0072] First, the number information of each control area in the diagram is identified, and the physical space range corresponding to the number is read one by one, for example, number R1 corresponds to 20cm to 60cm, and R2 corresponds to 60cm to 80cm. Each number is used as a key index item to call the field deployment account, and the control node number associated with the number is found in the account to confirm the existence state and running record validity of each control node in the system database. After excluding long-term no-data nodes, the signal record data is obtained, and a continuous time period sampling window is set according to the system unified time reference, for example, collecting once every 30 minutes, and 336 groups of time points can be obtained within 7 days. The system extracts all signal strength values of the corresponding time period for each control node number from the record. If the signal strength sequence of node N001 is [-73, -74, -75, -75, -76], the time sequence is read and stored in the temporary data buffer, and then the strength values between adjacent time points are calculated. The execution of the difference calculation is to subtract the previous time value from the current time value in order, and the change trend and amplitude between adjacent points are recorded. If the change value is negative, it means that the signal is weakened, and if the change value is positive, it means that the signal is enhanced. The difference sequence of the above node N001 is [-1, -1, 0, -1]. The system binds and stores all node numbers and their corresponding signal strength difference value sequences, finally generates all node difference value sequences, and forms a node signal strength difference value sequence set with clear structure and consistent time sequence.

[0073] The node anomaly discrimination submodule calls the node signal strength difference value sequence, counts the number of continuous decreases of the node, judges whether it exceeds the set decrease threshold, marks the node number with an abnormality, records the normal node number, and generates an abnormal node screening identification set.

[0074] The difference sequence corresponding to each node number needs to be traversed and analyzed, and the number of continuous decreases is counted as the basis for discrimination. In the execution, the difference values are read from the start of the sequence. If a certain item is negative, it is determined to be a decrease event. The system sets a cumulative decrease counter, which is incremented once when the current item is negative. If the next item is still negative, it continues to be counted. If a non-negative value appears, the accumulation is interrupted, the counter is reset, and a new accumulation segment is started. In a complete sequence, the system records the length of each continuous decrease, and after the analysis is complete, the maximum continuous decrease is extracted as the final determination value. This maximum value is compared with the set decrease threshold. If the maximum value exceeds the threshold, the node is marked as an abnormal node. If it does not exceed, it is a normal node. The decrease threshold can be set according to the communication stability in the actual deployment environment. For example, if most nodes have less than 3 consecutive decreases in daily fluctuations, the threshold can be set to 3. If node N001 has a maximum consecutive decrease of 4, it is determined to exceed the threshold and is marked as abnormal. If node N002 has a maximum decrease of 2, it is determined to be normal. After the discrimination is completed, the system stores all abnormal node numbers in the abnormal node screening identification set, and outputs the remaining normal node numbers. The abnormal and normal state division of all nodes in the current monitoring section is completed, and the abnormal node screening identification set with clear structure is generated.

[0075] The path configuration generation submodule reads the connection records between normal nodes according to the abnormal node screening identification set, selects adjacent node pairs with stable connection relationships, configures them as backup transmission paths, and arranges the node numbers and connection relationships of the paths to generate an irrigation communication path list.

[0076] First, all marked abnormal node numbers are excluded from the middle, only normal node numbers are retained into the subsequent processing flow, the system calls the node connection record data table on this basis, verifies the connection state between all normal nodes, and selects the node pairs that meet the stable connection relationship. In the screening process, the system needs to retrieve the connection log records of the node pairs in the historical period one by one, and evaluate the connection state, connection frequency and signal quality of each node pair. The criteria for determining whether it is a stable connection are: 1. The number of connection interruptions within 30 days does not exceed 2 times; 2. The average signal strength during communication is better than -75dBm; 3. The duration of each connection interruption should not exceed 1 hour. If node N002 and N005 interrupt only once within 30 days, the average signal strength is -70dBm, and there is no long-term disconnection, it is determined to be a stable connection. The system records the node pair, marks its number combination and connection strength data, and confirms the physical proximity and path smoothness of the two nodes in the network topology. All node pairs that meet the conditions will be configured as backup transmission paths and recorded in a structured format. For example, node pair (N002-N005) is marked as path P1, and node pair (N006-N008) is marked as path P2. Finally, all available backup path entries are integrated and summarized, and the output is a complete irrigation communication path list.

[0077] Specifically, as shown in Figure 2 、 6 The timing control module includes:

[0078] The response starting point extraction submodule calls the leaf temperature change starting point, soil moisture content change starting point and greenness change starting point corresponding to the control number for the irrigation communication path list, extracts the first change time record corresponding to the index, and pairs with the irrigation start time to calculate the time difference value, generating a control index starting point interval matrix.

[0079] Firstly, read the control number corresponding to each irrigation path listed in the list one by one, and call the various environmental sensing devices deployed under the number from the control number and monitoring device binding table in turn, obtain the data sequence recorded by the leaf temperature monitoring device, soil moisture monitoring device and greenness monitoring device through indexing, find the time position where the value first changes continuously in the three types of monitoring data, and take the time point at this position as the starting point of the change of the index, wherein the starting point of the change of the leaf temperature is the first recorded time when the temperature value changes from a stable state to a continuous rising state, the starting point of the change of the soil moisture is the time when the soil moisture first continuously increases after maintaining at a low level, and the starting point of the change of the greenness is the first position where the vegetation index in the image analysis result starts to show positive growth, for example, the temperature value under a certain control number continuously rises from 25.0 to 25.6 and continues to rise, and the first rising point is the starting point of the change, and the corresponding recorded time point needs to be extracted in a unified format. Similarly, if the moisture content increases from 13.0 to 13.5 and continues to increase, the corresponding time is also extracted as the starting point, and the greenness data is recorded as the starting point through the position where the NDVI value in the continuous image analysis result first shows an upward trend. After the system extracts the starting point time, it further reads the irrigation start time of the control number, and corresponds each irrigation start time with the three index starting points one by one, calculates the time difference value, and if the temperature starting point is 15 minutes before the start, the moisture starting point is 10 minutes before the start, and the greenness starting point is 5 minutes after the start, the difference values are 15, 10 and -5 respectively. All difference values are combined into a multi-dimensional data matrix with index dimension and number as index, and finally a control index starting point interval matrix is output.

[0080] The interval change judgment submodule calculates the time difference value change direction of the number in the continuous period based on the control index starting point interval matrix, judges whether there is a consistent deviation in the change direction, marks the control number that needs to be adjusted, and generates a start time point adjustment identification sequence.

[0081] Read the starting time difference value record of each control number corresponding to the three indicators of leaf temperature, soil moisture content and greenness one by one, and construct the interval sequence in the order of continuous irrigation period, perform difference operation between adjacent items of each sequence, judge the change direction of each indicator on the time axis, that is, if the interval of the current period is shorter than the last period, it is marked as negative change, if it is longer, it is marked as positive change, if there is no change, it is marked as zero, count all the change directions in the complete sequence and analyze whether there is a continuous directional shift, for example, the starting interval of soil moisture content of number C01 is 12, 10, 8 in three periods, which means that the interval value is continuously reduced, and the direction is negative, so the system marks this sequence as "consistent directional shift", on the contrary, if the interval sequence of a certain number is 10, 12, 11, it is marked as "no continuous shift" because the direction change is not consistent, in each control number, the system needs to judge the direction of the three indicator sequences respectively, if the change direction of two or more indicators is consistent and the change times exceed the set minimum continuous threshold, for example, the threshold is set to 2 times, then the system will mark this number as "need to adjust", otherwise it will be marked as "no adjustment", after executing all numbers, the system will arrange the numbers marked as "need to adjust" to form the starting point adjustment identification sequence, which is the basic input data for the subsequent module sorting adjustment.

[0082] The action timing adjustment submodule calls the starting point adjustment identification sequence, reorders all control numbers as needed, prioritizes the numbers marked as "need to adjust", maintains the original time interval structure, fills in the rest of the numbers without adjustment demand in the original order, and generates an irrigation timing plan;

[0083] First, read the complete irrigation plan table, including the starting order and time interval structure corresponding to each control number, then scan the adjustment identification sequence, and prioritize the numbers marked as "need to adjust", the system groups the numbers according to the true value state of the adjustment identification, first extracts the list of numbers that need to be adjusted, sorts them in the original plan according to the time sequence and places them at the front of the new plan table, the rest of the numbers not marked for adjustment maintain the original order and are filled in the sorting result in turn, pay attention to maintaining the original start time interval structure in the original plan during the sorting process, that is, if the original interval is set to 5 minutes, it should be maintained even after the adjustment, for example, if the original interval between numbers C01 and C02 is 5 minutes, and C02 is an adjustment number, it is placed before C01, the interval is still maintained at 5 minutes, after sorting and time offset updating of all numbers, the system binds the new sorted numbers with their reset start times, and finally outputs an updated irrigation timing plan table, in which all numbers marked as "need to adjust" get priority in the plan, while the time interval structure remains unchanged, meeting the system's continuous control constraints.

[0084] Specifically, as shown in Figure 2 , 7 , the irrigation control execution module includes:

[0085] The instruction configuration setting sub-module calls the water pump on-off action and the valve opening-closing action corresponding to the control number based on the irrigation timing plan, arranges the control action content of each group in the order of the number, and sets it as an independent execution instruction sequence to generate a numbered control execution instruction set;

[0086] First, the starting time and the number order corresponding to each control number in the plan are read, and then the execution action configuration data corresponding to the number is called, which contains the on-off action information of the water pump and the opening-closing action information of the electric valve. The system traverses the numbers one by one, extracts the water pump action type (water pump opening or closing) and the valve action type (valve opening or closing) corresponding to the number in each traversal, and binds the two action contents as a group of control action records, for example, the number C05 has a starting time of the 3rd period in the plan, and the configuration action is water pump opening and valve opening, then the action group is set as "C05-ON-OPEN", all actions are set as independent execution units, and there is no logical dependence with the front and rear number actions, which ensures that the action takes effect independently after the instruction is issued. The system needs to maintain the number order consistent with the timing plan during the arrangement process, arrange the instruction groups in the order of small to large or the original order according to the number, and assign each group of action records a unique action sequence number, while setting necessary action execution confirmation flag bits, for example, status code 0 indicates to be executed, 1 indicates to be executed, 2 indicates to fail and wait for retransmission, and finally all the arranged numbered control action groups are converted into execution instruction units with unified format and independent structure, each instruction includes number, action content, sequence identifier, execution status identifier and other fields, and after integration, the numbered control execution instruction set is formed.

[0087] The control path synchronization sub-module calls the numbered control execution instruction set and the irrigation communication path list, extracts the standby path information corresponding to each control number, and synchronizes the execution instruction to the control terminal device in the path, records the synchronization transmission state, and generates a standby path execution state table;

[0088] Firstly, the execution instruction content corresponding to each control number is extracted from the instruction set, and the node information of the control number in the standby communication path is found by establishing an index relationship with the path list. During the execution process, the system establishes a binding relationship between the instructions and the path terminal devices, performs identity verification and connection test on the terminals under each standby path, ensures that the communication link is unobstructed, and then synchronously transmits the instructions to the corresponding terminal node devices under the path. Each instruction is attached with a unique instruction number and an instruction hash check value for the terminal to confirm whether the reception is successful. After receiving the complete instruction, the terminal sends a response identifier to the upper system. The system records the synchronization state according to the received feedback identifier. If the terminal returns an "ACK" confirmation code, it is marked as successful synchronization. If it does not return or returns an error code, it is marked as failure. The system records the synchronization state of each control number item by item, and forms a complete state data table with all the records. The table contains fields such as number, instruction type, transmission state (success / failure), and number of attempts. All state records are sorted in number order and finally form a standby path execution state table.

[0089] The response state recording submodule detects the response signal of each control number according to the standby path execution state table, extracts the feedback action state of the number, and records the response completion and timestamp data in order of number to obtain the crop irrigation control scheme.

[0090] Firstly, read all control numbers and their corresponding synchronization states in the table, and start the response signal detection program for the numbers marked as "synchronization success". The system establishes a two-way listening with the terminal control device through the real-time communication interface. Within the specified time window after instruction execution, it detects whether the terminal returns the preset feedback action signal. The feedback signal mainly includes pump running state (running / stop), valve position state (open / close), current load signal and internal state code, etc. The system analyzes each feedback signal, extracts the core state field of whether the control action is executed, for example, if the pump action is expected to be turned on and the actual return state is "running", it is marked as completed, if the feedback is "stop", it is marked as not completed. The feedback action execution of each number needs to be compared and judged with the original instruction content, and the timestamp information of the response completion is recorded synchronously, for example, number C05 returns the feedback signal at the 2nd minute after the instruction is issued, the system records this time as the execution response time. All response records are arranged in order of number and time, and written into the irrigation response record table. Finally, after completing the response situation detection of all numbers, a data table containing the response state, execution situation, response time, etc. of each number is output, which is used as the final execution feedback basis of the crop irrigation control scheme.

[0091] The above merely illustrates the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A precision irrigation control system integrating crop water requirement model and multi-source data, characterized in that, The system includes: The humidity response identification module collects soil moisture data at different depths in agricultural areas, extracts the moisture content difference as the disturbance amplitude, determines whether the increase or decrease of adjacent sampling points has changed direction, filters the layers with disturbance trends, and generates the root layer distribution location. The root region division module, based on the distribution location of the root layer, matches the change direction of adjacent layers, extracts the markers of the disturbed layers, and groups the layers with consistent markers into the same segment, while those with inconsistent markers are divided into independent response segments, thereby generating a control region distribution map. Based on the control area distribution map, the communication path construction module extracts the control number, calls the corresponding node signal record, calculates the signal strength difference, judges the transmission anomaly by the number of descents, and generates an irrigation communication path list. Based on the irrigation communication path list, the timing control module calls the starting point of leaf temperature, humidity, and greenness changes for each number, compares the interval with the irrigation start time, determines whether the start time needs to be adjusted according to the direction of the interval change, and generates an irrigation timing plan. The irrigation control execution module calls the actions of each numbered water pump and valve according to the irrigation sequence plan and issues instructions, which are synchronized to the control terminal and backup path to generate a crop irrigation control plan.

2. The integrated crop water requirement model and multi-source data precision irrigation control system according to claim 1, characterized in that: The root layer distribution location includes disturbance amplitude change value, disturbance direction switching point, and continuous disturbance layer. The control area distribution map includes continuous response zone number, segment boundary mark, and layer affiliation label. The irrigation communication path list includes signal strength fluctuation value, number of transmission anomaly judgments, and backup node path. The irrigation timing plan includes irrigation start interval difference, action sequence number, and periodic change direction identifier. The crop irrigation control scheme includes control command type, corresponding pump and valve number, and control response record item.

3. The precision irrigation control system integrating crop water requirement model and multi-source data as described in claim 1, characterized in that: The layer of disturbance trend refers to the soil layer where the direction of increase or decrease changes during the soil moisture disturbance process.

4. The precision irrigation control system integrating crop water requirement model and multi-source data as described in claim 1, characterized in that: The transmission anomaly refers to an abnormal state in which the signal strength of a node in the irrigation communication path continuously drops more than a set number of times.

5. The integrated crop water requirement model and multi-source data precision irrigation control system according to claim 1, characterized in that, The humidity response recognition module includes: The humidity data acquisition submodule acquires soil humidity sensing data deployed at different depths in agricultural planting areas, collects the water content change value of each depth point within a preset time period, organizes the data in chronological order and depth labels, and generates a soil humidity time series dataset. The disturbance amplitude calculation submodule calculates the difference in water content of continuous time series at the same depth point as the disturbance amplitude based on the soil moisture time series dataset, records the disturbance direction marker, and organizes the disturbance amplitude and direction sequence data. The root layer location identification submodule calls the disturbance amplitude and direction sequence data, determines whether the direction of adjacent disturbances has changed, filters the depth layers where the continuous disturbance amplitude exceeds the preset disturbance threshold, counts the frequency and continuity, and generates the root layer distribution location.

6. The integrated crop water requirement model and multi-source data precision irrigation control system according to claim 1, characterized in that, The root region partitioning module includes: The layer orientation matching submodule, based on the root layer distribution location, calls the identified perturbation layer and its corresponding change direction marker, extracts adjacent layer pairs in layer order, compares the change direction markers one by one to see if they are consistent, records the comparison results, and generates a layer orientation consistency identifier matrix. The response segment classification submodule determines whether consecutively marked consistent layer pairs meet the merging condition based on the layer direction consistency identifier matrix. If they do, they are merged into one group; otherwise, they are divided into independent groups. The response segment set is constructed in sequence to generate a response segment structure list. The region numbering generation submodule calls the response segment structure list, numbers each response segment according to the layer arrangement order, constructs a number index mapping relationship, draws the control boundary in combination with the layer range corresponding to the number, and obtains the control area distribution map.

7. The integrated crop water requirement model and multi-source data precision irrigation control system according to claim 1, characterized in that, The communication path construction module includes: The number signal extraction submodule extracts the control number according to the control area distribution map, obtains the signal record data of the field control node corresponding to the number, extracts the signal strength value of each node in a continuous time period, calculates the strength difference between adjacent time points, and generates a node signal strength difference sequence. The node anomaly detection submodule calls the node signal strength difference sequence to count the number of consecutive drops in the node signal strength, determines whether it exceeds the set drop threshold, marks the node number with anomalies, records the normal node number, and generates an abnormal node screening identifier set. The path configuration generation submodule reads the connectivity records between normal nodes based on the abnormal node filtering identifier set, filters adjacent node pairs with stable connection relationships, configures them as backup transmission paths, and organizes the node numbers and connection relationships of the paths to generate an irrigation communication path list.

8. The integrated crop water requirement model and multi-source data precision irrigation control system according to claim 1, characterized in that, The timing control module includes: The response start point extraction submodule, for the irrigation communication path list, calls the leaf temperature change start point, soil moisture change start point and greenness change start point corresponding to the control number, extracts the first change time record corresponding to the index, and calculates the time difference by pairing it with the irrigation start time of each irrigation, and generates a control index start point interval matrix. The interval change judgment submodule calculates the direction of change of the time difference value of the number within a continuous period based on the starting interval matrix of the control index, judges whether there is a consistency deviation in the direction of change, marks the control number that needs to be adjusted, and generates the start-up time point adjustment identifier sequence. The action timing adjustment submodule calls the start-time point adjustment identifier sequence, reorders all control numbers as needed, prioritizes the numbers whose adjustment identifiers are in the adjustment state, keeps the original time interval structure unchanged, and fills in the numbers without adjustment needs in the original order to generate an irrigation timing plan.

9. The precision irrigation control system integrating crop water requirement model and multi-source data as described in claim 1, characterized in that, The irrigation control execution module includes: The instruction configuration setting submodule, based on the irrigation timing plan, calls the pump on / off action and valve opening / closing action corresponding to the control number, organizes the content of each group of control actions in numerical order, and sets them as an independent execution instruction sequence to generate a numbered control execution instruction set; The control path synchronization submodule calls the numbered control execution instruction set and the irrigation communication path list, extracts the backup path information corresponding to each control number, synchronizes the execution instructions to the control terminal equipment in the path, records the synchronization transmission status, and generates a backup path execution status table. The response status recording submodule detects the response signal of each control number according to the backup path execution status table, extracts the feedback action status of the number, and records the response completion status and timestamp data in numerical order to obtain the crop irrigation control scheme.

10. The integrated crop water requirement model and multi-source data precision irrigation control system according to claim 7, characterized in that: The drop threshold refers to the number of times the node signal strength continuously drops to determine whether it is abnormal.

Images