Wheat-corn annual water nitrogen content adjusting system

Abstract

The invention relates to the technical field of internet-of-things control, in particular to a wheat-corn annual water nitrogen content adjusting system, which comprises a data sensing module for acquiring soil humidity and nitrogen content data in a wheat-corn growth cycle, filtering the soil humidity and nitrogen content data and generating processed soil data; based on the processed soil data, a set standard threshold value is compared, and a soil state result is obtained. According to the method, the adjustment strategy is dynamically generated by further combining water and nitrogen demand standards of different growth stages, instant weather forecast data is synchronously analyzed when the strategy is generated, multi-dimensional environmental factors such as precipitation, wind speed, temperature and air humidity are considered, and an optimization adjustment scheme which better conforms to real-time demands of crops and environmental changes is generated. In the execution process, an instruction received by the irrigation and fertilization machinery is not a static single execution instruction any more, but is iteratively executed based on real-time feedback and environmental dynamics.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things control technology, and particularly to a wheat-corn annual water and nitrogen content regulation system. Background Art

[0002] The Internet of Things control technology field is a systematic technology that integrates sensors, actuators, communication networks, and data processing platforms to achieve automatic monitoring, analysis, decision-making, and control of devices and resources in the physical environment. The wheat-corn annual water and nitrogen content regulation system is an intelligent agricultural solution designed based on the Internet of Things control technology for the precise management of water and nitrogen resources in wheat and corn rotation farmland. The system uses soil sensors to monitor soil humidity and nitrogen content in real time, and combines the water and nitrogen requirements of the crop growth stage and weather forecast data to dynamically generate optimized irrigation and fertilization strategies.

[0003] In the prior art, without fully combining real-time meteorological parameters and relying only on historical meteorological means and fixed growth models, it is difficult to reflect the immediate impact of sudden weather on the water and nitrogen requirements of farmland. Especially in the case of drastic changes in precipitation, wind speed, and temperature, problems such as the mismatch between water supply and fertilization amount and actual demand are likely to occur. Therefore, improvements are needed. Summary of the Invention

[0004] The purpose of the present invention is to solve the deficiencies in the prior art and propose a wheat-corn annual water and nitrogen content regulation system.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions: The wheat-corn annual water and nitrogen content regulation system includes:

[0006] A perception data module that collects soil humidity and nitrogen content data during the growth cycle of wheat and corn, filters them, and generates processed soil data; based on the processed soil data, compares with set standard thresholds to obtain soil status results;

[0007] A decision logic module that receives the soil status results, calculates adjustment plans according to the water and nitrogen requirements of wheat and corn, and generates adjustment strategies; for the adjustment strategies, refers to real-time weather forecast data to obtain optimized adjustment plans;

[0008] An execution control module that uses the optimized adjustment plan to set commands for irrigation and fertilization machinery, generates control signal output results; based on the control signal output results, monitors the execution situation and records the execution results to obtain execution feedback;

[0009] A synchronous feedback module that analyzes the execution feedback, judges the irrigation and fertilization effects, and generates effect evaluation results.

[0010] Preferably, the steps for obtaining the processed soil data are as follows:

[0011] Continuously monitor the soil humidity and nitrogen content during the growth cycles of wheat and corn through soil sensors, collect data in real time and record it to obtain the original soil data set;

[0012] Based on the original soil data set, perform noise filtering and outlier detection to obtain the processed soil data.

[0013] Preferably, the steps for obtaining the soil status result are as follows:

[0014] Based on the processed soil data, extract the soil humidity and nitrogen content values in the data, classify each data point, and compare the timestamp information of each data point to obtain the time series soil data;

[0015] According to the time series soil data, calculate the soil status score. The calculation formula is:

[0016]

[0017] where Q is the soil status score, H j is the soil humidity at the jth time point, T H is the standard humidity threshold, N k is the soil nitrogen content at the kth time point, T N is the standard nitrogen content threshold, and m is the total number of data points;

[0018] According to the soil status score, compare it with the set standard threshold to judge the matching situation of the soil humidity and nitrogen content, and obtain the soil status result.

[0019] Preferably, the steps for obtaining the adjustment strategy are as follows:

[0020] According to the soil status result, calculate the water and nitrogen demand adjustment index. The calculation formula is:

[0021]

[0022] where I is the water and nitrogen demand adjustment index, X h is the current soil humidity, Y h is the standard water demand, X n is the current soil nitrogen content, Y n is the standard nitrogen demand, Z t is the monitoring timestamp difference, L m is the monitoring location coordinate difference, M s is the soil texture matching difference;

[0023] Select the adjustment range of water and fertilizer supply according to the water and nitrogen demand adjustment index, and generate an adjustment strategy.

[0024] Preferably, the steps for obtaining the optimized adjustment plan are as follows:

[0025] Based on the adjustment strategy, extract the water and nitrogen adjustment demand parameters, and at the same time call the real-time weather forecast data to extract the precipitation probability, temperature and wind speed, and generate an environmental impact factor set;

[0026] According to the environmental impact factor set, calculate the adjustment fitness, and the calculation formula is:

[0027]

[0028] Among them, F is the adjustment fitness, A r is the irrigation demand in the adjustment strategy, P r is the precipitation expectation in the weather forecast, B t is the forecast temperature, C h is the environmental humidity, D w is the wind speed, E s is the soil evapotranspiration rate;

[0029] According to the adjustment fitness, recalculate the irrigation amount, fertilization amount and application time to form an optimized adjustment plan.

[0030] Preferably, the steps for obtaining the control signal output result are as follows:

[0031] Based on the optimized adjustment plan, call the irrigation amount, fertilization amount, application time and frequency, and combine with the monitoring position coordinates to generate an irrigation and fertilization instruction parameter set;

[0032] According to the irrigation and fertilization instruction parameter set, parse the irrigation amount, fertilization amount, application time and frequency in each instruction item by item, match the operation interfaces of the irrigation machinery and fertilization machinery, and convert the parameter format according to the interface protocol to generate an execution instruction set;

[0033] According to the execution instruction set, send a control signal to the operation terminals of the irrigation machinery and fertilization machinery, monitor the mechanical reception status and response in real time, record the execution feedback of each instruction, and generate a control signal output result.

[0034] Preferably, the steps for obtaining the execution feedback are as follows:

[0035] Based on the control signal output result, collect the operation logs of each mechanical terminal in real time, sort out the execution status, operation instructions, equipment feedback and error code records in chronological order, and generate an execution monitoring data set;

[0036] Analyze the execution status parameters one by one according to the execution monitoring data set. By comparing the instruction requirements with the mechanical feedback results, parse the equipment startup delay, execution duration deviation, and abnormal interruption one by one, and screen the alarm events and terminal response situations during the equipment operation to generate an execution result report;

[0037] According to the execution result report, extract the execution success rate, abnormal frequency, and failure type of each mechanical terminal, and combine the monitoring timestamp, operation log, and equipment feedback code to classify and organize them according to equipment, time, and operation category to generate an execution feedback.

[0038] Preferably, the steps for obtaining the effect evaluation result are as follows:

[0039] According to the execution feedback, calculate the irrigation and fertilization effect scores. The calculation formula is:

[0040]

[0041] Among them, E is the irrigation and fertilization effect score, P a is the irrigation volume, Q a is the standard irrigation volume, R d is the fertilization execution deviation, S d is the standard fertilization deviation, T m is the operation time deviation, U f is the equipment failure frequency, V t is the number of successful executions, W l is the number of operation interruptions;

[0042] According to the irrigation and fertilization effect scores, evaluate the irrigation and fertilization effects to generate an effect evaluation result.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0044] In the present invention, by further combining the water and nitrogen demand standards at different growth stages, an adjustment strategy is dynamically generated, and real-time weather forecast data is synchronously analyzed when generating the strategy. Considering multi-dimensional environmental factors such as precipitation, wind speed, temperature, and air humidity, an optimized adjustment plan that better conforms to the real-time needs of crops and environmental changes is generated. During the execution process, the instructions received by the irrigation and fertilization machinery are no longer static single-execution instructions, but iterative executions based on real-time feedback and environmental dynamics. By continuously monitoring key parameters such as the execution status, operation duration, and abnormal alarms, the execution results are recorded in real time, and differential analysis is performed in combination with the crop growth status and soil water and nitrogen content to judge the execution effect. This dynamic closed-loop water and nitrogen regulation system, through the linkage of real-time monitoring, precise adjustment, and feedback correction, makes the water and fertilizer supply more precise and self-adaptive, can effectively reduce the waste of water and nitrogen resources, reduce the environmental stress of farmland caused by excessive fertilization and irrigation, and ensure the water and nitrogen balance of crops at different growth stages. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a system flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with 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 used to limit the present invention.

[0047] Please refer to Figure 1 , the present invention provides a technical solution: the annual water and nitrogen content regulation system for wheat and corn includes:

[0048] A perception data module, which collects soil humidity and nitrogen content data during the growth cycle of wheat and corn, filters them, and generates processed soil data; based on the processed soil data, compares with the set standard thresholds to obtain soil status results;

[0049] A decision logic module, which receives the soil status results, calculates an adjustment plan according to the water and nitrogen requirements of wheat and corn, and generates an adjustment strategy; for the adjustment strategy, refers to the real-time weather forecast data to obtain an optimized adjustment plan;

[0050] An execution control module, which adopts the optimized adjustment plan to set instructions for the irrigation and fertilization machinery, generates a control signal output result; based on the control signal output result, monitors the execution situation and records the execution results to obtain execution feedback;

[0051] A synchronous feedback module, which analyzes the execution feedback, judges the irrigation and fertilization effects, and generates an effect evaluation result.

[0052] The steps for obtaining the processed soil data are as follows:

[0053] Continuously monitor the soil humidity and nitrogen content during the growth cycles of wheat and corn through soil sensors, collect data in real time and record it to obtain the original soil data set;

[0054] Based on the original soil data set, perform noise filtering and outlier detection to obtain the processed soil data.

[0055] Specifically, in this process, it is necessary to first disassemble the range and measurement accuracy information of the sensor according to the sensor technical specification text provided by the manufacturer, and pre-determine the appropriate number and location of sensor installations in the wheat and corn growth areas. For example, one sensor can be set at each of the four corners and the center of each field, for a total of five sensors. The corresponding sensor measurement range can be determined by empirical data and on-site tests. For example, the soil humidity is limited to the range of 0% to 70%, and the nitrogen content is limited to the range of 0 to 1000 mg / kg. If a certain measurement data is lower than 0% or higher than 70%, it is marked as abnormal humidity. If the nitrogen content exceeds the range of 0 to 1000 mg / kg, it is marked as abnormal nitrogen value. After arranging the sensors in this way, it is necessary to conduct stability and consistency checks. For example, read the humidity and nitrogen content values output by the sensor every five minutes and record them, and then compare the recorded values with the reference readings of the reference device. If the average deviation between the two exceeds the preset range, it indicates that there may be problems with the installation or sensor calibration. The deviation threshold here can be set to 5% by experience. To determine the reasonableness of this value, a certain area can be selected for continuous monitoring for three days, and several data are randomly selected every day to calculate the deviation and take its average. When the average deviation is lower than 5% and stable, this 5% is used as the threshold to mark the abnormality degree of subsequent readings. After multiple rounds of confirmation, the long-term measurement process can be started. Continuously sample the sensors at the previously set time interval and record all readings. Each record includes a timestamp, sensor number, soil humidity value, and nitrogen content value. When the quantity accumulates to a certain scale, for example, when two thousand records are generated daily on average, summarize all real-time acquisition results to form the original soil data set.

[0056] In this process, it is necessary to disassemble three elements: timestamp, soil humidity, and nitrogen content from the original soil dataset obtained previously, and perform noise filtering and outlier detection for each record. For noise filtering, a preliminary noise discrimination range can be set first. For example, compare the humidity value with the range of 0% to 70%, and compare the nitrogen content with the range of 0 to 1000 mg / kg. If a record shows that the soil humidity is less than 0% or greater than 70%, it is judged as candidate noise data, and then further investigation is carried out in combination with the output characteristics of the sensor. Specifically, by comparing the measurement results of adjacent sensors in the same area at the same time period, if the measurement difference exceeds a certain proportional threshold, the record is continuously marked as high noise and screened out. The proportional threshold here can be estimated through testing. For example, take 10% as the initial setting value. If there are still too many abnormal entries meeting this proportion in the actual monitoring data, this value can be increased to 15% or 20% and re-evaluated. Furthermore, when detecting outliers for the remaining data, the method based on fixed standard deviation can be selected. First, calculate the mean and standard deviation of humidity and nitrogen content respectively. If the deviation of a record from the mean exceeds three times the standard deviation, it is regarded as an outlier and needs to be separately retained or excluded in subsequent links. Combining these steps, the cleaned results can be compared, marked, and recorded item by item, and finally the processed soil data is obtained.

[0057] The steps to obtain the soil state result are as follows:

[0058] Based on the processed soil data, extract the soil humidity and nitrogen content values in the data, classify each data point, and compare the timestamp information of each data point to obtain the time-series soil data;

[0059] According to the time-series soil data, calculate the soil state score. The calculation formula is:

[0060]

[0061] where Q is the soil state score, H j is the soil humidity at the j-th time point, T H is the standard humidity threshold, N k is the soil nitrogen content at the k-th time point, T N is the standard nitrogen content threshold, and m is the total number of data points;

[0062] According to the soil state score, compare the set standard threshold to judge the matching situation of soil humidity and nitrogen content, and obtain the soil state result.

[0063] Specifically, based on the processed soil data formed previously, it is necessary to first check the timestamps and corresponding soil moisture and nitrogen content values in each record, and arrange these records from the earliest to the latest timestamp. After the arrangement, group them to compare the value ranges of soil moisture and nitrogen content. For example, check each soil moisture value against the pre-established range of 0% to 70% one by one. If a value is lower than 0% or higher than 70%, mark that record as an abnormal situation. Then, compare each record's nitrogen content against the previously recorded range of 0 to 1000 milligrams per kilogram one by one. If the nitrogen content in a certain record is less than 0 or higher than 1000 milligrams per kilogram, mark that record as abnormal. Gather the remaining records within the normal range together to form a subsequent available dataset. To avoid losing key time information, it is also necessary to maintain the original timestamp order of the records during the grouping process. Then, classify and organize the soil moisture and nitrogen content of each record. Bucket the integer or decimal part of the moisture value according to the set sub-intervals for statistical purposes, and at the same time bucket the nitrogen content value according to the same principle and record the corresponding timestamps. Sort the records within each bucket again by timestamp, and extract the consecutive time period data after sorting as a group. If the number of records in some consecutive time periods within a group is too small, merge them into adjacent groups to form a more complete sequence for subsequent calculation processes. Finally, summarize the sequences of soil moisture and nitrogen content values arranged in chronological order in each group to obtain the time series soil data.

[0064] The advantage of the formula is that by simultaneously reflecting the deviation degrees of soil moisture and nitrogen content in the same expression, the differences between moisture and nitrogen content from the standard thresholds can be quantified synchronously in one calculation step, thereby reflecting the overall deviation amount between the soil state and the ideal threshold at the numerical level.

[0065] H j Parameter Introduction:

[0066] H j represents the soil moisture value measured at the j-th time point. To obtain H j , it is necessary to deploy moisture sensors in the field and collect data at fixed time intervals. The interval duration can be determined by referring to the local soil moisture change rate and management requirements. For example, set to record every ten minutes. Each data includes a timestamp and a moisture value. Up to 144 records can be obtained in a day, and more than 1000 records can be obtained in a week. To ensure the accuracy of the moisture data, multiple devices can be used and the data can be cross-compared, and the results can be summarized after calculating the deviations. The specific acquisition method of H j is not limited to automated detection only. Conventional agronomic testing methods can also be referred to, but usually, automated sensor data is more convenient for subsequent processing. If numerical examples are needed below, H j data within a certain time period can be selected for display.

[0067] T H Parameter Introduction:

[0068] T H represents the standard humidity threshold, which usually needs to be determined according to the growth requirements of the target crop and local climate conditions. It can be obtained by monitoring the yields and plant health conditions of multiple fields at different humidity levels, and then statistically analyzing the combined empirical data of planting experts. To obtain a more representative threshold range, it is necessary to continuously monitor the humidity situation within at least one growth cycle, collect samples and calculate the numerical range with relatively ideal growth conditions when the humidity is within a certain interval. Subsequently, the median or average value can be selected as the specific T H value. It can also be moderately optimized before the start of each round of planting, for example, making different degrees of adjustments in summer or winter to cope with climate differences. If it is to be used in the numerical example, T H can be set to 40%.

[0069] N k Parameter Introduction:

[0070] N k represents the soil nitrogen content value measured at the k-th time point. To accurately obtain N k , sensors capable of detecting nitrogen concentration need to be installed in the field or chemical detection methods need to be adopted, such as periodically collecting a small amount of soil samples and conducting laboratory analysis, and then correlating the nitrogen content obtained in the laboratory with the time stamp to form sequence data, so as to obtain multiple records on the unified time axis. If automated sensors are used, the sensitivity and drift rate of the sensors need to be calibrated before installation and regular maintenance needs to be carried out to reduce the measurement deviation caused by long-term use. The monitoring frequency can be set the same as or slightly lower than that of humidity, and it can be selected in combination with the fertilizer requirements of the crop. If this value is to be called in the numerical example, several groups of representative monitoring data can be selected for comparison.

[0071] T N Parameter Introduction:

[0072] T N represents the standard nitrogen content threshold, which is determined by observing the phased changes in nitrogen demand during the crop growth process and based on a certain amount of field test data. The interval with the most concentrated nitrogen content distribution or the interval corresponding to the most stable plant growth is used as the judgment basis. It can also refer to the suitable range of soil nitrogen published by relevant agricultural research institutions and be revised by comparing the local crop yield records to form a threshold value that matches the local soil characteristics and management requirements. Subsequently, the threshold can be moderately fine-tuned according to the actual measurement situation during the growing season, but it needs to be based on the trend analysis of the collected data and then make a comprehensive judgment. In the numerical example, T N can be set to 600 mg / kg to evaluate the actual degree of deviation of the nitrogen content.

[0073] Introduction to parameter m:

[0074] m represents the total number of data points. In this formula, it is used as the counting basis for summing the deviation values of humidity and nitrogen content. If the number of actually collected data records in the same time period is much higher than the average level, m will increase accordingly, which means a larger amount of computation is required. Usually, in actual monitoring, one day or one week can be taken as a calculation cycle, and all valid data within this cycle are counted and set as the value of m. The value of m can be obtained from an automatically recorded counter. If there is screening or abnormal data elimination after data collection, its value will also change. In the example, if one week's data is taken, assuming sampling once every ten minutes, there are 144 records in a day and approximately 1008 valid data in a week, then m = 1008.

[0075] Calculation process:

[0076] Here, m = 5 is selected as an example. There are humidity values and nitrogen content values at a total of five time points. Let H 1 = 36%, H 2 = 38%, H 3 = 42%, H 4 = 39%, H 5 = 45%, and for the nitrogen content values, let N 1 = 550, N 2 = 610, N 3 = 580, N 4 = 620, N 5 = 650, and let T H = 40%, T N = 600. When substituting into the formula, first calculate the humidity term: (H 1 -T H ) 2 =(36 - 40) 2 =(-4) 2 = 16

[0077] (H 2 -T H ) 2 =(38 - 40) 2 =(-2) 2 = 4

[0078] (H 3 -T H ) 2 =(42 - 40) 2 = 2 2 = 4

[0079] (H 4 -T H )2 =(39 - 40) 2 =(-1) 2 =1

[0080] (H 5 -T H ) 2 =(45 - 40) 2 =5 2 =25

[0081] Then calculate the nitrogen content term: (N 1 -T N ) 2 =(550 - 600) 2 =(-50) 2 =2500

[0082] (N 2 -T N ) 2 =(610 - 600) 2 =10 2 =100

[0083] (N 3 -T N ) 2 =(580 - 600) 2 =(-20) 2 =400

[0084] (N 4 -T N ) 2 =(620 - 600) 2 =20 2 =400

[0085] (N 5 -T N ) 2 =(650 - 600) 2 =50 2 =2500

[0086] Add up the above results: The sum of humidity deviation amounts = 16 + 4 + 4 + 1 + 25 = 50, the sum of nitrogen content deviation amounts = 2500 + 100 + 400 + 400 + 2500 = 5400. Therefore, the total is 50 + 5400 = 5450. Then take the square root:

[0087] The results show that there are certain degrees of deviation in soil moisture and nitrogen content simultaneously. The higher the value, the more significant the deviation. If Q is compared with the standard threshold in its subsequent process, for example, in agricultural practice, the range between Q = 40 and Q = 60 is regarded as a reasonable interval. If it exceeds 60, it is considered that the deviation of moisture and nitrogen content is large. At this time, 73.82 indicates an obvious difference in the current soil indicators, and further analysis or corresponding operations may be required for the subsequent soil state results.

[0088] Based on the previously obtained soil state scores, a standard threshold range applicable to this region needs to be given first. For example, through local field tests and long-term record statistics, it can be found that when the soil state score is in the range of 30 to 60, the soil moisture and nitrogen content are relatively stable. When it is between 60 and 80, the degree of deviation of moisture and nitrogen content increases. To determine this interval, continuous detection for a whole year is required on several plots in different seasons, and the daily or weekly soil state scores are collected. The section with the densest distribution and relatively stable corresponding plant conditions is selected as the reference value, and then the data is corrected with expert experience to obtain a comprehensive threshold range. Next, the previously calculated soil state scores can be compared with the above intervals item by item. If the value is found to be between 30 and 60, it indicates that the soil moisture and nitrogen content are not much different from the expected values. If it is greater than 60 but does not exceed 80, record its specific range and make targeted adjustments in combination with the growth period of the crop. If it exceeds 80, it means that there may be a more obvious deviation in moisture and nitrogen content, and re-detection or additional measures are required during this period, so as to obtain the soil state results after the comparison.

[0089] The steps to obtain the adjustment strategy are as follows:

[0090] According to the soil state results, calculate the water and nitrogen requirement adjustment index. The calculation formula is:

[0091]

[0092] Among them, I is the water and nitrogen requirement adjustment index, X h is the current soil moisture, Y h is the standard water requirement, X n is the current soil nitrogen content, Y n is the standard nitrogen requirement, Z t is the monitoring timestamp difference, L m is the monitoring location coordinate difference, M s is the soil texture matching difference;

[0093] Select the adjustment range of water and fertilizer supply according to the water and nitrogen requirement adjustment index to generate the adjustment strategy.

[0094] Specifically, the advantage of the formula is that by combining soil moisture, nitrogen content, timestamp difference, monitoring location coordinate difference, and soil texture matching difference into the same cube root operation expression, the influence of multiple factors on water and nitrogen requirement adjustment can be reflected in a single comprehensive calculation, thus more comprehensively reflecting the gap between the current soil environment and the standard requirements as a whole.

[0095] X h Parameter introduction:

[0096] X h is the current soil moisture, which is obtained by installing moisture sensors in the planting area and measuring at stable intervals. The measured data needs to exclude outliers caused by sensor failures or extreme environments before it can be confirmed as valid. To more accurately reflect the soil moisture status at different times, a reasonable measurement period should also be set in combination with weather conditions and soil layer depth. For example, increase the monitoring frequency before and after precipitation to grasp the dynamic changes of moisture. To quantify this parameter, it can be recorded as a percentage or mass water content. The humidity in common planting areas usually fluctuates within the range of 0% to 70%. In actual agricultural production, it is necessary to further subdivide according to the drought or waterlogging tolerance of the plants to obtain X. h When obtaining X, it can also be compared with the long-term observation records of farmers and calibrated. If X is cited in this formula. h the value is 30% or 35%.

[0097] Y h Parameter introduction:

[0098] Y h is the standard water requirement. This value usually needs to be determined by sorting out the optimal water requirements of crops at different growth stages. It can be extracted by long-term tracking of the corresponding relationships between different water application rates and indicators such as yield, plant height, and leaf color, or by referring to the crop water requirement recommended values provided by local agricultural research institutions and then correcting according to the actual situation of the farmland. Before obtaining this value, parallel tests are often carried out on multiple plots, recording the irrigation amount and soil moisture input at each stage during the planting process, and then comparing the comprehensive indicators of crop growth for regression analysis, so as to extract a relatively stable water requirement range, and then taking the median or average value of this range as Y. h , or corresponding values can be set in stages. If it is to be substituted into this formula, Y can be set. h = 40%.

[0099] X n Parameter introduction:

[0100] X nis the current soil nitrogen content, which needs to be obtained by testing soil samples in the field or using sensors equipped with nitrogen measurement functions. Usually, the nitrogen content in agricultural production can range from several hundred milligrams to over a thousand milligrams per kilogram. To determine whether this value is accurate, it is necessary to cross-compare the nitrogen content test results of multiple samplings and exclude anomalies caused by instrument errors or operational mistakes. The relatively stable value is defined as the effective X n , if using sensors for automatic monitoring, it is necessary to perform reference calibration before installation and conduct regular comparisons to prevent drift. In addition, the soil types and fertility bases vary greatly in some areas. Therefore, after obtaining X n , it is necessary to make further adjustments based on the monitoring sequence of the previous round or multiple rounds to determine whether it matches the actual crop requirements. When introducing the formula, X n can be set to 520 milligrams per kilogram.

[0101] Y n Parameter introduction:

[0102] Y n is the standard nitrogen requirement, which is the reference demand for nitrogen by the crop at the current growth stage. Through field trials and actual soil testing results, the general suitable range of nitrogen can be determined. For example, during the critical period of grain growth, the suitable nitrogen content ranges can be statistically calculated for the seedling stage, jointing stage, and filling stage respectively, and then corrected in combination with climate, soil quality, and historical fertilization amount. Finally, the specific value of the standard nitrogen requirement is obtained. If the nitrogen requirement varies greatly at different times, different thresholds can be set in stages. When substituting into this formula, Y n can be set to 600 milligrams per kilogram as the reference value for the current period.

[0103] Z t Parameter introduction:

[0104] Z t is the monitoring timestamp difference, which is used to measure the time span between different sampling or monitoring times. This difference needs to be extracted from the time information of each soil parameter collection. If one measurement is at 8 o'clock on the same day and another measurement is at 14 o'clock on the same day, the corresponding time difference can be defined as 6 hours, or it can be converted into finer-grained units such as minutes. If comparing on a longer time scale, the cumulative time difference is calculated sequentially according to seasonal or monthly sampling records. At the same time, for an automated monitoring system, the corresponding machine clock log can also be read to accurately judge the interval between collection points. When taking the square root of Z t in the formula, it is necessary to first map this difference to a reasonable numerical range. For example, hours or days can be used as the input. When the time span is large, this value will be larger, and the proportion it occupies in the subsequent adjustment index will also increase. In the calculation example, Z t can be set to 6, representing a difference of 6 hours.

[0105] L m Parameter Introduction:

[0106] L m is the difference in monitoring position coordinates, which is determined by GPS coordinates or position identifiers in a geographic information system. By calculating the differences in longitude and latitude or projected coordinates between different measurement points, a numerical position information difference can be obtained. If multiple sensors are distributed within the same plot, their respective coordinates need to be recorded, and the difference is calculated during comparison. It can also be converted based on the geographic distance formula, such as using spherical distance or planar projection distance, and combining the size of the reference plot to judge the relative position differences between specific measurement points. In practical applications, if two monitoring points are close to each other, L m is smaller; if they are far apart, L m is larger. Subsequently, this difference will affect the calculation of the comprehensive index. In the example calculation, L m can be set to 10, indicating that the monitoring point is approximately 10 meters away from the standard measurement point.

[0107] M s Parameter Introduction:

[0108] M s is the difference in soil texture matching, representing the difference in texture between the currently monitored soil and the standard soil sample. It can be extracted from the detection data of the particle composition of various types such as clay, loam, and sandy soil. To obtain this parameter, the soil needs to be taken back to the laboratory or measured using an on-line analysis instrument to determine the proportions of sand, silt, and clay particles, and then compared with the particle size distribution of the standard soil sample. The numerical difference degree is obtained by using the Euclidean distance or weighted distance method. In field management, it is usually calibrated according to regional experience and the formula fertilization recommendation letter to obtain a relatively accurate difference range. If the textures of the two are highly similar, this value should be close to 0; if there is a large difference, a relatively high value will appear. In the example calculation, M s can be set to 3, indicating that there is a certain difference in particle composition between the two.

[0109] Calculation Process:

[0110] Here, an example calculation is selected. Substitute X h = 30%, Y h = 40%, X n = 520 mg / kg, Y n = 600 mg / kg, Z t = 6, L m = 10, M s = 3 into the formula respectively:

[0111] Calculate the humidity difference term:

[0112]

[0113] Calculate the nitrogen content difference term:

[0114]

[0115] Calculate the square root of the timestamp difference term:

[0116]

[0117] Calculate the position coordinate difference term:

[0118]

[0119] Add the above results:

[0120] 0.32 + 0.0236 + 2.449 + 2.5 ≈ 5.2926

[0121] Finally, take the cube root:

[0122]

[0123] This result indicates that when the soil humidity is lower than the standard value and the nitrogen content is significantly low, combined with factors such as the sampling time interval and position difference, the calculated water and nitrogen requirement adjustment index I is approximately 1.75. A higher value means that a relatively greater adjustment of water and fertilizer supply should be prepared for implementation. If it is found in practice that the index is significantly higher than a certain threshold, the irrigation amount can be increased or more fertilizer can be applied in the subsequent links. If the index is low, it means that the gap from the target demand is not too large, and a smaller adjustment can be made.

[0124] After obtaining the water and nitrogen adjustment index, it is necessary to first read the historical distribution data of the index in the same area and combine the previously obtained soil status results to retrieve whether there is a situation that is significantly higher than the effective range statistically determined in advance. For example, it can be set that the index is within the range of 0 to 2.5 as a general deviation, 2.5 to 4 as an obvious deviation, and more than 4 as a large deviation. To obtain this range, it is necessary to collect the index at different seasons and growth stages in the same type of farmland multiple times and make horizontal comparisons. If the actually collected value has not appeared in the past monitoring or the data volume is very small, it can be marked as a relatively rare state and confirmed after discussion with planting experts. Then, according to which range the index is in, select an appropriate water and fertilizer supply amplitude to adjust the soil moisture and nitrogen content. The specific implementation process can be to first set up several optional gears to respectively match the index range. For example, when the index is less than 2.5, allocate the basic irrigation amount and the benchmark nitrogen application amount. When the index is between 2.5 and 4, increase the water and fertilizer input by a certain proportion, such as an increase between 30% and 50%, and arrange a detection after each application to record the soil changes. When the index is greater than 4, directly allocate a higher water and fertilizer supply amount and shorten the interval period between two irrigations or fertilizations. Correspond each irrigation and fertilization action with the index threshold one by one, so that there is reliable data basis for adjacent operations. After completing these steps, check again whether the soil moisture and nitrogen content have returned to a range closer to the standard water demand and nitrogen demand. If not, it is necessary to further increase the irrigation or fertilization amount and divide the operation time more finely. If it has returned to the reasonable range, record the corresponding specific timestamp, location coordinates, and the irrigation and fertilization amounts during this period to generate the final adjustment strategy.

[0125] The steps to obtain the optimized adjustment plan are as follows:

[0126] Based on the adjustment strategy, extract the water and nitrogen adjustment demand parameters, and at the same time call the real-time weather forecast data to extract the precipitation probability, temperature, and wind speed to generate an environmental impact factor set;

[0127] According to the environmental impact factor set, calculate the adjustment adaptability. The calculation formula is:

[0128]

[0129] Among them, F is the adjustment adaptability, A r is the irrigation demand in the adjustment strategy, P r is the precipitation expectation in the weather forecast, B t is the forecast temperature, C h is the environmental humidity, D w is the wind speed, E s is the soil evapotranspiration rate;

[0130] According to the adjustment adaptability, recalculate the irrigation amount, fertilization amount, and application time to form an optimized adjustment plan.

[0131] Specifically, after referring to the previously obtained adjustment strategy, it is necessary to disassemble and obtain the water-nitrogen adjustment demand parameters from this strategy first. These demand parameters often include the target humidity increment value and the target nitrogen content increment value. Subsequently, specific indicators such as precipitation probability, predicted temperature, and wind speed are extracted from the real-time forecast information of the meteorological service agency through the established meteorological data call process. Before this process, several data inspection conditions are usually set. For example, the precipitation probability is compared with the pre-set range of 0% to 100%, and abnormal items outside this range are excluded. The temperature is compared with the range of -30°C to 50°C, and the wind speed is compared with the range of 0 m / s to 30 m / s. After confirming that the values are compliant, the next step of statistics is carried out. In order to cover more content in the environmental impact factor set that affects irrigation and fertilization adjustment, historical meteorological data records are also combined to determine whether there are periods of sharp changes or extreme weather conditions. If the extracted precipitation probability or wind speed exceeds the previous record upper limit, an empirical threshold can be set here for comparison. For example, a precipitation probability higher than 80% is marked as a possible rainfall event, and a wind speed higher than 15 m / s is marked as a strong wind event, so as to accurately identify these sudden situations in subsequent operations. Then, information such as precipitation probability, temperature, and wind speed is associated with the water-nitrogen adjustment demand parameters obtained earlier, the positions of each value are listed item by item, and they are summarized into an environmental impact factor set. To ensure the same order, these data need to be spliced item by item according to the recording time sequence, and the current date, time period, and positioning information are attached to distinguish the environmental conditions of different plots and different times. Finally, an environmental impact factor set that matches the previous adjustment strategy is obtained.

[0132] The advantage of the formula is that it combines several factors such as irrigation demand, precipitation expectation, environmental humidity, temperature, wind speed, and soil evapotranspiration rate into a comprehensive measurement index inside the square root. It can not only reflect the difference between irrigation and precipitation, but also non-linearly quantify different intensities of meteorological factors in the environment through the cube of temperature, the square of wind speed, and the square root of evapotranspiration rate.

[0133] A r Parameter introduction:

[0134] A r is the irrigation demand in the adjustment strategy, and this value can be obtained through the water-nitrogen adjustment demand parameters obtained earlier or the calculation process of the previous link. Generally, it is estimated based on the actual water demand level of the crop, and it is necessary to repeatedly compare the soil humidity, the current growth stage of the crop, and the existing field test data. If the soil humidity is much lower than the standard value, the irrigation demand increases; otherwise, it slightly decreases. To determine A rIt is often necessary to collect environmental data and crop growth conditions over several days or weeks, summarize and statistically analyze them, select appropriate models such as linear regression or methods based on historical means to predict a reference irrigation amount, and then correct it by comparing the crop status and soil conditions on the same day to obtain a relatively accurate A r , and finally substitute it into this formula. When giving examples, the unit of A r can be set to millimeters or cubic meters per hectare and other measures. If the average soil humidity of a certain plot is lower than 30% within a week, and the crop is in the jointing stage, according to the comparison, A r can be set to about 35 millimeters.

[0135] P r Parameter introduction:

[0136] P r is the expected precipitation amount in the weather forecast, which needs to be obtained by connecting with the data of the local meteorological bureau or commercial meteorological API. Generally, it is expressed in millimeters. For example, it is forecasted that several millimeters of rainfall will occur in a day, and it can also be statistically analyzed in segments according to the cumulative situation of more time periods. To ensure the stability of the value, it is necessary to compare the latest forecast information with the forecast several hours ago or the previous day, exclude abnormal fluctuations in the statistics, and judge whether there is a deviation from the observed ground rainfall. When the deviation is too large for a long time, it is necessary to check the meteorological source or model. When obtaining P r , the rainfall measurement records of nearby automatic weather stations can also be referred to for appropriate revision. If the error between the precipitation forecast value and the actual value in the past 24 hours does not exceed two millimeters, the latest forecast value can be directly regarded as P r . For example, in a drought season area, there may be no rainfall or the rainfall is less than 1 millimeter on the same day, then P r is very small, while it will reach more than 20 millimeters during the precipitation concentration period.

[0137] B t Parameter introduction:

[0138] B t is the forecast temperature, generally referring to the daily maximum temperature or the temperature at key time periods provided by the weather forecast. According to the area where the cultivated land is located, the daytime temperature or the nighttime temperature may be selected for comparison. The temperature value can be directly read from the forecast products of the local meteorological agency, and generally ranges from -30°C to 50°C. If it is an outdoor farmland, the common peak temperature may fluctuate around 35°C. To uniformly incorporate the temperature into this formula, it is necessary to perform time comparison like other variables. If the temperature for the corresponding day has not arrived, the predicted temperature can be used as B in the formula t, if there is already measured temperature, it can be corrected again. If you want the temperature to occupy a larger proportion in the formula, you can amplify it by a certain multiple or perform other quantization conversions before input. Before setting, you should check the temperature fluctuation range in the local area in recent years and determine a reasonable value range. For example, the common predicted temperature in a certain area in summer is between 25°C and 40°C.

[0139] C h Parameter introduction:

[0140] C h Represents the environmental humidity, which mostly comes from the humidity detection of meteorological monitoring stations or portable measurement devices. The corresponding range usually varies from 0% to 100%. To obtain as accurate a value as possible, it is necessary to conduct multi-point comparisons during a relatively stable sampling time period. If the differences are large, further investigation can be carried out in subsequent steps. When obtaining C h , the ground dew point temperature or other humidity observation indicators can also be referred to as a calibration. The acquisition process can include comparing the morning and evening humidity at a fixed time every day, or continuous automatic sampling. Finally, the average humidity or predicted humidity value over a period of time is substituted into the formula. If the humidity in a certain area is relatively low all year round, it may be around 30%. In some areas, the humidity may be 40% at noon and rise to 80% or higher in the early morning and at night. Therefore, the selection of this value needs to correspond to the specific time period and be time-matched with other parameters. If the average humidity monitoring value of a certain plot during the daytime is 60%, it can be regarded as the current C h .

[0141] D w Parameter introduction:

[0142] D w is the wind speed, which is mostly measured in meters per second or kilometers per hour. In farmland management, it is generally obtained from automatic weather stations or handheld instruments. Each time a numerical value is collected, it is necessary to compare multiple measurements and select the average value or peak value. If it is necessary to evaluate the impact of wind on evaporation, the average wind speed within a day can be selected for calculation. If you are concerned about the impact of wind disasters or strong winds on the irrigation uniformity, the maximum value of gusts may be selected for reference. After obtaining it, it is often also necessary to check whether it is too high or too low compared with the historical average wind speed. If the deviation is too large, it will be marked and recorded separately at this stage for subsequent analysis. If the common wind speed in a certain area is around 3 meters per second, and the predicted wind speed on the monitored day reaches 10 meters per second, this 10 meters per second can be directly used for the value of D in the formula w .

[0143] E s Parameter introduction:

[0144] E sis the soil evapotranspiration rate, mainly used to characterize the evaporation consumption of water on the crop surface and the soil surface layer. It is generally obtained through a field evapotranspirometer or a professional estimation method, or can be extracted from the reference evapotranspiration published by the local meteorological station and corrected in combination with the actual crop type. To measure a more accurate evapotranspiration rate, parameters such as air temperature, wind speed, solar radiation, and crop leaf area index need to be monitored and time series analysis is performed. In agronomy or water conservancy research, it is often expressed as the water loss value per square kilometer, per hectare, or per cubic meter per unit time, and can also be converted into the form of millimeters per hour or millimeters per day. For example, if the daily average evapotranspiration of a certain plot is measured to be between 3 millimeters and 6 millimeters, and the sun shines extremely strongly and the wind speed is high on that day, then E s = 6 or a higher value can be input into the formula.

[0145] Calculation process:

[0146] Here, an example is selected: A r = 30 mm, P r = 10 mm, B t = 35 (°C), C h

[0147] = 60%, D w = 5 m / s, E s = 4 mm per day. First, substitute each item into the formula in turn:

[0148] Calculate the absolute difference |A r -P r |:

[0149] |30 - 10| = 20

[0150] Calculate the term related to the cube of temperature and environmental humidity

[0151]

[0152] C h +1 = 60 + 1 = 61,

[0153]

[0154] Calculate the square of the wind speed

[0155] 5 2 = 25

[0156] Calculate the square root of the soil evapotranspiration rate

[0157]

[0158] Add these values together:

[0159] 20 + 703.69 + 25 + 2 = 750.69

[0160] Take the square root of the result:

[0161]

[0162] The result shows that under the current environmental conditions, for parameters such as the given irrigation demand and precipitation expectation, the overall adjustment adaptability is approximately 27.39. If the system subsequently sets a corresponding threshold range for F, such as F < 15 indicating a high environmental matching degree, 15 ≤ F ≤ 30 indicating a certain degree of difference, and F > 30 indicating a significant deviation, then 27.39 falls within the middle interval, indicating that medium-intensity irrigation and fertilization optimization strategies may be required subsequently to cope with the impact of environmental variables.

[0163] After obtaining the previously calculated adjustment adaptability value, it is necessary to recompute the set irrigation amount and fertilization amount in combination with weather forecast information and the current growth cycle stage of the crop. For this purpose, the adjustment adaptability can be compared with historical similar data first. If it is found that a similar adaptability value once occurred in the same region when both the temperature and wind speed were high, then while referring to the irrigation and fertilization plan at that time, the current plan can be numerically revised. The specific approach is to first compare the daily irrigation amount with the previously recorded irrigation target amount and calculate the difference between the two. Then, check each item in the interval with a high correlation between the difference and the adaptability. For example, compare the records with an adaptability exceeding 20 with the previous one hundred historical records in the same range to find out how many millimeters more or how many kilograms more of fertilization amount were used for irrigation at that time. Then, confirm whether there is a deviation in the temperature and humidity under the current meteorological conditions compared with that period. If the temperature and evapotranspiration rate increase significantly, then proportionally increase the irrigation amount, such as increasing it by 10% to 20%, and slightly increase the fertilization amount to meet the nitrogen demand of the crop. At the same time, divide the application cycle more finely according to the monitoring timestamp to make the interval between each irrigation or fertilization operation shorter. If the wind speed is even greater than in the previous situation with a similar adaptability, then a relatively concentrated irrigation period can be set at the execution level to reduce evaporation losses. Arrange all these revised values in sequence with the expected execution time, form new entries, and correspond them one by one with the previously obtained environmental impact factors. Finally, summarize to obtain the optimized adjustment plan.

[0164] The steps to obtain the control signal output result are as follows:

[0165] Based on the optimized adjustment plan, call the irrigation amount, fertilization amount, application time, and frequency, and combine with the monitoring location coordinates to generate an irrigation and fertilization instruction parameter set;

[0166] According to the irrigation and fertilization instruction parameter set, parse the irrigation volume, fertilization volume, application time and frequency in each instruction one by one, match the operation interfaces of the irrigation machinery and fertilization machinery, convert the parameter format according to the interface protocol, and generate an execution instruction set;

[0167] According to the execution instruction set, send control signals to the operation terminals of the irrigation machinery and fertilization machinery, monitor the mechanical reception status and response in real time, record the execution feedback of each instruction, and generate the control signal output result.

[0168] Specifically, based on the obtained optimization and adjustment plan, it is necessary to first disassemble the key contents such as irrigation volume, fertilization volume, application time and frequency from it, and check whether these values are consistent with the previously obtained water and nitrogen adjustment requirements. If it is found that some entries in the irrigation volume exceed the specified value, it is necessary to clarify how the specified value is set. For example, it can be set by referring to the maximum allowable irrigation volume of the crop at the corresponding stage and combining the upper limit of soil moisture that can be tolerated, and set it to a range not exceeding 60 millimeters per day or 10 millimeters per hour. Then, verify each entry that exceeds this range one by one. If it is confirmed to be correct, it will be retained; otherwise, it will be revised with the previous data. Then, calculate the fertilization volume based on the previously obtained nitrogen demand and check whether it conforms to the pre-set fertilization safety threshold. This threshold can be obtained through empirical statistics. For example, in multiple comparative tests of the same variety of crops, it is observed that when the fertilization volume on a certain plot on the same day is greater than 20 kilograms per hectare, obvious leaf burns will appear, and this is summarized as an empirical upper limit. If the current plan exceeds this upper limit, a re-verification will be requested. After completing the debugging of the irrigation volume and fertilization volume entries, it is necessary to organize the application time and frequency into a unified time period table to ensure that there are no conflicts in the sequence of multiple irrigation and fertilization times within the same plot, and also to avoid overlapping time periods with the needs of surrounding plots to prevent unreasonable equipment allocation. When associating the monitoring position coordinates with each instruction, the coordinate grid of the plot can be queried first to determine the specific location that the irrigation machine or fertilization machine should go to, and the coordinates and operation content are numbered and stored in the same recording method to generate a collection of necessary information for each instruction. Once all the irrigation volume, fertilization volume, time and frequency are matched with the coordinates and passed the threshold verification, an irrigation and fertilization instruction parameter set can be formed.

[0169] According to the irrigation and fertilization instruction parameter set obtained previously, each instruction in it needs to be parsed in detail. The irrigation volume, fertilization volume, application time, and frequency contained in the instruction should be split out in a fixed format and compared one by one with the mechanical equipment interface protocol confirmed previously. For example, the required flow unit for irrigation machinery can be liters per minute or cubic meters per hour. If it is measured in millimeters of rainfall or milliliters in the instruction parameters, it must be converted to the corresponding unit for matching. Similarly, the fertilization machinery may require the fertilization volume to be expressed in kilograms, so the numbers in the parameter set need to be re-converted. If there are differences in the time period format of the application time, it should be unified into the form of hours and minutes. When comparing the frequency with the maximum operating frequency of the machinery, the safety threshold also needs to be considered. For example, it is limited that irrigation can be performed at most 3 times continuously or fertilization can be performed 2 times a day within one day. If it is detected that a certain instruction exceeds this range, it is marked as abnormal and returned for revision. When all instructions meet the requirements, the content is made into executable instruction entries, and the machinery equipment model and the corresponding operation interface identifier are attached. By referring to the pre-established interface protocol format, the text or digital data is arranged in order and converted into operation codes recognizable by the equipment item by item. For example, the irrigation volume is written into the first byte segment of the operation code, the fertilization volume is written into the second byte segment, the application time is converted into the hour-minute value and stored in the third byte segment, and the application frequency is placed in the fourth byte segment. After all are completed, they are summarized in sequence and an execution instruction set is generated.

[0170] According to the generated execution instruction set, the operation codes of each instruction entry need to be sent to the corresponding irrigation machinery or fertilization machinery terminal for actual actions. First, control signals are sent to the machinery in sequence according to the order of the entries in the instruction set, and the reception confirmation or error code returned by the machinery terminal is monitored. During this process, situations exceeding the set response time threshold need to be marked. This response time threshold can be obtained through on-site testing or equipment manufacturer parameter setting methods. For example, in the test, if the response duration of the same model of equipment in different network environments is within two seconds for ten consecutive times, the threshold can be set to three seconds. If timeouts occur frequently, it is judged that the equipment may have network congestion or mechanical failures, and it is listed as an abnormal event for subsequent investigation. For each instruction that has been successfully sent and confirmed, the execution period, irrigation and fertilization parameters, and the equipment return code also need to be summarized. If the return code records information such as partial completion or machinery being busy, these prompts need to be collected for subsequent scheduling. The immediate monitoring results during the entire sending process are centrally sorted into the execution feedback data corresponding to each instruction, and finally, they are arranged in sequence according to the sending order and confirmation order to form the control signal output result.

[0171] The steps for obtaining the execution feedback are as follows:

[0172] Based on the control signal output result, the operation logs of each machinery terminal are collected in real-time, and the execution status, operation instructions, equipment feedback, and error code records are sorted in chronological order to generate an execution monitoring data set;

[0173] Analyze the execution status parameters one by one according to the execution monitoring data set. By comparing the instruction requirements with the mechanical feedback results, analyze the equipment start-up delay, execution duration deviation, and abnormal interruption one by one, and screen the alarm events and terminal response situations during the equipment operation to generate an execution result report;

[0174] Extract the execution success rate, abnormal frequency, and fault type of each mechanical terminal according to the execution result report. Combine the monitoring timestamp, operation log, and equipment feedback code, and classify and organize them according to equipment, time, and operation category to generate an execution feedback.

[0175] Specifically, based on the previously obtained control signal output results, it is necessary to collect the operation logs of each mechanical terminal in chronological order and read the timestamp, instruction number, execution status, etc. information item by item. At the same time, disassemble the meaning corresponding to the equipment feedback code and error code according to the technical description provided by the equipment manufacturer. Arrange the collected log entries in chronological order and make a one-to-one match with the operation instructions in the control signal output. If it is found that the timestamp in some log entries is too different from the instruction issuance time, mark the start-up delay situation, and determine a delay threshold according to the empirical method. For example, in multiple tests, it is statistically found that the start-up delay of the same model of machinery is usually distributed between 1 second and 2 seconds. Considering network stability and mechanical response time, an additional 0.5 seconds can be added as redundancy, and 2.5 seconds is set as the delay threshold. Compare item by item and if the start-up delay of a certain instruction exceeds 2.5 seconds, highlight it with a separate mark. Then, combine the error code records in the equipment feedback item and refer to the fault quick check list to analyze the fault types. For example, map the error code 1001 to abnormal pump body current, and the error code 1002 to valve blockage, etc. Then, sort and record the normal or abnormal identifiers in the execution status field in combination with the timestamp. For log entries with alarm events or extremely short interruptions, additional summaries are required. If the alarm number appears on the previously established high-risk number list, continue to track and check the terminal response situation. Associate all instructions, equipment feedback, and error codes that meet the chronological order and make lists according to the machine type and plot. Finally, integrate these associated data to generate an execution monitoring data set.

[0176] After obtaining the execution monitoring dataset, a more detailed analysis needs to be done on the execution status of each instruction. First, read values such as the startup delay, execution duration, and end time in the execution status parameters, and then extract the corresponding expected duration and expected status from the instruction requirements. If the actual execution duration of a certain instruction is significantly higher than the expected duration, it is marked as a timeout situation. An overtime threshold can be set here to define whether it is abnormal. For example, referring to multiple tests, it is found that the duration of the same type of irrigation machinery when executing regular startup and stop instructions is mostly in the range of 1 minute to 2 minutes. So, 2.5 minutes is set as the overtime threshold for comparison. Then, combine the recorded error codes to identify possible interruption and alarm events. If the alarm number indicates that the hardware has an overheating temperature, special fields are added in the subsequent screening to record these warning messages. The text descriptions of the terminal responses also need to be classified and counted according to the operation categories. For example, the execution situations of irrigation instructions and fertilization instructions are recorded separately. Synchronously check whether there are duplicate instructions or overlapping operations for each device during this period. If detected, mark these overlapping operations and compare their delays and durations. After summarization, list the actual start and end times of each instruction, the startup delay values feedback by the device, and whether there are any abnormalities or alarms in the table. After sorting item by item, summarize the overall analysis results, and then store these results in the corresponding entries to form an execution result report.

[0177] Based on the previously obtained execution result report, it is necessary to calculate the execution success rate and abnormal frequency of each mechanical terminal and extract and classify the fault types. To complete this step, first sort the records in the execution result report by mechanical number and operation category. Use the timestamp as the vertical reference axis and compare the total number of instructions received by the device with the number of instructions successfully completed row by row. Divide the two to get the execution success rate. If 2 out of 10 instructions of a certain device have timeouts or abnormalities during a certain period, record a success rate data of 80%. Then, use the same method to count the number of times the error code appears and divide it by the total number of instructions to get the abnormal frequency. Combine the error code comparison table disassembled from the manufacturer's technical data previously to locate the fault type. For example, regard the error code 2001 as a pipeline blockage fault and the error code 2002 as a communication interruption fault. Then, classify and organize these fault types and the occurrence time periods together by device and time. If the same fault code appears 5 times and the proportion exceeds 20% for a certain mechanical device during a certain period, it can be marked as a high failure rate. Map the classified success rate, abnormal frequency, and fault type information together with the previous timestamp and operation log, so as to complete the induction of the execution situations of multiple devices and generate an execution feedback.

[0178] The steps to obtain the effect evaluation results are as follows:

[0179] According to the execution feedback, calculate the irrigation and fertilization effect scores. The calculation formula is:

[0180]

[0181] Among them, E is the irrigation and fertilization effect score, and P a is the irrigation volume, Q a is the standard irrigation volume, R d is the fertilization execution deviation, S d is the standard fertilization deviation, T m is the operation time deviation, U f is the equipment failure frequency, V t is the number of successful executions, W l is the number of operation interruptions;

[0182] According to the irrigation and fertilization effect score, evaluate the irrigation and fertilization effects to generate the effect evaluation result.

[0183] Specifically, the advantage of the formula is that by simultaneously incorporating the irrigation volume, standard irrigation volume, fertilization deviation, operation time deviation, equipment failure frequency, number of successful executions, and number of operation interruptions into the same comprehensive expression, it can more intuitively present the impacts of multiple important factors during the irrigation and fertilization processes and evaluate the execution deviation and equipment operation status in one calculation.

[0184] P a Parameter introduction:

[0185] P a is the irrigation volume, which is actually used to measure the water input value obtained by the field during one operation. The unit can be measured in millimeters or cubic meters per hectare, etc. If a more precise grasp of the humidity distribution is required, the water volume can also be split and recorded by block. To obtain P a it is necessary to combine the flowmeter or supporting counter of the irrigation machinery. Record the initial reading of the machinery before each irrigation starts, and then read the final measurement again after completion. The difference between the two is the irrigation volume for this period. If it is superimposed irrigation in multiple periods, the total irrigation volume for a day or a week can also be accumulated and summed up. Through multiple tests, the accuracy error of the machinery under different load conditions can be known, and only after correcting this error can a more accurate P a value be obtained. For example, through multiple consecutive irrigation measurements of the same machinery, if the reading error each time is within 2%, then take 2% as the correction reference and assign it to the final P a value.

[0186] Q a Parameter introduction:

[0187] Q a is the standard irrigation volume, which is usually formulated based on the water demand laws of local crops and the results of field tests, and can also be dynamically adjusted in combination with soil moisture content detection. If the crop growth stage is in the peak water demand period, Q aIt will relatively increase and be appropriately reduced if it enters the late growth slowdown stage. When determining the specific value, the yields, plant growth performances, and soil water conduction characteristics of multiple fields will be investigated to extract one or more reference intervals, and then the median or average value will be set within these intervals. For example, by statistically calculating the daily irrigation requirements of multiple demonstration fields during the jointing stage of cereals and taking the average value, a standard amount of several millimeters per day can be obtained, and this value is taken as Q a Use.

[0188] R d Parameter Introduction:

[0189] R d It represents the fertilization execution deviation, which needs to be obtained by comparing the actual fertilization amount with the expected fertilization amount in the instruction. A common way to obtain this deviation is to read the metering device on the fertilization machinery after the operation. If this measured value differs significantly from the value required by the instruction, it indicates an obvious deviation, which may be caused by reasons such as blockage, uneven fertilizer spreading, or mechanical damage. When setting this parameter, it is also necessary to observe whether the fertilizer application speed and duration match the instruction, and assist the judgment through the additionally recorded crop fertilizer absorption situation. If deviations occur repeatedly, they need to be included in the subsequent analysis as clues for troubleshooting. For example, if the cumulative expected total fertilization amount is 150 kg in three fertilizations, and the actual total amount only reaches 135 kg, it means the deviation accumulation is 15 kg.

[0190] S d Parameter Introduction:

[0191] S d It is the standard fertilization deviation, which is used to refer to the tolerance range between the actual fertilizer amount spread by the machinery and the instructed amount under normal working conditions. To obtain this tolerance range, multiple calibration tests need to be carried out on the machinery. For example, the machine should spread 5 kg of chemical fertilizer per unit time, and most of the results obtained through repeated measurements are distributed between 4.8 kg and 5.1 kg. Then, this interval of 0.2 kg to 0.3 kg can be used as the reference deviation value, and based on this, combined with the actual situation of the field and the fertilizer fluidity error, an average or maximum deviation value is determined as S d .

[0192] T m Parameter Introduction:

[0193] T mIt is the operation time deviation, which measures the difference between the actual execution duration and the expected duration. To obtain this value, it is necessary to first determine the expected operation time set by the instruction. For example, if the instruction requires the irrigation machine to run for 30 minutes, and the actual execution is recorded as 35 minutes after completion, then the deviation can be considered as 5 minutes. Here, the additional time consumption that may occur during the machine startup warm-up and stop buffer links also needs to be considered. Therefore, the judgment of the time deviation also needs to be combined with the machine instruction manual or the average startup and stop durations statistically obtained previously, and can only be determined as abnormal after comparison. If the average deviation within multiple operations in a week is within 2 minutes, then 2 minutes can be used as the monitoring threshold. If the deviation reaches 7 minutes after a certain execution, it can be concerned in subsequent dimensions and recorded in T m 。

[0194] U f Parameter introduction:

[0195] U f It is the equipment failure frequency, which is generally obtained by counting the number of times the machine fails within a certain period. Before obtaining this number, it is necessary to clarify the definition range of failures. For example, motor non-response, sensor offline, valve blockage, etc. are all counted as failures. If the machine reports an error three times during a startup process, it is counted as three failures and continues to accumulate in other operations on the same day. When a week or a month ends, the cumulative value of the machine's failure events at each time period can be obtained as the original input of U f If a device has 4 valve blockages or 2 motor overloads within seven days, then (4 + 2) = 6 can be recorded as the failure frequency.

[0196] V t Parameter introduction:

[0197] V t It is the number of successful executions, which refers to the total number of instructions that the machine has normally completed during a period of time or multiple instruction executions. To obtain this value, it is necessary to first count all the issued instruction entries together, and then check the final status of each operation one by one. If the machine completes the task within the specified time and does not experience mid-operation shutdown or error reporting, it is determined as a success. Finally, these success records are summarized. If a total of 50 instructions are issued within five days, and 45 of them are completed by the machine in a normal state, then it can be recorded as V t =45。

[0198] W l Parameter introduction:

[0199] W l It is the number of operation interruptions, which represents the frequency of abnormal stops or forced interruptions of the machine during the execution of instructions. It may be caused by machine body failures, emergency stop signals or human intervention, or may also be caused by accidental stops due to lost network instructions during operation. To obtain W lIt is necessary to systematically summarize the mechanical operation logs. Each time an instruction ends abnormally, add 1 to the record, and the accumulated number within a certain period is the result.

[0200] Calculation process:

[0201] Here is an example: Let P a = 25 mm, Q a = 20 mm, R d = 5 kg, S d = 2 kg, T m = 6 minutes, U f = 4 times, V t = 16 times, W l = 2. The calculation steps are as follows:

[0202] First, calculate

[0203]

[0204] (1.19) 2 ≈ 1.4161.

[0205] Calculate |R d - S d |:

[0206] |5 - 2| = 3.

[0207] Calculate

[0208]

[0209] Calculate

[0210] V t + 1 = 16 + 1 = 17,

[0211]

[0212] Calculate

[0213] (2) 2 = 4.

[0214] Add up the above results:

[0215] 1.4161 + 3 + 2.449 + 0.2353 + 4 ≈ 11.1004.

[0216] Finally, take the square root:

[0217]

[0218] The results show that after accumulating factors such as the deviation of current irrigation and fertilization operations, an E value of approximately 3.33 is generated. If several intervals are set for this score in advance, for example, when E ≤ 2, the irrigation and fertilization effects are considered good; when 2 < E ≤ 4, it is considered to be in a medium state; when E > 4, it is regarded as a significant deviation. Then 3.33 falls between 2 and 4, which can be considered to correspond to a medium level. It is necessary to continue to conduct a detailed analysis in combination with the operation scenario and make corresponding adjustments in the next cycle.

[0219] According to the irrigation and fertilization effect scores obtained previously, it is necessary to compare their values in combination with the current growth stage of the corresponding crop and the records of soil moisture changes. To achieve this process, the collected score information can be disassembled first. Read each sub-item such as the recorded irrigation volume, fertilization deviation, equipment failure frequency, and operation time deviation one by one, and list these sub-items on a time axis. Compare them with the previously statistically determined soil water content range of 0% to 70% and the target nitrogen content range of 0 to 1000 mg / kg. After the comparison, if it is found that the score is too high, it can be judged whether the failure is the main influencing factor according to the previously collected equipment failure frequency, or the problems of water shortage or over-irrigation can be determined according to the cumulative situation of fertilization deviation and irrigation volume deviation. Then mark the date and time stamp corresponding to the score on the crop growth progress table. At regular time intervals, such as every three days or one week, make a vertical comparison of the recent score situation with the data of the previous cycle to check whether the score fluctuates slightly or has a significant increase. Through this comparison, if it is confirmed that the score shows a continuous upward trend and exceeds the pre-set threshold, such as exceeding 4, then a special review can be conducted on the latest several irrigation or fertilization operations, check the operation logs and execution feedback of each machine. If there are multiple deviation alarms, correct the deviation centrally and continue to monitor the score curve in the subsequent stage. At the same time, conduct grouped statistics on the recorded score data, and make a mapping table of the corresponding irrigation and fertilization schemes and score values. For example, list the schemes with scores between 2 and 3 as the relatively normal range, list the schemes with scores higher than 4 as the range that needs to be reviewed key points, list the schemes with scores lower than 2 as the stable range, combine and summarize these mapping tables with the corresponding soil moisture and nitrogen conditions to obtain a set of complete data items for subsequent evaluation. Finally, at the end of this cycle, extract all the score entries that meet the threshold screening conditions and summarize the effect evaluation results.

Claims

1. Wheat-corn annual water and nitrogen content regulation system, characterized in that: The system comprises: The sensing data module collects soil moisture and nitrogen content data during the growth period of wheat and corn, filters them, and generates processed soil data; based on the processed soil data, it compares with the set standard threshold to obtain the soil status result; A decision logic module receives the soil state result, calculates an adjustment plan according to the water and nitrogen requirements of wheat and corn, and generates an adjustment strategy; and obtains an optimized adjustment plan with reference to the real-time weather forecast data for the adjustment strategy; The execution control module adopts the optimization adjustment scheme to set instructions for irrigation and fertilization machinery and generate control signal output results; based on the control signal output results, the execution status is monitored, and the execution results are recorded to obtain execution feedback; The synchronous feedback module analyzes the execution feedback, determines the irrigation and fertilization effects, and generates effect evaluation results.

2. The wheat-corn annual water and nitrogen content regulation system according to claim 1, characterized in that: The steps for obtaining the processed soil data are as follows: Soil moisture and nitrogen content are continuously monitored through soil sensors during the growth cycle of wheat and corn, and data is collected and recorded in real time to obtain the original soil data set; Based on the original soil data set, noise filtering and outlier detection are performed to obtain processed soil data.

3. The wheat-corn annual water and nitrogen content regulation system according to claim 1, characterized in that: The steps for obtaining the soil status results are: Based on the processed soil data, soil moisture and nitrogen content values ​​in the data are extracted, and each data point is classified, and the timestamp information of each data point is compared to obtain time series soil data; According to the time series soil data, the soil state score is calculated using the following formula: Among them, Q is the soil status score, H j is the soil moisture at the jth time point, T H is the standard humidity threshold, N k is the soil nitrogen content at the kth time point, T N is the standard nitrogen content threshold, m is the total number of data points; According to the soil state score, the set standard threshold is compared to determine the matching of soil moisture and nitrogen content, and the soil state result is obtained.

4. The wheat-corn annual water and nitrogen content regulation system according to claim 1, characterized in that: The steps for obtaining the adjustment strategy are: According to the soil state results, the water and nitrogen demand adjustment index is calculated using the following formula: Among them, I is the water and nitrogen demand adjustment index, X h is the current soil moisture, Y h is the standard water requirement, X n is the current soil nitrogen content, Y n is the standard nitrogen requirement, Z t To monitor the timestamp difference, L m To monitor the position coordinate difference, M s Match the difference for soil texture; The adjustment range of water and fertilizer supply is selected according to the water and nitrogen demand adjustment index to generate an adjustment strategy.

5. The wheat-corn annual water and nitrogen content regulation system according to claim 1, characterized in that: The steps for obtaining the optimization adjustment plan are: Based on the adjustment strategy, water and nitrogen adjustment demand parameters are extracted, and real-time weather forecast data is called to extract precipitation probability, temperature and wind speed to generate a set of environmental impact factors; According to the environmental impact factor set, the adjustment adaptability is calculated using the following formula: Among them, F is the adjustment adaptability, A r To adjust the irrigation demand in the strategy, P r is the expected amount of precipitation in the weather forecast, B t is the predicted temperature, C h is the ambient humidity, D w is the wind speed, E s is the soil evapotranspiration rate; According to the adjustment adaptability, the irrigation amount, fertilizer amount and application time are recalculated to form an optimized adjustment plan.

6. The wheat-corn annual water and nitrogen content regulation system according to claim 1, characterized in that: The steps of obtaining the control signal output result are: Based on the optimization adjustment scheme, the irrigation amount, fertilization amount, application time and frequency are called, and combined with the monitoring position coordinates, an irrigation and fertilization instruction parameter set is generated; According to the irrigation and fertilization instruction parameter set, the irrigation amount, fertilization amount, application time and frequency in each instruction are parsed one by one, the operation interface of the irrigation machinery and the fertilization machinery is matched, the parameter format is converted according to the interface protocol, and an execution instruction set is generated; According to the execution instruction set, control signals are sent to the operating terminals of the irrigation machinery and the fertilization machinery, the receiving status and response of the machinery are monitored in real time, the execution feedback of each instruction is recorded, and the control signal output result is generated.

7. The wheat-corn annual water and nitrogen content regulation system according to claim 1, characterized in that: The steps for obtaining the execution feedback are: Based on the control signal output result, the operation log of each mechanical terminal is collected in real time, and the execution status, operation instructions, equipment feedback and error code records are sorted in chronological order to generate an execution monitoring data set; According to the execution monitoring data set, the execution status parameters are analyzed one by one, and by comparing the instruction requirements with the mechanical feedback results, the equipment startup delay, execution time deviation and abnormal interruption are analyzed one by one, and the alarm events and terminal response conditions in the equipment operation are screened to generate an execution result report; According to the execution result report, the execution success rate, abnormal frequency and fault type of each mechanical terminal are extracted, combined with the monitoring timestamp, operation log and equipment feedback code, and classified by equipment, time and operation category to generate execution feedback.

8. The wheat-corn annual water and nitrogen content regulation system according to claim 1, characterized in that: The steps for obtaining the effect evaluation results are as follows: According to the execution feedback, the irrigation and fertilization effect scores are calculated, and the calculation formula is: Among them, E is the irrigation and fertilization effect score, P a is the irrigation amount, Q a is the standard irrigation amount, R d is the fertilization execution deviation, S d is the standard fertilization deviation, T m is the operation time deviation, U f is the equipment failure frequency, V t is the number of successful executions, W l is the number of operation interruptions; According to the irrigation and fertilization effect scores, the irrigation and fertilization effects are evaluated to generate effect evaluation results.

Images