Regulation method for deep loosening tillage of straw returning based on soil temperature and humidity

Abstract

The present invention discloses a regulation method for straw returning and subsoiling tillage based on soil temperature and humidity, which specifically relates to the technical field of tillage regulation; by arranging a plurality of soil temperature and humidity sensors and meteorological sensors in the field, real-time collection of soil temperature and humidity and meteorological environment data is carried out, and these data are preprocessed and feature extracted, and then a data prediction model is constructed to evaluate the accuracy of the monitoring data. If the accuracy of the monitoring data is high, the system will directly adjust the operation time and depth; if the accuracy is low, by analyzing the severity of the deviation, the data collection frequency is automatically increased, so as to optimize the monitoring accuracy and tillage decision-making, effectively avoid tillage mistakes caused by misjudgment of soil humidity, improve the tillage effect and crop yield, and at the same time improve the soil quality.

Description

Technical Field

[0001] The present invention relates to the technical field of tillage regulation, and particularly relates to a method for regulating straw returning and subsoiling tillage based on soil temperature and humidity. Background Art

[0002] Straw returning to the field and subsoiling tillage are important technical means to improve agricultural soil quality and crop yields. Straw returning to the field provides organic fertilizers for the soil by returning crop straws to the soil, improves soil structure, increases the organic matter content, and enhances the soil's water retention capacity. Subsoiling tillage, on the other hand, loosens the soil by mechanical means, improves soil aeration and drainage capacity, helps the deep growth of roots, enhances crop stress resistance, and promotes yield increase.

[0003] In traditional agriculture, the timing and methods of straw returning to the field and subsoiling tillage often rely on farmers' experience. However, changes in factors such as climate change, soil humidity, and temperature make it particularly complex to accurately grasp the optimal tillage timing. Therefore, a regulation method based on soil temperature and humidity has emerged, which relies on real-time monitoring of soil temperature and humidity changes to dynamically adjust tillage operations to achieve the best results. By real-time monitoring of soil temperature and humidity, the system can perform subsoiling tillage and straw returning to the field at appropriate times, avoiding operations when the soil is too wet or too dry, thereby improving tillage effects and reducing resource waste. For example, when the soil humidity is too high, early subsoiling tillage may exacerbate soil compaction, and straw returning to the field in dry conditions may not yield ideal results. Therefore, adjusting the operation time and depth in combination with soil temperature and humidity data is an important technical path to improve agricultural production efficiency and resource utilization rate.

[0004] The existing technologies have the following deficiencies:

[0005] The key to accurately monitoring soil temperature and humidity lies in high-precision sensors and monitoring equipment, but these devices may have errors, malfunctions, or improper calibrations. Sensors may drift due to long-term use, resulting in inaccurate data, which affects tillage decisions. In addition, if the soil humidity sensor reports that the soil is too dry or too wet, which does not match the actual situation, it may lead to incorrect tillage timing selection. For example, if the humidity sensor malfunctions, the system may wrongly think that the soil is too wet, thereby delaying subsoiling tillage, causing soil compaction, and affecting the growth of crop roots. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for regulating straw returning and subsoiling tillage based on soil temperature and humidity to solve the deficiencies in the background art.

[0007] To achieve the above purpose, the present invention provides the following technical solution: A method for regulating straw returning and subsoiling tillage based on soil temperature and humidity, comprising the following steps:

[0008] S1: Deploy several soil temperature and humidity sensors in different areas of the field to collect soil temperature and humidity data in real time for multiple time periods, and deploy different types of meteorological sensors to collect soil meteorological environment data in real time;

[0009] S2: Preprocess the collected soil temperature and humidity data and soil meteorological environment data, and extract features from the preprocessed soil temperature and humidity data and soil meteorological environment data, respectively extracting soil temperature and humidity mutation features and soil meteorological prediction deviation features;

[0010] S3: Construct a data prediction model based on the extracted soil temperature and humidity mutation features and soil meteorological prediction deviation features, and determine the accuracy of soil temperature and humidity monitoring data according to the output result of the data prediction model;

[0011] S4: If the accuracy of the soil temperature and humidity monitoring data is high, directly input it into the operation decision system to dynamically adjust the operation time and depth of straw returning to the field and deep loosening tillage. If the accuracy of the soil temperature and humidity monitoring data is low, judge the severity of the deviation of the soil temperature and humidity monitoring data. If the severity is high, increase the data collection frequency of the soil temperature and humidity sensors.

[0012] Preferably, in S2, after analyzing the variation trend characteristics of the extracted soil temperature and humidity over time, a soil temperature and humidity anomaly index is generated. The method for obtaining the soil temperature and humidity anomaly index is as follows:

[0013] Define the neighborhood for each soil temperature and humidity data point, select a parameter K to represent the neighborhood size of each data point. For a data point p, calculate the distance d(p,q) from point q, and the expression is: where p = (p1, p2, …, p n ) and q = (q1, q2, …, q n ) are the feature vectors of the soil temperature and humidity data points. For data point p and point q in the neighborhood, define the reachable distance r k (p,q) as: r k (p,q) = max(d(p,q), k dist (q)); where k dist (q) is the distance from point q to its K-neighborhood, that is, the distance from point q to its K-th nearest neighbor. The local reachable density LRD k (p) represents the density situation of data point p in its neighborhood, and the calculation formula is: where N k (p) are the K neighbors of data point p, r k(p, q) is the reachable distance between the data point p and its neighboring point q, and the local outlier factor is calculated, that is, the soil temperature and humidity anomaly index is calculated. The calculation formula is: In the formula, QSD is the soil temperature and humidity anomaly index.

[0014] Preferably, in S2, after analyzing the deviation characteristics between the extracted real-time meteorological data and the meteorological prediction data, a meteorological data deviation index is generated. The method for obtaining the meteorological data deviation index is as follows: Let x t represent the real-time meteorological data at time t, and represent the meteorological prediction data at the same time t. The difference between them is the error: where, e t is the absolute error at the t-th moment. The error e t at each moment is multiplied by a weight w t , and the weighted error is: For all the collected meteorological data points, the meteorological data deviation index is calculated. The expression is: where, N is the total number of data points, and WMAE is the meteorological data deviation index.

[0015] Preferably, in S3, according to the extracted soil temperature and humidity mutation characteristics and the soil meteorological prediction deviation characteristics, a data prediction model is constructed, and the accuracy of the soil temperature and humidity monitoring data is determined according to the output result of the data prediction model. Specifically:

[0016] The soil temperature and humidity anomaly index and the meteorological data deviation index are normalized, and the accuracy value of the soil temperature and humidity monitoring data is calculated through the normalized soil temperature and humidity anomaly index and the meteorological data deviation index.

[0017] Preferably, the accuracy value of the obtained soil temperature and humidity monitoring data is compared with the preset accuracy value reference threshold according to the historical data. If the accuracy value of the soil temperature and humidity monitoring data is greater than or equal to the preset accuracy value reference threshold, it indicates that the accuracy of the soil temperature and humidity monitoring data is high. At this time, a data normal signal is generated, and the soil temperature and humidity monitoring data is classified as accuracy monitoring data; if the accuracy value of the soil temperature and humidity monitoring data is less than the preset accuracy value reference threshold, it indicates that the accuracy of the soil temperature and humidity monitoring data is low. At this time, a data anomaly signal is generated, and the soil temperature and humidity monitoring data is classified as inaccurate monitoring data.

[0018] Preferably, in S4, if the accuracy of the soil temperature and humidity monitoring data is high, it is directly input into the operation decision system to dynamically adjust the operation time and depth of straw returning to the field and deep loosening tillage. Specifically:

[0019] Based on the real-time collected soil temperature and humidity data and meteorological prediction data, adjust the operation time T adjust and the operation depth D adjust dynamically. The expression is: T adjust = T base + k1·ΔT soil + k2·ΔH soil + k3·ΔQ; where T adjust is the adjusted operation time; T base is the preset operation time; ΔT soil is the difference between the current soil temperature and the preset temperature; ΔH soil is the difference between the current soil humidity and the preset humidity; ΔQ is the difference between the accuracy value of the soil temperature and humidity monitoring data and the accuracy threshold, which affects the fine-tuning of the operation time; k1, k2, k3 are adjustment coefficients;

[0020] The operation depth is adjusted according to the changes in soil humidity and soil temperature. The adjustment formula is: D adjust = D base + k4·ΔT soil + k5·ΔH soil + k6·ΔQ; where D adjust is the adjusted operation depth, D base is the preset operation depth; ΔT soil and ΔH soil are the differences between the soil temperature and humidity and the preset values; k4, k5, k6 are adjustment coefficients; Input the adjusted operation time and operation depth into the operation decision system and make dynamic adjustments according to the actual farmland conditions and tillage machinery.

[0021] Preferably, if the accuracy of the soil temperature and humidity monitoring data is low, that is, the accuracy value of the soil temperature and humidity monitoring data is less than the preset accuracy value reference threshold, judge the severity of the deviation of the monitoring data. The expression is: S severity = |Q - Q threshold | + A·ΔT soil + B·ΔH soil ; where S severity is the severity of the deviation, Q is the accuracy value of the current soil temperature and humidity monitoring data; Q threshold is the preset accuracy value reference threshold; ΔT soil is the difference between the soil temperature and the preset temperature; ΔH soil is the difference between the soil humidity and the preset humidity; A and B are adjustment coefficients;

[0022] According to the calculated severity of the deviation S severity , if its value exceeds the reference threshold S severity, it is considered that the deviation is serious and the data acquisition frequency of the soil temperature and humidity sensor needs to be increased. The expression is: ifS severity <S threshold ifS severity ≥S threshold ; f adjust is the adjusted data acquisition frequency, f base is the preset basic data acquisition frequency, γ is the adjustment coefficient, which is used to control the influence of the deviation severity on the acquisition frequency; S threshold is the reference threshold of the deviation severity.

[0023] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0024] 1. By arranging multiple soil temperature and humidity sensors and meteorological sensors in the field, the present invention can collect soil temperature and humidity and meteorological data in real time, and generate soil temperature and humidity anomaly indexes and meteorological data deviation indexes by using data processing and feature extraction methods. On this basis, the present invention evaluates the accuracy of soil temperature and humidity monitoring data through a data prediction model, and dynamically adjusts the tillage timing and depth according to the accuracy value, ensuring that the operation is adapted to soil and meteorological conditions and improving the accuracy of tillage decision-making.

[0025] 2. By analyzing the severity of the deviation of the monitoring data, when the accuracy of the monitoring data is low, the present invention automatically increases the data acquisition frequency of the soil temperature and humidity sensor, thereby improving the monitoring accuracy. This dynamic regulation mechanism based on real-time data feedback effectively avoids tillage mistakes caused by misjudgment of soil temperature and humidity, optimizes tillage operations, improves the growth environment and yield of crops, and at the same time improves soil quality, providing more intelligent and reliable technical support for precision agriculture. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0027] Figure 1 is the method flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0029] For the embodiments, please refer to Figure 1 As shown in the figure, the method for regulating straw returning to the field and deep loosening tillage based on soil temperature and humidity in this embodiment includes the following steps:

[0030] S1: Arrange a number of soil temperature and humidity sensors in different areas of the field to collect soil temperature and humidity data in real time for multiple time periods, and deploy different types of meteorological sensors to collect soil meteorological environment data in real time;

[0031] S2: Preprocess the collected soil temperature and humidity data and soil meteorological environment data, and extract features from the preprocessed soil temperature and humidity data and soil meteorological environment data, respectively extracting soil temperature and humidity mutation features and soil meteorological prediction deviation features;

[0032] S3: Build a data prediction model based on the extracted soil temperature and humidity mutation features and soil meteorological prediction deviation features, and determine the accuracy of soil temperature and humidity monitoring data according to the output results of the data prediction model;

[0033] S4: If the accuracy of the soil temperature and humidity monitoring data is high, directly input it into the operation decision system to dynamically adjust the operation time and depth of straw returning to the field and deep loosening tillage. If the accuracy of the soil temperature and humidity monitoring data is low, judge the severity of the deviation of the soil temperature and humidity monitoring data. If the severity is high, increase the data collection frequency of the soil temperature and humidity sensors.

[0034] Among them, in S1, arranging a number of soil temperature and humidity sensors in different areas of the field to collect soil temperature and humidity data in real time for multiple time periods, and deploying different types of meteorological sensors to collect soil meteorological environment data in real time, specifically:

[0035] Arrange multiple soil temperature and humidity sensors in different areas of the field to obtain more comprehensive and accurate soil temperature and humidity data. According to the area of the field, soil type, and crop variety, the number of sensors arranged may vary. Usually, dozens to hundreds of sensors are arranged to ensure the representativeness and coverage of the data.

[0036] Sensors need to be deployed at different locations and depths in the field. These locations should be representative and cover factors such as different soil types, crop varieties, growth stages, etc. Sensors can be deployed at different orientations in the field (such as southeast, northwest, etc.) to collect soil data under different environmental conditions.

[0037] Shallow sensors: Deployed at the soil surface layer (0 - 20 cm), mainly monitoring the temperature and humidity of the topsoil, which is important for the initial growth of crops and the evaluation of the effect of straw returning to the field.

[0038] Mid - layer sensors: Deployed at a soil depth of 20 - 40 cm, used to monitor the humidity and temperature of the mid - layer soil. This depth is crucial for the water supply to crop roots and the effect of subsoiling tillage.

[0039] Deep sensors: Deployed at a soil depth of 40 - 80 cm or deeper, mainly monitoring the humidity and temperature of the deep - layer soil. These data help analyze the distribution of water in the deep layer and judge the necessity and operation timing of subsoiling tillage.

[0040] Types of soil temperature and humidity sensors include: Capacitive sensors: Obtain soil humidity by measuring the dielectric constant of the soil, usually with high accuracy and stability, suitable for long - term monitoring. Resistive sensors: Calculate soil humidity by measuring resistance changes, usually with a lower price, but vulnerable to soil salinity and may not be suitable for some special soil types. TDR (Time Domain Reflectometry) sensors: Measure water content using the reflection characteristics of the soil, with very high accuracy and reliability, suitable for application scenarios with high - precision requirements. Thermal conductivity sensors: Measure the change in soil thermal conductivity to infer humidity and temperature, suitable for high - precision long - term monitoring.

[0041] Sensors need to have high - precision temperature and humidity measurement capabilities, capable of capturing details in small changes in soil temperature and humidity. Sensors should have strong anti - interference capabilities to resist the impact of external environmental changes (such as wind, rain, etc.) on data collection. Due to the complex soil environment and the need for sensors to be buried in the soil for a long time, they need to have strong durability and corrosion resistance. To facilitate the real - time transmission of field data, sensors usually need to have wireless communication functions such as LoRa, NB - IoT, Wi - Fi, etc., to ensure that data can be uploaded to the central processing system in a timely manner.

[0042] Meteorological sensors are used to monitor soil meteorological environment data in real time. These sensors collect meteorological data related to soil conditions and further assist in analyzing the moisture status and temperature changes of the soil. The purpose of deploying meteorological sensors is to obtain external meteorological factors that affect soil moisture and temperature, such as precipitation, temperature, humidity, wind speed, etc. Common types of meteorological sensors include: Temperature sensors: used to monitor air temperature in real time. Air temperature has a direct impact on soil temperature and humidity, so this data is crucial. Relative humidity sensors: used to monitor air humidity in real time. Air humidity has a significant impact on the soil evaporation rate, especially under high-temperature or drought conditions, where the water evaporation rate of the soil will accelerate. Precipitation sensors: used to monitor precipitation. The direct impact of precipitation on soil moisture is particularly important. By monitoring precipitation, it is possible to evaluate whether the soil is in a moist state and suitable for tillage. Wind speed sensors: monitor wind speed. Wind power has an impact on the evaporation rate of soil moisture, temperature fluctuations, etc. Especially in arid regions, strong winds will accelerate the evaporation of soil moisture. Atmospheric pressure sensors: monitor changes in atmospheric pressure. Atmospheric pressure changes are often related to meteorological changes and can provide auxiliary information for meteorological model predictions.

[0043] The combined use of meteorological data and soil data is very important because meteorological conditions directly affect the moisture status and temperature of the soil. For example: In the case of heavy precipitation, soil moisture often increases, and at this time, premature or overly deep subsoiling tillage should be avoided to prevent soil compaction. Under dry or high-temperature weather conditions, the evaporation of soil moisture is relatively fast. At this time, by adjusting irrigation or tillage depth in a timely manner, the water retention capacity and crop growth conditions can be effectively improved. Therefore, the combination of meteorological data and soil temperature and humidity data can improve the accuracy of tillage and help farmers better master the tillage timing and avoid unsuitable tillage conditions. All the data collected by the deployed soil temperature and humidity sensors and meteorological sensors usually need to be transmitted to the central data management platform in real time through wireless communication technologies (such as LoRa, NB-IoT, 4G / 5G, Wi-Fi, etc.).

[0044] S2: Preprocess the collected soil temperature and humidity data and soil meteorological environment data, and extract features from the preprocessed soil temperature and humidity data and soil meteorological environment data, respectively extracting soil temperature and humidity mutation features and soil meteorological prediction deviation features.

[0045] Data preprocessing is a process of organizing, cleaning, and standardizing the original collected data to ensure the accuracy and effectiveness of subsequent analysis. The main purpose of preprocessing is to remove noise, fill in missing data, smooth the data, handle outliers, etc., to ensure that the data is more accurate and reliable.

[0046] Since sensors may be affected by environmental interference, hardware failures, or data transmission errors, outliers (such as extremely high or low values) may appear in the collected soil temperature and humidity data or meteorological data. By setting a reasonable threshold range, data that is clearly not in line with the actual situation can be removed. During the data collection process, some missing values may occur. For example, the temperature and humidity data for a certain period may not be collected. Common methods include interpolation methods (such as linear interpolation, spline interpolation, etc.) or filling with historical data.

[0047] To remove high-frequency noise in the data, smoothing algorithms such as moving average and weighted average are usually used to reduce the interference of short-term fluctuations on the analysis results. Through low-pass filters or other filtering techniques, the frequent changes or noise in the data are removed, making the data more in line with the actual trend. Ensure that the units of all sensor data are unified. For example, soil temperature should be uniformly expressed in degrees Celsius, and humidity in percentage form. To eliminate the bias caused by different dimensions between different sensors, data is usually standardized (such as Z-score standardization) or normalized (such as Min-Max normalization) to convert all data into the same scale for subsequent processing.

[0048] The abrupt change characteristics of soil temperature and humidity refer to a large change in the soil temperature and humidity within a certain period. Such abrupt changes may be caused by external meteorological condition changes (such as precipitation, temperature changes) or human operations (such as irrigation, tillage, etc.). The extraction of these abrupt change characteristics can help the system promptly capture the drastic changes in the soil state, and then adjust the tillage strategy. The difference method or threshold method is used to detect the abrupt change points of soil temperature and humidity. For example, when the change in soil temperature or humidity exceeds a certain threshold (such as ±5%), it is regarded as an abrupt change. Through the difference method (the difference between the current moment and the previous moment) or analysis based on a sliding window, the moments with large change amplitudes can be found. Analyze the change trend of soil temperature and humidity over time, and capture whether the rising or falling trend of temperature and humidity continues, so as to identify short-term fluctuations and long-term change trends. The intensity of the abrupt change characteristics can be quantified by calculating the rate and amplitude of data changes, etc. For example, the change rate of soil temperature in a short period or the increase and decrease rate of humidity can be calculated, which helps to evaluate the potential impact of abrupt changes on crop growth.

[0049] The characteristics of soil meteorological prediction deviation refer to extracting the prediction deviation characteristics by comparing the actually collected meteorological data with the meteorological prediction model or historical data. These deviation characteristics can reveal the inaccuracy of meteorological prediction and its impact on soil temperature and humidity. By comparing real-time meteorological data (such as air temperature, precipitation, humidity, etc.) with meteorological prediction data, the deviation between the two is calculated. For example, calculate the error between the predicted air temperature and the actual air temperature, or the difference between the predicted precipitation and the actual precipitation. Extract characteristics such as deviation magnitude (the difference between the predicted value and the actual value), deviation change rate (the change of the prediction error over time), and deviation persistence (whether the error continuously increases or decreases). These characteristics can be used as the basis for analyzing the impact of soil meteorology. Data with large deviations may mean that the operation plan needs to be adjusted or more flexible measures need to be taken. Based on historical meteorological data and prediction errors, a deviation correction model (such as a model based on regression analysis) is constructed so that the meteorological prediction data can be corrected in the system, thereby improving the accuracy of subsequent analysis and decision-making.

[0050] After analyzing the extracted characteristics of the changing trend of soil temperature and humidity over time, a soil temperature and humidity anomaly index is generated. The method for obtaining the soil temperature and humidity anomaly index is as follows:

[0051] For each soil temperature and humidity data point, define its neighborhood, and select a parameter K, which represents the size of the neighborhood of each data point (i.e., the neighborhood contains K nearest neighbors).

[0052] Calculate the distance from each point in the dataset to all other points (for example, using the Euclidean distance). For a data point p, calculate the distance d(p,q) to all other points q, and the expression is: where p = (p1, p2, …, p n ) and q = (q1, q2, …, q n ) are the feature vectors of the soil temperature and humidity data points. For the data point p and the point q in the neighborhood, define the reachable distance r k (p,q) as: r k (p,q) = max(d(p,q), k dist (q)); where k dist (q) is the distance from the point q to its K-neighborhood, that is, the distance from the point q to its K-th nearest neighbor. The local reachability density LRD k (p) represents the density situation of the data point p in its neighborhood, which represents the reciprocal of the reachable distance of this point. The calculation formula is: where N k (p) are the K neighbors of the data point p, r k (p,q) is the reachable distance between the data point p and its neighborhood point q. Calculate the local outlier factor, that is, calculate the soil temperature and humidity anomaly index. The calculation formula is: In the formula, QSD is the soil temperature and humidity anomaly index.

[0053] The larger the value of the soil temperature and humidity anomaly index, generally, the greater the difference between the soil temperature and humidity data at this point and its neighborhood data, that is, this data point may be an outlier. Specifically, when the changes in soil temperature and humidity are drastic, or the temperature and humidity data in a certain area are significantly inconsistent with the surrounding environment, the LOF algorithm will calculate a higher anomaly index. This situation may indicate that there is a malfunction in the soil temperature and humidity sensor, abnormal data collection, or sudden atypical changes in the soil itself (such as a sudden increase in humidity or a sharp rise in temperature in a local area), resulting in inaccurate monitoring data at this point. Therefore, a larger anomaly index usually indicates errors in the monitoring data, and further verification and correction of the data accuracy are required.

[0054] On the contrary, the smaller the soil temperature and humidity anomaly index, the more consistent the soil temperature and humidity data at this point are with the surrounding data, that is, the data at this point conforms to the overall trend of its environment. A smaller anomaly index indicates that the soil temperature and humidity sensor is working properly, the data is relatively accurate, and there are no significant deviations. Therefore, a smaller anomaly index usually means that the soil temperature and humidity monitoring data is relatively accurate and can better reflect the actual soil temperature and humidity conditions. In this case, the data can be directly used in the tillage decision-making system to optimize the operation time and depth, thereby improving agricultural production efficiency.

[0055] After analyzing the deviation characteristics between the extracted real-time meteorological data and the meteorological prediction data, a meteorological data deviation index is generated. The method for obtaining the meteorological data deviation index is as follows:

[0056] Let x t represent the real-time meteorological data at time t, while represents the meteorological prediction data at the same time t, and their difference is the error: where, e t is the absolute error at the t-th moment, and the error e t at each moment will be multiplied by a weight w t , and this weight can be set according to the importance of different time periods. For example, a larger weight can be given to the critical period of crop growth or the time period with drastic changes in meteorological conditions. The weight w t is a non-negative value representing the relative importance of each moment. The weighted error is: For all collected meteorological data points, calculate the meteorological data deviation index, and the expression is: where, N is the total number of data points, and WMAE is the meteorological data deviation index.

[0057] The larger the meteorological data deviation index is, the greater the difference between the real-time meteorological data and the meteorological forecast data is, which usually means that the meteorological forecast is more inaccurate. For the soil temperature and humidity monitoring system based on meteorological data, the increase in the meteorological data deviation index will lead to a decrease in the accuracy of the soil temperature and humidity monitoring data. This is because the changes in soil temperature and humidity are greatly affected by meteorological conditions (such as precipitation, temperature, humidity, etc.). When the meteorological data deviation is large, the prediction and regulation of soil temperature and humidity will also produce errors, thereby affecting the accuracy of farming decisions. Therefore, when the meteorological data deviation index is large, the credibility of the monitoring data is low, which may lead to inappropriate tillage timing and depth selection, thereby affecting crop growth and soil quality.

[0058] On the contrary, the smaller the meteorological data deviation index is, the smaller the difference between the real-time meteorological data and the meteorological forecast data is, and the higher the accuracy of the prediction model is. At this time, the accuracy of the soil temperature and humidity monitoring data based on meteorological data is also high, because the impact of meteorological factors on soil is more precise. A lower deviation index indicates that the meteorological forecast is more accurate, thereby improving the reliability of the soil temperature and humidity prediction model, and can better control the optimal time and depth of straw return and deep tillage. This can not only effectively improve soil quality, but also improve agricultural production efficiency and ensure that crops grow in a suitable environment.

[0059] S3: Based on the extracted soil temperature and humidity mutation characteristics and soil meteorological forecast deviation characteristics, a data prediction model is constructed, and the accuracy of the soil temperature and humidity monitoring data is determined based on the output results of the data prediction model.

[0060] The soil temperature and humidity anomaly index and the meteorological data deviation index are normalized, and the accuracy value of the soil temperature and humidity monitoring data is calculated by the normalized soil temperature and humidity anomaly index and the meteorological data deviation index.

[0061] For example, the present invention can use the following formula to calculate the accuracy value of soil temperature and humidity monitoring data, and the calculation expression is: Where Q is the accuracy value of soil temperature and humidity monitoring data, QSD is the soil temperature and humidity anomaly index, WMAE is the meteorological data deviation index, a1 and a2 are the proportional coefficients of the soil temperature and humidity anomaly index and the meteorological data deviation index, and a2>a1>0.

[0062] Compare the accuracy value of the obtained soil temperature and humidity monitoring data with the preset accuracy value reference threshold based on historical data. If the accuracy value of the soil temperature and humidity monitoring data is greater than or equal to the preset accuracy value reference threshold, it indicates that the accuracy of the soil temperature and humidity monitoring data is high. At this time, generate a data normal signal and classify the soil temperature and humidity monitoring data as accurate monitoring data. If the accuracy value of the soil temperature and humidity monitoring data is less than the preset accuracy value reference threshold, it indicates that the accuracy of the soil temperature and humidity monitoring data is low. At this time, generate a data abnormal signal and classify the soil temperature and humidity monitoring data as inaccurate monitoring data.

[0063] S4: If the accuracy of the soil temperature and humidity monitoring data is high, directly input it into the operation decision system for dynamically adjusting the operation time and depth of straw returning to the field and deep loosening tillage. If the accuracy of the soil temperature and humidity monitoring data is low, judge the severity of the deviation of the soil temperature and humidity monitoring data. If the severity is high, increase the data acquisition frequency of the soil temperature and humidity sensor.

[0064] By deploying soil temperature and humidity sensors and meteorological sensors, real-time collect the soil temperature and humidity data and meteorological environment data. Calculate the accuracy value Q of the soil temperature and humidity monitoring data and compare it with the preset accuracy reference threshold Q. If the accuracy value of the soil temperature and humidity monitoring data is greater than or equal to the preset accuracy value reference threshold, it is considered that the data is accurate and can enter the next step. If it is inaccurate, the system will mark it as data abnormal and suspend tillage.

[0065] According to the real-time collected soil temperature and humidity data and meteorological prediction data, dynamically adjust the operation time T adjust and the operation depth D adjust , and the expression is: T adjust = T base + k1·ΔT soil + k2·ΔH soil + k3·ΔQ; where, T adjust is the adjusted operation time; T base is the preset operation time (for example, the conventional operation time or the best operation time in historical data); ΔT soil is the difference between the current soil temperature and the preset temperature (which can be obtained by real-time monitoring with a soil temperature sensor); ΔH soil is the difference between the current soil humidity and the preset humidity (also obtained in real-time through a humidity sensor); ΔQ is the difference between the accuracy value of the soil temperature and humidity monitoring data and the accuracy threshold, which affects the fine-tuning of the operation time. The higher the accuracy value, the smaller the adjustment range of the operation time; k1, k2, k3 are adjustment coefficients used to adjust the sensitivity of the operation time according to factors such as specific farmland and climate conditions.

[0066] The operation depth (i.e., the depth of subsoiling tillage) is adjusted according to the changes in soil moisture and soil temperature. Moist soil is suitable for shallower tillage, while dry soil is suitable for deeper tillage. The adjustment formula is: D adjust = D base + k4·ΔT soil + k5·ΔH soil + k6·ΔQ; where D adjust is the adjusted operation depth, D base is the preset operation depth (which can be determined by historical data or experience); ΔT soil and ΔH soil are the differences between the soil temperature and moisture and their preset values, similar to the operation time adjustment formula; k4, k5, and k6 are adjustment coefficients used to adjust the operation depth according to factors such as actual soil conditions and the goals of subsoiling tillage.

[0067] The adjusted operation time and operation depth are input into the operation decision system, and dynamic adjustment is performed according to the actual farmland conditions and tillage machinery to ensure the best effects of straw returning to the field and subsoiling tillage. During the operation, if it is found that the adjustment effects of the operation time or depth are not ideal, the system can perform optimization adjustment through further monitoring data feedback.

[0068] If the accuracy of the soil temperature and humidity monitoring data is low, that is, the accuracy value of the soil temperature and humidity monitoring data is less than the preset accuracy value reference threshold, the severity of the monitoring data deviation is judged. The expression is: S severity = |Q - Q threshold | + A·ΔT soil + B·ΔH soil ; where S severity is the severity of the deviation, and the larger the value, the more serious the deviation; Q is the accuracy value of the current soil temperature and humidity monitoring data; Q threshold is the preset accuracy value reference threshold; ΔT soil is the difference between the soil temperature and the preset temperature; ΔH soil is the difference between the soil humidity and the preset humidity; A and B are adjustment coefficients used to control the influence of temperature and humidity on the severity of the deviation.

[0069] According to the calculated severity of the deviation S severity , if its value exceeds the set reference threshold S severity of the severity of the deviation, it is considered that the deviation is serious, and the data acquisition frequency of the soil temperature and humidity sensors needs to be increased. The expression is: if S severity < S threshold if S severity ≥ S threshold ; fadjust is the adjusted data acquisition frequency, f base is the preset basic data acquisition frequency (for example, the initial acquisition frequency is once per hour); γ is the adjustment coefficient used to control the influence of the deviation severity on the acquisition frequency; S threshold is the reference threshold of the deviation severity. The purpose of increasing the data acquisition frequency is to obtain more accurate monitoring data and improve the accuracy of operation decisions. After increasing the data acquisition frequency, the system continues to monitor the data changes to ensure that the deviation is improved. If the deviation gradually decreases, the system can gradually return to the initial data acquisition frequency.

[0070] In this embodiment, first, a plurality of soil temperature and humidity sensors are arranged in different areas of the field, and meteorological sensors are deployed to collect soil temperature and humidity and meteorological environment data in real time. Then, the collected data is preprocessed, and the mutation characteristics of soil temperature and humidity and the prediction deviation characteristics of meteorological data are extracted. Next, a data prediction model is constructed based on the extracted characteristics, and the accuracy of soil temperature and humidity monitoring data is evaluated according to the model results. If the data accuracy is high, it will be directly input into the operation decision system to dynamically adjust the operation time and depth of straw returning to the field and deep loosening tillage; if the data accuracy is low, the deviation severity is evaluated. If it is severe, the data acquisition frequency of the sensor is increased to ensure the accuracy of subsequent data and optimize operation control.

[0071] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by software simulation of a large amount of collected data to get a formula closest to the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0072] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0073] It should be understood that the term "and / or" in this article is merely an association relationship describing associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context. Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0074] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application.

Claims

1. A method for regulating straw returning and subsoiling tillage based on soil temperature and humidity, characterized in that: It includes the following steps: S1: Deploy a number of soil temperature and humidity sensors in different areas of the field to collect soil temperature and humidity data in real time for multiple time periods, and deploy different types of meteorological sensors to collect soil meteorological environment data in real time; S2: Preprocess the collected soil temperature and humidity data and soil meteorological environment data, and extract features from the preprocessed soil temperature and humidity data and soil meteorological environment data, respectively extracting soil temperature and humidity mutation features and soil meteorological prediction deviation features; Specifically, it includes: analyzing the extracted change trend characteristics of soil temperature and humidity over time to generate a soil temperature and humidity anomaly index. The acquisition method of the soil temperature and humidity anomaly index is: Define the neighborhood for each soil temperature and humidity data point, and select a parameter K to represent the neighborhood size of each data point. For a data point p, calculate the distance d(p, q) to point q, and the expression is: ; where and are the feature vectors of the soil temperature and humidity data points. For data points p and q in the neighborhood, define the reachability distance as: ; where is the distance from point q to its K-neighborhood, that is, the distance from point q to its K-th nearest neighbor. The local reachability density represents the density situation of data point p in its neighborhood, and the calculation formula is: ; where are the K neighbors of data point p, is the reachability distance between data point p and its neighborhood point q. Calculate the local outlier factor, that is, calculate the soil temperature and humidity anomaly index, and the calculation formula is: ; in the formula, QSD is the soil temperature and humidity anomaly index; S3: Construct a data prediction model based on the extracted soil temperature and humidity mutation features and soil meteorological prediction deviation features, and determine the accuracy of soil temperature and humidity monitoring data according to the output result of the data prediction model; S4: If the accuracy of the soil temperature and humidity monitoring data is high, directly input it into the operation decision system to dynamically adjust the operation time and depth of straw returning to the field and deep loosening tillage. If the accuracy of the soil temperature and humidity monitoring data is low, judge the severity of the deviation of the soil temperature and humidity monitoring data. If the severity is high, increase the data collection frequency of the soil temperature and humidity sensors.

2. The straw returning deep loosening tillage regulation method based on soil temperature and humidity according to claim 1, characterized in that: In S2, after analyzing the deviation characteristics between the extracted real-time meteorological data and the meteorological prediction data, a meteorological data deviation index is generated. The method for obtaining the meteorological data deviation index is as follows: Let represent the real-time meteorological data at time t, while represents the meteorological prediction data at the same time t, and their difference is the error: ; where is the absolute error at the t-th moment, and the error at each moment will be multiplied by a weight , and the weighted error is: ; For all collected meteorological data points, calculate the meteorological data deviation index, and the expression is: ; where N is the total number of data points, and WMAE is the meteorological data deviation index.

3. The straw returning deep loosening tillage regulation method based on soil temperature and humidity according to claim 2, characterized in that: In S3, based on the extracted soil temperature and humidity mutation features and soil meteorological prediction deviation features, construct a data prediction model, and determine the accuracy of soil temperature and humidity monitoring data according to the output result of the data prediction model. Specifically: Normalize the soil temperature and humidity anomaly index and the meteorological data deviation index, and calculate the accuracy value of the soil temperature and humidity monitoring data through the normalized soil temperature and humidity anomaly index and meteorological data deviation index.

4. The straw returning and deep loosening tillage regulation method based on soil temperature and humidity according to claim 3, characterized in that: Compare the obtained accuracy value of the soil temperature and humidity monitoring data with the preset accuracy value reference threshold according to historical data. If the accuracy value of the soil temperature and humidity monitoring data is greater than or equal to the preset accuracy value reference threshold, it indicates that the accuracy of the soil temperature and humidity monitoring data is high. At this time, generate a data normal signal and classify the soil temperature and humidity monitoring data as accuracy monitoring data; if the accuracy value of the soil temperature and humidity monitoring data is less than the preset accuracy value reference threshold, it indicates that the accuracy of the soil temperature and humidity monitoring data is low. At this time, generate a data anomaly signal and classify the soil temperature and humidity monitoring data as inaccurate monitoring data.

5. The straw returning and subsoiling tillage regulation method based on soil temperature and humidity according to claim 1, characterized in that: In S4, if the accuracy of the soil temperature and humidity monitoring data is high, directly input it into the operation decision system to dynamically adjust the operation time and depth of straw returning to the field and deep loosening tillage. Specifically: Based on the real-time collected soil temperature and humidity data and meteorological prediction data, the operation time and the operation depth are dynamically adjusted. The expression is: ; In the formula, is the adjusted operation time; is the preset operation time; is the difference between the current soil temperature and the preset temperature; is the difference between the current soil humidity and the preset humidity; is the difference between the accuracy value of the soil temperature and humidity monitoring data and the accuracy threshold, which affects the fine-tuning of the operation time; is the adjustment coefficient; The operation depth is adjusted according to the changes in soil moisture and soil temperature, and the adjustment formula is: ; In the formula, is the adjusted operation depth, is the preset operation depth; and are the differences between the soil temperature and moisture and the preset values; is the adjustment coefficient; The adjusted operation time and operation depth are input into the operation decision system for dynamic adjustment according to the actual farmland conditions and tillage machinery.

6. The straw returning and subsoiling tillage regulation method based on soil temperature and humidity according to claim 5, characterized in that: If the accuracy of the soil temperature and humidity monitoring data is low, that is, the accuracy value of the soil temperature and humidity monitoring data is less than the preset accuracy value reference threshold, judge the severity of the deviation of the monitoring data. The expression is: ; In the formula, is the severity of the deviation, Q is the accuracy value of the current soil temperature and humidity monitoring data; is the preset accuracy value reference threshold; is the difference between the soil temperature and the preset temperature; is the difference between the soil humidity and the preset humidity; A and B are adjustment coefficients; According to the calculated deviation severity , if its value exceeds the reference threshold of the set deviation severity , it is considered that the deviation is serious and the data acquisition frequency of the soil temperature and humidity sensor needs to be increased. The expression is: ; is the adjusted data acquisition frequency, is the preset basic data acquisition frequency, is the adjustment coefficient, which is used to control the influence of the deviation severity on the acquisition frequency; is the reference threshold of the deviation severity.

Images