A tobacco field variable fertilization control system and method

Abstract

The present invention relates to the field of adaptive control technology, and specifically to a variable fertilization control system and method for tobacco fields. The system includes: an induction data acquisition module that collects soil humidity data through soil humidity sensors, monitors the growth data of crops at the same time, and synthesizes real-time soil and growth data; performs time series analysis on the real-time soil and growth data, identifies data trends and outliers, and generates trend analysis results. In the present invention, soil humidity and crop growth data are collected, combined with time series analysis, to achieve rapid identification of trends and anomalies, improving the accuracy of prediction. A convolutional neural network is used to identify and probabilistically evaluate pests and diseases, making pest and disease management more precise and the response more rapid. Through these technologies, the system can dynamically adjust irrigation parameters based on real-time data, optimize water distribution, and ensure that the irrigation strategy perfectly matches crop needs and potential risks.

Description

Technical Field

[0001] The present invention relates to the technical field of adaptive control, and particularly to a variable fertilization control system and method for tobacco fields. Background Art

[0002] Adaptive control technology is a control system technology that enables a system to automatically adjust its control parameters when the external environment or system parameters change, so as to maintain or improve system performance. The variable fertilization control system for tobacco fields is an intelligent control system designed specifically for the tobacco planting field, and its main purpose is to automatically adjust the fertilization or irrigation amount to adapt to the changing soil and environmental conditions in the tobacco field.

[0003] However, the existing technology relies on regular detection and manual intervention, which results in slow response and difficulty in effectively coping with short-term environmental changes or pest and disease emergencies. This processing lag may not only lead to waste of resources, but also increase the risk of crop diseases due to improper irrigation. In addition, the existing technology lacks highly integrated data analysis capabilities, making it difficult to achieve refined management, thus affecting the overall efficiency of crop production and environmental sustainability. Summary of the Invention

[0004] The purpose of the present invention is to solve the drawbacks existing in the prior art, and to propose a variable fertilization control system and method for tobacco fields.

[0005] To achieve the above purpose, the present invention adopts the following technical solution: A variable fertilization control system for tobacco fields includes:

[0006] An induction data acquisition module, which collects soil humidity data through a soil humidity sensor, monitors the growth data of crops at the same time, and synthesizes real-time soil and growth data; performs time series analysis on the real-time soil and growth data, identifies data trends and abnormal points, and generates a trend analysis result;

[0007] A pest and disease analysis module, which receives the trend analysis result, analyzes the pest and disease signs in the crop image through image recognition and convolutional neural network, identifies the types of pests and diseases, and generates a pest and disease identification result; evaluates the probability of the occurrence of pests and diseases according to the pest and disease identification result, and generates a pest and disease occurrence probability evaluation result;

[0008] An irrigation regulation module, which adjusts the irrigation amount according to the pest and disease occurrence probability evaluation result, optimizes the pump rate and switch time to match the current crop demand and predicted pest and disease risks, and generates adjusted irrigation parameters; combines the matching degree between the adjusted irrigation parameters and the real-time soil humidity data to optimize water distribution and obtain an optimized irrigation configuration;

[0009] The fertilization decision-making module, based on the optimized irrigation configuration, simultaneously collects real-time soil moisture data, determines the need for chemical fertilizer adjustment, and selects the type and application rate of chemical fertilizers with reference to the growth requirements of crops and the chemical properties of the soil.

[0010] Preferably, the steps for obtaining the real-time soil and growth data are as follows:

[0011] Monitor the soil moisture in the tobacco field area through a soil moisture sensor, and simultaneously capture the growth status of the crops through a camera to obtain preliminary soil moisture data and crop growth data;

[0012] Based on the preliminary soil moisture data and crop growth data, perform alignment and merging of timestamp markings to form a unified dataset;

[0013] Based on the unified dataset, analyze the interaction between soil moisture and crop growth, and generate real-time soil and growth data through fusion processing.

[0014] Preferably, the steps for obtaining the trend analysis result are as follows:

[0015] Based on the real-time soil and growth data, perform statistical analysis, including the calculation of mean, standard deviation, and extreme values, to obtain the statistical feature analysis result of the data;

[0016] Based on the statistical feature analysis result of the data, perform time series analysis and calculate the comprehensive score at each time point. The calculation formula is:

[0017] ;

[0018] where, is the comprehensive score at time t, is the observed value at time t, and are the mean and standard deviation of the dataset respectively, is the moving average, is the standard deviation of the moving average, and are the weight coefficients;

[0019] Based on the comprehensive score, evaluate the overall trend and identify outliers to obtain the trend analysis result.

[0020] Preferably, the steps for obtaining the pest and disease identification result are as follows:

[0021] Use the trend analysis result to select crop images for pest and disease sign analysis to obtain the selected crop images;

[0022] Based on the selected crop images, a convolutional neural network is deployed for image processing to identify the characteristics of pests and diseases in the images, and the analysis results of the characteristics of pests and diseases in the images are obtained;

[0023] Based on the analysis results of the characteristics of pests and diseases in the images, the types of pests and diseases are classified. By comparing the established database of pest and disease characteristics, the types of pests and diseases are identified and calibrated to obtain the pest and disease identification results.

[0024] Preferably, the steps for obtaining the pest and disease occurrence probability assessment results are as follows:

[0025] Receive the pest and disease identification results, including the types of each pest and disease and the distribution characteristics in the images, to obtain the pest and disease distribution data;

[0026] Based on the pest and disease distribution data, calculate the occurrence probabilities of various pests and diseases. The calculation formula is:

[0027] ;

[0028] Where, is the occurrence probability of the i-th pest and disease, represents the coverage area of the pest and disease in the identification results, represents the influence degree, represents the temperature sensitivity, represents the seasonality, is the adjustment coefficient, and n is the total number of pests and diseases;

[0029] Based on the occurrence probabilities, conduct a risk assessment, and combine environmental factors and past data to obtain the pest and disease occurrence probability assessment results.

[0030] Preferably, the steps for obtaining the adjusted irrigation parameters are as follows:

[0031] Analyze the probabilities of various pest and disease types provided by the pest and disease occurrence probability assessment results, conduct an impact analysis on the irrigation requirements according to the probabilities, including an investigation of the crop growth cycle and water requirements, and determine the preliminary irrigation volume range for adjustment;

[0032] Based on the preliminary irrigation volume range, calculate the pump rate. The calculation formula is:

[0033] ;

[0034] Where, is the adjusted pump rate, is the basic pump rate, is the water volume required by the current crop, is the standard water volume, is the pest and disease risk adjustment coefficient;

[0035] Reconfigure the irrigation parameters based on the water pump rate to obtain the adjusted irrigation parameters.

[0036] Preferably, the steps for obtaining the irrigation optimization configuration are as follows:

[0037] Calculate the matching degree index between the adjusted irrigation parameters and the soil humidity data. The calculation formula is:

[0038] ;

[0039] Wherein, is the matching degree index, is the ideal soil humidity at the i-th detection point, is the soil humidity at the i-th detection point, is the number of detection points;

[0040] Optimize the operating parameters of the water pump and the irrigation plan according to the matching degree index to obtain the irrigation optimization configuration.

[0041] The present invention provides a method for controlling variable fertilization in tobacco fields, including the following steps:

[0042] Collect soil humidity data through a soil humidity sensor, and at the same time collect crop growth data. Combine the collected data to perform time series analysis, identify data trends and abnormal points, and obtain a comprehensive analysis result of soil and growth data;

[0043] Based on the comprehensive analysis result of soil and growth data, analyze the pest and disease signs in the crop image through image recognition and convolutional neural network, identify the types of pests and diseases, evaluate the occurrence probability of pests and diseases, and generate a pest and disease recognition and risk assessment result;

[0044] Based on the pest and disease recognition and risk assessment result, adjust the irrigation amount, optimize the water pump rate and switch time to match the current crop demand and the predicted pest and disease risk, and generate the adjusted irrigation parameters;

[0045] Combine the matching degree of the adjusted irrigation parameters with the real-time soil humidity data to optimize the water distribution and obtain the irrigation optimization configuration result;

[0046] Utilize the irrigation optimization configuration result, and select the type and application rate of chemical fertilizers with reference to the growth requirements of the crops and the chemical properties of the soil.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0048] In the present invention, soil moisture and crop growth data are collected, and combined with time series analysis to achieve rapid identification of trends and anomalies, improving the accuracy of prediction. A convolutional neural network is used to identify and probabilistically evaluate pests and diseases, making pest and disease management more precise and responsive. Through these technologies, the system can dynamically adjust irrigation parameters based on real-time data, optimize water distribution, and ensure that the irrigation strategy perfectly matches crop needs and potential risks. This method not only reduces resource waste but also effectively mitigates the potential impact of pests and diseases, promotes the healthy growth of crops, and improves yield and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a system flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] In order to make the objectives, technical solutions and advantages of the present invention clearer, 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.

[0051] Please refer to Figure 1 , the present invention provides a technical solution: a variable fertilization control system for tobacco fields includes:

[0052] An induction data acquisition module that collects soil moisture data through soil moisture sensors, monitors the growth data of crops at the same time, and synthesizes real-time soil and growth data; performs time series analysis on the real-time soil and growth data, identifies data trends and anomaly points, and generates trend analysis results;

[0053] A pest and disease analysis module that receives the trend analysis results, analyzes the pest and disease signs in the crop images through image recognition and convolutional neural network, identifies the types of pests and diseases, and generates pest and disease identification results; evaluates the probability of the occurrence of pests and diseases according to the pest and disease identification results, and generates a pest and disease occurrence probability evaluation result;

[0054] An irrigation regulation module that adjusts the irrigation amount according to the pest and disease occurrence probability evaluation result, optimizes the pump speed and switch time to match the current crop needs and predicted pest and disease risks, and generates adjusted irrigation parameters; combines the matching degree of the adjusted irrigation parameters with the real-time soil moisture data to optimize water distribution and obtain an optimized irrigation configuration;

[0055] A fertilization decision-making module that, based on the optimized irrigation configuration, collects real-time soil moisture data at the same time, determines the need for chemical fertilizer adjustment, and selects the type and application rate of chemical fertilizers with reference to the growth needs of crops and the chemical properties of the soil;

[0056] Specifically, based on the optimized irrigation configuration and while collecting real-time soil moisture data, first read the prepared irrigation parameters and collate the corresponding specific irrigation time periods, irrigation amounts, etc. side by side with the real-time soil moisture records. For each soil moisture collection point, check its geographical location and sensor number, verify its current output based on the factory calibration value of the sensor and confirm whether there is transmission delay. If it is found that the output of a certain sensor is unstable, record this item and detect its circuit or power connection again. After confirming that the outputs of all sensors are within the normal range, compare the multiple soil moisture readings obtained with the current growth stage of the crop. The growth stage can be deduced from the sowing time to obtain sections such as the initial stage, elongation stage, or flowering and fruiting stage, and be confirmed in combination with the height and number of leaves of the plant. When summarizing the soil moisture and crop growth stage information, then retrieve the soil chemical property parameters of the field plot. These chemical property parameters are generally obtained by regular field sampling and measured in an agricultural laboratory, including pH value, organic matter content, and various nutrient indicators. Then, under the guidance of referring to the absorption ratios of chemical elements such as nitrogen, phosphorus, and potassium by the crop and the field fertilization specifications, retrieve the local accumulated fertilizer recommendation tables respectively. These fertilizer recommendation tables include common chemical fertilizer types and their applicable crop growth stages. The proportion of active ingredients and release duration of each chemical fertilizer are obtained by agricultural technicians summarizing the test data of previous experiments. Then, analyze the synchronous absorption of water and nutrients by the crop roots in combination with the collected real-time soil moisture data. If it is found that the soil moisture in some areas deviates from the threshold set by years of experience for a long time, reallocate the chemical fertilizer types and application methods in those areas. For example, change the original whole-bag application to batch input in multiple time periods, or adjust the ratio of granular and liquid fertilizers. When these targeted analyses are completed, obtain the required chemical fertilizer types and their respective application amounts, record the application amount as how many kilograms per hectare or how many grams per plant, and match it with the previously obtained optimized irrigation configuration. Summarize the results at each key stage of the crop growth cycle to form the final adjustment plan, and finally determine the chemical fertilizer type and application amount.

[0057] The steps for obtaining real-time soil and growth data are as follows:

[0058] Monitor the soil moisture in the tobacco field area through soil moisture sensors, and at the same time capture the growth status of the crop through cameras to obtain preliminary soil moisture data and crop growth data;

[0059] Based on the preliminary soil moisture data and crop growth data, align and merge the time stamp markings to form a unified data set;

[0060] Based on the unified data set, analyze the interaction between soil moisture and crop growth, and generate real-time soil and growth data through fusion processing.

[0061] Specifically, by deploying multiple soil moisture sensors in the tobacco field area and placing them at different depths, initially calibrate each sensor. When calibrating, select several reference values that meet the local planting conditions. These reference values are derived from historical planting experience and the average value of multiple actual measurements in similar soil environments. If the monitoring data of a certain sensor deviates from the reference value by more than the agreed comparison value, while recording the data, re-check the installation position of the sensor and the connection of the circuit. After calibration, install the camera at a fixed position that can cover the growth area of the target crop and set appropriate resolution and shooting frequency. Record the start time of the camera and the start time of the soil moisture sensor monitoring as the subsequent correlation basis. Subsequently, continuously obtain the soil moisture changes and the growth images of the crops in the picture and perform digital encoding to finally obtain preliminary soil moisture data and crop growth data.

[0062] Based on the preliminary soil moisture data and crop growth data obtained previously, select the time tags output by the sensors and the time tags of the camera shooting frames for pairing and alignment. For the parts with inconsistent timestamps, first confirm the inserted or discarded time periods by comparing the values at adjacent moments and the image frame numbers, and record the details of the alignment in the comparison table. This comparison table is comprehensively generated from the actual monitoring start time, the sensor output frequency, and the camera shooting frequency. If missing records or picture information are encountered, fill them by interpolation or repeated reading. The interpolation threshold is determined by the minimum update interval under the observation conditions and calibrated by manual experience. When there are still data segments that cannot be normally aligned after filling, make manual marks and retain the original data for subsequent inspection. After aligning and merging all time points, a unified data set is formed.

[0063] Based on a unified dataset, first extract the soil moisture readings in each record and growth indicators such as crop height and number of leaves in the corresponding crop images. Match these indicators in the same row record according to the reference arrangement order at the same moment, and use the linear regression algorithm to measure the correlation between soil moisture and each growth indicator. The linear regression algorithm splits the dataset by selecting the training ratio and validation ratio. For example, 70% of the data entries are selected as the training subset and 30% of the data entries are selected as the validation subset. In the training subset, set the independent variable of the regression equation as the soil moisture value and the dependent variable as a certain growth indicator of the target crop. Use the least squares method to iteratively fit the corresponding regression coefficients, and then apply these regression coefficients to each sample in the validation subset to check the output error situation. If the error exceeds the preset limit obtained by statistically analyzing the historical planting data, re-adjust the bias term of the regression equation and continue to iterate. Through multiple corrections, a relatively stable regression model is obtained. Subsequently, synchronously input this model, the real-time collected soil moisture readings, and crop growth indicators into the fusion module. When the fusion module receives new data, it calls the trained regression equation for one-time calculation and outputs the real-time soil and growth data, and finally generates the real-time soil and growth data.

[0064] The steps to obtain the trend analysis results are as follows:

[0065] Based on the real-time soil and growth data, conduct statistical analysis, including the calculation of mean, standard deviation, and extreme values, to obtain the statistical feature analysis results of the data;

[0066] Based on the statistical feature analysis results of the data, perform time series analysis and calculate the comprehensive score at each time point. The calculation formula is:

[0067] ;

[0068] where, is the comprehensive score at time t, is the observed value at time t, and are the mean and standard deviation of the dataset respectively, is the moving average, is the standard deviation of the moving average, and are the weight coefficients;

[0069] Based on the comprehensive score, evaluate the overall trend and identify outliers to obtain the trend analysis results.

[0070] Specifically, based on the real-time soil and growth data, first organize all records into a numerical sequence, then accumulate them item by item and divide the accumulated sum by the total number of records to obtain the mean value. The total number of records can be confirmed by counting each item of the collected soil and growth data one by one. Then calculate the difference between each record and the mean value and accumulate the squares. Divide the accumulated square value by the total number of records and take the square root to obtain the standard deviation. Then select the maximum and minimum record points in the entire numerical sequence to determine the extreme values. Finally, complete the statistical analysis of the real-time soil and growth data to obtain the statistical characteristic analysis results of the data.

[0071] The advantage of the formula is to obtain a comprehensive measurement value by introducing two sets of differences and weighting them respectively. and is used to balance the influence of different differences on the comprehensive score. and can be obtained by calculating the mean value and standard deviation of the previously obtained data sequence. For example, from 500 consecutive sets of soil moisture and growth parameters, count and . The measured value at time t can be recorded as 50. and are obtained by cumulative calculation item by item through a moving window. For example, the weighted average of the last 5 data is obtained. and its standard deviation . Take and . Then substituting these data into the formula in turn, the multi-level operation process is as follows:

[0072] First calculate , then let , and then let . Subsequently, use and to weight, that is . Through item-by-item operation, we can get . This result indicates that the comprehensive score at this moment is about 0.86. When this score is greater than a certain empirical reference value, it can be regarded as abnormal or significantly deviated. When it is lower than this reference value, it is in a relatively stable range.

[0073] Based on the comprehensive score, select the comprehensive score curve within a certain period of time and correspond it to the observation moment. First, set a threshold range from historical analysis, which is comprehensively determined by the statistical interval of past data and expert experience in the actual monitoring environment. If the comprehensive score continuously exceeds this threshold range at multiple consecutive moments, while recording these moments, list the corresponding soil moisture or growth indicators in the key attention list and track each record. If it is found that the comprehensive score drops below the threshold range within a certain period of time, observe the subsequent time period and maintain the original recording frequency. After comparing and marking the comprehensive scores of all moments, the trend analysis result is obtained.

[0074] The steps to obtain the pest and disease identification result are as follows:

[0075] Use the trend analysis result to select crop images for pest and disease sign analysis to obtain the selected crop images;

[0076] Based on the selected crop images, deploy a convolutional neural network for image processing to identify the pest and disease characteristics in the images and obtain the pest and disease characteristic analysis result in the images;

[0077] Based on the pest and disease characteristic analysis result in the images, classify the pest and disease types. By comparing the established pest and disease characteristic database, identify and calibrate the pest and disease types to obtain the pest and disease identification result.

[0078] Specifically, screen the crop images according to the previously obtained trend analysis result. First, lock the corresponding dates and shooting numbers in the image database according to the time period indicated in the trend analysis result. After summarizing all the images on these dates, detect their shooting focal lengths and lighting conditions. If it is found that the content of some images cannot be distinguished due to overexposure or perspective deviation, these images are excluded first. The remaining images are further grouped according to the proportion of the crop area in the picture, and this proportion is calculated by reading the proportion of green pixels in the image and combining the crop height information. The threshold is set by daily monitoring experience combined with the average plant height of local crops. If it is detected that the proportion of green pixels in some images is too low and the main position of the crop deviates from the center by more than the defined value given by expert measurement, then exclude this image from the subsequent analysis queue. After obtaining the images that meet the quality requirements, make a preliminary judgment on the pest and disease signs of these images. The judgment basis includes whether there are curly spots on the crop leaves and whether there are abnormal patterns on the stems. When the above signs are detected, record the image and mark the pixel coordinate range of the suspicious area. After multiple record comparisons and confirmations, complete the final screening and obtain the crop images that need to be deeply analyzed this time, and finally obtain the selected crop images.

[0079] Based on the selected crop images, perform in-depth processing. First, assign a corresponding multi-channel data input form to each image and specify the row and column dimensions. Perform numerical normalization on each pixel point in this row and column dimension. At the same time, remove the background part in the image that has nothing to do with the crop main body through the set color intervals. These color intervals are set by analyzing the pixel statistical results of several background images. If there is a situation where the pixel mean and chromaticity distribution deviate significantly from the crop main body, mark its pixel matrix as an invalid area. After the image data is preprocessed, deploy a convolutional neural network to train it. This network contains several convolutional layers and pooling layers, and enhances the recognition accuracy by extracting texture features formed on the surfaces of leaves, fruits, and stems layer by layer. The network weights are obtained by training with previously collected healthy crop and common pest and disease images. During training, perform forward operations on each batch of image samples and compare the output with the known labels, and update the weights iteratively through error backpropagation. After multiple rounds of training, fix these weights to form the final model. Then, perform convolution calculations on the selected crop images one by one in inference mode and output the feature response matrix. Then, perform classification analysis on these response matrices and mark the positions and significance levels of the suspicious pest and disease areas. Finally, obtain the pest and disease feature analysis results in the image.

[0080] After obtaining the pest and disease feature analysis results in the image, classify each pest and disease feature. First, call the key attribute indicators of pests and diseases collected from a reference document, decompose these key attribute indicators into multiple dimensions such as appearance morphology, color distribution, and pattern shape. Then, compare the pest and disease features in the current image respectively. If the matching degree in terms of morphology and pattern is relatively high with a certain pest and disease type, mark this pest and disease feature as this type. The determination of the matching degree comes from the query results of the established pest and disease feature database. This database contains tens of thousands of image feature vectors and known pest and disease labels. When retrieving, compare item by item with multiple fields, including spot shape indicators, leaf damage degree, etc. If there are multiple types coexisting, include this image in the composite type statistics at the same time. Through such step-by-step classification, identify and calibrate the actual pest and disease types that appear in turn, and finally obtain the pest and disease recognition results.

[0081] The steps to obtain the pest and disease occurrence probability assessment results are as follows:

[0082] Receive the pest and disease recognition results, including the type of each pest and disease and its distribution characteristics in the image, to obtain the pest and disease distribution data;

[0083] Based on the pest and disease distribution data, calculate the occurrence probability of each type of pest and disease. The calculation formula is:

[0084] ;

[0085] Among them, is the occurrence probability of the i-th type of pest and disease, represents the coverage area of the pest and disease in the recognition result, represents the degree of impact, represents the temperature sensitivity, represents the seasonality, is the adjustment coefficient, and n is the total number of pests and diseases;

[0086] Based on the occurrence probability, risk assessment is carried out. Combining environmental factors and past data, the assessment result of the occurrence probability of pests and diseases is obtained.

[0087] Specifically, receive the pest and disease recognition result and obtain all types and their distribution characteristics in the image in categories. First, according to the pest and disease type information indicated in the recognition result and the corresponding coverage range, mark the pixel coordinate areas in the image one by one. Archive the positions and area ranges of each identified pest and disease in each image and record their relative position relationships with the crop characteristics. Then, merge and organize the repeated distribution positions of the same pest and disease type. During this process, if it is found that the overlapping area of the repeated calibration boundaries of the same pest and disease type exceeds the determination value specified based on empirical investigation (for example, 10% of the covered pixel area), then merge these overlapping areas into a single distribution area and update its area parameter. When the fitting degree of the distribution area coordinates to the crop contour is lower than the comparison threshold specified by some expert experience (for example, 60% fitting degree), mark this place as a suspicious area and read the adjacent frame images again to confirm whether there is a recognition deviation. If it is determined that the recognition is normal after proofreading multiple frames of images, retain the coordinate and area information of the pest and disease distribution area. Classify and count all the pest and disease distribution areas after proofreading and merging according to the pest and disease types and distribution situations. In the statistics, the occurrence times and total coverage area of each pest and disease need to be recorded separately to avoid confusion between the area and the number of times in the later stage. Finally, summarize these statistical data to form complete pest and disease distribution data.

[0088] The benefit of the formula is that it integrates four indicators: the coverage area of pests and diseases, the degree of impact, temperature sensitivity, and seasonality, and performs normalization calculation in exponential form, so that impact factors of the same order of magnitude or different orders of magnitude can be reasonably quantified. The acquisition steps of each parameter are as follows: coverage area is obtained by matching the pixel area in the image recognition result with the field measurement data. Degree of impact is quantified by agricultural technicians for the yield loss ratio caused by pests and diseases and recorded as a score between 0 and 10. Temperature sensitivity is obtained from the monitoring of the corresponding temperature threshold during the local growth period of the pest and disease and converted into the range of 0 to 5. Seasonality is obtained through monthly or quarterly statistical analysis, indicating the reproduction intensity of pests and diseases in different seasons and reflected by a score between 0 and 5. Adjustment coefficient Then, it is corrected step by step according to the influence depth of various parameters in previous years on the final probability result, and the fluctuation range in the actual environment is recorded.

[0089] Calculation process:

[0090] Take the total number of pests and diseases , where the coverage area of the first type of pest and disease (in square centimeters, obtained by combining the recognition result and on-site measurement), the influence degree , the temperature sensitivity , the seasonality , and in addition, set , substitute these values into the exponential term to get:

[0091] ;

[0092] Multiply and add item by item to get:

[0093] ;

[0094] Then calculate , and perform the same numerical substitution and exponential calculation for other pests and diseases (the second and third types) to get and two results. Add the three results as the denominator, and divide by the denominator respectively to obtain the occurrence probabilities of the three pests and diseases;

[0095] This result shows the probability values corresponding to each pest and disease under the current environmental parameters and recognition data. If the probability value is higher than the reference boundary established through previous years' data (for example, 0.4), the diffusion risk in a specific plot should be focused on in subsequent processing. If it is lower than this boundary, it is regarded as the general occurrence probability interval.

[0096] When conducting risk assessment based on the previously obtained occurrence probabilities, first compare the probabilities of each type of pest and disease with the on-site monitoring data in the local area over the past three to five years. These historical data include annual average temperature, precipitation, soil pH, and density information of the crop population, etc. During the comparison stage, item by item, load the types of pests and diseases that occurred in different months of each year and their coverage areas, and compare this information with the newly calculated occurrence probabilities. If the actual past occurrence times of certain pests and diseases show significant increases or decreases under similar climatic conditions, then this difference will be incorporated into the assessment and the score will be adjusted. The adjustment of this score refers to the statistical standards obtained through multiple regular consultations by the local agricultural technology department. For example, set the additional risk score of pests and diseases brought about by a 1°C increase in temperature, or certain pest and disease derivative factors in the case of continuously low precipitation. Then, apply the adjusted probabilities of pests and diseases in turn to construct a unified risk level system, and correspond each type of pest and disease category with its comprehensive risk value. If the comprehensive risk value exceeds the threshold set by the expert group, it is marked as high risk; if it falls within the normal range, it is marked as regular risk. Finally, summarize the risk levels and corresponding occurrence probabilities of each pest and disease to obtain the assessment result of the occurrence probability of pests and diseases.

[0097] The steps for obtaining the adjusted irrigation parameters are as follows:

[0098] Analyze the probabilities of each type of pest and disease provided by the assessment result of the occurrence probability of pests and diseases, and conduct an impact analysis on the irrigation demand based on the probabilities, including investigations of the crop growth cycle and water requirements, and determine the preliminary range of adjusted irrigation volume;

[0099] Based on the preliminary range of irrigation volume, calculate the pump rate, and the calculation formula is:

[0100] ;

[0101] Where, is the adjusted pump rate, is the basic pump rate, is the water volume required by the current crop, is the standard water volume, is the pest and disease risk adjustment coefficient;

[0102] Based on the pump rate, reconfigure the irrigation parameters to obtain the adjusted irrigation parameters.

[0103] Specifically, the probability of each type of pest and disease provided by the pest and disease occurrence probability assessment results is analyzed. First, the probability value of each pest and disease is read from the assessment result and compared with the recent growth stage of the crop. The growth stage information is obtained through the key nodes such as crop sowing time, seedling stage to heading stage and corresponding physiological characteristics survey recorded in the planting archives. Then, according to the sensitivity of crops to water demand at different stages, the corresponding irrigation time and water consumption references are listed one by one. For the stage with a higher probability of pests and diseases, the relationship between irrigation amount and irrigation period is focused on. If the irrigation demand statistics for similar crops in the planting habits of the region show that low water volume in a certain period of time is likely to cause the spread of pests and diseases, then the irrigation amount for this period will be increased in combination with the crop water demand model accumulated by agricultural technicians. Conversely, if the probability of pests and diseases in a certain period of time is found to be low, the irrigation amount can be reduced until it is maintained. To ensure that crops absorb and maintain soil moisture normally above the standard threshold set by the agricultural technology department, this standard threshold is based on field monitoring and statistics of the local average precipitation over many years, crop root depth and soil water holding capacity. After completing the comparison of the relationship between the probability of pests and diseases and water demand in all stages, the water demand adjustment situation in each period is obtained and combined into a preliminary irrigation volume range. If the irrigation volume in some high-pest and disease risk stages is planned to be too high, it is necessary to verify again whether it exceeds the safety upper limit obtained from the actual irrigation conditions in multiple consecutive seasons. The safety upper limit is also summarized by the agricultural technology department through years of accumulated field observations. If there is no over-limit situation, the proposed amount can be maintained. If there is an over-limit, the period will be appropriately adjusted down without affecting crop absorption. Finally, after confirming the minimum, regular and maximum irrigation volume ranges for each period, a complete preliminary irrigation volume range is formed.

[0104] The benefit of the formula is that by multiplying the square root term and the risk coefficient increment term, the actual water requirement of the crop is compared with the standard water requirement, and then appropriately enlarged or reduced according to the risk of pests and diseases, so that the overall irrigation efficiency is more in line with the current field conditions. represents the adjusted pump speed, represents the basic water pump rate measured under normal environmental and soil moisture conditions, The current monitored crop water requirement is obtained by measuring the moisture content at the surface and root depth of the field and combining it with the survey results of the water absorption characteristics of the crop varieties. It is based on the standard amount of water required for each irrigation cycle of general crops calculated from local planting records over many years. It comes from the previous pest and disease occurrence probability assessment results, which represents the pest and disease risk adjustment coefficient corresponding to the percentage scale;

[0105] Calculation process:

[0106] Through previous measurements and statistics in the same area, (Unit: L / min), (Unit: L), (Unit: L), while referring to the pest and disease risk assessment, it is obtained that , substitute the above data item by item. First, calculate the part under the square root:

[0107] ;

[0108] Then calculate the increment term:

[0109] ;

[0110] Finally, multiply the three terms:

[0111] ;

[0112] After multi - level operations, it can be obtained that (about Unit: L / min);

[0113] This result shows that in this example, the new water pump rate is increased to about 19.71 L / min. If this value is further greater than the water pump safety upper limit set by the local agricultural technology department, it should be re - checked during actual operation. If it is within a reasonable range, this value can be directly put into use. And the larger the specific value, the more water output in the same time period, and it is necessary to monitor and adjust in combination with the soil infiltration capacity and drainage measures.

[0114] After obtaining the adjusted water pump rate, re - configure the irrigation parameters. First, compare the new water pump rate with the terrain slope, pipeline distribution, and drainage path of the crop field, and record the corresponding relationships between these elements and the water pump rate respectively. For example, in a field with a larger slope, it is necessary to correspondingly increase the water supply head to maintain a stable flow rate, record the farthest water supply distance from each area to the water pump interface. If it is found that the water flow rate at the end of some pipelines drops significantly, then set this field as a priority adjustment unit and allocate additional water supply duration. For the drainage path, it is also necessary to confirm point by point whether it can carry the new flow rate. If the local drainage volume is insufficient, then dig and dredge or transform the channel. When the adaptation adjustment of the pipeline information and water pump rate for each field is completed, conduct quantitative irrigation for the fields in stages, record the actual time consumed for the current round of irrigation and the change trend of soil humidity. If the soil humidity detection result is within the range determined by agricultural technicians and the potential pest and disease risk does not change significantly, then the current irrigation parameters can be continued to be applied to the subsequent stages. On the contrary, if it is monitored that the soil humidity seriously deviates from the normal value after irrigation or the pest and disease level shows an abnormal increase, then it is necessary to correct the water pump rate again in combination with the risk adjustment coefficient, and finally confirm the irrigation plan for all subdivided areas and output the corresponding irrigation time period, rate, and duration, thus obtaining the adjusted irrigation parameters.

[0115] The steps for obtaining the optimal irrigation configuration are as follows:

[0116] Based on the adjusted irrigation parameters and soil moisture data, calculate the matching degree index between the two. The calculation formula is:

[0117] ;

[0118] Where, is the matching degree index, is the ideal soil moisture at the i-th detection point, is the soil moisture at the i-th detection point, is the number of detection points;

[0119] Optimize the operating parameters of the water pump and the irrigation plan according to the matching degree index to obtain the optimal irrigation configuration.

[0120] Specifically, the advantage of the formula is that by normalizing and accumulating the difference between the ideal soil moisture and the actual soil moisture at each detection point, it is mapped to the interval between 0 and 1 in an exponential form, thereby summarizing the deviation degree between the irrigation parameters and the soil moisture data. Among them, The closer the value of

[0121] is to 1, the closer the two are; the closer it is to 0, the greater the deviation. represents the ideal soil moisture at the i-th detection point. This value is intercepted from the optimal interval obtained by long-term observation of the growth habits of the crops planted in the field and the local climate (for example, some vegetables may be suitable for soil moisture between 20% and 30%). In actual use, it can be further subdivided according to different crop types and growth stages. is collected in real time by a soil moisture sensor and the value is obtained by combining the depth position of each detection point and the sensor calibration parameters.

[0122] Calculation process:

[0123] In a certain plot, detection points are arranged, and the ideal humidity are 25, 26, 28, 28, 30 (in percentage) respectively, and the actual humidity are 24, 25, 30, 29, 31 (also in percentage) respectively. The deviation term corresponding to the i-th detection point is , and calculating item by item, we can get:

[0124] ;

[0125] ;

[0126] ;

[0127] ;

[0128] ;

[0129] Adding these deviations together gives:

[0130] ;

[0131] Then calculate the exponential term , so there is:

[0132] ;

[0133] This result shows that the matching degree index is approximately 0.55, indicating that there is a certain deviation between the current irrigation parameters and the soil moisture data. When the value is greater than 0.8, it can be regarded as a high matching degree. When it is less than 0.4, it can be regarded as a low matching degree. The intermediate range is further analyzed according to the intervals subdivided by the agricultural technology department.

[0134] When analyzing the parameters in the matching degree index, first collect the information of all detection point positions and check their coordinate distribution in the field. Record the ideal humidity sources of each detection point as a column of values and label that the ideal humidity comes from the optimal water content of the crop at the current growth stage. When reading the actual humidity later, it is necessary to focus on checking the output stability of the sensor. In the calibration process, adjust the reference value of each sensor, which is usually obtained by measuring a control sample. When it is determined that all sensor outputs are consistent, compare the current multiple detection point records with the ideal humidity, calculate the relative deviation, and separately count the single-point deviation and the comprehensive deviation. If it is found that the deviation of an individual detection point has always been higher than the threshold (e.g., 0.25) formed by years of monitoring, record this detection point and conduct a line check or replacement again. At the same time, investigate the local topography, drainage channel smoothness, and crop root zone depth in the field to ensure that there is no local ground hardening or blockage affecting the data collection of the sensor. When it is confirmed that the detection error is within the normal range, add up all the deviation values one by one and perform an index conversion to obtain the matching degree index of the entire field. Immediately compare this matching degree index with the previously determined irrigation parameters. If the matching degree index continues to be low, check whether there are extreme temperatures, heavy precipitation, or other abnormal crop growth conditions during this period, and adjust the irrigation water volume and irrigation duration in the next stage according to the specific situation. Finally, summarize these processes into a systematic record to facilitate historical tracing when calling soil humidity data in subsequent stages, complete the verification of the matching degree index and the analysis of its coincidence with the adjusted irrigation parameters, and thus save the ideal humidity, actual humidity, and corresponding difference results in the same comparison to obtain more comprehensive soil humidity dynamic data.

[0135] When further optimizing the operating parameters of the water pump and the irrigation plan based on the previously obtained matching degree index, it is necessary to first extract the change trend of the matching degree index at different dates or time periods, list its differences during high-temperature periods, precipitation periods or night periods one by one, and make corresponding arrangements with the previously configured irrigation duration, irrigation flow rate, etc. If it is found that the matching degree index increases significantly at night or in rainy weather, it may mean that there is redundancy in the existing irrigation parameters in some periods, or it may indicate that the soil has better water retention in these periods. At this time, fine-tune the operating power of the water pump and lower the power to a smaller range to avoid wasting excess water and ensure uniform humidity distribution between detection points. Then, check the rotation irrigation sequence and duration of each field block in the irrigation plan. If the matching degree index of some field blocks is high for a long time, it means that they are in a state closer to the ideal humidity, and the water supply period can be appropriately reduced. If it is found that the matching degree index of some field blocks continuously decreases in dry weather, it means that they require more precise irrigation scheduling. By comparing the historical meteorological records and the actual needs of the crops one by one, increase the water supply period of this field block or increase the water pump output rate in the existing plan. After the matching degree indexes of all field blocks are uniformly updated and recorded again, incorporate these updated values into the subsequent irrigation data statistics, and finally generate an optimized irrigation configuration and compare it with the previous plan for reference.

Claims

1. A variable fertilization control system for tobacco fields, characterized in that: The system comprises: The sensing data acquisition module collects soil moisture data through a soil moisture sensor and monitors crop growth data to synthesize real-time soil and growth data; performs time series analysis on the real-time soil and growth data to identify data trends and abnormal points and generate trend analysis results; a pest and disease analysis module, receiving the trend analysis result, analyzing the pest and disease signs in the crop image through image recognition and convolutional neural network, identifying the pest and disease type, and generating a pest and disease identification result; evaluating the probability of pest and disease occurrence according to the pest and disease identification result, and generating a pest and disease occurrence probability evaluation result; The irrigation control module adjusts the irrigation amount according to the pest occurrence probability assessment result, optimizes the water pump rate and switching time to match the current crop demand and the predicted pest risk, and generates adjusted irrigation parameters; combines the matching degree of the adjusted irrigation parameters with the real-time soil moisture data, optimizes water distribution, and obtains the optimal irrigation configuration; A fertilization decision module, based on the irrigation optimization configuration, collects real-time soil moisture data, determines the need for chemical fertilizer adjustment, and selects the type and amount of chemical fertilizer to be applied based on the growth requirements of the crop and the chemical properties of the soil; The steps for obtaining the adjusted irrigation parameters are: Analyze the probability of each type of pests and diseases provided by the pest and disease occurrence probability assessment results, and conduct an impact analysis on irrigation demand based on the probability, including investigation of crop growth cycle and water demand, to determine the range of initial irrigation volume to be adjusted; Based on the preliminary irrigation volume range, calculate the pump rate using the formula: ； in, is the adjusted pump speed, is the basic pump speed, is the amount of water currently required by crops, is the standard water volume, is the pest risk adjustment factor, Derived from the previous assessment results of the probability of occurrence of pests and diseases; Based on the water pump rate, the irrigation parameters are reconfigured to obtain adjusted irrigation parameters.

2. The variable fertilization control system for tobacco fields according to claim 1, characterized in that: The steps for obtaining the real-time soil and growth data are as follows: Soil moisture sensors are used to monitor soil moisture in tobacco fields, and cameras are used to capture crop growth status to obtain preliminary soil moisture data and crop growth data. Based on the preliminary soil moisture data and crop growth data, aligning and merging time stamps to form a unified data set; Based on the unified data set, the interaction between soil moisture and crop growth is analyzed, and real-time soil and growth data are generated through fusion processing.

3. The variable fertilization control system for tobacco fields according to claim 1, characterized in that: The steps for obtaining the trend analysis results are: Based on the real-time soil and growth data, statistical analysis is performed, including calculation of mean, standard deviation and extreme value, to obtain statistical characteristic analysis results of the data; Based on the statistical characteristics analysis results of the data, time series analysis is performed to calculate the comprehensive score at each time point. The calculation formula is: ； in, is the comprehensive score at time t, is the observed value at time t, and are the mean and standard deviation of the data set, respectively. is the moving average, is the standard deviation of the moving average, and is the weight coefficient; Based on the comprehensive score, the overall trend is evaluated and abnormal points are identified to obtain trend analysis results.

4. The variable fertilization control system for tobacco fields according to claim 1, characterized in that: The steps for obtaining the pest identification result are as follows: Using the trend analysis results, selecting crop images for pest and disease sign analysis to obtain selected crop images; Based on the selected crop image, deploy a convolutional neural network to perform image processing, identify pest and disease characteristics in the image, and obtain a pest and disease characteristic analysis result in the image; Based on the analysis results of the pest and disease characteristics in the image, the pest and disease types are classified, and by comparing with an established pest and disease characteristic database, the pest and disease types are identified and calibrated to obtain pest and disease identification results.

5. The variable fertilization control system for tobacco fields according to claim 1, characterized in that: The steps for obtaining the pest occurrence probability assessment result are as follows: Receiving the pest identification result, including the type of each pest and the distribution characteristics in the image, and obtaining pest distribution data; Based on the pest distribution data, the occurrence probability of various pests is calculated, and the calculation formula is: ； in, is the occurrence probability of the i-th pest, Represents the coverage area of ​​pests and diseases in the identification results, Represents the degree of impact, represents temperature sensitivity, Represents seasonality, is the adjustment coefficient, n is the total number of pests and diseases; Based on the occurrence probability, a risk assessment is conducted, and combined with environmental factors and past data, an assessment result of the occurrence probability of pests and diseases is obtained.

6. The variable fertilization control system for tobacco fields according to claim 1, characterized in that: The steps for obtaining the optimal irrigation configuration are: Based on the adjusted irrigation parameters and soil moisture data, the matching index between the two is calculated, and the calculation formula is: ； in, is the matching index, is the ideal soil moisture at the i-th detection point, is the soil moisture at the i-th detection point, is the number of detection points; The operating parameters of the water pump and the irrigation plan are optimized according to the matching index to obtain an optimal irrigation configuration.

7. A variable fertilization control method for tobacco fields, characterized in that: The variable fertilization control system for tobacco fields according to any one of claims 1 to 6 comprises the following steps: Soil moisture data is collected through soil moisture sensors, and crop growth data is collected at the same time. Time series analysis is performed on the collected data to identify data trends and anomalies, and obtain comprehensive analysis results of soil and growth data; Based on the comprehensive analysis results of soil and growth data, image recognition and convolutional neural networks are used to analyze the signs of pests and diseases in crop images, identify the types of pests and diseases, assess the probability of pest and disease occurrence, and generate pest and disease identification and risk assessment results; Based on the results of pest and disease identification and risk assessment, adjust the irrigation amount, optimize the pump rate and switching time to match the current crop demand and predicted pest and disease risks, and generate adjusted irrigation parameters; Combine the adjusted irrigation parameters with the real-time soil moisture data to optimize water distribution and obtain the optimal irrigation configuration result; Using the irrigation optimization configuration results, the type and application amount of chemical fertilizers are selected according to the growth requirements of crops and the chemical properties of the soil.

Images