Automatic fertilization and watering control method and system for landscape garden seedlings

Abstract

The invention relates to the technical field of fertilization and watering control, in particular to an automatic fertilization and watering control method and system for landscape garden seedlings. The method comprises the following steps: acquiring and analyzing morphological data and soil state characteristics of landscape garden seedlings through an electronic monitoring and soil monitoring sensor; then, the nursery stock root system density is predicted through the nursery stock form data, abnormal fluctuation of the soil salt content is monitored, and the root system absorption limitation condition is estimated; then, according to root system absorption limited data and soil salt content abnormal fluctuation, the dosage of a slow release fertilizer is matched, and sectional type drip irrigation proportion fusion is carried out; and finally, constructing a fertilization and watering control model based on a strategy gradient algorithm, and implementing automatic fertilization and watering control through a terminal. The fertilization and watering control technology is optimized, so that the fertilization and watering control technology is more perfect.

Description

Technical Field

[0001] The present invention relates to the technical field of fertilization and watering control, and in particular to a method and system for controlling automatic fertilization and watering of landscape garden seedlings. Background Art

[0002] By real-time monitoring of seedling morphology, soil characteristics and environmental changes, more accurate data can be obtained to help achieve precise fertilization and watering control. In particular, with the support of multi-dimensional data such as soil moisture, temperature, and salt content, it can effectively avoid human operation errors, reduce resource waste, and improve the scientificity and stability of plant growth. Factors such as the root growth of seedlings, soil moisture and nutrient supply will affect the health of plants, and the previous manual judgment and regular fertilization and watering cannot respond to the real-time needs of plants in a timely manner. Through the automatic control system, combined with the monitoring of soil, air and plant conditions by intelligent sensors, the amount of fertilization and watering can be accurately adjusted according to the actual needs of seedlings. For example, abnormal fluctuations in soil salt content may affect the root absorption capacity, and the intelligent system can adjust the amount of fertilizer according to these changes to avoid soil salt accumulation caused by excessive fertilization, thereby achieving the purpose of precise regulation. In addition, the intelligent control system can realize segmented drip irrigation and fertilizer delivery to meet the needs of different plants and maximize maintenance efficiency and quality. However, a traditional method for controlling the automated fertilization and watering of landscape seedlings has the problem of inaccurate analysis of the limited root nutrient absorption caused by soil salinization, which results in large errors in automated fertilization and watering. Summary of the invention

[0003] Based on this, it is necessary to provide a method and system for controlling the automated fertilization and watering of landscape seedlings to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above purpose, a method for controlling automatic fertilization and watering of landscape garden seedlings is provided, the method comprising the following steps: Step S1: collecting seedling morphological data of landscape garden seedlings through electronic monitoring to obtain seedling morphological data; deploying soil monitoring sensors under each landscape garden seedling, and monitoring soil state characteristics through soil detection sensors to obtain soil state characteristic monitoring characteristic data; Step S2: predicting the density of seedling root systems based on seedling morphology data to generate seedling root density prediction data; analyzing abnormal fluctuations in soil salinity on soil state characteristic monitoring feature data to obtain abnormal fluctuations in soil salinity; estimating root absorption limitation on seedling root density prediction data based on abnormal fluctuations in soil salinity to obtain root absorption limitation estimation data; Step S3: According to the root absorption limitation estimation data, the abnormal fluctuation data of soil salt content is matched with the slow-release fertilizer dosage control to obtain the slow-release fertilizer dosage control data; based on the slow-release fertilizer dosage control data, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio of fertilization and watering; Step S4: construct a fertilization and watering control model for the segmented drip irrigation ratio of fertilization and watering based on the strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; send the garden seedling fertilization and watering control model to the terminal to execute automated fertilization and watering control of landscape seedlings.

[0005] Preferably, step S1 comprises the following steps: Step S11: collecting seedling morphological data of landscape garden seedlings through electronic monitoring to obtain seedling morphological data; Step S12: deploying soil monitoring sensors under each landscape garden seedling, and monitoring soil state characteristics through the soil detection sensors to obtain soil state characteristic monitoring data; Step S13: performing data cleaning on the soil state characteristic monitoring data to obtain soil state characteristic monitoring cleansing data; Step S14: performing state characteristic feature analysis on the soil state characteristic monitoring cleaning data to obtain soil state characteristic monitoring characteristic data.

[0006] Preferably, step S2 comprises the following steps: Step S21: predicting the density of seedling root systems based on the seedling morphology data to generate seedling root density prediction data; Step S22: analyzing abnormal fluctuation of soil salt content on the soil state characteristic monitoring characteristic data to obtain abnormal fluctuation data of soil salt content; Step S23: analyzing the soil nutrient imbalance effectiveness loss on the soil state characteristic monitoring characteristic data based on the soil salt content abnormal fluctuation data to obtain soil nutrient imbalance effectiveness loss data; Step S24: performing root absorption limitation estimation on the seedling root density prediction data according to the soil salt content abnormal fluctuation data and the soil nutrient imbalance effectiveness loss data to obtain root absorption limitation estimation data.

[0007] Preferably, step S23 includes the following steps: Step S231: performing ion concentration decomposition processing on the abnormal fluctuation data of soil salt content to generate time series variation data of salt ion concentration; Step S232: extracting organic matter content fluctuation from soil state characteristic monitoring feature data to obtain soil organic matter content fluctuation data; Step S233: performing nonlinear intensity regression analysis of cell membrane damage based on the salt ion concentration time series variation data to obtain cell membrane damage intensity regression data; Step S234: performing an equal amount simulation prediction of organic nutrient leakage on the soil organic matter content fluctuation data according to the cell membrane damage intensity regression data to obtain equal amount prediction data of nutrient leakage; Step S235: performing ion competition intensity increment analysis on soil organic matter content fluctuation data according to salt ion concentration time series variation data to generate ion competition intensity increment data; Step S236: Perform soil nutrient imbalance effectiveness loss analysis based on the nutrient leakage equal amount prediction data and the ion competition intensity increment data to obtain soil nutrient imbalance effectiveness loss data.

[0008] Preferably, step S235 includes the following steps: Perform mutation increment index calculation on the time series variation data of salt ion concentration to obtain the mutation increment index of salt ion concentration; Based on the salt ion concentration mutation increment index, the salt ion concentration time series change data is combined by continuous concentration mutation approximation to generate concentration continuous mutation approximation combination data; Based on the concentration continuity mutation approximation combination data, the soil organic matter content fluctuation data was simulated by the preemptive continuous intensity of salt ion absorption sites, and the preemptive continuous intensity data of salt ion absorption sites were obtained. The preemptive continuous intensity data of the absorption site are processed by calculating the time interval difference of the ion flux to obtain the difference of the salt ion flux at the absorption site; The ion competition intensity increment analysis is performed based on the difference in salt ion flux at the absorption site to generate ion competition intensity increment data.

[0009] Preferably, step S24 comprises the following steps: Step S241: performing fluctuation amplitude deviation rate calculation on the abnormal fluctuation data of soil salt content to obtain salt content fluctuation amplitude difference data; Step S242: converting and quantifying the soil solution osmotic pressure increase ratio according to the salt content fluctuation amplitude difference data to obtain the soil solution osmotic pressure increase ratio; Step S243: integrating the fluctuation of the microbial nutrient conversion inhibition value based on the soil solution osmotic pressure increase ratio to obtain the nutrient conversion inhibition fluctuation value; Step S244: estimating the soil compaction cumulative gradient according to the salt content fluctuation amplitude difference data and the soil solution osmotic pressure increase ratio to obtain soil compaction cumulative gradient data; Step S245: performing root growth / respiration restriction gradient evaluation on the seedling root density prediction data based on the soil compaction cumulative gradient data to obtain root growth / respiration restriction gradient data; Step S246: Root absorption limitation estimation is performed on the seedling root density prediction data according to the nutrient conversion inhibition fluctuation value, root growth / respiration restriction gradient data and soil nutrient imbalance effectiveness loss data to obtain root absorption limitation estimation data.

[0010] Preferably, step S3 comprises the following steps: Step S31: performing slow-release fertilizer dosage control matching on the abnormal fluctuation data of soil salt content according to the root absorption limitation estimation data to obtain slow-release fertilizer dosage control data; Step S32: performing dosage control logic learning on the slow-release fertilizer dosage control data to obtain slow-release fertilizer dosage control learning data; Step S33: Based on the slow-release fertilizer dosage control learning data and the root absorption limitation estimation data, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio for fertilization and watering.

[0011] Preferably, step S33 includes the following steps: Step S331: Deducing the respiratory conversion demand space based on the root absorption limitation estimation data to obtain the root respiratory conversion demand space; Step S332: performing nutrient absorption rate interval calculation per unit time on the root absorption limitation estimation data to obtain the root nutrient absorption rate interval per unit time; Step S333: analyzing the fertilizer effect release rate of the slow-release fertilizer dosage control learning data to obtain the fertilizer effect dosage release rate; Step S334: dividing the fertilizer efficiency dosage release rate into dosage absorption extreme value points per unit time according to the root respiration conversion demand space and the root nutrient absorption rate interval to obtain root fertilizer efficiency release absorption extreme value data; Step S335: performing an absorption-release load ratio analysis on the root fertilizer efficiency release absorption extreme value data to obtain the root fertilizer efficiency absorption-release load ratio; Step S336: Based on the root fertilizer absorption and release load ratio, the root nutrient absorption rate interval and the root respiration conversion demand space, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio for fertilization and watering.

[0012] Preferably, step S4 comprises the following steps: Step S41: normalizing the fertilization and watering segmented drip irrigation ratio to obtain a fertilization and watering segmented drip irrigation normalized ratio; Step S42: constructing a fertilization and watering control model for the segmented drip irrigation normalized ratio of fertilization and watering based on a strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; Step S43: Sending the garden seedling fertilization and watering control model to the terminal to execute the landscape garden seedling automated fertilization and watering control.

[0013] Preferably, the present invention also provides a landscape garden seedling automatic fertilization and watering control system, which is used to execute the landscape garden seedling automatic fertilization and watering control method as described above, and the landscape garden seedling automatic fertilization and watering control system comprises: The soil state characteristic monitoring module is used to collect the seedling morphology data of landscape garden seedlings through electronic monitoring to obtain the seedling morphology data; deploy soil monitoring sensors under each landscape garden seedling, and monitor the soil state characteristics through soil detection sensors to obtain soil state characteristic monitoring characteristic data; The root absorption limitation estimation module is used to predict the density of seedling root systems based on seedling morphological data and generate seedling root density prediction data; to analyze the abnormal fluctuation of soil salt content on the soil state characteristic monitoring characteristic data and obtain the abnormal fluctuation data of soil salt content; to estimate the root absorption limitation of the seedling root density prediction data based on the abnormal fluctuation data of soil salt content and obtain the root absorption limitation estimation data; The segmented drip irrigation ratio fusion module is used to control and match the abnormal fluctuation data of soil salt content with the slow-release fertilizer dosage according to the root absorption limitation estimation data to obtain the slow-release fertilizer dosage control data; the segmented drip irrigation ratio is fused based on the slow-release fertilizer dosage control data to obtain the segmented drip irrigation ratio for fertilization and watering; The fertilization and watering control model construction module is used to construct a fertilization and watering control model for the fertilization and watering segmented drip irrigation ratio based on the strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; the garden seedling fertilization and watering control model is sent to the terminal to execute automated fertilization and watering control of landscape seedlings.

[0014] The beneficial effect of the present invention is that the morphological data and soil state characteristics of landscape garden seedlings are collected by electronic monitoring and soil monitoring sensors, and the detailed information of the growth state of the seedlings and the soil environment in which they are located can be accurately obtained. The seedling morphological data obtained by electronic monitoring provides a basis for the subsequent root density prediction, and the soil data provided by the soil monitoring sensor can help to grasp the important characteristics such as soil moisture, temperature, and salinity in real time, ensure that the entire maintenance process is more accurate, avoid errors caused by manual intervention, and improve the maintenance effect and resource utilization efficiency. Based on the seedling morphological data, the root density is predicted, and the distribution and density of the root system can be estimated by scientifically analyzing the seedling morphological characteristics, thereby providing a basis for the evaluation of the root absorption status. At the same time, after the abnormal fluctuation analysis of the salt content, the soil state characteristic monitoring data can identify the salt changes in the soil and help determine whether the soil has unfavorable conditions such as salinization. Combined with these data, the root absorption restriction estimation can effectively grasp the growth status of the seedlings, provide a scientific basis for subsequent fertilization and watering, and avoid excessive or insufficient fertilization. By using the root absorption restriction estimation data to control the abnormal fluctuation of soil salt content, the amount of slow-release fertilizer can be matched. According to the soil salt content and the root absorption status, the amount of fertilizer to be applied can be accurately calculated to avoid excessive salt from damaging the root system. The segmented drip irrigation ratio fusion technology can more reasonably manage water and fertilizer integration through the precise control of the amount of slow-release fertilizer, ensuring that seedlings can grow healthily in a suitable environment, while avoiding the waste of resources and improving the scientificity and efficiency of garden maintenance. Based on the policy gradient algorithm, a fertilization and watering control model is constructed. Through the intelligent algorithm, the ratio of fertilization and watering can be automatically adjusted to ensure that the needs of seedlings at different growth stages are accurately met. The policy gradient algorithm continuously optimizes and adjusts the strategy, improves the adaptability and decision-making accuracy of the system, and realizes dynamic management and real-time feedback. This automated control model can not only greatly improve the efficiency of garden fertilization and watering, but also achieve environmentally friendly goals such as water and fertilizer saving, and can be executed through terminal equipment to achieve accurate and automated maintenance of landscape garden seedlings. Therefore, the present invention is an optimization treatment of a traditional landscape seedling automatic fertilization and watering control method, which solves the problem of inaccurate analysis of the limited root nutrient absorption caused by soil salinization in the traditional landscape seedling automatic fertilization and watering control method, thereby causing large errors in automatic fertilization and watering, improves the accuracy of the analysis of the limited root nutrient absorption caused by soil salinization, and reduces the errors caused by automatic fertilization and watering. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of the steps of a method for controlling the automatic fertilization and watering of landscape garden seedlings; Figure 2 for Figure 1Detailed implementation steps of step S2 in the flowchart; Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG. DETAILED DESCRIPTION

[0016] See also Figures 1 to 3 , a landscape gardening seedling automatic fertilization and watering control method, the method comprising the following steps: Step S1: collecting seedling morphological data of landscape garden seedlings through electronic monitoring to obtain seedling morphological data; deploying soil monitoring sensors under each landscape garden seedling, and monitoring soil state characteristics through soil detection sensors to obtain soil state characteristic monitoring characteristic data; Step S2: predicting the density of seedling root systems based on seedling morphology data to generate seedling root density prediction data; analyzing abnormal fluctuations in soil salinity on soil state characteristic monitoring feature data to obtain abnormal fluctuations in soil salinity; estimating root absorption limitation on seedling root density prediction data based on abnormal fluctuations in soil salinity to obtain root absorption limitation estimation data; Step S3: According to the root absorption limitation estimation data, the abnormal fluctuation data of soil salt content is matched with the slow-release fertilizer dosage control to obtain the slow-release fertilizer dosage control data; based on the slow-release fertilizer dosage control data, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio of fertilization and watering; Step S4: construct a fertilization and watering control model for the segmented drip irrigation ratio of fertilization and watering based on the strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; send the garden seedling fertilization and watering control model to the terminal to execute automated fertilization and watering control of landscape seedlings.

[0017] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a method for controlling the automatic fertilization and watering of landscape garden seedlings of the present invention. In this example, the method for controlling the automatic fertilization and watering of landscape garden seedlings includes the following steps: Step S1: collecting seedling morphological data of landscape garden seedlings through electronic monitoring to obtain seedling morphological data; deploying soil monitoring sensors under each landscape garden seedling, and monitoring soil state characteristics through soil detection sensors to obtain soil state characteristic monitoring characteristic data; In the embodiment of the present invention, the morphological data of landscape seedlings are collected by a 4K high-definition camera installed 3 meters above the nursery area. The camera takes pictures once every 2 hours, the shooting angle is 45° looking down, and the resolution is set to 3840×2160 pixels. The obtained image is processed by an image segmentation algorithm to extract three key morphological indicators of seedling height, crown diameter, and leaf color index to form a seedling morphological data matrix; at the same time, a plurality of soil monitoring sensors are arranged in a regular pentagon around the root of each landscape seedling. The sensor is buried at a depth of 15 cm, and the distance between each sensor is 25 cm. The sensor collects soil pH value (precision) every 30 minutes. The collected raw data are transmitted to the data processing terminal via the wireless transmission module using the ZigBee protocol; the data processing terminal first removes outliers based on the 3σ principle, then performs linear interpolation to fill in missing data points, and then removes high-frequency noise through wavelet transform, and finally extracts the temporal fluctuation characteristics, spatial distribution characteristics, and correlation characteristics of soil state characteristics, forming a soil state characteristic monitoring feature data set containing 15 feature dimensions, which provides a data basis for subsequent root system prediction and soil analysis.

[0018] Step S2: predicting the density of seedling root systems based on seedling morphology data to generate seedling root density prediction data; analyzing abnormal fluctuations in soil salinity on soil state characteristic monitoring feature data to obtain abnormal fluctuations in soil salinity; estimating root absorption limitation on seedling root density prediction data based on abnormal fluctuations in soil salinity to obtain root absorption limitation estimation data; In the embodiment of the present invention, firstly, based on the seedling morphological data obtained in step S1, the morphological-root system association algorithm is applied to calculate the density of the seedling root system. The algorithm uses the ratio of the crown diameter to the seedling height as the root lateral expansion index (value range 0.5-2.5), and the leaf color index as the root vitality coefficient (value range 0.6-1.2), and the product of the two is multiplied by the root system coefficient corresponding to the seedling type (1.0 for trees, 1.2 for shrubs, and 0.8 for vines) to obtain a predicted value of the root system density (value range 0.24-3.0); at the same time, the conductivity data in the soil state characteristic monitoring characteristic data is subjected to time-frequency analysis, and the main frequency and amplitude of the conductivity fluctuation are extracted by Fourier transform. When the fluctuation amplitude exceeds 25% of the baseline value and the duration exceeds 72 hours, it is determined that the soil salt content fluctuates abnormally, and the abnormal fluctuation index (fluctuation amplitude percentage × duration / 24) is calculated; then the ion concentration is decomposed and the conductivity is converted into The rate data were decomposed into ion concentration time series data according to the contribution rates of the five main ions Na+, Cl-, K+, Ca2+, and Mg2+ (0.35, 0.30, 0.15, 0.12, and 0.08, respectively). Then, the cell membrane damage intensity was calculated based on the ion concentration data, and the exponential regression equation D=0.15×e^(0.28×S) (D is the damage intensity, S is the salinity value) was used. At the same time, the nutrient leakage was analyzed based on the fluctuation data of soil organic matter content, and the linear relationship model L=0.2×D×O (L is the leakage, D is the damage intensity, and O is the organic matter content) was used. Taking into account the predicted data of root density, the incremental data of ion competition intensity (calculated by the mutation detection algorithm), and the soil nutrient imbalance data, the weighted summation method (weights were 0.4, 0.3, and 0.3, respectively) was used to obtain the estimated data of root absorption restriction, with a value range of 0-1, where 0 means no restriction and 1 means complete restriction.

[0019] Step S3: According to the root absorption limitation estimation data, the abnormal fluctuation data of soil salt content is matched with the slow-release fertilizer dosage control to obtain the slow-release fertilizer dosage control data; based on the slow-release fertilizer dosage control data, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio of fertilization and watering; In the embodiment of the present invention, based on the root absorption limitation estimation data and the soil salt content abnormal fluctuation data obtained in step S2, the slow-release fertilizer dosage control matching calculation is performed, and a piecewise linear function is specifically used: when the root absorption limitation estimation value R<0.3, the slow-release fertilizer dosage is the reference amount (0.5kg / m²)×(1-R / 0.3); when 0.3≤R<0.7, the slow-release fertilizer dosage is the reference amount×0.5×(0.7-R) / 0.4; when R≥0.7, the slow-release fertilizer dosage is 0; at the same time, combined with the fluctuation index W in the soil salt content abnormal fluctuation data, when W< When W is 10, the amount of slow-release fertilizer remains unchanged; when 10≤W<30, the amount of slow-release fertilizer is reduced by W-10%; when W≥30, the amount of slow-release fertilizer is reduced by 20%; the minimum value of the two is taken as the final slow-release fertilizer dosage control data; then the slow-release fertilizer dosage control data is used for dosage control logic learning, and a decision tree is constructed using historical data. The input features are the estimated value of root absorption limitation, the abnormal fluctuation index of soil salt content, the seasonal factor (spring 1.2, summer 1.0, autumn 0.8, winter 0.6) and the seedling type factor (tree 1.0, shrub 0.8, vine 1.2), and the output is the slow-release fertilizer dosage control data. Fertilizer dosage correction coefficient K (value range 0.7-1.3); then calculate the respiratory conversion demand space T=0.8×(1-R)^2 (T is respiratory oxygen demand, R is the limited estimated value) based on the root absorption restriction estimate data, calculate the nutrient absorption rate interval A=[0.05×(1-R), 0.15×(1-R)] per unit time (A is the absorption rate interval, unit is mg / h·g root), and calculate the fertilizer efficiency release rate F=0.025×K (F is the release rate, unit is mg / h·g fertilizer); finally, T, A, and F are added together. According to the values ​​of 365 time points in a 24-hour period obtained by time series interpolation, the drip irrigation fertigation concentration ratio B=F / A median at each time point is determined according to the ratio of A to F, and a segmented setting is performed: 24 hours are divided into 6 time periods, namely 0:00-4:00, 4:00-8:00, 8:00-12:00, 12:00-16:00, 16:00-20:00, and 20:00-24:00. The average value of the B value in each time period is taken as the drip irrigation ratio value of that period, forming a segmented drip irrigation ratio plan for fertilization and watering.

[0020] Step S4: construct a fertilization and watering control model for the segmented drip irrigation ratio of fertilization and watering based on the strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; send the garden seedling fertilization and watering control model to the terminal to execute automated fertilization and watering control of landscape seedlings.

[0021] In the embodiment of the present invention, the fertilization and watering segmented drip irrigation ratio obtained in step S3 is first normalized, and the drip irrigation ratio value of each period is mapped to the [0.1,1] interval using the Min-Max normalization method. The formula is B'=(B-Bmin) / (Bmax-Bmin)×0.9+0.1 (B' is the normalized ratio value, B is the original ratio value, Bmin is the minimum ratio value, and Bmax is the maximum ratio value). The fertilization and watering segmented drip irrigation normalized ratio is obtained; then, based on the policy gradient algorithm, the The fertilization and watering control model was constructed. The algorithm first defined the state space as four dimensions: the current time period index, soil moisture, soil conductivity, and root absorption limit estimation value. The action space was two dimensions: drip irrigation switch state (0 or 1) and drip irrigation flow size (ranging from 10% to 100% of the baseline flow). The reward function was the matching degree between root absorption and release (calculated as 1-|Aactual-Factual| / max(Aactual,Factual)). Then a two-layer neural network was constructed as a policy function approximator. The first layer contained 16 The second layer contains 8 neurons with ReLU as the activation function, and the output layer contains 2 neurons corresponding to the two dimensions of the action space respectively. Through 500 rounds of training in the simulation environment, each round contains 24 periods, the learning rate is set to 0.01, and the discount factor is set to 0.95, the fertilization and watering control strategy π(a|s) (indicating the probability of selecting action a under state s) is obtained. The strategy function is converted into a decision table form, which contains all discrete combinations of the state space (6 types of period indexes, 5 levels of soil moisture, 5 levels of soil conductivity, and 5 levels of root absorption limitation estimation) and the corresponding optimal actions, a total of 750 decision rules, which constitute the garden seedling fertilization and watering control model. Finally, the control model is packaged into a binary file and transmitted to the drip irrigation control terminal equipment on site through the 4G network. The control terminal executes the corresponding drip irrigation control instructions according to the real-time monitoring data to complete the automatic fertilization and watering control of landscape garden seedlings. The control terminal detects the environmental status every 10 minutes and updates the control decision to achieve all-weather precision fertilization and watering.

[0022] Step S1 includes the following steps: Step S11: collecting seedling morphological data of landscape garden seedlings through electronic monitoring to obtain seedling morphological data; Step S12: deploying soil monitoring sensors under each landscape garden seedling, and monitoring soil state characteristics through the soil detection sensors to obtain soil state characteristic monitoring data; Step S13: performing data cleaning on the soil state characteristic monitoring data to obtain soil state characteristic monitoring cleansing data; Step S14: performing state characteristic feature analysis on the soil state characteristic monitoring cleaning data to obtain soil state characteristic monitoring characteristic data.

[0023] In an embodiment of the present invention, seedling morphology data is collected through an electronic monitoring system set up in a landscape garden seedling planting area. The electronic monitoring system consists of 8 high-definition digital cameras fixed on a 3.5-meter-high bracket. Each camera covers an area of ​​100 square meters. The camera resolution is 2560×1440 pixels and the frame rate is 25 frames per second. The camera uses a 120° wide-angle lens and is equipped with an infrared fill light device to meet the needs of all-weather monitoring. The seedlings are photographed at four fixed time points at 6:00, 10:00, 14:00, and 18:00 every day to obtain image data of 8 angles of the top view and side view of the seedlings; at the same time, the system is also equipped with a multispectral imaging device, which uses 5 bands (blue light 450nm, green light 550nm, The multispectral images are collected once a week with a resolution of 1280×960 pixels. All image data are transmitted to the data processing server in real time through the 5G network. After being processed by the image segmentation algorithm, morphological parameters such as seedling height (accurate to 0.1 cm), crown diameter (accurate to 0.5 cm), stem thickness (accurate to 0.05 cm), number of leaves (accurate to 10 leaves), leaf color index (RGB value range 0-255), and crown volume (accurate to 0.01 cubic meter) are extracted to form a seedling morphological data matrix containing 6 dimensions and 4 time points. Each element in the data matrix is ​​marked with a timestamp to track changes in the growth status of the seedlings.The soil area under each landscape seedling is deployed with detection sensors and data monitoring. Specifically, a soil monitoring sensor network is arranged within the root range of each seedling (usually 1.2 times the crown projection area). The network consists of a central main sensor and four sub-sensors, which are distributed in a "cross" shape. The central main sensor is located 10 cm below the trunk of the seedling, and the four sub-sensors are located 30 cm in the north, east, south and west directions of the trunk. The sensor is buried at two levels of 5 cm and 20 cm. The soil monitoring sensor used is a multi-parameter integrated digital sensor module with a sensor probe length of 25 cm and a diameter of 2 cm. It measures soil pH (measurement range 3-9, accuracy ±0.1), soil moisture (measurement range 5%- The sensor collects data every 15 minutes according to the preset program. The data is transmitted to the on-site data collector via the ZigBee wireless transmission protocol (2.4GHz frequency band, transmission rate 250kbps, transmission distance 100 meters). The data collector uploads the data packet to the cloud server for storage via the 4G network. Within 24 hours, 5 parameter data are collected at 96 time points and 10 spatial points, forming a soil status characteristic monitoring data set with 4,800 original data points per seedling per day.

[0024] The soil state characteristic monitoring data obtained in step S12 is subjected to comprehensive data cleaning processing. First, a data integrity check is performed, and the data quality is evaluated by detecting the data missing rate (the calculation method is the number of missing data points / total number of data points × 100%). When the missing rate is lower than 5%, the linear interpolation method is used to fill in the missing data. The specific operation is to take the monitoring values ​​of the two time points before and after the missing point to calculate the average change rate, and then calculate the interpolation value according to the time interval; when the missing rate is between 5% and 15%, the polynomial interpolation method is used, and the monitoring values ​​of the three time points before and after the missing point are taken to fit the cubic polynomial curve, and the function value corresponding to the missing time point is calculated; when the missing rate exceeds 15%, the monitoring point is marked as an invalid point; then an outlier detection is performed, and the 3σ criterion is used (that is, when the data deviates from the mean by more than 3 times the standard deviation, it is judged to be abnormal). The calculation method is to first calculate the number of consecutive 24 hours of the same parameter and the same spatial point. The mean μ and standard deviation σ of the data are then determined, and each data point x is tested to see if |x-μ|>3σ. The detected outliers are replaced by median filtering, and the replacement value is the median of the two time points before and after the outlier. Then, data smoothing is performed using the sliding window averaging method, with the window size set to 5 time points (i.e., 75 minutes), and the arithmetic mean of the data in the window is taken as the smoothing value of the center point. Finally, data normalization is performed, and the Min-Max normalization method is used for different parameters to map the data to the [0,1] interval. The calculation formula is x'=(x-xmin) / (xmax-xmin), where x is the original data, xmin and xmax are the minimum and maximum values ​​of the parameter, respectively, and x' is the normalized data value. After the above four steps, the soil state characteristic monitoring cleaning data obtained is more regular and coherent, laying the foundation for subsequent feature analysis.The soil state characteristic monitoring cleaning data obtained in step S13 is subjected to multi-dimensional feature extraction and analysis. First, time series feature analysis is performed to extract the time variation characteristics of each monitoring parameter, including five time series features: intraday fluctuation amplitude (maximum value minus minimum value), intraday volatility (fluctuation amplitude / average value × 100%), peak value occurrence time point (value range 0-23 hours), valley value occurrence time point (value range 0-23 hours), and fluctuation periodicity (calculated by autocorrelation function, value range 0-1); then spatial distribution feature analysis is performed to calculate the spatial distribution characteristics of 10 spatial point data, including three spatial features: horizontal gradient (parameter difference / distance between monitoring points in the east-west direction, unit is parameter unit / meter), vertical gradient (parameter difference / depth difference between monitoring points at different depths at the same location, unit is parameter unit / meter), and spatial non-uniformity (standard deviation / mean × 100%); then parameter correlation is performed Analysis, the correlation between different monitoring parameters was calculated by the Pearson correlation coefficient, which was calculated by dividing the covariance of two parameters by the product of the standard deviations of the two parameters. The obtained correlation coefficient matrix reflects the correlation strength between the five parameters of pH value, moisture content, temperature, conductivity, and organic matter content (value range -1 to 1); finally, all the extracted features were integrated to form the soil state characteristic monitoring feature data of each seedling, which included time series characteristics (5 parameters × 5 time series characteristics = 25 feature dimensions), spatial distribution characteristics (5 parameters × 3 spatial characteristics = 15 feature dimensions), parameter correlation characteristics (5 parameters combined in pairs for a total of 10 pairs × 1 correlation coefficient = 10 feature dimensions), a total of 50 feature vectors of feature dimensions, each dimension in the feature vector was given a clear physical meaning, which provided comprehensive and accurate feature data support for the subsequent prediction of seedling root density and analysis of abnormal fluctuations in soil salinity.

[0025] Step S2 includes the following steps: Step S21: predicting the density of seedling root systems based on the seedling morphology data to generate seedling root density prediction data; Step S22: analyzing abnormal fluctuation of soil salt content on the soil state characteristic monitoring characteristic data to obtain abnormal fluctuation data of soil salt content; Step S23: analyzing the soil nutrient imbalance effectiveness loss on the soil state characteristic monitoring characteristic data based on the soil salt content abnormal fluctuation data to obtain soil nutrient imbalance effectiveness loss data; Step S24: performing root absorption limitation estimation on the seedling root density prediction data according to the soil salt content abnormal fluctuation data and the soil nutrient imbalance effectiveness loss data to obtain root absorption limitation estimation data.

[0026] As an example of the present invention, refer to Figure 2As shown, in this example, step S2 includes: Step S21: predicting the density of seedling root systems based on the seedling morphology data to generate seedling root density prediction data; In the embodiment of the present invention, the seedling root density is predicted based on the seedling morphological data obtained in step S1. First, the six parameters of seedling height, crown diameter, stem thickness, number of leaves, leaf color index, and crown volume in the seedling morphological data are preprocessed, and each parameter is standardized to a standard normal distribution form with a mean of 0 and a standard deviation of 1 by the Z-Score standardization method. The calculation formula is z=(x-μ) / σ, where x is the original parameter value, μ is the mean of the parameter, and σ is the standard deviation of the parameter; then a root density prediction algorithm is constructed. The algorithm is based on the correlation between morphological parameters and root development, and adopts a weighted summation calculation method, specifically: root density index , where D is the crown diameter (in centimeters), H is the height of the seedling (in centimeters), S is the stem thickness (in centimeters), L is the number of leaves (in pieces), V is the crown volume (in cubic centimeters), and C is the leaf color index (0-100 dimensionless value, converted from RGB). to is the weight coefficient, and the values ​​are 0.35, 0.25, 0.15, 0.15, and 0.10 respectively; for different types of seedlings, the species adjustment coefficient k is introduced, coniferous seedlings k=0.8, broad-leaved seedlings k=1.0, mixed seedlings k=0.9, and the final root density prediction value is k×RDI; according to the calculated root density prediction value, it is divided into five levels: very low (0-0.2), low (0.2-0.4), medium (0.4-0.6), high (0.6-0.8), and very high (0.8-1.0). At the same time, a root distribution prediction map is generated to show the root distribution density in the horizontal and vertical directions, and the density is represented by the depth of color, forming seedling root density prediction data containing numerical values ​​and visualization results.

[0027] Step S22: analyzing abnormal fluctuation of soil salt content on the soil state characteristic monitoring characteristic data to obtain abnormal fluctuation data of soil salt content; In the embodiment of the present invention, the soil state characteristic monitoring characteristic data is subjected to abnormal fluctuation analysis of soil salt content. First, parameters related to soil salt content are extracted from the soil state characteristic monitoring characteristic data, mainly including three indicators: soil conductivity, pH value and organic matter content. Then, the soil conductivity data is subjected to time series analysis. The sliding window method is adopted. The window size is set to 72 hours (i.e., 288 consecutive data points) and the window sliding step is 6 hours (i.e., 24 data points). The conductivity mean, standard deviation, change rate (first and last difference / mean × 100%) and fluctuation frequency (window) in each window are calculated. The number of peak occurrences / 3); then the salt content abnormality judgment criteria are introduced, and it is judged as an abnormal fluctuation when any of the following conditions are met: ① The conductivity continues to rise within 72 hours and the change rate exceeds 25%; ② The conductivity suddenly increases by more than 30% of the baseline value (the average value of the previous 7 days) within 24 hours; ③ The conductivity fluctuation frequency is abnormal (the number of fluctuations within 24 hours exceeds 6 times and the amplitude of each fluctuation exceeds 10%); ④ The conductivity and pH value show opposite and drastic changes (the conductivity rises and the pH value decreases, and the amplitude of both changes exceeds 20%); then the detected abnormal fluctuations are quantified and the abnormal fluctuation index is calculated. , where ΔEC is the conductivity change rate, T is the abnormal duration (in hours), F is the fluctuation frequency, and ΔpH is the pH change rate. to are weight coefficients, with values ​​of 0.4, 0.3, 0.2, and 0.1 respectively; finally, according to the abnormal fluctuation index AWI, the abnormal fluctuation is divided into four levels: mild (AWI<30), moderate (30≤AWI<60), severe (60≤AWI<90), and very severe (AWI≥90), and the abnormal fluctuation data of soil salt content is generated, including abnormal fluctuation time period, abnormal fluctuation index, fluctuation level and fluctuation trend chart, which provides a basis for the subsequent estimation of root absorption limitation.

[0028] Step S23: analyzing the soil nutrient imbalance effectiveness loss on the soil state characteristic monitoring characteristic data based on the soil salt content abnormal fluctuation data to obtain soil nutrient imbalance effectiveness loss data; In the embodiment of the present invention, the soil nutrient imbalance effectiveness loss analysis is performed on the soil state characteristic monitoring characteristic data based on the abnormal fluctuation data of soil salt content. First, the conductivity data in the abnormal fluctuation data of soil salt content is converted into salt concentration data. The conversion formula is S=k×EC, where S is the salt concentration (unit g / L), EC is the conductivity (unit mS / cm), and k is the conversion coefficient, which is 0.64; then, the salt concentration data is decomposed into ion composition, and the total salt concentration is decomposed into according to the typical soil salt composition ratio. (35%) (accounting for 27%), (accounting for 15%), (accounting for 12%), (8%) and (3%) The concentrations of six major ions are used to form the time series change data of salt ion concentration; then the organic matter content data in the soil state characteristic monitoring characteristic data is extracted to calculate the change in organic matter content within 72 consecutive hours. ,in is the organic matter content at the initial time, The organic matter content after 72 hours. When ΔOM < 0, it means that the organic matter is degraded or lost. The data of these degradation or loss periods are extracted to form the soil organic matter content fluctuation data. Then, a nonlinear regression model of cell membrane damage is established based on the time series change data of salt ion concentration, using an exponential function form. , where MD is the degree of membrane damage (a dimensionless value between 0 and 1), S is the total salt concentration, a and b are regression coefficients, which are 0.12 and 1.35 respectively. The model is used to calculate the degree of cell membrane damage under different salt concentrations to form the regression data of cell membrane damage intensity. Then, the cell membrane damage intensity regression data is combined with the soil organic matter content fluctuation data to calculate the nutrient leakage NL=MD×OM×c, where NL is the nutrient leakage (in mg / kg), MD is the degree of membrane damage, OM is the organic matter content (in g / kg), and c is the conversion coefficient, which is 8.5. The nutrient leakage equivalent prediction data is obtained through this calculation. Finally, the soil nutrient imbalance effectiveness loss index is calculated by combining the results of various analyses. , where NL is the nutrient leakage, IC is the ion competition intensity increment data (obtained from step S235), pH_dev is the degree of pH value deviation from the optimal range (|pH-6.5| / 6.5×100%), NP is the total nutrient content of soil (unit: mg / kg), to is the weight coefficient, which takes values ​​of 0.5, 0.3, and 0.2 respectively. The final NIL value range is 0-100%, which represents the percentage of soil nutrient effectiveness loss, forming the soil nutrient imbalance effectiveness loss data.

[0029] Step S24: performing root absorption limitation estimation on the seedling root density prediction data according to the soil salt content abnormal fluctuation data and the soil nutrient imbalance effectiveness loss data to obtain root absorption limitation estimation data.

[0030] In the embodiment of the present invention, the root absorption limitation estimation is performed on the seedling root density prediction data according to the abnormal fluctuation data of soil salt content and the loss data of soil nutrient imbalance effectiveness. First, the fluctuation amplitude deviation rate calculation is performed on the abnormal fluctuation data of soil salt content, and the calculation formula is SDF=(EC_max-EC_base) / EC_base×100%, wherein SDF is the difference in the fluctuation amplitude of salt content (unit %), EC_max is the highest conductivity value during the monitoring period, and EC_base is the reference conductivity value (the average value of the previous 15 days is taken); then, the increase ratio of the osmotic pressure of the soil solution is calculated according to the difference in the fluctuation amplitude of salt content, and the van Hoff equation π=i×C×R×T is adopted, wherein π is the osmotic pressure (unit MPa), i is the degree of dissociation (the average salt is 1.8), C is the salt concentration (unit mol / L, obtained by converting the conductivity, and the conversion coefficient is 0.011), and R is the gas constant (0.00831 L·MPa / (mol·K)), T is the absolute temperature (taken as 293K), and the osmotic pressure increase ratio OPI=(π_abnormal-π_normal) / π_normal×100% is calculated, where π_abnormal is the osmotic pressure under abnormal conditions and π_normal is the osmotic pressure under normal conditions; then the microbial nutrient conversion inhibition evaluation is performed based on the soil solution osmotic pressure increase ratio, and the microbial activity inhibition rate ,in is the proportionality coefficient, the value is 0.022, is an exponential coefficient with a value of 1.25. The cumulative inhibition effect during the entire monitoring period is calculated by numerical integration ∫MIR(t)dt to obtain the nutrient transformation inhibition fluctuation value NTI. At the same time, the soil structure change is estimated based on the salt content fluctuation amplitude difference data and the soil solution osmotic pressure increase ratio. The soil compaction cumulative gradient SSG=α×SDF+β×OPI+γ×(SDF×OPI), where α, β, and γ are coefficients with values ​​of 0.015, 0.025, and 0.0005, respectively. The soil compaction cumulative gradient data is calculated. Then, the root growth and respiration are constrained and evaluated based on the soil compaction cumulative gradient data. The constraint gradient calculation formula is RIG=SSG×(1+δ×RDI), where RDI is the root density (obtained from step S21), and δ is the adjustment coefficient with a value of 0.8. The root growth / respiration constraint gradient data is obtained. Finally, the root absorption restriction coefficient is calculated by comprehensively considering the nutrient transformation inhibition fluctuation value, the root growth / respiration constraint gradient, and the soil nutrient imbalance effectiveness loss data. ,in is the weight coefficient, with values ​​of 0.02, 0.03, and 0.025 respectively. The RAL value range is 0-1, 0 represents no restriction, and 1 represents complete restriction. By standardizing RAL, we finally generate root absorption restriction estimation data containing five levels (slight, mild, moderate, severe, and extremely severe).

[0031] Step S23 includes the following steps: Step S231: performing ion concentration decomposition processing on the abnormal fluctuation data of soil salt content to generate time series variation data of salt ion concentration; Step S232: extracting organic matter content fluctuation from soil state characteristic monitoring feature data to obtain soil organic matter content fluctuation data; Step S233: performing nonlinear intensity regression analysis of cell membrane damage based on the salt ion concentration time series variation data to obtain cell membrane damage intensity regression data; Step S234: performing an equal amount simulation prediction of organic nutrient leakage on the soil organic matter content fluctuation data according to the cell membrane damage intensity regression data to obtain equal amount prediction data of nutrient leakage; Step S235: performing ion competition intensity increment analysis on soil organic matter content fluctuation data according to salt ion concentration time series variation data to generate ion competition intensity increment data; Step S236: Perform soil nutrient imbalance effectiveness loss analysis based on the nutrient leakage equal amount prediction data and the ion competition intensity increment data to obtain soil nutrient imbalance effectiveness loss data.

[0032] In the embodiment of the present invention, the acquisition of abnormal fluctuation data of soil salt content is based on the conductivity time series monitoring curve, which is collected at an interval of 10 minutes within 120 hours to form 720 sample points, and the unit is milliSiemens per centimeter. When the data is subjected to ion concentration decomposition processing, the common soil ion types, including five main ion components such as sodium ions, potassium ions, chloride ions, calcium ions and magnesium ions, are set, and the ion activity factor estimation method based on the extended Debye-Hückel equation is applied to convert the total conductivity value into the concentration change value of each type of ion. The process uses a data window with a sliding window length of 12, and uses the conductivity fluctuation amplitude between the minimum and maximum values ​​in each window as the standard offset of the concentration change distribution. The ion concentration is fitted in combination with local weighted regression (LOWESS), and the output format is the concentration sequence of each type of ion at each 10-minute time point, in milligrams per liter, which constitutes the salt ion concentration time series change data. The soil state characteristic monitoring characteristic data is used to extract organic matter content fluctuations. First, relying on the installed soil sensors, the data related to soil health and nutrient activity are obtained by monitoring the basic state characteristics of soil moisture content, pH value, oxidation-reduction potential (ORP), etc. On this basis, the data is first preprocessed to eliminate noise data for the fluctuation of soil organic matter content. Then, through dynamic time series analysis, the characteristic data of each period are smoothed to remove the interference of seasonal change factors and highlight the short-term volatility of organic matter in the soil. The wavelet transform method is used to perform multi-scale analysis on soil data to capture the high-frequency components of soil organic matter content fluctuations. By comparing the change trends of organic matter in each time period and combining the fluctuation characteristics of the data, the soil organic matter content fluctuation data is generated. In each time window, the output of organic matter fluctuation data includes key features such as fluctuation amplitude, fluctuation period, and fluctuation speed. All data are stored in the form of time series for subsequent analysis and decision-making. For example, during the monitoring period from March 21 to March 23, the soil ORP value increased from -120 mV to -70mV, accompanied by fluctuations in moisture content, indicating that the decomposition and conversion rate of organic matter has changed. This fluctuation data can be used for subsequent soil improvement and fertilization program adjustment.

[0033] Based on the time series data of salt ion concentration, nonlinear intensity regression analysis of cell membrane damage was performed. First, a relationship model between ion concentration and cell membrane damage was constructed, using an exponential function form. , where MD is the membrane damage strength (dimensionless value, range 0-1), C_Na is Concentration (mmol / L), C_Cl is concentration (unit: mmol / L), α, β, and γ are nonlinear regression coefficients, which are 0.95, 0.068, and 0.072, respectively; then, for each time point and The concentration data is substituted into the model to calculate the corresponding membrane damage intensity value and form a membrane damage intensity time series data set; then the interaction between membrane damage intensity and other ion concentrations is analyzed and the ion antagonism factor is introduced. , where δ is the antagonistic coefficient, which is 0.15, and C_K, C_Ca, and C_Mg are The concentration of the membrane damage intensity is corrected. ; Then the cumulative effect of membrane damage strength is calculated using the weighted moving sum method, , where i represents the time interval (in hours), w_i is the time weight coefficient, satisfying Σw_i=1, w_i decreases as i increases, and the specific values ​​are w_0=0.3, w_1=0.25, w_2=0.2, w_3=0.15, w_4=0.1; then the cumulative membrane damage intensity data are divided into intervals, and the membrane damage intensity is divided into five levels: slight (0-0.2), mild (0.2-0.4), moderate (0.4-0.6), severe (0.6-0.8) and extremely severe (0.8-1.0); finally, according to the interval division results, the proportion of each level in each time period is calculated to generate a time series diagram of the membrane damage intensity level distribution, and the changing trend of the membrane damage intensity is calculated through a sliding window (the window size is 24 hours), forming a cell membrane damage intensity regression data containing original data, level distribution and trend analysis. In the process of simulating and predicting the leakage of organic nutrients based on the fluctuation data of soil organic matter content according to the regression data of cell membrane damage intensity, the regression data of cell membrane damage intensity is first matched one by one with the fluctuation data of organic matter concentration in soil, and the differential differential recursive algorithm is used to establish the leakage intensity equivalent derivation relationship. During this operation, the time span is set to one cycle every 60 minutes, and the mass concentration data of soluble organic carbon (DOC) in the soil is collected and linked with the cell membrane damage intensity data. By performing time series difference processing on the DOC data, the DOC change rate per unit area between two adjacent cycles is extracted, and leakage is inferred based on the cell membrane damage regression coefficient. The leakage inference algorithm used is a quantitative segmented extrapolation method. For example, in a certain test area, the initial concentration of DOC in the surface soil is 65 mg / L, the membrane damage intensity is 0.78, and the concentration drops to 54 mg / L after one cycle. The extrapolation logic judges that it is an effective leakage rather than a contribution from the mineralization process. This method is based on the fixed time difference, and couples the DOC concentration changes in different period intervals with the cell membrane strength regression data according to the discrete integral method. The final output is the total DOC leakage per unit area per unit time as the organic nutrient leakage equivalent prediction data, in milligrams per square meter per hour. When analyzing the ion competition intensity increment of the soil organic matter content fluctuation data based on the salt ion concentration time series change data, firstly, based on each type of main cation (including , , , ) concentration change curve to establish a time series data matrix, and construct an analytic function for the incremental ion mutual repulsion rate. This function is based on the charge pair effect and the soil cation exchange capacity rule. The concentration change rate is the first driving factor, The concentration is the response factor, and the Beta distribution dependency regression method is used to derive the competition increment intensity. Among them, the data preprocessing method is standardized time series difference and peak calibration to eliminate the instantaneous fluctuation interference caused by environmental water input. In the specific experiment, the initial soil The concentration was 1.8 mmol / L, which increased to 3.1 mmol / L after 48 hours. The concentration dropped from 2.4 to 1.6 mmol / L, and the corresponding ion competition intensity increment value in the regression coupling curve was 0.58, which represents the adsorption priority transfer intensity between sodium and potassium. This process does not involve manual setting of thresholds, but is completely analyzed based on the time trend of ion concentration and the mutually exclusive comparison logic, and finally outputs the ion competition intensity increment data, in units of dimensionless relative gain ratio. In the operation of analyzing the effectiveness loss of soil nutrient imbalance based on nutrient leakage equivalent prediction data and ion competition intensity increment data, the leakage equivalent data of the three main nutrient elements (nitrogen, phosphorus, and potassium) are first used as the basic indicator items, corresponding to the ion competition intensity increment data of the same time period, and the imbalance loss function is constructed by the weighted integration method. The weight distribution logic adopts the standard three-factor evaluation model, in which the leakage equivalent weight is set to 0.6, and the competition intensity increment is set to 0.4. After integrating the data, the total effectiveness loss per unit area is obtained by solving the weighted difference within the time period. In this operation process, the data resolution is controlled to one grid area per 10 square meters, and one evaluation cycle is every 24 hours. The analysis indicators include nitrogen transfer ratio (according to concentration change), phosphorus release ratio (based on concentration and hydrolysis rate), potassium effective concentration inhibition ratio (based on and Relative change derivation). Taking the real collected data as an example, in a certain sampling, the nitrogen leakage was 13 mg / m2 per hour, and the ion competition increment value was 0.52. After the weighted function processing, the nitrogen effectiveness loss in this period was 10.14 mg / m2. This value was written into the database as the final generated soil nutrient imbalance effectiveness loss data and used in the subsequent irrigation and fertilization strategy optimization process. The entire operation process did not use neural networks or machine learning models, but was completed using static rule algorithms and mathematical derivation based on data distribution logic.

[0034] Step S235 includes the following steps: Perform mutation increment index calculation on the time series variation data of salt ion concentration to obtain the mutation increment index of salt ion concentration; Based on the salt ion concentration mutation increment index, the salt ion concentration time series change data is combined by continuous concentration mutation approximation to generate concentration continuous mutation approximation combination data; Based on the concentration continuity mutation approximation combination data, the soil organic matter content fluctuation data was simulated by the preemptive continuous intensity of salt ion absorption sites, and the preemptive continuous intensity data of salt ion absorption sites were obtained. The preemptive continuous intensity data of the absorption site are processed by calculating the time interval difference of the ion flux to obtain the difference of the salt ion flux at the absorption site; The ion competition intensity increment analysis is performed based on the difference in salt ion flux at the absorption site to generate ion competition intensity increment data.

[0035] In the embodiment of the present invention, for the operation of "calculating the mutation increment index of the salt ion concentration time series change data to obtain the mutation increment index of the salt ion concentration", it is necessary to first clarify the acquisition path of the salt ion concentration. In this embodiment, a multi-point ion selective electrode sensor in the deep soil layer is used to collect the ion concentration in the range of 0 to 30 cm. Typical salt ion concentration data are collected once an hour with a period of 72 hours to obtain a time-series data sequence of ion concentration changes. Before processing the original sequence, each ion concentration curve is first subjected to a 5th-order sliding window denoising process using a Savitzky-Golay smoothing filter, and the first-order derivative information is retained for subsequent analysis. Subsequently, the time series difference increment method is used to successively calculate the amplitude of the change in ion concentration at adjacent moments, and to construct an instantaneous mutation rate sequence of ion concentrations. In order to measure the nonlinear degree of mutation intensity, an exponential calculation is performed using the deviation accumulation method between the increase ratio and the historical change mean. In the specific processing, the historical window length is set to 12 hours, and the ratio difference between the current change value and the change mean within the window is accumulated, and the mutation increment exponential curve is updated hour by hour. In actual operation, when the soil When the concentration rapidly increased from 23.4mmol / L to 41.2mmol / L from the 48th hour to the 51st hour and remained at a high level, the cumulative value of the mutation increment index in the window exceeded the threshold standard of 5.6 and was marked as a first-level concentration mutation node, and included in the subsequent approximation combination processing flow. For the operation of "continuous concentration mutation approximation combination of salt ion concentration time series change data based on the salt ion concentration mutation increment index to generate concentration continuity mutation approximation combination data", the time interval marked as the first-level mutation node in the mutation increment index calculation is first segmented. Subsequently, the curve continuity reconstruction method based on B-spline interpolation is used to perform fitting and approximation operations on the ion concentration curve in the mutation section. In order to ensure that the fitted mutation morphology truly reflects the rapid change process, the third-order B-spline interpolation function is used, and the mutation rate boundary constraints are set for the control nodes to ensure that there is an obvious discontinuous jump in the first-order derivative of the interpolated function at the node, and the jump amount is consistent with the original mutation increment. In actual processing, In the concentration mutation section from 48 to 51 hours, a fitting sequence with one fitting point every 30 minutes is generated by interpolation, forming a concentration mutation combined data set containing the original concentration points and the fitted complement points. This combined data not only retains the original mutation information, but also provides compensation for the change trend of non-sampling time points for the next step of absorption site intensity simulation. For the operation of "simulating the preemptive continuous intensity of salt ion absorption sites on the soil organic matter content fluctuation data based on the concentration continuity mutation approximation combined data to obtain the preemptive continuous intensity data of salt ion absorption sites", it is necessary to simultaneously introduce the soil organic matter content change data in the same sampling time and space area. The near-infrared reflectance spectroscopy method and the soil high-frequency mixing sampling method are used to obtain the organic matter content value once an hour. The data is aligned with the above-generated concentration mutation combined data by timestamp, and a comparative relationship map of ion concentration and organic matter fluctuation is established on the same time axis. In operation, the salt ion preemption model is set as an interval superposition model based on the sequence of concentration fluctuations, that is, in the mutation area, if the time of ion concentration increase is earlier than the starting point of organic matter content decrease, it is marked as the occurrence of potential absorption site preemption. For each comparison interval, the mutation duration and the concentration increase rate are set as weighting factors, and the intensity fitting function is constructed by linear superposition. In the mutation section, the organic matter content dropped from the original 2.3% to 1.7%, and the starting point of the decline lagged by about 0.5 hours. This area constitutes a high-intensity preemptive absorption event, and the calculated intensity fitting value reaches the upper limit set by the preemptive warning model. Finally, the continuous intensity simulation results form a time series image with the time axis as the horizontal axis and the absorption intensity per unit time as the vertical axis, which is used as the triggering basis for the organic matter compensation strategy in the subsequent control system callback fertilization plan. No model prediction was introduced in the whole process, only the actual observation data interpolation reconstruction and causal time series superposition judgment were used. All processing algorithms are based on known mathematical analysis methods and signal processing functions, and do not involve non-physical basic modeling operations.

[0036] For the operation of "calculating and processing the time-series interval difference of ion flux on the preemptive continuous intensity data of the absorption site to obtain the difference of salt ion flux at the absorption site", it is necessary to perform the difference calculation of flux change at fixed time intervals based on the preemptive continuous intensity data of the absorption site obtained in the previous processing link. The core logic of flux difference processing is to estimate the difference in the change in the number of absorbed ions per unit area at two consecutive time points at the same absorption site within a fixed time step. In this embodiment, the absorption site is defined as a soil structure unit with significant ion exchange activity in the 1cm×1cm area of ​​the rhizosphere, and the flux per unit time is in millimoles per square centimeter per hour as the basic unit. The time step is set to 1 hour, and the continuous intensity data comes from The simulated absorption intensity time series values ​​of the three main ions are as follows. From the 48th hour to the 51st hour, the absorption values ​​of the ions are extracted from the intensity sequence one by one, and the difference is calculated. The difference calculation method is: the absorption intensity at the current moment minus the absorption intensity at the previous moment, that is, the flux difference. For example, if the absorption intensity is 3.1 mmol per square centimeter per hour at the 48th hour and 4.6 mmol per square centimeter per hour at the 49th hour, the flux difference in this time period is 1.5 mmol per square centimeter per hour. The above difference calculation operation is performed simultaneously on all identified absorption sites and all key ion channels. After the flux data processing is completed, a flux difference matrix is ​​generated that is saved in a three-dimensional structure, and its dimensions are time point, absorption site number and ion type. The flux difference data provides a time series change basis for subsequent ion competition analysis, and realizes the numerical characterization of absorption dynamics through strict difference processing.

[0037] For the operation of "performing incremental analysis of ion competition intensity based on the difference in salt ion flux at the absorption site to generate incremental ion competition intensity data", it is necessary to extract the flux difference vectors between ions at the same absorption site at the same time node from the flux difference matrix, and deduce the competition relationship for the relative change trend between the vectors. In this embodiment, a fixed absorption site is used as the basis, and a multi-ion flux difference ratio map is constructed hourly to determine whether absorption competition occurs within a certain period of time. The competition relationship judgment criterion is: if the flux difference of a certain ion increases significantly, and the flux difference of another ion decreases synchronously, and the ratio change of the two exceeds the static competition threshold of 0.8, then it is considered that competitive absorption behavior exists. At the 49th hour, The flux difference is +1.5, The flux difference is -0.4, The flux difference is -0.6, the three change directions are inconsistent, and and The absolute value of the flux difference ratio is 2.5, which exceeds the static competition threshold and is judged as right Absorption inhibition competition occurs. During the execution process, the ratio of the ion pairs at each time node is calculated, and the results are retained in the ion competition incremental intensity table. The fields in the table include time node, absorption site number, ion pair combination, flux difference ratio, competition direction and competition intensity level. The final ion competition intensity incremental data is presented in a three-dimensional tensor structure, which is used as the basis for adjusting the arrangement order of ion types in subsequent drip irrigation and fertilization control instructions. No predictive algorithm is introduced in the entire analysis process, which only relies on mathematical difference operations and proportional relationship judgments. All incremental data are derived based on the measured flux difference, excluding external inferences and non-causal data interference, to ensure the physical consistency of the data and the traceability of the calculation logic.

[0038] Step S24 includes the following steps: Step S241: performing fluctuation amplitude deviation rate calculation on the abnormal fluctuation data of soil salt content to obtain salt content fluctuation amplitude difference data; Step S242: converting and quantifying the soil solution osmotic pressure increase ratio according to the salt content fluctuation amplitude difference data to obtain the soil solution osmotic pressure increase ratio; Step S243: integrating the fluctuation of the microbial nutrient conversion inhibition value based on the soil solution osmotic pressure increase ratio to obtain the nutrient conversion inhibition fluctuation value; Step S244: estimating the soil compaction cumulative gradient according to the salt content fluctuation amplitude difference data and the soil solution osmotic pressure increase ratio to obtain soil compaction cumulative gradient data; Step S245: performing root growth / respiration restriction gradient evaluation on the seedling root density prediction data based on the soil compaction cumulative gradient data to obtain root growth / respiration restriction gradient data; Step S246: Root absorption limitation estimation is performed on the seedling root density prediction data according to the nutrient conversion inhibition fluctuation value, root growth / respiration restriction gradient data and soil nutrient imbalance effectiveness loss data to obtain root absorption limitation estimation data.

[0039] In the embodiment of the present invention, for the operation of step S241 "calculating the fluctuation amplitude offset rate of the abnormal fluctuation data of the soil salt content to obtain the fluctuation amplitude difference data of the salt content", it is necessary to extract the intensity difference caused by the abnormal fluctuation by calculating the fluctuation amplitude offset rate within the reference time window based on the soil profile salt content data sequence collected in multiple consecutive time periods. The sampling depth is set to the 0-20 cm area of ​​the soil surface, the sampling frequency is once every 2 hours, the observation period is 72 hours, and the conductivity sensor is used to record the conductivity value of the unit volume of soil solution in real time, with Siemens per centimeter as the unit. In the implementation process, the conductivity value of each sampling point is first converted into an approximate equivalent salt content value (grams per kilogram), and the conversion factor used is based on the previous field calibration experimental data. Then, a sliding window with a unit of 12 hours is constructed for the continuous time series, and the fluctuation amplitude of the difference between the maximum and minimum values ​​in each window is estimated, and then the fluctuation amplitude difference between adjacent windows is compared, and the fluctuation amplitude of the initial window is used as a reference standard to calculate the offset rate. The calculation formula of the offset rate is: the fluctuation amplitude of the latter window minus the fluctuation amplitude of the previous window divided by the fluctuation amplitude of the previous window. For example, in the observation area where the seedling number is M-23, the fluctuation amplitude of the first time window is 1.6 grams per kilogram, and the fluctuation amplitude of the second window is 2.3 grams per kilogram, then the offset rate is 0.4375. The offset rate data is used to determine whether there is a nonlinear aggravated salt disturbance phenomenon in the area. The offset rate results of all observation points are finally summarized to generate a two-dimensional grid structure of salt content fluctuation amplitude difference data map, whose rows and columns correspond to the sampling point number and time index number. For the operation of step S242 "converting and quantifying the increase ratio of soil solution osmotic pressure according to the salt content fluctuation amplitude difference data to obtain the increase ratio of soil solution osmotic pressure", it is necessary to use the quantitative relationship between salt concentration and osmotic pressure, and complete the conversion based on the physical relationship that osmotic pressure and total solute concentration are approximately linearly positively correlated without introducing an empirical model. In the specific operation, first, from the salt content fluctuation difference data extracted in step S241, the time period and sampling point combination with an offset rate greater than 0.25 are screened, and then the corresponding salt change is substituted into the osmotic pressure conversion factor for direct conversion. When sodium chloride is the main salt ion, the osmotic pressure conversion factor is 0.036 MPa per gram per kilogram, which is obtained by repeatedly measuring the artificial proportion of NaCl solution in a standard constant temperature chamber. In the same area where the seedlings are numbered M-23, if the salt increases by 0.9 grams per kilogram in a certain time period, the increase in the osmotic pressure of the soil solution at this point is 0.0324 MPa. The osmotic pressure increment value is divided by the initial average osmotic pressure at this point (set to 0.142 MPa), and the osmotic pressure increase ratio can be obtained, and the result is 0.228.This operation is performed in parallel on all the points where the offset rate exceeds the standard, and a tensor data set of the increase ratio with sampling points and time periods as dimensions is generated. This data is used to describe the potential osmotic pressure change trend caused by salt fluctuations in the soil. All calculation processes rely on the quantitative correspondence between the measured conductivity, salt content and osmotic pressure to avoid structural errors introduced by undefined empirical models. For the operation of step S243 "integrating the fluctuation of the microbial nutrient conversion inhibition value based on the soil solution osmotic pressure increase ratio to obtain the nutrient conversion inhibition fluctuation value", based on the experimental evidence of the sensitivity of soil microorganisms to high osmotic pressure environments, a response inhibition function based on weight factors is used to map the osmotic pressure increase ratio and accumulate the integral to construct a dynamic weakening trend of the nutrient conversion rate. In the operation, the main soil functional fungi represented by nitrifying bacteria, phosphomonas bacteria and cellulose degrading bacteria are first set, and the metabolic activity decrease ratio under different osmotic pressure levels is measured under experimental conditions. It is known that when the osmotic pressure increase ratio is 0.2, the nitrification rate of nitrifying bacteria decreases by 22% and the phosphatase activity decreases by 17%. In this embodiment, this relationship is used to assign the conversion inhibition weight. Then, the inhibition response of the corresponding bacterial community in each time period is numerically integrated based on the proportional tensor of the increase in soil solution osmotic pressure per hour. The integration operation uses time as the horizontal axis and the inhibition intensity as the vertical axis. The integration result is the nutrient conversion inhibition fluctuation value. Taking the data within the 72-hour period of point M-23 as an example, the overall integral value is 6.72, indicating that the average inhibition intensity caused by the increase in osmotic pressure during the entire period has accumulated to 6.72 units. This result is used to fine-tune the target nutrient release rate in subsequent fertilization instructions to ensure that the stability of the control system response is maintained under the condition of limited bacterial community function. In the whole process, each bacterial community inhibition factor is determined through field cultivation and control variable experiments. No predictive reasoning or hypothetical model is introduced. All calculation links are clear numerical logic and integral accumulation processes.

[0040] For the operation of step S244 "estimate the soil compaction cumulative gradient according to the salt content fluctuation amplitude difference data and the soil solution osmotic pressure increase ratio to obtain the soil compaction cumulative gradient data", based on two core input variables, namely the salt content fluctuation amplitude difference data obtained in step S241 and the osmotic pressure increase ratio tensor generated in step S242, a synchronous pairing relationship is established with the spatial position index and the time series index respectively, and the influence trend of the salt fluctuation amplitude and osmotic pressure change on the soil structure is synchronously analyzed in the same time period and sampling point. During the operation, the historical salt fluctuation value sequence of each grid is positively accumulated with a grid area of ​​0.5 meters by 0.5 meters as a unit to generate the fluctuation intensity accumulation; at the same time, the osmotic pressure increase ratio multiplied by the time span is used as the unit inhibition index, which is superimposed on the grid node to represent the induced trend of long-term osmotic pressure changes on the capillary water migration and particle structure convergence process. Subsequently, the salt fluctuation accumulation and the osmotic pressure inhibition index are standardized and linearly weighted, with the weights of 0.6 and 0.4 respectively, and the result is the initial estimation of the soil compaction cumulative gradient. Taking the garden nursery monitoring area numbered Z-11 as an example, the cumulative value of the salt content fluctuation at this point in a 7-day cycle is 7.2 grams per kilogram, the average osmotic pressure increase ratio is 0.27, and the inhibition index obtained by combining the time span of 168 hours is 45.36. The two values ​​after standardization are 0.81 and 0.63 respectively, and the hardening gradient obtained after linear weighting is 0.732. This operation is performed synchronously at all monitoring points, and finally a two-dimensional space-time hardening gradient tensor data set is formed, in which each unit contains the hardening trend data driven by salt accumulation and osmotic pressure, which is directly used as the input basis for subsequent root growth restriction analysis. For the operation of step S245 "based on the soil hardening cumulative gradient data, the root density prediction data of the seedlings is evaluated for the root growth / respiration restriction gradient, and the root growth / respiration restriction gradient data is obtained", it is necessary to first perform structured mapping on the seedling root density prediction data so that it maintains a spatial correspondence with the hardening cumulative gradient tensor obtained in step S244. The root density prediction data is the root volume data per unit volume (in grams per cubic decimeter) generated by the root shadow distribution range and underground biomass conversion ratio previously fused through image features. The spatial resolution is 0.5 meters, and the depth levels are divided from 0-10 cm, 10-20 cm, and 20-30 cm. In implementation, the root density value of each spatial unit is cross-analyzed with the cumulative gradient of the hardening of the unit where it is located, and the root growth capacity is converted by setting the growth inhibition coefficient. The inhibition coefficient is segmented based on the root density threshold and the hardening gradient threshold. If the root density is higher than 3.5 grams per cubic decimeter and the hardening gradient exceeds 0.7, the point is given a growth inhibition intensity of 0.6, indicating that 60% of the root tissue faces resistance; if the root density is lower than 1.5 grams and the hardening gradient is lower than 0.4, the inhibition intensity is 0.1.In addition, the respiration inhibition gradient is normalized by multiplying the oxygen permeability coefficient per unit root density by the ratio of the degree of compaction. Taking the 20-30 cm soil layer at point M-45 as an example, the root density is 2.8 g / dm3, the corresponding compaction gradient is 0.65, the oxygen permeability coefficient is set to 1.3 mg / h / dm3, and the degree of compaction is 1.5 MPa, then the normalized result is 0.43. Finally, the root growth inhibition gradient and the respiration inhibition gradient are linearly averaged to form the comprehensive constraint gradient value of the point. This operation is calculated in parallel on all grid cells and depth layers to generate a complete three-dimensional gradient structure for input into the root absorption analysis in the next stage. For the operation of step S246 "estimate root absorption restriction based on nutrient conversion inhibition fluctuation value, root growth / respiration restriction gradient data and soil nutrient imbalance effectiveness loss data, and obtain root absorption restriction estimation data", it is necessary to construct the seedling root density prediction data as the main index, and pair and integrate the nutrient conversion inhibition fluctuation value (unit dimensionless) generated from step S243, the root growth and respiration inhibition gradient (value range 0 to 1) generated from step S245, and the soil nutrient imbalance effectiveness loss data (expressed as effectiveness loss ratio, unit percentage) according to the root density unit. In the operation, the root density unit is taken as the minimum estimation unit, and the three data of the unit are first normalized, and the standard interval is set to [0,1], where the maximum value of the conversion inhibition value is set to 10, and the maximum value of the imbalance loss ratio is set to 90%. Then, the weighted linear combination method is used to weight the restricted valuation, and the weights are 0.3 for the conversion inhibition item, 0.5 for the growth / respiration restriction item, and 0.2 for the imbalance loss item. For example, in the 20-30 cm root layer of the M-45 area, the predicted root density is 2.4 grams per cubic decimeter, the corresponding conversion inhibition fluctuation value is 7.8, and the normalized value is 0.78; the corresponding growth respiration restriction gradient is 0.63; the soil effectiveness loss ratio is 58%, and the normalized value is 0.644. The weighted comprehensive valuation is 0.7154, which indicates the relative absorption restriction degree of the root system in the unit under the current conditions. The higher the value, the more severe the restriction. Finally, all the root-dense unit valuations are backfilled into the three-dimensional root distribution grid to form a complete root absorption restriction estimation data structure, which is used for the trigger judgment conditions of the subsequent fertilization and watering fine-tuning control instructions. This operation does not involve the simulation process throughout the process, and only relies on the original data and the logical weight relationship to realize the restriction degree analysis.

[0041] Step S3 includes the following steps: Step S31: performing slow-release fertilizer dosage control matching on the abnormal fluctuation data of soil salt content according to the root absorption limitation estimation data to obtain slow-release fertilizer dosage control data; Step S32: performing dosage control logic learning on the slow-release fertilizer dosage control data to obtain slow-release fertilizer dosage control learning data; Step S33: Based on the slow-release fertilizer dosage control learning data and the root absorption limitation estimation data, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio for fertilization and watering.

[0042] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes: Step S31: performing slow-release fertilizer dosage control matching on the abnormal fluctuation data of soil salt content according to the root absorption limitation estimation data to obtain slow-release fertilizer dosage control data; In the embodiment of the present invention, for the operation of step S31 "matching the slow-release fertilizer dosage control with the abnormal fluctuation data of soil salt content according to the root absorption restriction estimation data to obtain the slow-release fertilizer dosage control data", it is first necessary to establish a corresponding relationship between the root absorption restriction estimation data and the abnormal fluctuation data of soil salt content according to the spatial position. The root absorption restriction estimation data comes from step S246, and the unit is the dimensionless absorption restriction coefficient (between 0 and 1). The abnormal fluctuation data of soil salt content comes from step S241, which is the standard deviation of salt content within 7 days in unit volume of soil, and the unit is grams per kilogram. During the operation, the soil unit of 0.5 m × 0.5 m × 0.3 m is taken as the basic matching unit. First, the slow-release fertilizer dosage is initially limited according to the absorption restriction estimation. When the maximum restriction coefficient is 1.0, the corresponding slow-release fertilizer is completely suspended. When the minimum restriction coefficient is 0, the basic fertilization amount is maintained. The basic fertilization amount is set to 6 grams of slow-release nitrogen and potassium mixed particles per unit per day. Subsequently, the basic value is adjusted according to the fluctuation range of soil salinity. If the fluctuation value exceeds 3.0 grams per kilogram, the basic fertilizer amount will be reduced by 30%. If it is lower than 1.0 grams per kilogram, no adjustment will be made. If it is in the middle range, it will be reduced linearly in proportion. The specific adjustment ratio is the current fluctuation value minus 1.0, divided by 2.0, and then multiplied by 0.3 to get the reduction ratio. For example, in the nursery area numbered A-12, the absorption limit estimate is 0.67, the salt fluctuation is 2.3 grams per kilogram, and the corresponding basic fertilizer amount is 6 grams. According to the restriction coefficient reduction ratio of 0.67, the initial fertilizer amount is 6 times (1 minus 0.67) to get 1.98 grams. Combined with the salt fluctuation adjustment amount, because the fluctuation value is between 1.0 and 3.0, it is calculated as (2.3 minus 1.0) divided by 2.0 and then multiplied by 0.3, the result is 0.195, which is a further reduction of 19.5%. The final slow-release fertilizer application amount is 1.98 times (1 minus 0.195) to get about 1.59 grams. This operation is performed synchronously for all monitoring units to form a complete slow-release fertilizer dosage control data grid, which is output for subsequent learning and fusion.

[0043] Step S32: performing dosage control logic learning on the slow-release fertilizer dosage control data to obtain slow-release fertilizer dosage control learning data; In the embodiment of the present invention, for the operation of step S32 "performing dosage control logic learning on the slow-release fertilizer dosage control data to obtain the slow-release fertilizer dosage control learning data", it is necessary to perform sliding analysis and dynamic adjustment factor extraction on the slow-release fertilizer dosage control data in the time series dimension to form regular control logic data. The operation steps include data normalization processing, trend factor extraction, upper and lower fluctuation threshold determination, local abnormal increase identification and gradient change feature statistics. In the specific implementation process, the slow-release fertilizer dosage control data for 14 consecutive days is first normalized daily, and the normalization range is 6 grams of daily fertilizer upper limit and 0 grams of lower limit. Then, the cycle is set to 5 days in a sliding window manner, and the maximum value, minimum value, mean value and range in each 5-day window are extracted to judge the local fluctuation trend. The fluctuation determination threshold is set to be an unstable section if the range exceeds 0.4 of the normalized value. If two unstable sections appear consecutively, the logical mark "fluctuation overload" is inserted at this position. Taking the area numbered C-9 as an example, the normalized values ​​from the 1st to the 5th day are [0.30, 0.36, 0.42, 0.50, 0.46], with a range of 0.20, which is less than the threshold of 0.4 and is not marked as unstable; from the 6th to the 10th day, it is [0.41, 0.67, 0.72, 0.49, 0.75], with a range of 0.34, which is still within the threshold; but from the 11th to the 15th day, it is [0.35, 0.89, 0.94, 0.81, 0.87], with a range of 0.59, which exceeds the threshold and marks the unstable segment. In the above way, the time series fertilization data of all spatial units are converted into a data structure containing trend, stability and abnormality labels. This structure is the slow-release fertilizer dosage control learning data to support the next step of drip irrigation ratio structure reorganization.

[0044] Step S33: Based on the slow-release fertilizer dosage control learning data and the root absorption limitation estimation data, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio for fertilization and watering.

[0045] In the embodiment of the present invention, for the operation of step S33 "based on the slow-release fertilizer dosage control learning data and the root absorption limitation estimation data, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio of fertilization and watering", a segmented control logic table needs to be constructed to jointly map the fluctuation trend segment extracted from the slow-release fertilizer dosage control learning data with the absorption limitation intensity in the root absorption limitation estimation data, and determine the drip irrigation ratio of each control time slice. In the operation, the daily drip irrigation process is divided into three time periods: 06:00 to 10:00, 12:00 to 15:00, and 17:00 to 20:00, and the initial irrigation ratios of the three sections are respectively assigned to be 20%, 30%, and 50%. For the units with absorption limitation intensity in the range of 0.0 to 0.3, no adjustment is made. For the units with intensity in the range of 0.3 to 0.6, the morning and evening ratios are reduced by 10% each, and the middle section ratio is increased by 20%; for the units with intensity higher than 0.6, the morning and evening ratios are reduced by 15% each, and the middle section ratio is increased by 30%. Subsequently, further fine-tuning is performed based on the fluctuation trend mark in the learning data. If there are more than two consecutive unstable marks, the total daily drip irrigation volume is reduced by 20% and redistributed in proportion in each period. Taking the monitoring unit numbered B-6 as an example, the root absorption limitation estimate is 0.71. The control learning data has a "fluctuation overload" mark in two consecutive sliding cycles. The initial segment ratio is adjusted to: 20% minus 15% in the early stage is 5%, 30% plus 30% in the middle stage is 60%, and 50% minus 15% in the late stage is 35%, the total is 100%, and then uniformly compressed by 20%, the total drip irrigation volume is 80% of the original drip volume, then the actual early stage is 4%, the middle stage is 48%, and the late stage is 28%. Finally, after this operation is performed on all units, a complete space-time segmented drip irrigation ratio data set is formed, which serves as the instruction parameter data of the terminal drip irrigation control device.

[0046] Step S33 includes the following steps: Step S331: Deducing the respiratory conversion demand space based on the root absorption limitation estimation data to obtain the root respiratory conversion demand space; Step S332: performing nutrient absorption rate interval calculation per unit time on the root absorption limitation estimation data to obtain the root nutrient absorption rate interval per unit time; Step S333: analyzing the fertilizer effect release rate of the slow-release fertilizer dosage control learning data to obtain the fertilizer effect dosage release rate; Step S334: dividing the fertilizer efficiency dosage release rate into dosage absorption extreme value points per unit time according to the root respiration conversion demand space and the root nutrient absorption rate interval to obtain root fertilizer efficiency release absorption extreme value data; Step S335: performing an absorption-release load ratio analysis on the root fertilizer efficiency release absorption extreme value data to obtain the root fertilizer efficiency absorption-release load ratio; Step S336: Based on the root fertilizer absorption and release load ratio, the root nutrient absorption rate interval and the root respiration conversion demand space, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio for fertilization and watering.

[0047] In the embodiment of the present invention, for the operation of step S331 "deducing the respiratory conversion demand space based on the root absorption restriction estimation data to obtain the root respiratory conversion demand space", the root absorption restriction estimation data is first rasterized in the form of a two-dimensional space grid, the grid unit is 0.5 meters × 0.5 meters, and the absorption restriction estimate of each grid unit is between 0.0 and 1.0. The higher the value, the more serious the root absorption restriction. In this operation, the basic root respiratory oxygen demand is set to 45 grams of oxygen per cubic meter of root zone per day, and the basic amount is modulated by the absorption restriction estimate. A linear correction method is used, and the absorption restriction value is set to α. The root unit volume oxygen demand is the basic oxygen demand multiplied by 1 plus the square of α to obtain the oxygen demand intensity at different positions. Subsequently, based on the spatial distribution map, the oxygen demand intensity of all grid cells was reconstructed by high-density contour interpolation. The discrete bidirectional B-spline interpolation method was used to reconstruct the continuous space of oxygen demand distribution by Laplace interpolation. The root distribution depth layer was distinguished by combining the seedling type information. Three depth layers were set for deciduous trees and two depth layers were set for evergreen shrubs, with a thickness of 0.2 meters for each layer. The root respiration conversion demand space was constructed in a three-dimensional coordinate system, and the output structure was a three-dimensional tensor array. The axial directions corresponded to the X and Y plane coordinates and the depth direction, respectively, and the tensor value represented the oxygen demand density per unit volume space. Taking a nursery unit numbered Z-7 as an example, the average absorption restriction estimate of the area was 0.53, and the basic oxygen demand was 45 grams. After modulation, it was 45 times 1 plus 0.53 squared, and the oxygen demand was about 57.65 grams. In this way, the root respiration conversion demand space based on the soil spatial structure and oxygen demand density as an indicator was completely established for subsequent drip irrigation control space mapping. For the operation of step S332 "calculating the interval of nutrient absorption rate per unit time for the root absorption limited estimation data to obtain the root nutrient absorption rate interval per unit time", firstly, a standardized root activity function is constructed based on the absorption limited estimation data. The function takes the absorption limited estimation value α as input, and corresponds to the reduction coefficients of the lower and upper limits of the root absorption rate per unit time in the interval [0,1]. Taking the total amount of fertilizer applied per day as a reference, the standard maximum absorption rate is defined as 0.45 grams of nitrogen, 0.38 grams of potassium, and 0.22 grams of phosphorus per square meter of root zone per day, which are respectively used as the maximum absorption rates under unrestricted conditions. In the specific operation, the nitrogen, phosphorus, and potassium absorption capacity of each spatial unit is scaled to the maximum absorption rate multiplied by (1 minus α) based on the absorption limited estimation value, and the lower limit of the absorption rate interval is generated as the maximum absorption rate multiplied by (1 minus α minus 0.1), and the lower limit is not less than zero. When α is greater than 0.9, the absorption rate interval is directly assigned to [0, 0]. All data are stored in a structured manner as a three-dimensional array, where the array dimensions correspond to the spatial position, element type, and upper and lower limits.Taking the nursery unit numbered D-5 as an example, the absorption limit estimate of this area is 0.37, then the upper limit of nitrogen absorption is 0.45 times (1 minus 0.37) to about 0.2835 grams, and the lower limit is 0.45 times (1 minus 0.47) to about 0.2385 grams; potassium is 0.38 times (1 minus 0.37) to about 0.2394 grams, and the lower limit is 0.38 times (1 minus 0.47) to about 0.2014 grams; phosphorus is 0.22 times (1 minus 0.37) to about 0.1386 grams, and the lower limit is 0.22 times (1 minus 0.47) to about 0.1166 grams. Through the above calculations, a complete space-time-nutrient three-dimensional absorption rate range is formed, which is used for the boundary constraint setting of fertilization control ratio and drip irrigation frequency. For the operation of step S333 "analyzing the fertilizer release rate of the slow-release fertilizer dosage control learning data to obtain the fertilizer release rate", firstly, the daily application amount in the slow-release fertilizer dosage control learning data is used as the basis, and the time-sharing release rate is deduced in combination with the fertilization period and soil environmental data. The release of slow-release fertilizer is mainly controlled by soil temperature, soil moisture and microbial activity. During implementation, daily environmental parameters are synchronously sampled and recorded, and the sampling time is set to once every two hours. The sampling parameters include the temperature (in degrees Celsius) and moisture content (in volume percentage) at 5 cm below the surface, and the soil microbial activity index is introduced as an auxiliary factor. The basic release model is set to a piecewise linear release model. When the soil temperature is 25 degrees Celsius, the humidity is 22%, and the microbial index is medium, the 24-hour release rate of each gram of slow-release fertilizer is 18%. Under different environmental conditions, the release coefficient is corrected according to the following adjustment rules: the release rate increases by 1.2 percentage points for every 1 degree Celsius increase in temperature, 0.6 percentage points for every 1 percentage point increase in humidity, and 1.5 percentage points for every level increase in microbial activity (divided into weak, medium, and strong). Taking the F-4 area as an example, the average temperature of the day is 27 degrees Celsius, the average humidity is 24%, and the microbial activity is "strong". The increase brought by temperature is 1.2 times 2, which is 2.4 percentage points, the increase brought by humidity is 0.6 times 2, which is 1.2 percentage points, and the increase brought by microbial factors is 1.5 percentage points. The total release rate is 18 plus 2.4 plus 1.2 plus 1.5, which is equal to 23.1%. If the daily fertilizer application rate of this unit is 2.5 grams, the daily release amount is 2.5 times 23.1%, which is about 0.5775 grams. In this way, the cumulative release of fertilizer application in all time periods is calculated, a time series of release amount every 2 hours within 24 hours is constructed, and the release rate of fertilizer effectiveness is summarized and output to form a structured time series matrix for the subsequent dynamic coordination of drip irrigation flow and fertilizer application amount.

[0048] For the operation of step S334 "slicing the fertilizer use release rate per unit time according to the root respiration conversion demand space and the root nutrient absorption rate interval to obtain the root fertilizer release absorption extreme value data", firstly, the root respiration conversion demand space tensor and the root nutrient absorption rate interval tensor per unit time are synchronously registered in the spatial dimension, and a unified three-dimensional space grid system is used as the index structure. The grid size is set to 0.25 meters × 0.25 meters × 0.2 meters, and all tensors are uniformly interpolated to the spatial resolution. Then, the fertilizer use release rate time series matrix is ​​sliced ​​synchronously in time, and the sliding segmentation is performed according to the time step per hour. The release rate value in each time step is compared point by point with the upper limit of the root absorption rate in this area, and the intersection time point where the difference is lower than the set threshold is extracted. The threshold is set to the fertilizer release rate is less than or equal to the upper limit of the nutrient absorption plus 5%. At the same time, according to the oxygen density value converted from root respiration, the above intersection time points are weighted and screened, and only the release-absorption intersection points in the area with oxygen density greater than 50 grams per cubic meter per day are retained. It is considered that the root activity in this area meets the high absorption requirements. All screening results are organized into a structured four-dimensional array, the dimensions include X coordinate, Y coordinate, depth index and time index. The array value records the release rate, absorption rate and oxygen density corresponding to the extreme absorption time point, which is defined as the root fertilizer release and absorption extreme data. Taking the unit area numbered F-12 in zone F as an example, in the third layer depth position, the nitrogen release rate is 0.12 grams per hour, the absorption upper limit is 0.13 grams per hour, and the oxygen density is 62 grams per cubic meter. If all conditions are met, then this time point is recorded as an extreme point, and the extreme time step is further recorded as the 6th hour, the corresponding coordinates are X=6.25, Y=8.00, and the depth index is 2.

[0049] For the operation of step S335 "analyzing the absorption and release load ratio of the extreme value data of root fertilizer release and absorption to obtain the root fertilizer absorption and release load ratio", the extreme value data extracted in step S334 are firstly processed by element, and analyzed separately according to the three nutrient categories of nitrogen, phosphorus and potassium. For each extreme value data point, its absorption rate and release rate values ​​are extracted, and the absorption and release load ratio is calculated by direct ratio method, that is, the absorption rate is used as the numerator and the release rate is used as the denominator to obtain the absorption and release load ratio of the spatial unit per unit time. To avoid extreme data disturbance, all data points with a release rate less than 0.01 grams per hour are excluded. All ratio results are reconstructed in the spatial dimension to generate a three-dimensional space field. The dimension corresponds to X, Y and depth indexes. The tensor value is the average absorption and release load ratio of the corresponding position. At the same time, the extreme value interval is recorded, and the maximum and minimum thresholds are set in the extreme value distribution area. Points with a comparison value greater than 1.5 or less than 0.3 are set as high load or low load areas and marked in a separate logical mask matrix. Taking the area numbered E-9 in the nursery as an example, at the 4th hour, the corresponding position is X=10.5, Y=5.25, the depth index is 1, the absorption rate is 0.18 grams per hour, the release rate is 0.15 grams per hour, and the calculated load ratio is 0.18 divided by 0.15, which is 1.2, falling into the normal load range. Taking the area numbered H-3 as an example, the absorption rate is 0.08 grams, the release rate is 0.2 grams, and the ratio is 0.4. It is determined to be a low-load area, and the point value is set to "L" in the mask matrix. In this way, a structured absorption-release load ratio data field is formed to provide quantitative indicator support for the next step of proportional fusion. For the operation of step S336 "based on the root fertilizer efficiency absorption and release load ratio, the root nutrient absorption rate interval and the root respiration conversion demand space, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio of fertilization and watering", a unified spatial data framework is first established to uniformly map the absorption and release load ratio tensor, the nutrient absorption rate interval tensor and the respiration conversion demand space tensor to the same spatial grid system. The dimension of each grid is 0.25 meters × 0.25 meters × 0.2 meters, and the three groups of data are spatially aligned. Subsequently, three types of control factors are defined: load factor, absorption factor, and oxygen demand factor, which are determined by the load ratio value, the upper limit of the absorption rate and the oxygen demand density value, respectively. For each spatial unit, the control weight coefficients are set to 0.4, 0.35, and 0.25, respectively, and the weighted linear combination method is used to fuse and generate the drip irrigation control index within the unit space. The value range of the drip irrigation control index is between 0 and 1. The higher the value, the higher the fertilization drip irrigation ratio. Based on the index value, it is divided into five level intervals, and the drip irrigation ratios are set to 0.6, 0.75, 0.9, 1.05, and 1.2 respectively.Taking the cell numbered G-6 as an example, the load ratio is 1.15, the upper limit of the absorption rate is 0.14 grams per hour, and the oxygen density is 55 grams per cubic meter. The control index is 1.15 times 0.4 plus 0.14 times 0.35 plus 0.055 times 0.25, and the result is about 0.689, which falls into the range of 0.6 to 0.75. The corresponding drip irrigation ratio is set to 0.75 times the benchmark drip irrigation flow rate, and the benchmark flow rate is 2 liters per hour. The final drip irrigation ratio of the unit is 1.5 liters per hour. The drip irrigation ratios of all units form a three-dimensional spatial structure matrix, which is matched and mapped with the number of the ground drip irrigation execution equipment to generate a spatial drip irrigation ratio control map, which is finally used to generate a segmented fertilization and watering control instruction sequence.

[0050] Step S4 includes the following steps: Step S41: normalizing the fertilization and watering segmented drip irrigation ratio to obtain a fertilization and watering segmented drip irrigation normalized ratio; Step S42: constructing a fertilization and watering control model for the segmented drip irrigation normalized ratio of fertilization and watering based on a strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; Step S43: Sending the garden seedling fertilization and watering control model to the terminal to execute the landscape garden seedling automated fertilization and watering control.

[0051] In the embodiment of the present invention, for the specific operation of step S41 "normalizing the segmented drip irrigation ratio of fertilization and watering to obtain the normalized segmented drip irrigation ratio of fertilization and watering", first collect the three-dimensional spatial structured drip irrigation ratio matrix data generated by step S336, the matrix covers all drip irrigation unit points in the area, and its drip irrigation ratio value is a value that has been weighted and fused according to the load ratio, oxygen density and absorption rate, and the unit is liter per hour. In order to ensure that the drip irrigation ratio of each area is uniformly scheduled and managed under the same control scale, a normalized operation benchmark is set, and the maximum and minimum values ​​of all drip irrigation ratio values ​​in the current monitoring period are used as boundaries, and linear normalization is used to map all drip irrigation ratio values ​​to the [0,1] interval. The normalization method is: subtract the global minimum value from the drip irrigation ratio of each unit, and then divide it by the difference between the maximum and minimum values. Taking the cell numbered F-8 as an example, if its original drip irrigation ratio is 1.05 liters per hour, the minimum drip irrigation ratio in the current area is 0.6 liters per hour, and the maximum drip irrigation ratio is 1.2 liters per hour, then the normalized value is (1.05 minus 0.6) divided by (1.2 minus 0.6), and the result is 0.75. All normalized values ​​retain their three-dimensional grid index structure unchanged according to the spatial position, and finally form a normalized drip irrigation ratio space tensor. The tensor dimension is exactly the same as the original drip irrigation ratio matrix, which serves as the input data basis for subsequent strategy optimization and control model construction. All normalized values ​​are retained to three decimal places to ensure accuracy consistency. All calculation processes are carried out in a fixed parameter control environment, and the normalized results are written into an independent control parameter structure and bound to the independent identification code of each control area to ensure that the subsequent model construction process has the integrity of the corresponding relationship.

[0052] For the specific operation of step S42 "Based on the policy gradient algorithm, the fertilization and watering control model of the segmented drip irrigation normalized ratio is constructed to obtain the garden seedling fertilization and watering control model", the policy gradient algorithm based on the advantage function is used to realize the construction of the control strategy. First, the drip irrigation normalized ratio tensor generated in step S41 is used as the state input, and the historical job execution record data and soil moisture content change data are used as the feedback data set to establish a state-behavior-feedback triple data set. The state space is defined as the normalized drip irrigation ratio tensor at the current moment, the behavior space is defined as the two-dimensional vector composed of the combination of the drip irrigation opening time and the fertilizer concentration of each cell, and the feedback is the product function value of the soil moisture content change value and the root activity change value in the area within two hours. For this triple, the behavior strategy function is designed to construct a probability strategy function in the form of Gaussian distribution, and the behavior advantage function is used to enhance the current strategy. The advantage function is composed of the return value minus the state value function expectation, where the state value function is obtained by fitting the historical feedback data. The Monte Carlo sampling method is used to generate the state-behavior path for each round of update. The number of samples is set to 1000 trajectories per round. In each iteration, the gradient direction of the current policy under the normalized ratio is calculated according to the policy gradient formula. The Adam optimizer with a learning rate of 0.001 is used for gradient ascent. The policy function parameters are updated once after each round of iteration, and the number of iterations is set to 500. All training processes are executed under the CPU-GPU collaborative architecture, and the state tensors are sent to the memory in batches. The single-round calculation time is controlled within 20 seconds. Taking the area numbered G-10 as an example, in the 200th round of training, its drip irrigation normalization ratio is 0.78, and the optimal behavior output by the policy network is to turn on drip irrigation for 7 minutes and the fertilizer concentration is 0.06 grams per liter. Finally, all state-behavior mapping relationships are organized into a set of continuous function expressions, and a control mapping function group is established under the corresponding spatial index to form a parameter set of the garden seedling fertilization and watering control model. For the specific operation of step S43 "sending the garden seedling fertilization and watering control model to the terminal to perform the automated fertilization and watering control of landscape garden seedlings", firstly, the control strategy data structure corresponding to each cell in the fertilization and watering control model constructed by step S42 is packaged. The format adopts the binary compressed structure format. The file structure includes six items: area number, three-dimensional coordinate index, normalized ratio, corresponding fertilizer concentration, drip irrigation opening time and strategy evaluation score. The batch number of all structures is T-SFQ-0327. In order to ensure data compatibility with the terminal control system, all structure field encodings follow the preset control protocol standard, using 32-bit floating point numbers to represent numerical fields, and the coordinate index field is 8-bit integer encoding. After the data is packaged, it is sent to the control receiving module corresponding to each terminal number through the standard LAN communication module in UDP communication protocol packets. The communication port number is set to DIO-13, and the communication message cycle is set to update every 120 seconds.After receiving the policy model data packet, the terminal device calls the drip irrigation and fertilization control driver in the local control execution module, restores the received normalized ratio to a specific drip irrigation flow, and controls the pump pressure and the opening rate of the fertilizer solenoid valve according to the fertilizer concentration and the opening time. Taking the seedling area numbered D-4 as an example, the fertilizer concentration in the received policy control structure is 0.07 grams per liter, and the opening time is 8 minutes. The terminal controller calculates the total drip irrigation flow rate as 1.6 liters and the total fertilizer amount as 0.112 grams based on this data. The system execution cycle completes all operation processes within 1 minute after the data is received, and records the execution status code and uploads it to the control center for closed-loop control status feedback. The above method completes the deployment of the control model at the terminal and the precise implementation of the automated operation of fertilization and watering.

[0053] The present invention also provides a landscape garden seedling automatic fertilization and watering control system, which is used to execute the landscape garden seedling automatic fertilization and watering control method as described above. The landscape garden seedling automatic fertilization and watering control system comprises: The soil state characteristic monitoring module is used to collect the seedling morphology data of landscape garden seedlings through electronic monitoring to obtain the seedling morphology data; deploy soil monitoring sensors under each landscape garden seedling, and monitor the soil state characteristics through soil detection sensors to obtain soil state characteristic monitoring characteristic data; The root absorption limitation estimation module is used to predict the density of seedling root systems based on seedling morphological data and generate seedling root density prediction data; to analyze the abnormal fluctuation of soil salt content on the soil state characteristic monitoring characteristic data and obtain the abnormal fluctuation data of soil salt content; to estimate the root absorption limitation of the seedling root density prediction data based on the abnormal fluctuation data of soil salt content and obtain the root absorption limitation estimation data; The segmented drip irrigation ratio fusion module is used to control and match the abnormal fluctuation data of soil salt content with the slow-release fertilizer dosage according to the root absorption limitation estimation data to obtain the slow-release fertilizer dosage control data; the segmented drip irrigation ratio is fused based on the slow-release fertilizer dosage control data to obtain the segmented drip irrigation ratio for fertilization and watering; The fertilization and watering control model construction module is used to construct a fertilization and watering control model for the fertilization and watering segmented drip irrigation ratio based on the strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; the garden seedling fertilization and watering control model is sent to the terminal to execute automated fertilization and watering control of landscape seedlings.

[0054] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for controlling automatic fertilization and watering of landscape garden seedlings, characterized in that: The following steps are involved: Step S1: collecting seedling morphological data of landscape garden seedlings through electronic monitoring to obtain seedling morphological data; Soil monitoring sensors are deployed under each landscape garden seedling, and soil state characteristics are monitored through soil detection sensors to obtain soil state characteristic monitoring characteristic data; Step S2: predicting the density of seedling root systems based on seedling morphology data to generate seedling root density prediction data; analyzing abnormal fluctuations in soil salinity on soil state characteristic monitoring feature data to obtain abnormal fluctuations in soil salinity; estimating root absorption limitation on seedling root density prediction data based on abnormal fluctuations in soil salinity to obtain root absorption limitation estimation data; Step S3: performing slow-release fertilizer dosage control matching on the abnormal fluctuation data of soil salt content according to the root absorption limitation estimation data to obtain slow-release fertilizer dosage control data; Based on the slow-release fertilizer dosage control data, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio for fertilization and watering; Step S4: construct a fertilization and watering control model for the segmented drip irrigation ratio of fertilization and watering based on the strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; send the garden seedling fertilization and watering control model to the terminal to execute automated fertilization and watering control of landscape seedlings.

2. The method for controlling the automatic fertilization and watering of landscape garden seedlings according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: collecting seedling morphological data of landscape garden seedlings through electronic monitoring to obtain seedling morphological data; Step S12: deploying soil monitoring sensors under each landscape garden seedling, and monitoring soil state characteristics through the soil detection sensors to obtain soil state characteristic monitoring data; Step S13: performing data cleaning on the soil state characteristic monitoring data to obtain soil state characteristic monitoring cleansing data; Step S14: performing state characteristic feature analysis on the soil state characteristic monitoring cleaning data to obtain soil state characteristic monitoring characteristic data.

3. The automatic fertilization and watering control method for landscape gardening seedlings according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: predicting the density of seedling root systems based on the seedling morphology data to generate seedling root density prediction data; Step S22: analyzing abnormal fluctuation of soil salt content on the soil state characteristic monitoring characteristic data to obtain abnormal fluctuation data of soil salt content; Step S23: analyzing the soil nutrient imbalance effectiveness loss on the soil state characteristic monitoring characteristic data based on the soil salt content abnormal fluctuation data to obtain soil nutrient imbalance effectiveness loss data; Step S24: performing root absorption limitation estimation on the seedling root density prediction data according to the soil salt content abnormal fluctuation data and the soil nutrient imbalance effectiveness loss data to obtain root absorption limitation estimation data.

4. The automatic fertilization and watering control method for landscape gardening seedlings according to claim 3 is characterized in that: Step S23 includes the following steps: Step S231: performing ion concentration decomposition processing on the abnormal fluctuation data of soil salt content to generate time series variation data of salt ion concentration; Step S232: extracting organic matter content fluctuation from soil state characteristic monitoring feature data to obtain soil organic matter content fluctuation data; Step S233: performing nonlinear intensity regression analysis of cell membrane damage based on the salt ion concentration time series variation data to obtain cell membrane damage intensity regression data; Step S234: performing an equal amount simulation prediction of organic nutrient leakage on the soil organic matter content fluctuation data according to the cell membrane damage intensity regression data to obtain equal amount prediction data of nutrient leakage; Step S235: performing ion competition intensity increment analysis on soil organic matter content fluctuation data according to salt ion concentration time series variation data to generate ion competition intensity increment data; Step S236: Perform soil nutrient imbalance effectiveness loss analysis based on the nutrient leakage equal amount prediction data and the ion competition intensity increment data to obtain soil nutrient imbalance effectiveness loss data.

5. The method for controlling the automatic fertilization and watering of landscape garden seedlings according to claim 4, characterized in that: Step S235 includes the following steps: Perform mutation increment index calculation on the time series variation data of salt ion concentration to obtain the mutation increment index of salt ion concentration; Based on the salt ion concentration mutation increment index, the salt ion concentration time series change data is combined by continuous concentration mutation approximation to generate concentration continuous mutation approximation combination data; Based on the concentration continuity mutation approximation combination data, the soil organic matter content fluctuation data was simulated by the preemptive continuous intensity of salt ion absorption sites, and the preemptive continuous intensity data of salt ion absorption sites were obtained. The preemptive continuous intensity data of the absorption site are processed by calculating the time interval difference of the ion flux to obtain the difference of the salt ion flux at the absorption site; The ion competition intensity increment analysis is performed based on the difference in salt ion flux at the absorption site to generate ion competition intensity increment data.

6. The method for controlling the automatic fertilization and watering of landscape garden seedlings according to claim 3, characterized in that: Step S24 includes the following steps: Step S241: performing fluctuation amplitude deviation rate calculation on the abnormal fluctuation data of soil salt content to obtain salt content fluctuation amplitude difference data; Step S242: converting and quantifying the soil solution osmotic pressure increase ratio according to the salt content fluctuation amplitude difference data to obtain the soil solution osmotic pressure increase ratio; Step S243: integrating the fluctuation of the microbial nutrient conversion inhibition value based on the soil solution osmotic pressure increase ratio to obtain the nutrient conversion inhibition fluctuation value; Step S244: estimating the soil compaction cumulative gradient according to the salt content fluctuation amplitude difference data and the soil solution osmotic pressure increase ratio to obtain soil compaction cumulative gradient data; Step S245: performing root growth / respiration restriction gradient evaluation on the seedling root density prediction data based on the soil compaction cumulative gradient data to obtain root growth / respiration restriction gradient data; Step S246: Root absorption limitation estimation is performed on the seedling root density prediction data according to the nutrient conversion inhibition fluctuation value, root growth / respiration restriction gradient data and soil nutrient imbalance effectiveness loss data to obtain root absorption limitation estimation data.

7. The automatic fertilization and watering control method for landscape gardening seedlings according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: performing slow-release fertilizer dosage control matching on the abnormal fluctuation data of soil salt content according to the root absorption limitation estimation data to obtain slow-release fertilizer dosage control data; Step S32: performing dosage control logic learning on the slow-release fertilizer dosage control data to obtain slow-release fertilizer dosage control learning data; Step S33: Based on the slow-release fertilizer dosage control learning data and the root absorption limitation estimation data, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio for fertilization and watering.

8. The method for controlling the automatic fertilization and watering of landscape garden seedlings according to claim 7, characterized in that: Step S33 includes the following steps: Step S331: Deducing the respiratory conversion demand space based on the root absorption limitation estimation data to obtain the root respiratory conversion demand space; Step S332: performing nutrient absorption rate interval calculation per unit time on the root absorption limitation estimation data to obtain the root nutrient absorption rate interval per unit time; Step S333: analyzing the fertilizer effect release rate of the slow-release fertilizer dosage control learning data to obtain the fertilizer effect dosage release rate; Step S334: dividing the fertilizer efficiency dosage release rate into dosage absorption extreme value points per unit time according to the root respiration conversion demand space and the root nutrient absorption rate interval to obtain root fertilizer efficiency release absorption extreme value data; Step S335: performing an absorption-release load ratio analysis on the root fertilizer efficiency release absorption extreme value data to obtain the root fertilizer efficiency absorption-release load ratio; Step S336: Based on the root fertilizer absorption and release load ratio, the root nutrient absorption rate interval and the root respiration conversion demand space, the segmented drip irrigation ratio is integrated to obtain the segmented drip irrigation ratio for fertilization and watering.

9. The method for controlling the automatic fertilization and watering of landscape garden seedlings according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: normalizing the fertilization and watering segmented drip irrigation ratio to obtain a fertilization and watering segmented drip irrigation normalized ratio; Step S42: constructing a fertilization and watering control model for the segmented drip irrigation normalized ratio of fertilization and watering based on a strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; Step S43: Sending the garden seedling fertilization and watering control model to the terminal to execute the landscape garden seedling automated fertilization and watering control.

10. A landscape garden seedling automatic fertilization and watering control system, characterized in that: Used to execute the landscape gardening seedling automatic fertilization and watering control method as claimed in claim 1, the landscape gardening seedling automatic fertilization and watering control system comprises: The soil state characteristic monitoring module is used to collect the seedling morphology data of landscape garden seedlings through electronic monitoring to obtain the seedling morphology data; deploy soil monitoring sensors under each landscape garden seedling, and monitor the soil state characteristics through soil detection sensors to obtain soil state characteristic monitoring characteristic data; The root absorption limitation estimation module is used to predict the density of seedling root systems based on seedling morphological data and generate seedling root density prediction data; to analyze the abnormal fluctuation of soil salt content on the soil state characteristic monitoring characteristic data and obtain the abnormal fluctuation data of soil salt content; to estimate the root absorption limitation of the seedling root density prediction data based on the abnormal fluctuation data of soil salt content and obtain the root absorption limitation estimation data; The segmented drip irrigation ratio fusion module is used to control and match the abnormal fluctuation data of soil salt content with the slow-release fertilizer dosage according to the root absorption limitation estimation data to obtain the slow-release fertilizer dosage control data; the segmented drip irrigation ratio is fused based on the slow-release fertilizer dosage control data to obtain the segmented drip irrigation ratio for fertilization and watering; The fertilization and watering control model construction module is used to construct a fertilization and watering control model for the fertilization and watering segmented drip irrigation ratio based on the strategy gradient algorithm to obtain a garden seedling fertilization and watering control model; the garden seedling fertilization and watering control model is sent to the terminal to execute automated fertilization and watering control of landscape seedlings.

Images