Orchard nitrogen and phosphorus interception method and system based on intelligent monitoring

Abstract

The invention relates to the technical field of data processing, and discloses an orchard nitrogen and phosphorus interception method and system based on intelligent monitoring. The method comprises the steps of collecting orchard multi-sensor data and performing standardization processing to generate a monitoring data set; learning a nitrogen and phosphorus distribution rule by using a variational auto-encoder to generate an interception parameter relation model; calculating regulation and control instructions of the groove, the PRB and the sedimentation tank through a multi-objective optimization algorithm; all the intercepting devices are controlled to operate cooperatively to collect nitrogen-phosphorus-rich sediments; and recycling sediment to prepare slow-release fertilizer to be applied to the orchard again. Through data enhancement of the variational auto-encoder and a multi-objective optimization algorithm, the problems of insufficient training samples and single parameter optimization in orchard nitrogen and phosphorus interception are solved, and the prediction precision of intelligent monitoring and the cooperative control efficiency of an interception system are remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and system for intercepting nitrogen and phosphorus in an orchard based on intelligent monitoring. Background Art

[0002] In fruit-growing areas, nitrogen and phosphorus loss is a major cause of eutrophication in some water bodies. Existing nitrogen and phosphorus interception technologies primarily include centralized treatment facilities, constructed wetlands, and ecological buffer zones, which remove nitrogen and phosphorus pollutants from runoff through physical, chemical, or biological means. Traditional intelligent monitoring systems typically use a single sensor network to collect environmental parameters and apply basic data analysis algorithms to adjust these parameters and monitor the operating status of interception devices. Some advanced systems have begun applying machine learning algorithms to predict nitrogen and phosphorus concentrations and, combined with automated control technologies, adjust the operating parameters of interception equipment.

[0003] However, existing technologies have significant shortcomings: the high procurement and operating costs of nitrogen and phosphorus sensors lead to insufficient monitoring data samples, which seriously restricts the training effect and prediction accuracy of intelligent algorithms; traditional methods lack the ability to deeply learn the distribution patterns of nitrogen and phosphorus in orchard runoff, and cannot generate sufficient training samples to optimize interception parameters; existing single-objective optimization methods find it difficult to find a balance between nitrogen and phosphorus removal efficiency and operating costs, affecting the economic feasibility of the system.

[0004] Based on the above analysis, the core problem of existing technologies lies in the contradiction between data scarcity and the need for algorithm optimization. High sensor costs limit the scale of data collection, resulting in a lack of sufficient training samples for intelligent algorithms. This in turn affects the accuracy of parameter optimization, ultimately restricting the intelligence level and operational efficiency of the entire interception system. Therefore, there is an urgent need to develop an intelligent nitrogen and phosphorus interception method that can address the sample shortage issue through data augmentation technology and achieve multi-objective collaborative optimization. Summary of the Invention

[0005] The present application provides a method and system for intercepting nitrogen and phosphorus in orchards based on intelligent monitoring, which solves the problems of insufficient training samples and single parameter optimization in nitrogen and phosphorus interception in orchards through variational autoencoder data enhancement and multi-objective optimization algorithm, and significantly improves the prediction accuracy of intelligent monitoring and the collaborative control efficiency of the interception system.

[0006] In a first aspect, the present application provides an orchard nitrogen and phosphorus interception method based on intelligent monitoring, the orchard nitrogen and phosphorus interception method based on intelligent monitoring comprising: step S101, collecting data from an orchard nitrogen and phosphorus concentration sensor, a soil moisture sensor, a rainfall sensor, and a flow sensor, and generating an orchard nitrogen and phosphorus monitoring dataset through outlier detection and standardization processing; Step S101: Collect data from orchard nitrogen and phosphorus concentration sensors, soil moisture sensors, rainfall sensors, and flow sensors, and generate an orchard nitrogen and phosphorus monitoring data set through outlier detection and standardization. Step S102: Based on the orchard nitrogen and phosphorus monitoring dataset, a variational autoencoder algorithm is used to learn the distribution pattern of nitrogen and phosphorus in orchard runoff, and a relationship model between nitrogen and phosphorus content and interception parameters is generated; Step S103: using the relationship model between nitrogen and phosphorus content and interception parameters, a multi-objective optimization algorithm is used to calculate control instructions for the water flow velocity of the stepped groove, the filling amount of the PRB adsorption material, and the flocculant dosage of the sedimentation tank; Step S104: controlling the operation of the trench regulating valve, the PRB filling system, and the sedimentation tank stirring device according to the control instructions to intercept nitrogen and phosphorus in the runoff and collect nitrogen and phosphorus-rich sediments; Step S105: The nitrogen- and phosphorus-rich sediment is recovered and processed by a suction device to prepare slow-release nitrogen fertilizer and phosphorus fertilizer for re-application to the orchard.

[0007] Optionally, step S101 includes: The real-time nitrogen and phosphorus content data in the orchard soil and runoff are collected by nitrogen and phosphorus concentration sensors to obtain the original nitrogen and phosphorus concentration data series; The soil moisture sensor and rainfall sensor are used to collect environmental parameter data, and the flow sensor is used to obtain runoff flow change data to obtain the orchard environmental monitoring parameter set. Performing 3σ criterion outlier detection processing on the original nitrogen and phosphorus concentration data sequence, eliminating data points outside the normal range, and obtaining cleaned nitrogen and phosphorus concentration data; The cleaned nitrogen and phosphorus concentration data and the orchard environmental monitoring parameter set are normalized and standardized, and the data dimensions are unified to the [0, 1] interval to obtain an orchard nitrogen and phosphorus monitoring data set.

[0008] Optionally, step S102 includes: Constructing a variational autoencoder encoder network, taking the orchard nitrogen and phosphorus monitoring dataset as input for feature extraction, and outputting a potential feature vector of orchard nitrogen and phosphorus loss; Constructing a variational autoencoder decoder network, reconstructing the original data using the potential feature vector of nitrogen and phosphorus loss in the orchard as input, and training the complete variational autoencoder model in combination with the KL divergence loss function; Running the variational autoencoder model to learn the spatiotemporal distribution characteristics of nitrogen and phosphorus in orchard runoff, generating synthetic samples of nitrogen and phosphorus concentrations that conform to the orchard environmental characteristics, and expanding them to form an enhanced nitrogen and phosphorus training dataset; Based on the enhanced nitrogen and phosphorus training data set, a nonlinear mapping relationship between nitrogen and phosphorus content input and stepped groove water velocity, PRB adsorption material filling amount, and sedimentation tank flocculant dosage output is established to form a nitrogen and phosphorus content and interception parameter relationship model.

[0009] Optionally, step S103 includes: A multi-objective optimization algorithm framework is established, with the relationship model between nitrogen and phosphorus content and interception parameters as constraints, and nitrogen and phosphorus removal efficiency and operating cost as dual objective functions; The current real-time nitrogen and phosphorus concentration data of the orchard is input into the nitrogen and phosphorus content and interception parameter relationship model to calculate the theoretical processing load parameters of the three-stage interception system of stepped trenches, PRB reaction walls, and sedimentation tanks; Running a multi-objective optimization algorithm to search for a Pareto solution set, and calculating a combination of a stepped trench water velocity range, a PRB adsorption material filling amount range, and a sedimentation tank flocculant dosage range under the constraints of the theoretical treatment load parameters; The balance point between nitrogen and phosphorus removal efficiency and operating cost is selected from the combination scheme, and the specific values ​​of the water flow velocity in the stepped groove, the filling amount of the PRB adsorption material, and the flocculant dosage in the sedimentation tank are determined as control instructions.

[0010] Optionally, step S104 includes: parsing the stepped groove water flow velocity parameters in the control instructions, driving the groove regulating valve to perform an opening adjustment action, and controlling the flow rate of runoff through the groove and the contact time between nitrogen and phosphorus and the iron oxide coated sand; parsing the PRB adsorption material filling amount parameter in the control instruction, starting the PRB filling system to perform adsorption material replenishment or replacement operations, and maintaining the active adsorption capacity of the calcium-based material and the iron oxide coated sand in the PRB reaction wall; Analyze the flocculant dosage parameter of the sedimentation tank in the control instruction, control the rotation speed of the sedimentation tank stirring device and the flow rate of the polyaluminum chloride dosing pump, and promote the formation of flocculation and precipitation of residual nitrogen and phosphorus in the runoff and sedimentation to the bottom of the tank; Through three-stage coordinated interception treatment of stepped trenches, PRB reaction walls, and sedimentation tanks, the nitrogen and phosphorus in the runoff are gradually enriched into insoluble sediments and gathered at the bottom of the sedimentation tank to form nitrogen and phosphorus-rich sediments.

[0011] Optionally, parsing the sedimentation tank flocculant dosage parameter in the control instruction, controlling the rotation speed of the sedimentation tank stirring device and the flow rate of the polyaluminum chloride dosing pump, and promoting the residual nitrogen and phosphorus in the runoff to form flocculation precipitation and settle to the bottom of the tank, includes: Reading the numerical parameter of the flocculant dosage of the sedimentation tank in the control instruction, calculating the target flow rate and dosage duration of the polyaluminium chloride dosing pump, and generating a flocculant dosage control signal; Start the polyaluminum chloride dosing pump based on the flocculant dosing control signal, and quantitatively dose the polyaluminum chloride solution into the sedimentation tank according to the calculated target flow rate to form a flocculation reaction environment in the tank; According to the stirring device speed parameter in the regulation instruction, the rotation speed of the stirring blade in the sedimentation tank is adjusted to control the water flow shear force and mixing intensity in the tank, so that the residual nitrogen and phosphorus and the polyaluminum chloride occur flocculation reaction to generate large particle flocs; Stop the stirring device to enter the standing and settling stage, and use gravity to make the large particle flocs sink to the bottom of the sedimentation tank to separate the clear upper water body and the nitrogen and phosphorus-rich sediment at the bottom.

[0012] Optionally, step S105 comprises: Start the sedimentation tank bottom suction equipment to perform negative pressure extraction operation on the nitrogen and phosphorus-rich sediment, transport the sediment from the bottom of the sedimentation tank to the dewatering treatment device, and separate the solid and liquid phases to obtain concentrated nitrogen and phosphorus sludge; Transport the concentrated nitrogen and phosphorus sludge to the intelligent granulation system, add biochar carriers and slow-release coating materials for mixing and granulation treatment, control the granulation temperature and pressure parameters, and form uniform slow-release fertilizer particles; Detect and analyze the nitrogen and phosphorus content of the slow-release fertilizer particles, adjust the nitrogen and phosphorus ratio in the particles according to the detection results, and generate slow-release nitrogen and phosphorus fertilizer products that meet the orchard soil nutrient requirements; Use the intelligent fertilization machinery to apply the slow-release nitrogen and phosphorus fertilizer products according to the fruit tree growth stage and soil nutrient condition to complete the closed loop cycle process of nitrogen and phosphorus nutrients from interception and recovery to reuse.

[0013] In a second aspect, the present application provides an orchard nitrogen and phosphorus interception system based on intelligent monitoring, comprising: The acquisition module is configured to acquire fruit orchard nitrogen and phosphorus concentration sensor, soil humidity sensor, rainfall sensor and flow sensor data, and generate fruit orchard nitrogen and phosphorus monitoring data set after abnormal value detection and standardization processing; The generation module is configured to learn the distribution rule of orchard runoff nitrogen and phosphorus based on the fruit orchard nitrogen and phosphorus monitoring data set by using a variational autoencoder algorithm, and generate a nitrogen and phosphorus content and interception parameter relationship model; The calculation module is configured to calculate the regulation instruction of the stepped trench flow velocity, the PRB adsorption material filling amount, and the sedimentation tank flocculant dosing amount by using the nitrogen and phosphorus content and interception parameter relationship model through a multi-objective optimization algorithm; The control module is configured to control the trench adjusting valve, the PRB filling system and the sedimentation tank stirring device to run according to the regulation instruction, intercept the nitrogen and phosphorus in the runoff, and collect the nitrogen and phosphorus-rich sediment; The processing module is used to recover and process the nitrogen- and phosphorus-rich sediment through suction equipment to prepare slow-release nitrogen fertilizer and phosphorus fertilizer for re-application to the orchard.

[0014] In a third aspect, an orchard nitrogen and phosphorus interception device based on intelligent monitoring is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the orchard nitrogen and phosphorus interception device based on intelligent monitoring executes the above-mentioned orchard nitrogen and phosphorus interception method based on intelligent monitoring.

[0015] In a fourth aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, which, when executed on a computer, enables the computer to execute the above-mentioned orchard nitrogen and phosphorus interception method based on intelligent monitoring.

[0016] In the technical solution provided by this application, by collecting data from orchard nitrogen and phosphorus concentration sensors, soil moisture sensors, rainfall sensors and flow sensors, and generating an orchard nitrogen and phosphorus monitoring data set through outlier detection and standardization, high-quality basic data support is provided for subsequent intelligent algorithms, avoiding the algorithm training bias problem caused by inconsistent data quality and dimensional differences in traditional methods. The variational autoencoder algorithm is used to learn the distribution law of nitrogen and phosphorus in orchard runoff, and the generated nitrogen and phosphorus content and interception parameter relationship model effectively solves the problem of insufficient training samples caused by the high cost of nitrogen and phosphorus sensors. The data enhancement technology significantly expands the sample size that can be used for model training, and improves the model's adaptability and prediction accuracy to complex changes in the orchard environment. Using the nitrogen and phosphorus content and interception parameter relationship model, the multi-objective optimization algorithm is used to calculate the control instructions for the stepped ditch water flow velocity, the PRB adsorption material filling amount, and the sedimentation tank flocculant dosage, thereby achieving an intelligent balance between nitrogen and phosphorus removal efficiency and operating costs, overcoming the technical defects of the traditional single-objective optimization method that cannot take into account both economy and environmental protection.

[0017] The coordinated control mechanism of controlling the operation of the trench regulating valve, PRB filling system and sedimentation tank stirring device according to the control instructions to intercept nitrogen and phosphorus in the runoff and collect nitrogen and phosphorus-rich sediments has broken through the limitations of the traditional manual management method with delayed response and low adjustment accuracy, and realized the real-time linkage and precise control of multi-stage interception devices. The closed-loop utilization method based on the recovery and treatment of nitrogen and phosphorus-rich sediments through suction equipment to prepare slow-release nitrogen fertilizer and phosphorus fertilizer for re-application to the orchard not only solves the problem of difficult sediment disposal in traditional interception methods, but also realizes the on-site recycling of nutrients, organically combining environmental governance with resource recovery. The specific application of the variational autoencoder algorithm in the field of nitrogen and phosphorus interception in orchards in the entire technical solution fully considers the seasonal changes and spatial distribution characteristics of orchard runoff. Through the targeted optimization of the algorithm structure, the data enhancement effect is more in line with the actual application scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0019] Figure 1 This is a schematic diagram of an embodiment of a method for intercepting nitrogen and phosphorus in an orchard based on intelligent monitoring in an embodiment of the present application; Figure 2 This is a schematic diagram of an embodiment of the nitrogen and phosphorus interception treatment process in an orchard based on intelligent monitoring in an embodiment of the present application; Figure 3 This is a schematic diagram of an embodiment of an orchard nitrogen and phosphorus interception system based on intelligent monitoring in an embodiment of the present application; Figure 4 It is a schematic block diagram of the structure of an orchard nitrogen and phosphorus interception device based on intelligent monitoring in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The embodiments of the present application provide a method and system for intercepting nitrogen and phosphorus in an orchard based on intelligent monitoring. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices.

[0021] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiments of the present application, an embodiment of the orchard nitrogen and phosphorus interception method based on intelligent monitoring includes: Step S101: Collect data from orchard nitrogen and phosphorus concentration sensors, soil moisture sensors, rainfall sensors, and flow sensors, and generate an orchard nitrogen and phosphorus monitoring data set through outlier detection and standardization. Step S102: Based on the orchard nitrogen and phosphorus monitoring dataset, a variational autoencoder algorithm is used to learn the distribution patterns of nitrogen and phosphorus in orchard runoff, and a relationship model between nitrogen and phosphorus content and interception parameters is generated; Step S103: using the relationship model between nitrogen and phosphorus content and interception parameters, a multi-objective optimization algorithm is used to calculate the control instructions for the water flow velocity of the stepped groove, the filling amount of the PRB adsorption material, and the flocculant dosage of the sedimentation tank; Step S104: controlling the operation of the groove regulating valve, the PRB filling system, and the sedimentation tank stirring device according to the control instructions to intercept nitrogen and phosphorus in the runoff and collect nitrogen and phosphorus-rich sediments; Step S105: The nitrogen- and phosphorus-rich sediment is recovered and processed by a suction device to prepare slow-release nitrogen fertilizer and phosphorus fertilizer for re-application to the orchard.

[0022] It is understandable that the execution subject of this application can be an orchard nitrogen and phosphorus interception system based on intelligent monitoring, or a terminal or a server, which is not limited here. The embodiment of this application is described by taking the server as the execution subject as an example.

[0023] Specifically, the implementation process of nitrogen and phosphorus interception in orchards begins with the coordinated data collection of multi-source sensors. Nitrogen and phosphorus concentration sensors are deployed in soil profiles and runoff channels to monitor in real time. 、 and To monitor changes in nitrogen and phosphorus content, soil moisture sensors were embedded in the active root zone of fruit trees, rainfall sensors were installed in open areas, and flow sensors were placed at runoff collection points. Each sensor uploaded data synchronously at 5-minute intervals via IoT nodes. The raw monitoring data entered anomaly detection, using a dynamic threshold method to identify and remove outliers outside a reasonable range. Missing data was interpolated through spatiotemporal correlation analysis. The processed data stream underwent min-max normalization, converting parameters of different dimensions to the [0, 1] range, forming a structured orchard nitrogen and phosphorus monitoring dataset.

[0024] A standardized dataset is fed into a variational autoencoder for feature learning. The encoder network compresses the 6-dimensional input into an 8-dimensional latent space, capturing the spatiotemporal distribution of nitrogen and phosphorus loss. The decoder network reconstructs the original data distribution from the latent space. During training, the KL divergence term constrains the latent space to follow a standard normal distribution. The trained model generates synthetic samples by randomly sampling in the latent space, expanding the size of the original dataset. This augmented data is used to train a feedforward neural network, mapping nitrogen and phosphorus content and environmental parameters to interception parameters. The model outputs three key parameters: water flow velocity in the stepped trench, PRB adsorption material loading, and flocculant dosage in the sedimentation tank.

[0025] The multi-objective optimization algorithm takes nitrogen and phosphorus removal efficiency and operation cost as optimization objectives, takes the theoretical treatment load output by the relational model as a constraint condition, and searches for a Pareto optimal solution set by using an improved NSGA-II algorithm. The algorithm initializes a population to randomly generate parameter combinations in the constraint range, screens high-quality individuals by non-dominated sorting and congestion degree calculation, and outputs optimal parameter combinations after 200 generations of evolution. The optimization results are selected by decision preference analysis to balance points to generate control instructions containing specific numerical values.

[0026] The control instructions are issued to the field execution equipment through the bus. The trench regulating valve adjusts the opening according to the water flow velocity set value, the PRB filling system supplements the adsorption material according to the calculated amount, and the sedimentation tank injection pump injects flocculants according to the optimization results. The three-stage interception system operates cooperatively. The trench removes particulate nitrogen and phosphorus by physical interception, the PRB reaction wall removes dissolved pollutants by chemical adsorption, and the sedimentation tank removes colloidal residues by flocculation and sedimentation, finally forming a nitrogen and phosphorus-rich sediment at the bottom of the tank.

[0027] After the sediment accumulates to a set thickness, the recovery program is started. The screw pump transports the sludge to the dewatering device, and the concentrated sludge is obtained after centrifugal separation. The sludge is mixed with biochar and coated materials in a certain proportion, and granulated by a granulator to form slow-release fertilizer particles. A near-infrared detector monitors the nutrient content of the particles in real time and adjusts the formula. The finished fertilizer is variable applied by an intelligent fertilizer applicator according to the fruit tree's fertilizer requirement characteristics and soil moisture conditions, completing the material cycle from nitrogen and phosphorus interception to nutrient recycling. The whole system realizes efficient interception and resource utilization of nitrogen and phosphorus loss in the orchard through closed-loop control of the sensor network, optimization algorithm and execution device.

[0028] In a specific embodiment, the process of performing step S101 can specifically include the following steps: Collect real-time nitrogen and phosphorus content data in the orchard soil and runoff by nitrogen and phosphorus concentration sensors to obtain an original nitrogen and phosphorus concentration data sequence; Collect environmental parameter data based on soil humidity sensors and rainfall sensors, and obtain runoff flow change data by combining with flow sensors to obtain an orchard environmental monitoring parameter set; Perform 3σ rule outlier detection processing on the original nitrogen and phosphorus concentration data sequence to remove data points outside the normal range to obtain cleaned nitrogen and phosphorus concentration data; Perform normalization standardization processing on the cleaned nitrogen and phosphorus concentration data and the orchard environmental monitoring parameter set to unify the data dimension to the [0, 1] interval to obtain an orchard nitrogen and phosphorus monitoring data set.

[0029] Specifically, the construction of the orchard nitrogen and phosphorus monitoring data set starts with the synchronous data collection of the nitrogen and phosphorus concentration sensors, the soil humidity sensors, the rainfall sensors, and the flow sensors. The nitrogen and phosphorus concentration sensors measure the real-time nitrogen and phosphorus content in the soil and runoff by ion-selective electrodes, the soil humidity sensors measure the real-time soil humidity, the rainfall sensors measure the real-time rainfall, and the flow sensors measure the real-time runoff flow. , and Concentration outputs a raw nitrogen and phosphorus concentration data sequence, which includes a timestamp, monitoring point number, and the three ion concentration values. Soil moisture sensors are installed at a depth of 0-30 cm in the fruit tree root zone to measure volumetric moisture content and generate a soil moisture data sequence. Precipitation sensors use tipping bucket rain gauges to record accumulated rainfall and output a rainfall intensity data sequence. Flow sensors install ultrasonic flow meters in runoff collection ditches to record runoff velocity and cross-sectional flow, generating a flow data sequence. After analog-to-digital conversion, the analog signals from each sensor are synchronously collected with a 5-minute sampling period to ensure time series data alignment.

[0030] The original nitrogen and phosphorus concentration data series 、 and Concentration data may introduce outliers due to electrode drift or electrochemical noise. The 3σ criterion is used to detect outliers for each ion concentration. The arithmetic mean μ and standard deviation σ of the historical data are calculated, and data points falling outside the interval (μ-3σ, μ+3σ) are eliminated. Soil moisture data with values ​​exceeding 110% of field water holding capacity or below 50% of the wilting coefficient are marked as outliers. Extreme records exceeding 100 mm per hour in rainfall data and negative values ​​in flow data or values ​​exceeding 120% of the ditch design flow are discarded. After outliers are eliminated, cleaned nitrogen and phosphorus concentration data are generated. Missing values ​​are supplemented by linear interpolation of adjacent time points of the same sensor to ensure the continuity of the time series.

[0031] After cleaning, the nitrogen and phosphorus concentration data are aligned with the soil moisture, rainfall intensity and flow data by time stamp and spliced ​​into the orchard environmental monitoring parameter set. Each row of the parameter set contains the data at the same time. 、 and concentration, soil volume moisture content, rainfall intensity and runoff flow constitute a 6-dimensional vector set. Due to the significant differences in the dimensions of each parameter, min-max normalization is used to linearly map each dimension of data to the [0,1] interval. The normalization formula is ( x - min) / (max - min), where min and max are the historical extreme values ​​of the parameter under typical orchard operating conditions: 0-15 mg / L, 0-10 mg / L, The soil moisture content ranges from 0 to 5 mg / L, the soil volumetric moisture content ranges from 15% to 45%, the rainfall intensity ranges from 0 to 50 mm / h, and the runoff flow rate ranges from 0 to 0.5 m³ / s. The normalized 6-dimensional vector set forms the orchard nitrogen and phosphorus monitoring dataset. Each row of data represents a standardized monitoring parameter at a time point, with the timestamp serving as an implicit index.

[0032] In a specific embodiment, the process of executing step S102 may specifically include the following steps: A variational autoencoder network was constructed, which took the orchard nitrogen and phosphorus monitoring dataset as input for feature extraction and output the potential feature vector of orchard nitrogen and phosphorus loss. A variational autoencoder decoder network was constructed, which used the potential feature vectors of nitrogen and phosphorus loss in the orchard as input to reconstruct the original data. The complete variational autoencoder model was trained using the KL divergence loss function. The variational autoencoder model was run to learn the spatiotemporal distribution characteristics of nitrogen and phosphorus in orchard runoff, generating synthetic samples of nitrogen and phosphorus concentrations that matched the orchard's environmental characteristics, and expanding them to form an enhanced nitrogen and phosphorus training dataset. Based on the enhanced nitrogen and phosphorus training data set, a nonlinear mapping relationship between nitrogen and phosphorus content input and stepped trench water velocity, PRB adsorption material filling amount, and sedimentation tank flocculant dosage output was established to form a relationship model between nitrogen and phosphorus content and interception parameters.

[0033] Specifically, the encoder network of the variational autoencoder receives the orchard nitrogen and phosphorus monitoring dataset as input, which contains the standardized 、 and The encoder network consists of three fully connected layers. The first layer maps the 6-dimensional input to a 32-dimensional latent space. The second layer maintains the 32-dimensional latent space. The output layer is divided into two parallel branches, generating an 8-dimensional mean vector μ and an 8-dimensional logarithmic variance vector logσ², respectively. The latent feature vector z is obtained by sampling using the reparameterization technique. The sampling process multiplies random noise ε from a standard normal distribution by exp(logσ² / 2) and adds it to μ to ensure that the gradient can be back-propagated.

[0034] The decoder network takes the 8-dimensional latent feature vector z as input, and after two layers of 32-dimensional fully connected layers, it is finally mapped back to the 6-dimensional output space. 、 and Concentration, soil volumetric moisture content, rainfall intensity, and runoff flow maintain the same dimensionality and physical meaning as the original input. During training, the reconstruction loss uses the mean squared error (MSE) to calculate the difference between the original 6-dimensional vector and the reconstructed 6-dimensional vector. The KL divergence loss constrains the latent space distribution to approximate a standard normal distribution. The weighted sum of these two losses constitutes the overall loss function. Optimization is completed over 200 training epochs using a batch size of 64. The Adam optimizer with a learning rate of 0.001 automatically adjusts parameter updates.

[0035] The trained variational autoencoder generates new latent feature vectors by randomly sampling in the 8-dimensional latent space, and the decoder converts these vectors into synthetic samples. 、 and The concentration data have the same statistical properties as the real monitoring data, and the soil volume water content, rainfall intensity, and runoff flow data maintain reasonable physical correlations. The synthetic samples are merged with the original monitoring data to form an enhanced nitrogen and phosphorus training dataset, and the sample quantity is expanded to 5-8 times that of the original data.

[0036] The enhanced nitrogen and phosphorus training dataset is used to train a three-layer feedforward neural network to establish the mapping relationship between the nitrogen and phosphorus contents and the environmental parameters to the interception parameters. The input layer receives a 6-dimensional feature vector, including 、 and concentrations, soil volume water content, rainfall intensity, and runoff flow. The hidden layer uses 32 neurons and a Tanh activation function, and the output layer generates three interception parameters, i.e., the stepped trench water flow velocity, the PRB adsorption material filling amount, and the coagulant dosage of the sedimentation tank. The training process uses an early stopping strategy to prevent overfitting, and the finally formed nitrogen and phosphorus content and interception parameter relationship model can dynamically predict the optimal interception parameter combination according to real-time monitoring data. The stepped trench water flow velocity output by the model ranges from 0.1 to 0.5 m / s, the PRB adsorption material filling amount ranges from 10 to 50 kg / m³, and the coagulant dosage of the sedimentation tank is controlled in the range of 5-20 mg / L.

[0037] In a specific embodiment, the process of performing step S103 can specifically include the following steps: A multi-objective optimization algorithm framework is established, taking the nitrogen and phosphorus content and interception parameter relationship model as the constraint condition, and setting the nitrogen and phosphorus removal efficiency and the operation cost as the double objective functions; The current real-time nitrogen and phosphorus concentration data of the orchard are input into the nitrogen and phosphorus content and interception parameter relationship model to calculate the theoretical treatment load parameters of the three-level interception system of the stepped trench, the PRB reaction wall, and the sedimentation tank; The multi-objective optimization algorithm is run to search for the Pareto solution set, and the combination scheme of the stepped trench water flow velocity range, the PRB adsorption material filling amount range, and the coagulant dosage range of the sedimentation tank is calculated under the constraint of the theoretical treatment load parameters; The balance point of the nitrogen and phosphorus removal efficiency and the operation cost is selected from the combination scheme to determine the specific values of the stepped trench water flow velocity, the PRB adsorption material filling amount, and the coagulant dosage of the sedimentation tank as the regulation and control instructions.

[0038] Specifically, the multi-objective optimization algorithm framework takes the nitrogen and phosphorus content and interception parameter relationship model as the core constraint condition, which is input into the real-time monitoring 、 and The 6-dimensional data of concentration, soil volumetric water content, rainfall intensity and runoff flow rate are input into the model, and the 3-dimensional parameters of stepped trench water flow velocity, PRB adsorption material filling amount and sedimentation tank flocculant dosage are output. The multi-layer perception network inside the model has been trained, and the weight matrix and bias vector are fixed to form a mathematical expression of parameter mapping. The optimization framework maintains two objectives of nitrogen and phosphorus removal efficiency function and operation cost function at the same time. The removal efficiency function is calculated by the measured removal rate of the three interception systems, and the weight coefficient is calculated by the real-time flow distribution; the operation cost function comprehensively calculates the energy consumption of the trench regulating valve, the replacement cost of the PRB adsorption material and the consumption cost of the flocculant in the sedimentation tank.

[0039] After the real-time nitrogen and phosphorus concentration data is input into the model of the relationship between nitrogen and phosphorus content and interception parameters, the model outputs three theoretical treatment load parameters of stepped trench maximum hydraulic load, PRB adsorption capacity limit and sedimentation tank treatment capacity threshold. These parameters are used as hard constraints to limit the search space of the NSGA-II algorithm, ensuring that all candidate solutions are within the feasible range of the project. The algorithm initializes a population of 100 individuals, each encoded as a real triple representing the trench water flow velocity, PRB adsorption material filling amount and sedimentation tank flocculant dosage, with the value range strictly limited by the theoretical treatment load parameters.

[0040] During the operation of the NSGA-II algorithm, the crossover operation uses the simulated binary crossover operator with a crossover probability of 0.9, and the mutation operation uses the polynomial mutation operator with a mutation probability of 0.1. During each generation evolution process, the algorithm evaluates the performance of each individual on the two objective functions: the nitrogen and phosphorus removal efficiency is calculated by the physical and chemical equations of the three interception systems, and the operation cost is calculated based on the historical operation data of the orchard. After 200 iterations, the algorithm outputs the non-dominated solution set on the Pareto frontier, which forms a continuous distribution within the range of 0.05-0.25 m / s for the trench water flow velocity, 5-30 kg / m³ for the PRB adsorption material filling amount, and 2-12 mg / L for the sedimentation tank flocculant dosage.

[0041] When selecting the final control instruction from the Pareto solution set, the weighted Euclidean distance method is used for decision-making. After normalizing the nitrogen and phosphorus removal efficiency and the operation cost to the [0, 1] interval, the distance of each solution to the ideal point is calculated, and the solution with the smallest distance is selected as the optimal compromise scheme. The specific values of the stepped trench water flow velocity, PRB adsorption material filling amount and sedimentation tank flocculant dosage are packaged into the control instruction frame, which includes four fields: device identifier, parameter value, execution duration and check code, and is sent to the on-site PLC controller through the wireless communication module. After the execution of the instruction, the trench regulating valve adjusts the opening degree according to the water flow velocity parameter, the PRB filling system supplements the adsorption material according to the calculated amount, and the sedimentation tank dosing pump injects the flocculant according to the set value, realizing the coordinated operation of the three interception systems.

[0042] In a specific embodiment, the process of executing step S104 may specifically include the following steps: Analyze the stepped groove water flow velocity parameters in the control instructions, drive the groove regulating valve to perform opening adjustment action, control the flow rate of runoff through the groove and the contact time between nitrogen and phosphorus and iron oxide coated sand; Analyze the PRB adsorption material filling amount parameters in the control instructions, start the PRB filling system to perform adsorption material replenishment or replacement operations, and maintain the active adsorption capacity of calcium-based materials and iron oxide coated sand in the PRB reaction wall; Analyze the flocculant dosage parameters of the sedimentation tank in the control instructions, control the speed of the sedimentation tank agitator and the flow rate of the polyaluminum chloride dosing pump, and promote the formation of flocculation and precipitation of residual nitrogen and phosphorus in the runoff and sedimentation to the bottom of the tank; Through three-stage coordinated interception treatment of stepped trenches, PRB reaction walls, and sedimentation tanks, the nitrogen and phosphorus in the runoff are gradually enriched into insoluble sediments and gathered at the bottom of the sedimentation tank to form nitrogen and phosphorus-rich sediments.

[0043] Specifically, the control instruction frame is transmitted to the on-site PLC control system through the wireless communication network. The instruction parsing module extracts the three parameter values ​​of the stepped ditch water flow velocity, the PRB adsorption material filling amount, and the sedimentation tank flocculant dosage according to the predefined frame structure format. The ditch water flow velocity parameter is input into the PID control loop, and compared with the actual measured value of the electromagnetic flowmeter at the ditch outlet. A 4-20mA analog signal is output to drive the electric regulating valve. The valve opening is adjusted according to the preset water flow velocity-opening characteristic curve to stabilize the flow velocity of the runoff in the trapezoidal cross-section ditch within the design range of 0.05-0.25m / s. A 0.3m thick iron oxide coated sand filter layer is laid at the bottom of the ditch. The water flow velocity adjustment directly changes the contact time between nitrogen and phosphorus pollutants and the filter material. When the flow velocity is reduced to 0.1m / s, The coordination exchange time with the iron oxide surface was extended to 90 seconds, and the adsorption efficiency was increased by 15%.

[0044] The PRB adsorption material filling parameter triggers the automatic feeding program. The PLC calculates the amount of calcium-based material and iron oxide coating sand to be added based on the difference between the current weight sensor reading in the reaction wall and the set value. The two materials are added to the reaction zone through the screw conveyor at a mass ratio of 1:2. During the transportation process, the pressure difference sensor built into the wall monitors the change of permeability coefficient in real time. When the pressure difference growth rate exceeds 0.5kPa / min, the feeding is stopped to ensure that the permeability coefficient is maintained at After replenishment is completed, the spray system automatically starts according to the feedback from the wall humidity sensor to control the moisture content of the adsorption material within the optimal activity range of 15-20%.

[0045] The sedimentation tank flocculant dosing system receives a dosage parameter from the PLC, divides it by the inlet flowmeter reading, and uses the output signal to control the speed of a variable-frequency screw pump, injecting a 10% concentration polyaluminium chloride solution into the mixing zone at a flow rate of 0.5-2.5 L / min. Simultaneously, the agitator speed is adjusted in tandem with the dosage. Initially, high-intensity mixing is maintained at 120 rpm for 30 seconds, then reduced to 60 rpm for 90 seconds to allow floc formation. Finally, agitation is stopped and the sedimentation phase begins. A turbidity sensor installed in the mixing zone monitors floc formation. If the NTU value decreases by less than 5% / min, the agitation intensity is automatically increased by 10%. After sedimentation is complete, the supernatant is discharged through an overflow weir. The nitrogen and phosphorus-rich sediment formed at the bottom is monitored in real time by an ultrasonic mud level meter. When the sediment layer reaches 30 cm thick, the sludge pump is activated.

[0046] The operating status of the three-level interception system is monitored in real time via the Modbus RTU protocol. The trench outlet, PRB outlet and sedimentation tank overflow are installed separately. 、 and Online analyzer, monitoring data is uploaded to the edge computing node in a 5-minute cycle. The node runs a sliding window algorithm to analyze the trend of the last 6 sets of data. When the concentration of any monitoring point exceeds the threshold (such as >1.0mg / L, >1.0mg / L, >0.2mg / L), the system automatically triggers parameter re-optimization and generates new control instructions. Once sediment accumulates to a set thickness at the bottom of the sedimentation tank, the sludge pump activates according to a pre-set program, transporting nitrogen- and phosphorus-rich sediments with a solids content of 15-20% to the dewatering unit, completing the entire process from nitrogen and phosphorus interception to sediment collection.

[0047] In a specific embodiment, the process of executing the step of parsing the sedimentation tank flocculant dosage parameter in the control instruction may specifically include the following steps: Read the numerical parameters of the flocculant dosage in the sedimentation tank in the control instruction, calculate the target flow rate and dosage duration of the polyaluminium chloride dosing pump, and generate the flocculant dosage control signal; Based on the flocculant dosing control signal, the polyaluminium chloride dosing pump is started, and the polyaluminium chloride solution is quantitatively added into the sedimentation tank according to the calculated target flow rate to form a flocculation reaction environment in the tank; According to the stirring device speed parameter in the control instruction, the rotation speed of the stirring blade in the sedimentation tank is adjusted to control the shear force and mixing intensity of the water flow in the tank, so as to promote the flocculation reaction between the residual nitrogen and phosphorus and polyaluminum chloride to form large particle flocs; Stop the stirring device and enter the static sedimentation stage, use gravity to make the large particle flocs sink to the bottom of the sedimentation tank, separate and clarify the upper water and the nitrogen and phosphorus rich sediments at the bottom.

[0048] Specifically, during the flocculant dosing control process in the sedimentation tank, the flocculant dosage parameter in the control instruction frame is extracted by the PLC parsing module. This numerical parameter, along with the real-time monitoring data from the sedimentation tank inlet flowmeter, is input into the dosage calculation module. The dosage calculation module performs a division operation to convert the flocculant dosage concentration in mg / L into the polyaluminum chloride solution dosage flow rate in L / min, taking into account the fixed parameter of 10% concentration of the commercial polyaluminum chloride solution. The calculation results generate a flocculant dosage control signal, which contains two key parameters: the target flow value and the dosage duration. The target flow rate range is limited to between 0.5 and 2.5 L / min, corresponding to the sedimentation tank inlet flow rate of 20 to 100 m³ / h.

[0049] The flocculant dosing control signal drives the variable-frequency screw pump control system. The pump speed and target flow rate are correlated using a preset flow-speed curve. An electromagnetic flowmeter installed at the screw pump outlet provides real-time feedback on the actual flow rate, which is compared with the target flow rate to form a closed-loop control. The PID adjustment algorithm keeps the flow rate error within ±2%. The polyaluminum chloride solution is evenly injected into the sedimentation tank inlet channel through a perforated pipe. The injection point is located 30 cm above the channel bottom to ensure thorough mixing of the agent and the water flow. During the dosing process, the PLC records the cumulative dosage and automatically stops the pump when the preset dosing duration is reached.

[0050] The stirring device speed parameter is extracted from the control instructions and converted into an inverter control signal, which is output to the stirring motor. The stirring blade speed is adjusted in two stages: the first stage is set to 120 rpm for 30 seconds, generating high-intensity turbulence that allows rapid diffusion of polyaluminum chloride and full contact with residual nitrogen and phosphorus. The second stage is reduced to 60 rpm for 90 seconds, reducing shear forces and promoting the growth of micro-flocs into larger flocs of 0.5-1 mm. The stirring intensity is dynamically optimized based on the data monitored by the online turbidity meter in the tank. The speed is automatically increased by 10% when the NTU value decreases by less than 5% / min.

[0051] After the flocculation reaction is completed, the stirring device stops running and enters the static sedimentation stage. The sedimentation tank is designed as a horizontal flow type with an effective water depth of 2.5m and a hydraulic retention time of 2 hours. During the sedimentation process, the flocs sink at a speed of 0.3-0.5mm / s under the action of gravity, and the upper water body gradually clarifies. The ultrasonic mud level meter installed at the bottom of the tank monitors the height of the sediment interface in real time. When the interface rises to a position 30cm from the bottom of the tank, the mud discharge signal is triggered. The clarified supernatant is evenly discharged through the zigzag overflow weir, and the effluent NTU value is controlled below 5. The solid content of the nitrogen and phosphorus-rich sediment at the bottom reaches 15-20%, and is regularly sucked and transported to the dehydration treatment unit by the sludge pump, completing the whole process control from flocculation and dosing to solid-liquid separation.

[0052] In a specific embodiment, the process of executing step S105 may specifically include the following steps: Start the suction equipment at the bottom of the sedimentation tank to perform negative pressure extraction on the nitrogen and phosphorus-rich sediments, and transport the sediments from the bottom of the sedimentation tank to the dewatering treatment device to separate the solid and liquid phases to obtain concentrated nitrogen and phosphorus sludge; The concentrated nitrogen and phosphorus sludge is transported to the intelligent granulation system, where biochar carrier and slow-release coating material are added for mixed granulation treatment. The granulation temperature and pressure parameters are controlled to form uniform slow-release fertilizer particles. Analyze the nitrogen and phosphorus content of slow-release fertilizer granules, and adjust the nitrogen and phosphorus ratio in the granules based on the test results to produce slow-release nitrogen and phosphorus fertilizer products that meet the nutrient needs of orchard soils; Through intelligent fertilization machinery, slow-release nitrogen fertilizer and phosphorus fertilizer products are applied at targeted locations according to the growth stage of fruit trees and soil nutrient conditions, completing a closed-loop cycle of nitrogen and phosphorus nutrients from interception and recovery to reuse.

[0053] Specifically, when the ultrasonic mud level meter at the bottom of the sedimentation tank detects that the accumulated thickness of nitrogen-phosphorus-rich sediments reaches 30cm, it triggers the start signal of the suction equipment. The screw pump transports the sediment to the screw dewatering machine through a DN80 pipe under a negative pressure of -0.06MPa. The pipeline flow rate is controlled within 1.2m / s to prevent flocs from breaking. The dewatering machine drum speed is dynamically adjusted within the range of 2800-3200rpm according to the torque signal, and the back pressure plate gap is adjusted synchronously to reduce the moisture content of the sediment from 96% to 75-78%, forming a concentrated nitrogen and phosphorus sludge containing 42g / kg of total nitrogen and 8g / kg of total phosphorus. During the dehydration process, the PLC records the dry mud production data in real time and uploads it to the cloud server through the 4G module.

[0054] Concentrated nitrogen and phosphorus sludge is metered by a conveyor scale and then fed into a twin-shaft paddle mixer with a metering accuracy of ±0.5kg. The mixer simultaneously receives biochar carrier and slow-release coated urea-formaldehyde resin from a loss-in-weight scale. The biochar has an 8% moisture content and is added at a ratio of 30% to the dry sludge mass. The urea-formaldehyde resin has a particle size of 0.6mm and is added at a ratio of 4% to the finished pellet mass. The paddles mix for 180 seconds at 30rpm to ensure uniform distribution. The mixture then falls into a flat die pelletizer, where the die temperature is maintained at 105±2°C and the roller pressure is 8MPa. Cylindrical pellets with a diameter of 4mm and a length of 6mm are extruded. Data from the pelletizer's torque and temperature sensors are fed back to the control system in real time. Feed rate is automatically reduced when torque exceeds a set threshold, and heating power is increased when temperature falls below a set value, ensuring a stable pellet density of 1.28g / cm³.

[0055] After granulation, the particles are graded by a vibrating screen, and qualified particles enter the near-infrared detection station. A spectrometer scans the particle surface in the 1200-2400nm band. The collected absorbance data is input into a pre-trained partial least squares regression model, which outputs the nitrogen and phosphorus content test results. The detection system uses a nitrogen-phosphorus ratio of 3:1 as the control target. When the test value deviates from the target, the resin coating thickness and dry mud blending ratio of subsequent batches are automatically adjusted. The coating thickness is fine-tuned by the pelletizer mold temperature, reducing the coating amount by 0.2% for every 1°C decrease; the dry mud blending ratio is adjusted by the belt scale speed control, ultimately ensuring that the nitrogen and phosphorus content of the finished particles is controlled within an error of ±3%.

[0056] Qualified slow-release fertilizer granules are loaded into the storage bin of the intelligent fertilizer spreader, which then receives prescription map data from the orchard management system. The prescription map is generated based on soil conductivity, pH test results, and the fertilizer requirements of the fruit trees during their growth period. It controls the fertilizer spreader to precisely distribute fertilizer 20 cm below the root zone projection of the fruit trees. The speed of the spiral fertilizer pipe is proportional to the fertilizer application rate required by the prescription map. The fertilizer application rate is dynamically adjusted based on the measured values ​​of available nitrogen and phosphorus in the soil, increasing the nitrogen fertilizer ratio during the budding period and increasing the phosphorus fertilizer application during the fruit expansion period. Fertilization operation records, including location coordinates, fertilizer application amount, and timestamp, are uploaded to the cloud database via the IoT module, forming a complete closed loop with the previous nitrogen and phosphorus interception monitoring data.

[0057] The above describes the orchard nitrogen and phosphorus interception method based on intelligent monitoring in the embodiment of the present application. The following describes the orchard nitrogen and phosphorus interception treatment process based on intelligent monitoring in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the orchard nitrogen and phosphorus interception treatment process based on intelligent monitoring includes: The intelligent monitoring-based nitrogen and phosphorus interception and treatment process for orchards includes five core steps: data preprocessing, variational autoencoder modeling, multi-objective optimization and decision-making, three-level interception and coordinated control, and sediment resource utilization. This entire process forms a closed-loop system from data perception to intelligent decision-making and physical interception, achieving efficient removal and resource recovery of nitrogen and phosphorus pollutants in orchard runoff.

[0058] During the data preprocessing process, nitrogen and phosphorus concentration sensors, soil moisture sensors, rainfall sensors, and flow sensors were deployed in the orchard soil profile, runoff channel, and surrounding environment to collect data synchronously with a 5-minute cycle. 、 、 concentration, soil volume moisture content, rainfall intensity, and runoff flow to form the original data sequence; then the 3σ criterion outlier detection is performed to remove abnormal readings, and the missing values ​​are filled by linear interpolation, and then min-max normalization is performed to convert Mapped to 0–15 mg / L, Mapped to 0–10 mg / L, The nitrogen and phosphorus monitoring datasets for orchards were obtained by mapping nitrogen to 0–5 mg / L, soil volumetric moisture to 15%–45%, rainfall intensity to 0–50 mm / h, and runoff flow to 0–0.5 m³ / s.

[0059] In the variational autoencoder modeling process, the above 6-dimensional orchard nitrogen and phosphorus monitoring dataset is input into the encoder network and compressed into an 8-dimensional latent feature vector through three fully connected layers; the decoder network reconstructs the original 6-dimensional data with the latent feature vector as input. During training, the mean square error and KL divergence loss are combined, and the latent space sampling generates synthetic samples and merges them with the real data. The sample size is expanded to 5-8 times the original to form an enhanced nitrogen and phosphorus training dataset; based on the enhanced nitrogen and phosphorus training dataset, the feedforward neural network is trained, and the network input is 、 and concentration, soil volumetric moisture content, rainfall intensity, runoff flow, output stepped ditch water velocity of 0.1–0.5 m / s, PRB adsorption material filling amount of 10–50 kg / m³, sedimentation tank flocculant dosage of 5–20 mg / L, and a relationship model between nitrogen and phosphorus content and interception parameters was established.

[0060] A multi-objective optimization decision-making process was conducted with nitrogen and phosphorus removal efficiency and operating cost as the dual objective functions and the theoretical processing load output by the nitrogen and phosphorus content and interception parameter relationship model as the constraint. The NSGA-II algorithm initialized 100 individuals and evolved for 200 generations with a simulated binary crossover probability of 0.9 and a polynomial mutation probability of 0.1. The Pareto solution set with groove water velocity of 0.05–0.25 m / s, PRB adsorption material filling amount of 5–30 kg / m³, and sedimentation tank flocculant dosage of 2–12 mg / L was output. The comprehensive satisfaction index was calculated using fuzzy membership, and the compromise solution was selected and encapsulated as an optimization control instruction frame.

[0061] During the three-level interception coordinated control process, the PLC analyzes and controls the water velocity parameters of the stepped groove in the instruction frame, and controls the opening of the electric regulating valve through the PID and electromagnetic flowmeter in a closed loop, so that the flow velocity in the trapezoidal cross-section groove is maintained at the instruction value ±0.02m / s, and the thickness of the iron oxide coating sand is 0.3m; the PRB adsorption material filling amount parameter is analyzed, and the supplementary amount of calcium-based material and iron oxide coating sand is calculated by the difference between the weight sensor feedback and the set value. The materials are supplemented through the screw conveyor at a mass ratio of 1:2, and the differential pressure sensor monitors the permeability coefficient ≥ ; Analyze the flocculant dosage parameters of the sedimentation tank. The variable frequency screw pump is adjusted according to the inlet flow rate to 10% polyaluminium chloride solution 0.5–2.5 L / min. The stirring device speed is 120 rpm / 30 s→60 rpm / 90 s. When the turbidity sensor NTU value decreases at a rate of <5% / min, the speed is increased by 10%. The trench, PRB and sedimentation tank are operated in a coordinated manner. 、 、 The analyzer returns data every 5 minutes, and the edge node sliding window algorithm recalculates the control instructions when the concentration exceeds the threshold.

[0062] During the sediment resource utilization process, when the thickness of nitrogen- and phosphorus-rich sediments at the bottom of the sedimentation tank reaches 30 cm, as monitored by an ultrasonic mud level meter, a screw pump is used to pump the sediment to a spiral dehydrator via a DN80 pipe at a negative pressure of −0.06 MPa. The water content is reduced from 96% to 75%, resulting in concentrated nitrogen and phosphorus sludge containing 42 g / kg total nitrogen and 8 g / kg total phosphorus. The concentrated nitrogen and phosphorus sludge is then granulated with 30% biochar (8% moisture content) and 4% urea-formaldehyde resin (0.6 mm particle size) at 105°C and 8 MPa on a flat die, forming slow-release fertilizer granules with a diameter of 4 mm, a length of 6 mm, and a density of 1.28 g / cm³. A near-infrared detector scans the 1200–2400 nm wavelength range, and a partial least squares model outputs the nitrogen and phosphorus content. The algorithm adjusts the coating thickness and dry mud ratio to maintain a nitrogen-to-phosphorus ratio of 3:1 within an error of ±3%. An intelligent fertilizer applicator receives soil conductivity, pH, and a fruit tree growth period prescription map. The spiral fertilizer delivery pipe is positioned 20 meters below the root zone of the fruit tree. Fertilizer is applied precisely at a depth of 1 cm. The amount of fertilizer applied is inversely proportional to the measured value of effective nitrogen and phosphorus in the soil and directly proportional to the fertilizer requirement curve of the fruit trees. Operation records are uploaded to the cloud via a 4G module, completing a closed-loop cycle from interception and recovery of nitrogen and phosphorus to reuse.

[0063] The above describes the orchard nitrogen and phosphorus interception method based on intelligent monitoring in the embodiment of the present application. The following describes the orchard nitrogen and phosphorus interception system 300 based on intelligent monitoring in the embodiment of the present application. Figure 3 In the embodiment of the present application, an orchard nitrogen and phosphorus interception system 300 based on intelligent monitoring includes: The acquisition module 301 is used to collect data from orchard nitrogen and phosphorus concentration sensors, soil moisture sensors, rainfall sensors, and flow sensors, and generate orchard nitrogen and phosphorus monitoring data sets through outlier detection and standardization processing; A generation module 302 is configured to use a variational autoencoder algorithm to learn the distribution patterns of nitrogen and phosphorus in orchard runoff based on the orchard nitrogen and phosphorus monitoring dataset, and generate a relationship model between nitrogen and phosphorus content and interception parameters; The calculation module 303 is used to calculate the control instructions of the water flow velocity of the stepped groove, the filling amount of the PRB adsorption material, and the flocculant dosage of the sedimentation tank by using the relationship model between the nitrogen and phosphorus content and the interception parameter through a multi-objective optimization algorithm; The control module 304 is used to control the operation of the trench regulating valve, the PRB filling system and the sedimentation tank stirring device according to the control instructions, intercept nitrogen and phosphorus in the runoff and collect nitrogen and phosphorus rich sediments; The processing module 305 is used to recover the nitrogen and phosphorus-rich sediment through a suction device to prepare slow-release nitrogen fertilizer and phosphorus fertilizer for re-application to the orchard.

[0064] above Figure 3 The orchard nitrogen and phosphorus interception system based on intelligent monitoring in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The orchard nitrogen and phosphorus interception equipment based on intelligent monitoring in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0065] Reference Figure 4 In the embodiment of the present invention, there is also provided an orchard nitrogen and phosphorus interception device 400 based on intelligent monitoring. The orchard nitrogen and phosphorus interception device based on intelligent monitoring can be a server, and its internal structure can be as follows: Figure 4 As shown. The orchard nitrogen and phosphorus interception equipment based on intelligent monitoring includes a processor 402, a memory 403, a display screen 404, an input device 405, a network interface 406 and a database 407 connected via a system bus 401. Among them, the computer-designed processor is used to provide computing and control capabilities. The memory of the orchard nitrogen and phosphorus interception equipment based on intelligent monitoring includes a non-volatile storage medium 4031 and an internal memory 4032. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database 407 of the orchard nitrogen and phosphorus interception equipment based on intelligent monitoring is used to store the corresponding data in this embodiment. The network interface 406 of the orchard nitrogen and phosphorus interception equipment based on intelligent monitoring is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0066] Those skilled in the art will understand that Figure 4 The structure shown in is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the orchard nitrogen and phosphorus interception equipment based on intelligent monitoring to which the solution of the present invention is applied.

[0067] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the computer executes the steps of the orchard nitrogen and phosphorus interception method based on intelligent monitoring.

[0068] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0069] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling an orchard nitrogen and phosphorus interception device based on intelligent monitoring (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc., various media that can store program code.

[0070] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for intercepting nitrogen and phosphorus in orchards based on intelligent monitoring, characterized in that: The method comprises: Step S101: Collect data from orchard nitrogen and phosphorus concentration sensors, soil moisture sensors, rainfall sensors, and flow sensors, and generate an orchard nitrogen and phosphorus monitoring data set through outlier detection and standardization. Step S102: Based on the orchard nitrogen and phosphorus monitoring dataset, a variational autoencoder algorithm is used to learn the distribution pattern of nitrogen and phosphorus in orchard runoff, and a relationship model between nitrogen and phosphorus content and interception parameters is generated; Step S103: using the relationship model between nitrogen and phosphorus content and interception parameters, a multi-objective optimization algorithm is used to calculate control instructions for the water flow velocity of the stepped groove, the filling amount of the PRB adsorption material, and the flocculant dosage of the sedimentation tank; Step S104: controlling the operation of the trench regulating valve, the PRB filling system, and the sedimentation tank stirring device according to the control instructions to intercept nitrogen and phosphorus in the runoff and collect nitrogen and phosphorus-rich sediments; Step S105: The nitrogen- and phosphorus-rich sediment is recovered and processed by a suction device to prepare slow-release nitrogen fertilizer and phosphorus fertilizer for re-application to the orchard.

2. The orchard nitrogen and phosphorus interception method based on intelligent monitoring according to claim 1, characterized in that: The step S101 includes: The real-time nitrogen and phosphorus content data in the orchard soil and runoff are collected by nitrogen and phosphorus concentration sensors to obtain the original nitrogen and phosphorus concentration data series; The soil moisture sensor and rainfall sensor are used to collect environmental parameter data, and the flow sensor is used to obtain runoff flow change data to obtain the orchard environmental monitoring parameter set. Performing 3σ criterion outlier detection processing on the original nitrogen and phosphorus concentration data sequence, eliminating data points outside the normal range, and obtaining cleaned nitrogen and phosphorus concentration data; The cleaned nitrogen and phosphorus concentration data and the orchard environmental monitoring parameter set are normalized and standardized, and the data dimensions are unified to the [0, 1] interval to obtain an orchard nitrogen and phosphorus monitoring data set.

3. The orchard nitrogen and phosphorus interception method based on intelligent monitoring according to claim 1, characterized in that: The step S102 includes: Constructing a variational autoencoder encoder network, taking the orchard nitrogen and phosphorus monitoring dataset as input for feature extraction, and outputting a potential feature vector of orchard nitrogen and phosphorus loss; Constructing a variational autoencoder decoder network, reconstructing the original data using the potential feature vector of nitrogen and phosphorus loss in the orchard as input, and training the complete variational autoencoder model in combination with the KL divergence loss function; Running the variational autoencoder model to learn the spatiotemporal distribution characteristics of nitrogen and phosphorus in orchard runoff, generating synthetic samples of nitrogen and phosphorus concentrations that conform to the orchard environmental characteristics, and expanding them to form an enhanced nitrogen and phosphorus training dataset; Based on the enhanced nitrogen and phosphorus training data set, a nonlinear mapping relationship between nitrogen and phosphorus content input and stepped groove water velocity, PRB adsorption material filling amount, and sedimentation tank flocculant dosage output is established to form a nitrogen and phosphorus content and interception parameter relationship model.

4. The orchard nitrogen and phosphorus interception method based on intelligent monitoring according to claim 1, characterized in that: The step S103 includes: A multi-objective optimization algorithm framework is established, with the relationship model between nitrogen and phosphorus content and interception parameters as constraints, and nitrogen and phosphorus removal efficiency and operating cost as dual objective functions; The current real-time nitrogen and phosphorus concentration data of the orchard is input into the nitrogen and phosphorus content and interception parameter relationship model to calculate the theoretical processing load parameters of the three-stage interception system of stepped trenches, PRB reaction walls, and sedimentation tanks; Running a multi-objective optimization algorithm to search for a Pareto solution set, and calculating a combination of a stepped trench water velocity range, a PRB adsorption material filling amount range, and a sedimentation tank flocculant dosage range under the constraints of the theoretical treatment load parameters; The balance point between nitrogen and phosphorus removal efficiency and operating cost is selected from the combination scheme, and the specific values ​​of the water flow velocity in the stepped groove, the filling amount of the PRB adsorption material, and the dosage of the flocculant in the sedimentation tank are determined as control instructions.

5. The orchard nitrogen and phosphorus interception method based on intelligent monitoring according to claim 1, characterized in that: The step S104 includes: parsing the stepped groove water flow velocity parameters in the control instructions, driving the groove regulating valve to perform an opening adjustment action, and controlling the flow rate of runoff through the groove and the contact time between nitrogen and phosphorus and the iron oxide coated sand; parsing the PRB adsorption material filling amount parameter in the control instruction, starting the PRB filling system to perform adsorption material replenishment or replacement operations, and maintaining the active adsorption capacity of the calcium-based material and the iron oxide coated sand in the PRB reaction wall; Analyze the flocculant dosage parameter of the sedimentation tank in the control instruction, control the rotation speed of the sedimentation tank stirring device and the flow rate of the polyaluminum chloride dosing pump, and promote the formation of flocculation and precipitation of residual nitrogen and phosphorus in the runoff and sedimentation to the bottom of the tank; Through three-stage coordinated interception treatment of stepped trenches, PRB reaction walls, and sedimentation tanks, the nitrogen and phosphorus in the runoff are gradually enriched into insoluble sediments and gathered at the bottom of the sedimentation tank to form nitrogen and phosphorus-rich sediments.

6. The orchard nitrogen and phosphorus interception method based on intelligent monitoring according to claim 5, characterized in that: The method of analyzing the flocculant dosage parameter of the sedimentation tank in the control instruction, controlling the rotation speed of the sedimentation tank stirring device and the flow rate of the polyaluminum chloride dosing pump, and promoting the residual nitrogen and phosphorus in the runoff to form flocculation precipitation and settle to the bottom of the tank includes: Reading the numerical parameter of the flocculant dosage of the sedimentation tank in the control instruction, calculating the target flow rate and dosage duration of the polyaluminium chloride dosing pump, and generating a flocculant dosage control signal; Based on the flocculant addition control signal, the polyaluminium chloride addition pump is started to quantitatively add the polyaluminium chloride solution into the sedimentation tank according to the calculated target flow rate to form a flocculation reaction environment in the tank; According to the stirring device speed parameter in the control instruction, the rotation speed of the stirring blade in the sedimentation tank is adjusted to control the shear force and mixing intensity of the water flow in the tank, thereby promoting the flocculation reaction between the residual nitrogen and phosphorus and the polyaluminum chloride to form large particle flocs; The stirring device is stopped and the static sedimentation stage is entered, and the large particle flocs are made to sink to the bottom of the sedimentation tank by gravity, so as to separate and clarify the upper water body and the nitrogen and phosphorus-rich sediments at the bottom.

7. The orchard nitrogen and phosphorus interception method based on intelligent monitoring according to claim 1, characterized in that: The step S105 includes: Starting the suction equipment at the bottom of the sedimentation tank to perform a negative pressure extraction operation on the nitrogen-phosphorus-rich sediment, transporting the sediment from the bottom of the sedimentation tank to a dewatering treatment device, separating the solid and liquid phases to obtain concentrated nitrogen-phosphorus sludge; The concentrated nitrogen and phosphorus sludge is transported to an intelligent granulation system, biochar carrier and slow-release coating material are added for mixed granulation treatment, and granulation temperature and pressure parameters are controlled to form uniform slow-release fertilizer particles; Testing and analyzing the nitrogen and phosphorus content of the slow-release fertilizer granules, and adjusting the nitrogen and phosphorus ratio in the granules according to the test results to produce slow-release nitrogen fertilizer and phosphorus fertilizer products that meet the nutrient requirements of the orchard soil; The slow-release nitrogen fertilizer and phosphate fertilizer products are applied at specific locations according to the growth stage of the fruit trees and the nutrient status of the soil through intelligent fertilizing machinery, completing a closed-loop cycle process of nitrogen and phosphorus nutrients from interception and recovery to reuse.

8. An orchard nitrogen and phosphorus interception system based on intelligent monitoring, characterized in that: For implementing the orchard nitrogen and phosphorus interception method based on intelligent monitoring according to any one of claims 1 to 7, the orchard nitrogen and phosphorus interception system based on intelligent monitoring comprises: The acquisition module is used to collect data from orchard nitrogen and phosphorus concentration sensors, soil moisture sensors, rainfall sensors, and flow sensors, and generates orchard nitrogen and phosphorus monitoring data sets through outlier detection and standardization processing; A generation module is used to use a variational autoencoder algorithm to learn the distribution law of nitrogen and phosphorus in orchard runoff based on the orchard nitrogen and phosphorus monitoring dataset, and generate a relationship model between nitrogen and phosphorus content and interception parameters; A calculation module, for calculating control instructions for the water flow velocity of the stepped groove, the filling amount of the PRB adsorption material, and the dosage of the flocculant in the sedimentation tank by using the relationship model between the nitrogen and phosphorus content and the interception parameter through a multi-objective optimization algorithm; A control module is used to control the operation of the groove regulating valve, the PRB filling system and the sedimentation tank stirring device according to the control instructions, intercept nitrogen and phosphorus in the runoff and collect nitrogen and phosphorus rich sediments; The processing module is used to recover and process the nitrogen- and phosphorus-rich sediment through suction equipment to prepare slow-release nitrogen fertilizer and phosphorus fertilizer for re-application to the orchard.

9. An orchard nitrogen and phosphorus interception device based on intelligent monitoring, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the orchard nitrogen and phosphorus interception method based on intelligent monitoring according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the processor is enabled to execute the orchard nitrogen and phosphorus interception method based on intelligent monitoring according to any one of claims 1 to 7.

Images