Soil quality intelligent detection system fusing environmental monitoring data

Abstract

The invention discloses a soil quality intelligent detection system fusing environmental monitoring data, and belongs to the technical field of soil quality detection. The system comprises the following modules: a multi-source environment sensing module used for realizing accurate sensing and data fusion of multi-dimensional environment parameters; the intelligent soil detection module is used for realizing rapid quantitative analysis and intelligent quality evaluation of soil physicochemical biological indexes; the intelligent factor processing module is used for realizing environmental factor screening, weight optimization and baseline pattern recognition; the multi-temporal-scale coupling analysis module is used for quantifying the dynamic interaction influence between the multi-temporal-scale environmental factors and the soil quality indexes; the prediction decision module is used for generating a soil quality change trend prediction result and outputting a targeted intelligent decision suggestion; and the intelligent management and control execution module is used for realizing intelligent soil quality management and man-machine cooperative control.

Description

Technical Field

[0001] This application relates to the field of soil quality testing technology, and more specifically, to an intelligent soil quality testing system that integrates environmental monitoring data. Background Technology

[0002] With the rapid development of modern agriculture and ecological environment construction, soil quality has become a crucial factor affecting plant growth and the ecological environment. Traditional soil testing mainly relies on manual sampling and laboratory analysis, a cumbersome and time-consuming process that fails to meet the real-time and accuracy requirements of modern soil management. In recent years, although some sensor-based soil testing technologies have emerged, most focus on monitoring single indicators, lacking comprehensive analysis of multiple environmental factors and an understanding of their dynamic changes. Furthermore, the soil environment is complex and variable, with interactions between different factors. Existing technologies remain insufficient in factor screening, weight allocation, and data fusion, making it difficult to accurately reflect the overall condition of the soil. In addition, methods for predicting and managing soil quality are also limited, hindering effective early warning and intelligent control.

[0003] In summary, how to achieve integrated monitoring of multi-source environmental data, scientifically analyze soil quality influencing factors, improve the dynamic response capability of detection, and combine it with prediction and intelligent management methods has become an urgent technical problem to be solved. Summary of the Invention

[0004] To overcome a series of shortcomings in existing technologies, the purpose of this application is to provide an intelligent soil quality detection system that integrates environmental monitoring data, comprising the following modules:

[0005] The multi-source environmental sensing module is used to achieve accurate sensing and data fusion of multi-dimensional environmental parameters;

[0006] The intelligent soil testing module is used to achieve rapid quantitative analysis and intelligent quality assessment of soil physicochemical and biological indicators.

[0007] The intelligent factor processing module is used to screen environmental factors, optimize weights, and identify baseline patterns.

[0008] The multi-temporal-scale coupling analysis module is used to quantify the dynamic interaction between environmental factors and soil quality indicators at multiple temporal scales.

[0009] The prediction and decision-making module is used to generate predictions of soil quality change trends and output targeted intelligent decision-making suggestions.

[0010] The intelligent control and execution module is used to realize intelligent soil quality management and human-machine collaborative control.

[0011] Furthermore, the multi-source environmental sensing module includes a meteorological environmental sensing unit, a soil environmental sensing unit, and an air quality sensing unit, wherein:

[0012] The meteorological environment sensing unit is equipped with a high-precision temperature and humidity sensor array, and integrates an ultraviolet intensity sensor, a photosynthetically active radiation sensor, an ultrasonic wind speed and direction integrated sensor, and a tipping bucket rain gauge to acquire multi-dimensional meteorological parameters in the area in real time.

[0013] The soil environment sensing unit adopts a multi-layer buried sensor network structure. Within the soil surface to a depth of 1.5 meters, sensing nodes are arranged at 25-centimeter intervals. Each node integrates sensors for monitoring soil temperature, moisture content, conductivity, and pH value. In addition, it is equipped with a portable soil nutrient rapid detection device for supplementing sampling analysis and data verification.

[0014] The atmospheric quality sensing unit is equipped with a multi-parameter air quality monitoring device to simultaneously detect the concentration levels of PM2.5, PM10, carbon dioxide, carbon monoxide, sulfur dioxide, and nitrogen oxides.

[0015] Furthermore, the intelligent soil detection module includes an electrochemical index detection unit, a soil physical property detection unit, a microbial activity detection unit, and a near-infrared spectroscopy analysis unit, wherein:

[0016] The electrochemical index detection unit is equipped with a multi-parameter soil sensor array, including a pH sensor, a conductivity sensor, a redox potential sensor, and an ion-selective electrode. The pH sensor has a measurement range of 0-14 pH, an accuracy of ±0.1 pH, and a temperature compensation range of 0-50°C. The conductivity sensor has a measurement range of 0-20 mS / cm, an accuracy of ±2%FS, and a temperature coefficient of 2% / °C. The redox potential sensor has a measurement range of -1000 to +1000 mV, with an accuracy of ±5 mV. The ion-selective electrode includes four types: nitrate nitrogen, ammonium nitrogen, phosphate, and potassium ions, with detection limits of 0.1 mg / L, 0.5 mg / L, 0.2 mg / L, and 1 mg / L, respectively, and a linear range covering 3 to 4 orders of magnitude.

[0017] The soil physical property detection unit measures soil moisture content using a three-wire probe at a depth of 0–30 cm, with a measurement accuracy of ±2% and a measurement range of 0%–50%. It also integrates a soil bulk density sensor and a porosity calculation module. The bulk density sensor measures 0.8–2.0 g / cm³ with an accuracy of ±0.1 g / cm³. Porosity is calculated from bulk density and particle density using the formula: Porosity = (1 - Bulk Density / Particle Density) × 100%. Particle density is automatically matched from a built-in database based on soil texture type.

[0018] The microbial activity detection unit employs a combination of enzyme activity assay and microbial respiration rate assay. Enzyme activity detection includes four key enzymes: urease, phosphatase, sucrase, and catalase. Quantitative analysis is performed using spectrophotometry at detection wavelengths of 578 nm, 410 nm, 540 nm, and 240 nm, with a detection cycle of 2–6 hours and a reproducibility CV value not exceeding 8%. Microbial respiration rate is determined by CO2 release rate, with a measurement range of 0–50 μmol CO2·m³. -2 ·s -1 Accuracy ±2%, resolution 0.1 μmol m -2 ·s -1 Data were collected after incubation in a static chamber at a constant temperature of 20±2°C for 24 h.

[0019] The near-infrared spectroscopy analysis unit employs a Fourier transform near-infrared spectrometer, operating in the 1000-2500 nm wavelength range with a spectral resolution of 8 cm⁻¹. -1 The signal-to-noise ratio is not less than 8000:1, and the scan count is 32 times with cumulative averaging.

[0020] Furthermore, the intelligent factor processing module includes the following components:

[0021] The environmental factor preprocessing unit identifies and removes highly redundant or lowly correlated environmental factors based on correlation analysis between factors, and constructs an optimized factor set.

[0022] The information entropy assessment unit uses information entropy to measure the uncertainty contribution of each environmental factor to soil quality indicators and determines the importance ranking of factors based on the preprocessed factor set.

[0023] The baseline pattern recognition unit analyzes long-term typical change patterns based on historical environmental data to determine the baseline vector for factor weight adjustment under different seasons and management cycles.

[0024] The plant type feature mapping unit matches environmental factors with the corresponding plant type requirements based on the root characteristics and nutrient requirements of the target plant, generating a plant preference coefficient.

[0025] The weight optimization calculation unit combines information entropy, baseline pattern, and plant preference to dynamically solve for the optimal weight allocation and output normalized weight parameters.

[0026] Furthermore, the multi-timescale coupling analysis module includes the following components:

[0027] The short-term time window adaptive unit adaptively divides the short-term analysis window based on optimized factor weights to capture dynamic time lag effects.

[0028] The coupling relationship modeling unit comprehensively calculates the mutual information, optimal lag time, and correlation intensity between each factor and soil index in different short-term windows for the preprocessed key factors, and applies nonlinear regression for dynamic modeling.

[0029] The coupling relationship fusion unit weights and fuses the modeling results of each time scale and factor to generate a comprehensive dynamic coupling relationship matrix.

[0030] The results output and visualization unit provide time-series curves of coupling strength and key time-delay thermograms, intuitively presenting the dynamic multi-timescale impact of environmental factors on soil quality.

[0031] Furthermore, the implementation process of the short-term time window adaptive unit is as follows: The stratified window length is set according to the response characteristics of soil factors; soil temperature is set to a 6-24 hour window, soil moisture to a 12-72 hour window, soil nutrients to a 7-30 day window, and soil microbial activity to a 3-14 day window; the stability of data changes is monitored, and the window length is automatically adjusted to adapt to the data characteristics when the standard deviation change rate exceeds 20%; the abrupt changes in soil environmental conditions are identified, and the impact nodes of management measures and natural events are detected; the window parameters are dynamically optimized according to the root mean square error and the mean absolute percentage error to ensure the optimal balance between computational efficiency and analytical accuracy suitable for soil characteristics is achieved.

[0032] Furthermore, the prediction and decision-making module includes the following components:

[0033] The scenario definition unit automatically generates various typical application scenarios and corresponding data subsets based on the park's location, soil type, and management objectives.

[0034] The model library management unit stores and dynamically updates multiple prediction models, and automatically selects alternative models based on the scenario.

[0035] The trend prediction unit combines the prediction performance of each model under different scenarios to generate prediction results of soil quality change trends.

[0036] The decision-making rules and optimization units prioritize and rank different control measures triggered by the prediction results based on land management strategies and cost-benefit models.

[0037] The decision instruction generation unit automatically generates standardized control instructions, including operation type, execution parameters, priority, and expected results.

[0038] Furthermore, the intelligent control execution module includes the following components:

[0039] The instruction receiving and parsing unit receives and parses standardized control instructions from the predictive decision module, converting them into executable operation instructions and parameters;

[0040] The execution mode management unit dynamically switches and controls the three levels of execution modes—automatic, semi-automatic, and manual—based on preset settings to ensure that the execution process complies with the control strategy.

[0041] The equipment interface and command unit send operation commands to irrigation valves, fertilizer applicators and soil amendment devices through standardized communication protocols, and monitor the status of command issuance;

[0042] The job scheduling and coordination unit, in combination with the location of the plot, equipment load and priority, schedules and coordinates multiple control tasks in parallel to avoid resource conflicts and duplicate triggering.

[0043] The anomaly detection and safety protection unit can identify equipment malfunctions or abnormal operations in real time and automatically trigger alarms or switch to safe mode.

[0044] The operation log and audit unit logs all executed instructions, mode switches, and feedback data, supporting retrospective auditing and subsequent optimization analysis.

[0045] Furthermore, the execution mode management unit adapts to different soil management scenarios through an intelligent mode switching strategy. Specific mode management mechanisms include: In automatic execution mode, it autonomously executes routine maintenance instructions based on soil monitoring data and preset management strategies, suitable for low-risk daily management operations; in supervised execution mode, it sends a confirmation request to management personnel before executing important management measures, and executes only after obtaining authorization, suitable for operations with significant impact; in manual execution mode, all instructions require management personnel to manually trigger execution after confirming soil conditions on-site, suitable for initial management of new areas or handling of abnormal situations; a mode switching condition judgment mechanism is established to automatically select the most suitable execution mode based on the degree of soil quality deviation, operational risk level, historical management results, and current weather conditions; a mode switching permission management function is designed, with different levels of management personnel having different mode switching permissions; and an emergency response mechanism is established to automatically switch to manual mode when soil pollution or extreme weather is detected.

[0046] Compared with the prior art, this application has the following beneficial effects:

[0047] This application innovatively constructs an intelligent soil quality detection and control system that combines multi-source environmental data fusion with multi-timescale dynamic coupling analysis, realizing closed-loop management of the entire process of precise monitoring, factor optimization, trend prediction and intelligent execution. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the structure of an intelligent soil quality detection system that integrates environmental monitoring data, as disclosed in an embodiment of this application. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are some embodiments of this invention, but not all embodiments.

[0050] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0052] Figure 1 This is a schematic diagram of the structure of an intelligent soil quality detection system that integrates environmental monitoring data, as disclosed in an embodiment of this application.

[0053] like Figure 1 As shown, a soil quality intelligent detection system integrating environmental monitoring data includes the following modules:

[0054] The multi-source environmental sensing module is used to achieve accurate sensing and data fusion of multi-dimensional environmental parameters;

[0055] The intelligent soil testing module is used to achieve rapid quantitative analysis and intelligent quality assessment of soil physicochemical and biological indicators.

[0056] The intelligent factor processing module is used to screen environmental factors, optimize weights, and identify baseline patterns.

[0057] The multi-temporal-scale coupling analysis module is used to quantify the dynamic interaction between environmental factors and soil quality indicators at multiple temporal scales.

[0058] The prediction and decision-making module is used to generate predictions of soil quality change trends and output targeted intelligent decision-making suggestions.

[0059] The intelligent control and execution module is used to realize intelligent soil quality management and human-machine collaborative control.

[0060] This system achieves rapid, accurate detection and intelligent assessment of soil physicochemical and biological indicators through precise perception and fusion of multi-source environmental data. The screening and weight optimization of environmental factors enhances the scientific rigor of data processing, while multi-timescale coupling analysis further reveals the dynamic relationship between environmental factors and soil quality. Trend analysis based on the predictive decision-making module provides early warnings of soil quality changes and offers targeted management recommendations, enhancing the initiative and precision of soil management. The intelligent control and execution module automates the management process and enables human-machine collaboration, effectively improving the efficiency and reliability of soil quality maintenance.

[0061] Furthermore, the multi-source environmental sensing module includes a meteorological environmental sensing unit, a soil environmental sensing unit, and an air quality sensing unit, wherein:

[0062] The meteorological environment sensing unit is equipped with a high-precision temperature and humidity sensor array, and integrates an ultraviolet intensity sensor, a photosynthetically active radiation sensor, an ultrasonic wind speed and direction integrated sensor, and a tipping bucket rain gauge to acquire multi-dimensional meteorological parameters in the area in real time.

[0063] The soil environment sensing unit adopts a multi-layer buried sensor network structure. Within the soil surface to a depth of 1.5 meters, sensing nodes are arranged at 25-centimeter intervals. Each node integrates sensors for monitoring soil temperature, moisture content, conductivity, and pH value. In addition, it is equipped with a portable soil nutrient rapid detection device for supplementing sampling analysis and data verification.

[0064] The atmospheric quality sensing unit is equipped with a multi-parameter air quality monitoring device to simultaneously detect the concentration levels of PM2.5, PM10, carbon dioxide, carbon monoxide, sulfur dioxide, and nitrogen oxides.

[0065] Furthermore, the intelligent soil detection module includes an electrochemical index detection unit, a soil physical property detection unit, a microbial activity detection unit, and a near-infrared spectroscopy analysis unit, wherein:

[0066] The electrochemical index detection unit is equipped with a multi-parameter soil sensor array, including a pH sensor, a conductivity sensor, a redox potential sensor, and an ion-selective electrode. The pH sensor has a measurement range of 0-14 pH, an accuracy of ±0.1 pH, and a temperature compensation range of 0-50°C. The conductivity sensor has a measurement range of 0-20 mS / cm, an accuracy of ±2%FS, and a temperature coefficient of 2% / °C. The redox potential sensor has a measurement range of -1000 to +1000 mV, with an accuracy of ±5 mV. The ion-selective electrode includes four types: nitrate nitrogen, ammonium nitrogen, phosphate, and potassium ions, with detection limits of 0.1 mg / L, 0.5 mg / L, 0.2 mg / L, and 1 mg / L, respectively, and a linear range covering 3 to 4 orders of magnitude.

[0067] The soil physical property detection unit measures soil moisture content using a three-wire probe at a depth of 0–30 cm, with a measurement accuracy of ±2% and a measurement range of 0%–50%. It also integrates a soil bulk density sensor and a porosity calculation module. The bulk density sensor measures 0.8–2.0 g / cm³ with an accuracy of ±0.1 g / cm³. Porosity is calculated from bulk density and particle density, with particle density automatically matched from a built-in database based on soil texture type.

[0068] The microbial activity detection unit employs a combination of enzyme activity assay and microbial respiration rate assay. Enzyme activity detection includes four key enzymes: urease, phosphatase, sucrase, and catalase. Quantitative analysis is performed using spectrophotometry at detection wavelengths of 578 nm, 410 nm, 540 nm, and 240 nm, with a detection cycle of 2–6 hours and a reproducibility CV value not exceeding 8%. Microbial respiration rate is determined by CO2 release rate, with a measurement range of 0–50 μmol CO2·m³. -2 ·s -1 Accuracy ±2%, resolution 0.1 μmol m -2 ·s -1 Data were collected after incubation in a static chamber at a constant temperature of 20±2°C for 24 h.

[0069] The near-infrared spectroscopy analysis unit employs a Fourier transform near-infrared spectrometer, operating in the 1000-2500 nm wavelength range with a spectral resolution of 8 cm⁻¹. -1 The signal-to-noise ratio is not less than 8000:1, and the scan count is 32 times with cumulative averaging.

[0070] Furthermore, the intelligent factor processing module includes the following components:

[0071] The environmental factor preprocessing unit identifies and removes highly redundant or lowly correlated environmental factors based on correlation analysis between factors, and constructs an optimized factor set.

[0072] The information entropy assessment unit uses information entropy to measure the uncertainty contribution of each environmental factor to soil quality indicators and determines the importance ranking of factors based on the preprocessed factor set.

[0073] The baseline pattern recognition unit analyzes long-term typical change patterns based on historical environmental data to determine the baseline vector for factor weight adjustment under different seasons and management cycles.

[0074] The plant type feature mapping unit matches environmental factors with the corresponding plant type requirements based on the root characteristics and nutrient requirements of the target plant, generating a plant preference coefficient.

[0075] The weight optimization calculation unit combines information entropy, baseline pattern, and plant preference to dynamically solve for the optimal weight allocation and output normalized weight parameters.

[0076] This intelligent factor processing module effectively filters out redundant and irrelevant environmental factors through preprocessing, ensuring the scientific rationality of the factor set. Information entropy assessment further quantifies the uncertain impact of each factor on soil quality, improving the accuracy of factor ranking. Baseline pattern recognition, combined with historical data, dynamically adjusts factor weights to adapt to changes in different seasons and management cycles. Plant type feature mapping achieves precise matching between environmental factors and plant needs, improving the targeted application of factors. Finally, the weight optimization calculation unit integrates multiple aspects of information and dynamically allocates weights, ensuring that the assessment results better reflect the actual soil and plant growth environment.

[0077] The environmental factor preprocessing unit filters data by establishing a factor correlation matrix. Specifically, it includes: calculating the Pearson correlation coefficient between environmental factors, identifying highly redundant factors when the absolute value of the correlation coefficient is greater than 0.75; further verifying multicollinearity using the variance inflation factor test, setting a threshold of 5, and marking factors exceeding this threshold as objects to be removed; determining the contribution of each environmental factor to soil target indicators through stepwise regression analysis, retaining key factors with a contribution greater than 3%; and comprehensively considering the statistical significance, biological significance, and practical measurability of the factors to construct a set of environmental factors that is both concise and practically instructive.

[0078] The implementation process of the information entropy assessment unit is as follows: dividing continuous environmental factor data into 5 to 8 meaningful intervals; establishing a discretized system based on soil fertility grading standards for soil quality grades, divided into five levels: excellent, good, moderate, poor, and very poor; using the conditional entropy formula to calculate the degree of reduction of soil quality uncertainty by each environmental factor, while introducing a time weight factor to reflect the importance of recent data; combining mutual information and Gini impurity indexes to correct the limitations of a single entropy value index, and finally outputting a standardized factor importance score.

[0079] The implementation of the baseline pattern recognition unit includes: decomposing historical environmental data into three components: trend, seasonal, and irregular; identifying major periodic patterns based on soil thermodynamic properties, including daily temperature fluctuation cycles, weekly rainfall cycles, monthly plant growth cycles, and annual climate cycles; establishing a seasonal adjustment factor database to record the typical variation amplitude and phase difference of various environmental factors under the four seasons and different phenological stages of plants; identifying similar soil environmental pattern types and establishing corresponding baseline weight vectors for each pattern type to achieve a dynamic factor weight adjustment mechanism that conforms to soil ecological laws.

[0080] The implementation process of the plant type feature mapping unit is as follows: A plant characteristic database is constructed to record the root depth range, root activity layer thickness, optimal pH range, nutrient requirements, and salt tolerance parameters of different target plants; a plant-soil adaptability evaluation matrix is ​​established to quantify the sensitivity and suitability range of each plant type to environmental factors; a combination of trapezoidal and Gaussian membership functions is used to describe the plant's preference for environmental factors, setting the optimal interval membership degree to 0.9-1.0, the suitable interval membership degree to 0.6-0.8, and the barely suitable interval membership degree to 0.3-0.5; the preference coefficient is corrected using plant physiology expert knowledge and years of field observation data to generate personalized environmental factor weight adjustment parameters for different plant types.

[0081] The weight optimization calculation unit employs a multi-objective optimization algorithm to solve for the optimal weight allocation. Specifically, it includes: establishing a multi-objective optimization mathematical model with the objective functions of maximizing information entropy score, maximizing baseline pattern matching degree, and optimizing plant adaptability; setting weight constraints, including non-negativity constraints for each factor weight, normalization constraints that the sum of weights equals 1, and reasonable constraints that the difference in weights of similar factors does not exceed 30%; solving for the Pareto optimal solution set, setting the population size to 30, the maximum number of iterations to 100, the crossover probability to 0.7, and the mutation probability to 0.1; selecting the optimal compromise solution from the Pareto front, outputting the final weight parameters of each environmental factor, and performing normalization processing.

[0082] Furthermore, the multi-timescale coupling analysis module includes the following components:

[0083] The short-term time window adaptive unit adaptively divides the short-term analysis window based on optimized factor weights to capture dynamic time lag effects.

[0084] The coupling relationship modeling unit comprehensively calculates the mutual information, optimal lag time, and correlation intensity between each factor and soil index in different short-term windows for the preprocessed key factors, and applies nonlinear regression for dynamic modeling.

[0085] The coupling relationship fusion unit weights and fuses the modeling results of each time scale and factor to generate a comprehensive dynamic coupling relationship matrix.

[0086] The results output and visualization unit provide time-series curves of coupling strength and key time-delay thermograms, intuitively presenting the dynamic multi-timescale impact of environmental factors on soil quality.

[0087] The multi-timescale coupling analysis module effectively captures the dynamic time-lag relationship between environmental factors and soil quality by adaptively adjusting short-term time windows. The coupling relationship modeling unit comprehensively calculates the mutual information and correlation between key factors and soil indicators, employing nonlinear regression to achieve dynamic modeling, thus improving the model's accuracy and flexibility. The fusion unit integrates modeling results from different time scales and factors to construct a comprehensive coupling relationship matrix, fully reflecting the complex interactions between factors. The results output module graphically displays coupling strength and time-lag characteristics, helping users intuitively understand the multi-time-scale impact of environmental changes on soil quality.

[0088] The implementation process of the short-term time window adaptive unit is as follows: The stratified window length is set according to the response characteristics of soil factors; soil temperature is set to a 6-24 hour window, soil moisture to a 12-72 hour window, soil nutrients to a 7-30 day window, and soil microbial activity to a 3-14 day window; the stability of data changes is monitored, and the window length is automatically adjusted to adapt to the data characteristics when the standard deviation change rate exceeds 20%; the abrupt changes in soil environmental conditions are identified, and the impact points of management measures and natural events are detected; the window parameters are dynamically optimized based on the root mean square error and the mean absolute percentage error to ensure the optimal balance between computational efficiency and analytical accuracy suitable for soil characteristics is achieved.

[0089] The implementation process of the coupling relationship modeling unit is as follows: calculate the monotonic relationship strength between each environmental factor and soil quality index, and factors with an absolute value of correlation coefficient greater than 0.4 are included in the modeling candidate set; determine the optimal lag time through partial correlation analysis; and establish a soil factor coupling relationship model to capture the nonlinear response characteristics in the soil.

[0090] The coupling relationship fusion unit integrates multi-timescale analysis results through a confidence-based integration strategy, specifically including: calculating confidence weights based on the prediction accuracy and stability of the model within each time window; establishing a timescale importance assessment mechanism, where the short-term scale focuses on emergency response, the medium-term scale focuses on seasonal changes, and the long-term scale reflects cumulative effects; constructing a dynamic coupling relationship matrix, where matrix elements represent the coupling strength and direction of different environmental factors and soil indicators at specific time scales; and using factor analysis to reduce the dimensional complexity of the coupling relationship matrix, extracting the main coupling patterns and key influence paths, and retaining the principal factors with a cumulative contribution rate of 85%.

[0091] Furthermore, the prediction and decision-making module includes the following components:

[0092] The scenario definition unit automatically generates various typical application scenarios and corresponding data subsets based on the park's location, soil type, and management objectives.

[0093] The model library management unit stores and dynamically updates multiple prediction models, and automatically selects alternative models based on the scenario.

[0094] The trend prediction unit combines the prediction performance of each model under different scenarios to generate prediction results of soil quality change trends.

[0095] The decision-making rules and optimization units prioritize and rank different control measures triggered by the prediction results based on land management strategies and cost-benefit models.

[0096] The decision instruction generation unit automatically generates standardized control instructions, including operation type, execution parameters, priority, and expected results.

[0097] The predictive decision-making module can automatically define diverse application scenarios based on the actual conditions of the park, ensuring the relevance and accuracy of data analysis. By dynamically managing and selecting the most suitable predictive model, the accuracy of predicting soil quality change trends is improved. Combining management strategies and cost-effectiveness, possible control measures are rationally prioritized, optimizing resource allocation. Finally, standardized control instructions are automatically generated, clearly defining operational details and priorities, achieving efficient execution of soil management.

[0098] The scenario definition unit automatically generates typical application scenarios by combining soil classification and soil application characteristics. The specific scenario construction steps include: obtaining basic soil characteristics based on the Chinese Soil System Classification and the World Soil Resources Reference Base; identifying different functional positions by combining application functional zoning; classifying according to management intensity, including refined management, standardized management, and extensive management; classifying similar feature combinations into typical scenarios, each scenario including basic soil attributes, plant configuration patterns, management intensity characteristics, and historical performance data; calculating the matching degree between the current park characteristics and each typical scenario, and automatically selecting the scenario template with the highest similarity.

[0099] The specific prediction steps of the trend prediction unit include: selecting 2-4 prediction models from the model library that perform best under the current soil type and management scenario as base learners; assigning dynamic weights to each base learner based on its performance on data from the most recent 3 months, with the weight update formula being w. i =exp(-α i ×RMSE i ) / Σ j exp(-α j ×RMSE j ), where α iThe attenuation coefficient is used; a prediction interval estimation mechanism is designed to provide a 95% confidence interval for the prediction results; physical boundary conditions for soil quality changes are set, and prediction results that exceed a reasonable range will be marked and manually reviewed; the output is a multi-time span prediction result that conforms to the actual situation of soil management, including the soil quality change trend for the next 1 week, 2 weeks, 1 month and 3 months.

[0100] The decision instruction generation unit generates executable instructions for soil management using a standardized format. These instructions include: a management measure type field specifying irrigation replenishment, organic fertilizer application, chemical fertilizer topdressing, pH adjustment, and soil loosening operations; an application parameter field detailing irrigation volume, fertilizer type and dosage, operation depth, and treatment area parameters; a target area field precisely locating the operation area using the park's grid number; an execution time field determining the optimal application time window based on soil moisture conditions, weather forecasts, and plant growth stages; an expected target field describing the expected improvement in key soil indicators after execution; and a precautions field including operational safety requirements, environmental protection measures, and quality control points.

[0101] Furthermore, the intelligent control execution module includes the following components:

[0102] The instruction receiving and parsing unit receives and parses standardized control instructions from the predictive decision module, converting them into executable operation instructions and parameters;

[0103] The execution mode management unit dynamically switches and controls the three levels of execution modes—automatic, semi-automatic, and manual—based on preset settings to ensure that the execution process complies with the control strategy.

[0104] The equipment interface and command unit send operation commands to irrigation valves, fertilizer applicators and soil amendment devices through standardized communication protocols, and monitor the status of command issuance;

[0105] The job scheduling and coordination unit, in combination with the location of the plot, equipment load and priority, schedules and coordinates multiple control tasks in parallel to avoid resource conflicts and duplicate triggering.

[0106] The anomaly detection and safety protection unit can identify equipment malfunctions or abnormal operations in real time and automatically trigger alarms or switch to safe mode.

[0107] The operation log and audit unit logs all executed instructions, mode switches, and feedback data, supporting retrospective auditing and subsequent optimization analysis.

[0108] The intelligent control and execution module can translate prediction results into specific operation instructions and flexibly adapt to various execution modes such as automatic, semi-automatic, and manual, enhancing its adaptability. It controls various terminal devices through standardized interfaces, enabling efficient collaborative operations. The scheduling unit can perform reasonable scheduling based on task priority and equipment status to avoid resource conflicts. It also features anomaly detection and safe switching mechanisms to ensure stable and reliable execution. All operations are recorded, facilitating subsequent traceability and management optimization, thus enhancing the accuracy and safety of soil control.

[0109] The instruction receiving and parsing unit processes soil management instructions through a multi-level parsing mechanism. Specific parsing steps include: checking the completeness and parameter range rationality of the management instructions to ensure key parameters are within safe operating limits; converting soil improvement goals into specific operational parameters; verifying whether operational parameters meet soil carrying capacity and environmental protection requirements; identifying mutually influencing management measures within the same area; adjusting the execution sequence in real time based on soil urgency, weather conditions, and equipment availability; and predicting possible outcomes of instruction execution through soil process simulation to identify potential problems in advance.

[0110] The execution mode management unit adapts to different soil management scenarios through an intelligent mode switching strategy. Specific mode management mechanisms include: In automatic execution mode, it autonomously executes routine maintenance instructions based on soil monitoring data and preset management strategies, suitable for low-risk daily management operations; in supervised execution mode, it sends a confirmation request to management personnel before executing important management measures, and executes only after obtaining authorization, suitable for operations with significant impact; in manual execution mode, all instructions require management personnel to manually trigger execution after confirming soil conditions on-site, suitable for initial management of new areas or handling of abnormal situations; a mode switching condition judgment mechanism is established to automatically select the most suitable execution mode based on the degree of soil quality deviation, operational risk level, historical management results, and current weather conditions; a mode switching permission management function is designed, with different levels of management personnel having different mode switching permissions; and an emergency response mechanism is established to automatically switch to manual mode when soil pollution or extreme weather is detected.

[0111] The device interface and command unit achieve unified control of soil management equipment through a standardized communication protocol. Specific communication mechanisms include: using the RS485 bus protocol to communicate with traditional irrigation controllers and solenoid valve control cabinets, supporting a maximum transmission distance of 1200 meters and 32 device nodes; using the LoRaWAN protocol to communicate with distributed soil sensors and fertilization equipment, achieving low-power, long-distance communication; establishing a device authentication and registration mechanism, ensuring communication security through unique device identifiers and digital certificates; designing a command confirmation and status feedback mechanism, requiring the device to respond to the execution status within 30 seconds after each control command is sent, recording an anomaly and attempting to resend if the timeout occurs; establishing a device health status monitoring function, periodically querying key device parameters; designing a communication load balancing strategy, dynamically adjusting the command sending frequency based on network congestion and device responsiveness; and establishing a communication log recording mechanism, recording in detail the time, content, response status, and anomaly information of each communication.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A soil quality intelligent detection system integrating environmental monitoring data, characterized in that, Includes the following modules: The multi-source environmental sensing module is used to achieve accurate sensing and data fusion of multi-dimensional environmental parameters; The intelligent soil testing module is used to achieve rapid quantitative analysis and intelligent quality assessment of soil physicochemical and biological indicators. The intelligent factor processing module is used to screen environmental factors, optimize weights, and identify baseline patterns. The multi-temporal-scale coupling analysis module is used to quantify the dynamic interaction between environmental factors and soil quality indicators at multiple temporal scales. The prediction and decision-making module is used to generate predictions of soil quality change trends and output targeted intelligent decision-making suggestions. The intelligent control and execution module is used to realize intelligent soil quality management and human-machine collaborative control.

2. The intelligent soil quality detection system integrating environmental monitoring data according to claim 1, characterized in that, The multi-source environmental sensing module includes a meteorological environmental sensing unit, a soil environmental sensing unit, and an air quality sensing unit, wherein: The meteorological environment sensing unit is equipped with a high-precision temperature and humidity sensor array, and integrates an ultraviolet intensity sensor, a photosynthetically active radiation sensor, an ultrasonic wind speed and direction integrated sensor, and a tipping bucket rain gauge to acquire multi-dimensional meteorological parameters in the area in real time. The soil environment sensing unit adopts a multi-layer buried sensor network structure. Within the soil surface to a depth of 1.5 meters, sensing nodes are arranged at 25-centimeter intervals. Each node integrates sensors for monitoring soil temperature, moisture content, conductivity, and pH value. In addition, it is equipped with a portable soil nutrient rapid detection device for supplementing sampling analysis and data verification. The atmospheric quality sensing unit is equipped with a multi-parameter air quality monitoring device to simultaneously detect the concentration levels of PM2.5, PM10, carbon dioxide, carbon monoxide, sulfur dioxide, and nitrogen oxides.

3. The intelligent soil quality detection system integrating environmental monitoring data according to claim 2, characterized in that, The intelligent soil detection module includes an electrochemical index detection unit, a soil physical property detection unit, a microbial activity detection unit, and a near-infrared spectroscopy analysis unit, wherein: The electrochemical index detection unit is equipped with a multi-parameter soil sensor array, including a pH sensor, a conductivity sensor, a redox potential sensor, and an ion-selective electrode. The pH sensor has a measurement range of 0-14 pH, an accuracy of ±0.1 pH, and a temperature compensation range of 0-50°C. The conductivity sensor has a measurement range of 0-20 mS / cm, an accuracy of ±2%FS, and a temperature coefficient of 2% / °C. The redox potential sensor has a measurement range of -1000 to +1000 mV, with an accuracy of ±5 mV. The ion-selective electrode includes four types: nitrate nitrogen, ammonium nitrogen, phosphate, and potassium ions, with detection limits of 0.1 mg / L, 0.5 mg / L, 0.2 mg / L, and 1 mg / L, respectively, and a linear range covering 3 to 4 orders of magnitude. The soil physical property detection unit measures soil moisture content using a three-wire probe at a depth of 0–30 cm, with a measurement accuracy of ±2% and a measurement range of 0%–50%. It also integrates a soil bulk density sensor and a porosity calculation module. The bulk density sensor measures 0.8–2.0 g / cm³ with an accuracy of ±0.1 g / cm³. Porosity is calculated from bulk density and particle density, with particle density automatically matched from a built-in database based on soil texture type. The microbial activity detection unit employs a combination of enzyme activity assay and microbial respiration rate assay. Enzyme activity detection includes four key enzymes: urease, phosphatase, sucrase, and catalase. Quantitative analysis is performed using spectrophotometry at detection wavelengths of 578 nm, 410 nm, 540 nm, and 240 nm, with a detection cycle of 2–6 hours and a reproducibility CV value not exceeding 8%. Microbial respiration rate is determined by CO2 release rate, with a measurement range of 0–50 μmol CO2·m³. -2 ·s -1 Accuracy ±2%, resolution 0.1 μmol m -2 ·s -1 Read the values ​​after culturing in a static chamber at a constant temperature of 20±2°C for 24 hours; The near-infrared spectroscopy analysis unit employs a Fourier transform near-infrared spectrometer, operating in the 1000-2500 nm wavelength range with a spectral resolution of 8 cm⁻¹. -1 The signal-to-noise ratio is not less than 8000:1, and the scan count is 32 times with cumulative averaging.

4. The intelligent soil quality detection system integrating environmental monitoring data according to claim 1, characterized in that, The intelligent factor processing module includes the following components: The environmental factor preprocessing unit identifies and removes highly redundant or lowly correlated environmental factors based on correlation analysis between factors, and constructs an optimized factor set. The information entropy assessment unit uses information entropy to measure the uncertainty contribution of each environmental factor to soil quality indicators and determines the importance ranking of factors based on the preprocessed factor set. The baseline pattern recognition unit analyzes long-term typical change patterns based on historical environmental data to determine the baseline vector for factor weight adjustment under different seasons and management cycles. The plant type feature mapping unit matches environmental factors with the corresponding plant type requirements based on the root characteristics and nutrient requirements of the target plant, generating a plant preference coefficient. The weight optimization calculation unit combines information entropy, baseline pattern, and plant preference to dynamically solve for the optimal weight allocation and output normalized weight parameters.

5. The intelligent soil quality detection system integrating environmental monitoring data according to claim 1, characterized in that, The multi-timescale coupling analysis module includes the following components: The short-term time window adaptive unit adaptively divides the short-term analysis window based on optimized factor weights to capture dynamic time lag effects. The coupling relationship modeling unit comprehensively calculates the mutual information, optimal lag time, and correlation intensity between each factor and soil index in different short-term windows for the preprocessed key factors, and applies nonlinear regression for dynamic modeling. The coupling relationship fusion unit weights and fuses the modeling results of each time scale and factor to generate a comprehensive dynamic coupling relationship matrix. The results output and visualization unit provide time-series curves of coupling strength and key time-delay thermograms, intuitively presenting the dynamic multi-timescale impact of environmental factors on soil quality.

6. The intelligent soil quality detection system integrating environmental monitoring data according to claim 5, characterized in that, The implementation process of the short-term time window adaptive unit is as follows: The stratified window length is set according to the response characteristics of soil factors; soil temperature is set to a 6-24 hour window, soil moisture to a 12-72 hour window, soil nutrients to a 7-30 day window, and soil microbial activity to a 3-14 day window; the stability of data changes is monitored, and the window length is automatically adjusted to adapt to the data characteristics when the standard deviation change rate exceeds 20%; the moment of abrupt change in soil environmental conditions is identified, and the impact nodes of management measures and natural events are detected; the window parameters are dynamically optimized according to the root mean square error and the mean absolute percentage error to ensure the best balance between computational efficiency and analytical accuracy suitable for soil characteristics.

7. The intelligent soil quality detection system integrating environmental monitoring data according to claim 1, characterized in that, The prediction and decision-making module includes the following components: The scenario definition unit automatically generates various typical application scenarios and corresponding data subsets based on the park's location, soil type, and management objectives. The model library management unit stores and dynamically updates multiple prediction models, and automatically selects alternative models based on the scenario. The trend prediction unit combines the prediction performance of each model under different scenarios to generate prediction results of soil quality change trends. The decision-making rules and optimization units prioritize and rank different control measures triggered by the prediction results based on land management strategies and cost-benefit models. The decision instruction generation unit automatically generates standardized control instructions, including operation type, execution parameters, priority, and expected results.

8. The intelligent soil quality detection system integrating environmental monitoring data according to claim 1, characterized in that, The intelligent control and management execution module includes the following components: The instruction receiving and parsing unit receives and parses standardized control instructions from the predictive decision module, converting them into executable operation instructions and parameters; The execution mode management unit dynamically switches and controls the three levels of execution modes—automatic, semi-automatic, and manual—based on preset settings to ensure that the execution process complies with the control strategy. The equipment interface and command unit send operation commands to irrigation valves, fertilizer applicators and soil amendment devices through standardized communication protocols, and monitor the status of command issuance; The job scheduling and coordination unit, in combination with the location of the plot, equipment load and priority, schedules and coordinates multiple control tasks in parallel to avoid resource conflicts and duplicate triggering. The anomaly detection and safety protection unit can identify equipment malfunctions or abnormal operations in real time and automatically trigger alarms or switch to safe mode. The operation log and audit unit logs all executed instructions, mode switches, and feedback data, supporting retrospective auditing and subsequent optimization analysis.

9. The intelligent soil quality detection system integrating environmental monitoring data according to claim 8, characterized in that, The execution mode management unit adapts to different soil management scenarios through an intelligent mode switching strategy. Specific mode management mechanisms include: In automatic execution mode, it autonomously executes routine maintenance instructions based on soil monitoring data and preset management strategies, suitable for low-risk daily management operations; in supervised execution mode, it sends a confirmation request to management personnel before implementing important management measures, and executes only after obtaining authorization, suitable for operations with significant impact; in manual execution mode, all instructions require management personnel to manually trigger execution after confirming soil conditions on-site, suitable for initial management of new areas or handling of abnormal situations; a mode switching condition judgment mechanism is established to automatically select the most suitable execution mode based on the degree of soil quality deviation, operational risk level, historical management results, and current weather conditions; a mode switching permission management function is designed, with different levels of management personnel having different mode switching permissions; and an emergency response mechanism is established to automatically switch to manual mode when soil pollution or extreme weather is detected.

Images