Multi-sensor networking soil parameter high-precision detection device and method

Abstract

The application belongs to the technical field of intelligent soil monitoring, and particularly relates to a multi-sensor networking soil parameter high-precision detection device and method. The device adopts a four-layer architecture of sensing, transmission, data processing and application. The sensing layer collects soil texture, temperature and salt parameters, constructs an environmental interference factor matrix, and outputs non-interference data through a layered interference calibration model. The transmission layer completes data transmission in wired and wireless dual modes. The data processing layer extracts spatial and temporal characteristics after preprocessing, and obtains a soil parameter fusion value through dynamic weight fusion. The application layer matches the fusion value with a preset template, realizes scenario generation and visual rendering. The application can solve the problems of the existing multi-sensor networking system, such as easy influence of precision by external factors and insufficient intelligence.

Description

Technical Field

[0001] This invention belongs to the field of intelligent soil monitoring technology, and particularly relates to a high-precision soil parameter detection device and method using a multi-sensor network. Background Technology

[0002] As the foundation of terrestrial ecosystems, the accurate measurement of soil parameters is crucial for agricultural production, ecological environmental protection, and geological research. In agriculture, soil parameters such as moisture, nutrients, pH, and electrical conductivity directly affect crop growth, yield, and quality. Precise understanding of these parameters enables precise fertilization and rational irrigation, improving fertilizer and water utilization efficiency, reducing agricultural non-point source pollution, and promoting sustainable agricultural development. In ecological environmental protection, soil parameter testing helps assess changes in soil quality, monitor soil pollution and degradation, and provide a scientific basis for ecological restoration and environmental protection. For example, by detecting indicators such as heavy metal content and organic pollutants in the soil, soil pollution problems can be identified in a timely manner, allowing for appropriate remediation measures to protect the ecological environment and human health.

[0003] Domestic research has made certain progress in soil parameter detection devices and multi-sensor networking technology. Regarding soil parameter detection devices, various types of sensors have been developed, such as soil moisture sensors, soil nutrient sensors, soil pH sensors, and soil temperature sensors. These sensors have been continuously optimized and improved in terms of detection principles and performance indicators, and the accuracy and stability of some sensors have reached international advanced levels. For example, some soil moisture sensors based on frequency domain reflectance (FDR) can accurately measure soil moisture content with errors controlled within a small range. In terms of multi-sensor networking technology, domestic scholars have conducted extensive research, exploring different networking methods and data transmission protocols, achieving self-organization, self-healing of sensor nodes, and reliable data transmission. Simultaneously, certain achievements have been made in data fusion algorithms and sensor network optimization, improving the performance and efficiency of multi-sensor networking systems.

[0004] However, current research still has some shortcomings. On the one hand, the performance of some sensors is not stable enough and is easily affected by environmental factors, leading to errors in the detection data. For example, the accuracy of some sensors will decrease in harsh environments such as high temperature and high humidity. On the other hand, the integration and intelligence level of multi-sensor network systems need to be improved, and their data processing and analysis capabilities are relatively weak, making it difficult to meet the needs of large-scale, complex soil parameter monitoring. Summary of the Invention

[0005] The technical problem solved by this invention is to provide a high-precision soil parameter detection device and method using a multi-sensor network, so as to solve the problems of existing multi-sensor network systems having accuracy that is easily affected by external factors and insufficient intelligence.

[0006] The basic solution provided by this invention is a high-precision soil parameter detection device with multi-sensor networking, comprising a sensing layer, a transmission layer, a data processing layer, and an application layer, wherein:

[0007] The perception layer includes a multi-parameter integrated sensor module and an interference calibration module. The multi-parameter integrated sensor module is used to construct an environmental interference factor matrix based on real-time acquisition of soil environmental parameters by multiple sensor units. The interference calibration module calls a preset interference calibration model to dynamically adjust the detection parameters of each sensor unit according to the environmental interference factor matrix and outputs a set of interference-free soil data from each sensor.

[0008] The transport layer is used to transmit the interference-free soil data set generated by the sensing layer to the data processing layer;

[0009] The data processing layer includes a preprocessing module and a soil parameter extraction module. The preprocessing module is used to preprocess the received set of interference-free soil data to generate preprocessed soil data. The soil parameter extraction module is used to call a preset soil parameter extraction model to perform spatial feature extraction and temporal feature extraction based on the preprocessed soil data, and output the fused soil parameter fusion value.

[0010] The application layer is used to receive the fused soil parameter values ​​and call the preset soil parameter template to generate and render soil data.

[0011] Furthermore, in the perception layer, the multi-parameter integrated sensor module includes a multi-parameter sensor unit and an environmental perception unit. The multi-parameter sensor unit is used to collect soil texture parameters, soil temperature, and soil salinity data.

[0012] The environmental sensing unit is used to acquire soil texture parameters, soil temperature, and soil salinity data, and to construct an environmental interference factor matrix.

[0013] Furthermore, the environmental sensing unit acquires soil texture parameters, soil temperature, and soil salinity data, and constructs an environmental interference factor matrix as follows:

[0014] Obtain any sensor node of the multi-parameter sensor unit During the detection time Soil texture parameters Soil temperature Soil salinity data ;

[0015] Based on the hardware specifications of the multi-parameter sensor unit, inherent sensor detection parameters and spatiotemporal characteristic parameters of the sensor network are extracted. The inherent sensor detection parameters include the... Sensor detection sensitivity coefficient for soil parameters Sensor basic detection error threshold The spatiotemporal characteristic parameters of sensor networking include the time decay coefficient. Spatial networking coefficient Soil parameter background values ;

[0016] Based on sensor detection sensitivity coefficient and sensor basic detection error threshold For a single sensor node m in time The original parameters are normalized, and the expression is:

[0017]

[0018] in, The normalized interference value of the original sensor parameters. The original values ​​of the sensor for the i-th type of parameter are... , Let be the extreme value of the sensor's detection range for the i-th type of parameter;

[0019] Based on the real-time detection value of the i-th type of parameter, determine its relationship with the soil parameter background value. The deviation is used to obtain the environmental deviation coefficient. Based on the detection principle of the multi-parameter sensor unit and the types of interference from soil parameters to sensor detection, the foundation contribution coefficient was calibrated. ;

[0020] Based on the basic contribution coefficient Environmental deviation coefficient Time decay coefficient Spatial networking coefficient Calculate the contribution coefficient of spatiotemporal coupling interference The expression is:

[0021]

[0022] in, ; Characterizing soil texture parameters Characterizing soil temperature, Characterizing soil salinity;

[0023] Based on normalized interference value Contribution coefficient of spatiotemporal coupling interference The independent interference of a single environmental parameter on the sensor is calculated using the following expression:

[0024]

[0025] in, For the single-factor interference value of the i-th type parameter;

[0026] The cooperative interference generated by the interaction of two types of parameters is calculated based on the single-factor interference value, and the expression is as follows:

[0027]

[0028] in, , These represent different soil parameters. ; The coupling interference values ​​of the two types of soil parameters; This represents the single-factor interference value of the i-th type of parameter. This represents the single-factor interference value of the j-th type parameter. Indicates the parameter coupling coefficient; For spatial networking coefficients;

[0029] Based on the above single-factor interference values ​​and coupled interference values, an environmental interference factor matrix is ​​constructed. The expression is:

[0030]

[0031] Based on the detection time update, repeat the above steps to generate a new environmental interference factor matrix.

[0032] Furthermore, in the interference calibration module, a preset interference calibration model is invoked to dynamically adjust the detection parameters of each sensor unit according to the environmental interference factor matrix, and the output of the interference-free soil data set of each sensor is specifically as follows:

[0033] Based on the structural characteristics of the environmental interference factor matrix, an interference calibration model including a single error correction layer and a coupled error correction layer is constructed.

[0034] Obtain the environmental interference factor matrix The raw soil detection parameter set collected by the multi-parameter sensor unit is input into the interference calibration model, and preset calibration parameters are calibrated based on the hardware calibration of each sensor; the calibration parameters include calibration baseline coefficients. Parameter coupling coefficient Error feedback coefficient ;

[0035] The single-factor interference values ​​on the diagonal of the environmental interference factor matrix are mapped to a single error correction layer of the interference calibration model to obtain the single-corrected detection parameters, expressed as:

[0036]

[0037] in, This represents the value of the detection parameter after a single correction. This represents the original sensor reading of the i-th soil parameter. These are the single-factor interference values ​​on the diagonal of the environmental interference factor matrix. To calibrate the baseline coefficients, This is the error feedback coefficient of the previous detection node;

[0038] The off-diagonal coupling interference values ​​of the environmental interference factor matrix are mapped to the coupling error correction layer of the interference calibration model to obtain the coupling-corrected detection parameters, which characterize the final soil data after single-sensor interference cancellation. The expression is as follows:

[0039]

[0040] in, , This represents the detected parameter after coupling correction for the i-th soil parameter. This represents the value of the detection parameter after a single correction. Indicates the coupling interference value. Indicates the parameter coupling coefficient;

[0041] The final interference-free soil data obtained after coupling correction are integrated to obtain an interference-free soil data set.

[0042] Furthermore, the soil parameter extraction module is used to call a preset soil parameter extraction model based on the preprocessed soil data to perform spatial feature extraction and temporal feature extraction respectively, and output the fused soil parameter fusion value as follows:

[0043] Acquire preprocessed soil data Based on sensor spatial networking coefficients As the core weight, calculate The global spatial weighted mean of the soil parameters of type i at time i is expressed as:

[0044]

[0045] in, for Spatial characteristic values ​​of the i-th type of soil parameter at time i, For sensor nodes, This represents the total number of sensor nodes.

[0046] With time decay coefficient For time weighting, calculate The time-weighted difference eigenvalue of the i-th type of soil parameter at time i is expressed as:

[0047]

[0048]

[0049] in, , express The global mean of the soil parameter of type i at time i. express The global mean of the soil parameter of type i at time i; express The temporal characteristic value of the i-th type of soil parameter at time i;

[0050] Based on spatial eigenvalues and time eigenvalues Calculate the spatial variation coefficient separately and coefficient of variation over time The expression is:

[0051]

[0052]

[0053] in, Let be the spatial mean of the i-th type of soil parameter. The time characteristic mean of the i-th type of soil parameter;

[0054] Based on spatial variability coefficient and coefficient of variation over time Computational Spatial Fusion Weights Weighting with time The expression is:

[0055]

[0056]

[0057] in, ;

[0058] Based on spatial fusion weights Weighting with time We obtain the weighted sum of the spatial and temporal eigenvalues. The fused value of the soil parameter of type i at time i is expressed as:

[0059]

[0060] The values ​​of soil texture, soil temperature, and soil salinity are integrated to generate a set of soil parameter fusion values, which are then transmitted to the application layer.

[0061] Furthermore, the application layer includes a data receiving module, a data parsing module, a soil parameter template management module, and a soil data generation and rendering module, wherein:

[0062] The data receiving module receives the set of fused soil parameter values ​​transmitted by the data processing layer based on a standard interface;

[0063] The data parsing module is used to standardize and parse the received set of fused soil parameter values ​​to generate standardized fused soil parameter values.

[0064] The soil parameter template management module presets soil parameter templates based on different application scenarios and needs, and provides operations for modifying, adding and deleting soil parameter templates.

[0065] The soil data generation and rendering module is used to generate and visualize the fused values ​​of the parsed standardized soil parameters and the soil parameter template.

[0066] A high-precision soil parameter detection method using a multi-sensor network, applied to the aforementioned high-precision soil parameter detection device using a multi-sensor network, includes:

[0067] S1: Real-time collection of soil environmental parameters to construct an environmental disturbance factor matrix;

[0068] S2: Call the preset interference calibration model to dynamically adjust the detection parameters of each sensor unit according to the environmental interference factor matrix, and output the set of interference-free soil data of each sensor.

[0069] S3: Preprocess the undisturbed soil dataset to generate preprocessed soil data;

[0070] S4: Based on the preprocessed soil data, call the preset soil parameter extraction model to perform spatial feature extraction and temporal feature extraction respectively, and output the fused soil parameter fusion value;

[0071] S5: Receive the fused soil parameter values ​​and call the preset soil parameter template to generate and render soil data.

[0072] The principle and advantages of this invention are as follows: The multi-sensor networked high-precision soil parameter detection device of this invention is based on a four-layer architecture of perception layer, transmission layer, data processing layer and application layer. The perception layer collects soil texture, temperature, salinity and other parameters through multi-parameter integrated sensor modules. Combining the inherent detection parameters of the sensors and the spatiotemporal characteristic parameters of the network, it calculates single-factor interference values ​​and coupled interference values ​​through normalization, deviation analysis and other steps to construct an environmental interference factor matrix. Then, the interference calibration model dynamically adjusts the sensor detection parameters through a single error correction layer and a coupled error correction layer to output an interference-free soil data set. After the transmission layer completes the data transmission, the data processing layer first preprocesses the data, and then extracts the spatial and temporal feature values ​​of the soil parameters through a soil parameter extraction model, with spatial network coefficient and time decay coefficient as the core weights. It calculates the spatial and temporal variation coefficients and assigns fusion weights to complete the spatiotemporal feature weighted fusion to obtain the fused soil parameter value. Finally, the application layer matches the fused value with a preset soil parameter template to realize the scene generation and visualization rendering of soil data.

[0073] The advantages are as follows: This application effectively solves the problems of sensor accuracy being easily affected by the environment and the low intelligence and integration of multi-sensor network systems in traditional soil parameter detection. By constructing an environmental interference factor matrix and performing hierarchical error correction, it achieves accurate cancellation of single-factor environmental interference and multi-parameter coupling interference, significantly reducing the error of the detection data and improving the accuracy and stability of soil parameter detection. At the same time, by combining spatiotemporal feature extraction and dynamic weight fusion, it fully explores the spatial and temporal dimension value of multi-sensor network data, improves data processing and analysis capabilities, meets the needs of soil parameter monitoring in large-scale and complex environments, and provides high-precision and high-reliability soil data support for fields such as precision agricultural planting and ecological environmental protection, helping to achieve scientific decision-making in precision fertilization, rational irrigation, and soil ecological restoration. Attached Figure Description

[0074] Figure 1 This is a device architecture diagram according to an embodiment of the present invention;

[0075] Figure 2 This is a flowchart of an embodiment of the present invention. Detailed Implementation

[0076] The following detailed description illustrates the specific implementation method:

[0077] The basic implementation examples are as follows: Figure 1 As shown: A high-precision soil parameter detection device with multi-sensor networking, comprising a sensing layer, a transmission layer, a data processing layer, and an application layer, wherein:

[0078] The sensing layer includes a multi-parameter integrated sensor module and an interference calibration module. The multi-parameter integrated sensor module is used to construct an environmental interference factor matrix based on real-time acquisition of soil environmental parameters from multiple sensor units. Specifically, the multi-parameter integrated sensor module consists of multi-parameter sensor units and an environmental sensing unit, which work together to complete parameter acquisition and environmental interference factor matrix construction.

[0079] The multi-parameter sensor unit integrates an FDR humidity sensor, an electrochemical nutrient sensor, a glass electrode pH sensor, a platinum resistance temperature sensor, and a conductivity sensor, enabling real-time data acquisition from any sensor node within the monitoring area. During the detection time The original soil parameters include:

[0080] Soil texture parameters : Dimensionless parameter, values The data is collected and generated collaboratively by an FDR humidity sensor, an electrochemical nutrient sensor, and a glass electrode pH sensor. By separately acquiring core indicators of soil moisture, nutrient content, and pH, a comprehensive parameter characterizing soil texture is obtained. The larger the value, the higher the clay content and the stronger the physical interference to the sensor probe.

[0081] Soil temperature : Generated by a platinum resistance temperature sensor;

[0082] Soil salinity data : Generated by a conductivity sensor, which detects the conductivity value of soil fusion and converts it into specific data on soil salinity;

[0083] The raw data collected is then transmitted to the environmental sensing unit in real time.

[0084] Next, based on the hardware specifications of the multi-parameter sensor unit, the inherent detection parameters of the sensor and the spatiotemporal characteristic parameters of the sensor network are extracted. The inherent detection parameters of the sensor include:

[0085] No. Sensor detection sensitivity coefficient for soil parameters : , representing soil texture parameters, soil temperature, and soil salinity data, respectively; sensor detection sensitivity coefficient. Value The larger the value, the more sensitive the sensor is to environmental interference for this type of parameter;

[0086] Sensor basic detection error threshold Dimensionless, with values ; represents the inherent error of the sensor when there is no environmental interference, and serves as the benchmark value for interference quantification;

[0087] The spatiotemporal characteristic parameters of sensor networking include:

[0088] Time decay coefficient Dimensionless, with values This characterizes the attenuation characteristics of interference due to detection time, such as the interference attenuation caused by sensor thermal stability during continuous detection.

[0089] Spatial networking coefficient Dimensionless, with values This characterizes the spatial location interference differences of network nodes, for example, the interference coefficients of near-ground nodes and deep soil nodes are different;

[0090] Soil parameter background values It serves as a benchmark value for soil surveys in the detection area, characterizing the baseline state of the soil environment and used to calculate the actual environmental deviation of parameters.

[0091] Next, based on the sensor detection sensitivity coefficient and sensor basic detection error threshold For a single sensor node m in time The original parameters are normalized, and the expression is:

[0092]

[0093] in, The normalized interference value of the original sensor parameters satisfies The larger the value, the stronger the environmental interference detected by the sensor for this type of parameter; The original values ​​of the sensor for the i-th type of parameter are... , The extreme value of the sensor detection range for the i-th type of parameter; in this embodiment, the normalization method is an extreme value normalization algorithm designed to adapt to the sensor characteristics. By incorporating the sensor detection sensitivity coefficient and the basic error threshold into the normalization process, the differences in the dimensions and value ranges of the above three types of soil parameters are eliminated, while the normalization result directly reflects the actual degree of interference perception of each environmental parameter by the sensor. This provides basic data that fits the sensor hardware characteristics for subsequent interference quantification, and achieves accurate adaptation between interference quantification and sensor characteristics.

[0094] Then, based on the real-time detection value of the i-th type of parameter, determine its relationship with the soil parameter background value. The deviation is used to obtain the environmental deviation coefficient. The expression is:

[0095]

[0096] like This makes This ensures the continuity of deviation quantification;

[0097] Based on the detection principle of the multi-parameter sensor unit and the types of interference from soil parameters to sensor detection, such as physical interference, thermal interference, and electrochemical interference, the basic contribution coefficient is calibrated. , ; Characterizing soil texture parameters Characterizing soil temperature, Characterizing soil salinity; specifically:

[0098] Soil texture is the dominant physical disturbance, and it has the most significant impact on sensor probe contact and signal reflection, thus having the highest weight.

[0099] Soil temperature is affected by both thermal and ion activity interference, which impacts the performance of sensor electronic components and soil ion migration; therefore, it is given secondary weight.

[0100] Soil salinity is an electrochemical interference, mainly affecting the electrochemical detection of the sensor, and has a low weight.

[0101] Based on the above settings, the basic contribution coefficient is integrated. Environmental deviation coefficient Time decay coefficient Spatial networking coefficient Calculate the contribution coefficient of spatiotemporal coupling interference The expression is:

[0102]

[0103] in, ; Characterizing soil texture parameters Characterizing soil temperature, Characterizing soil salinity; In this embodiment, the design process of the weights incorporates the temporal and spatial characteristics of the multi-sensor network into the quantification of interference contribution, allowing the interference weights to change dynamically with detection time and network nodes, adapting to the large-scale and dynamic detection needs of multi-sensor networks.

[0104] Next, we performed hierarchical quantification of single factors and coupled interference values. In the traditional environmental interference quantification process, we only considered the independent interference of a single parameter and ignored the coupled interference effect between soil parameters. For example, increased temperature will accelerate the migration of soil ions, making the electrochemical interference of salt stronger; heavy texture will reduce the soil temperature conduction speed, making the thermal interference of temperature exhibit superposition characteristics, resulting in incomplete interference quantification.

[0105] Therefore, this step proposes an interference classification and quantification strategy, dividing environmental interference into single-factor independent interference and multi-parameter coupled interference, and performing quantification calculations separately. The single-factor interference value corresponds to the independent influence of a single environmental parameter on the sensor, while the coupled interference value corresponds to the synergistic influence of the interaction between two types of parameters. Both types of interference values ​​incorporate the spatiotemporal characteristics of the network, providing core elements for subsequent matrix construction, and are precisely adapted to the hierarchical correction logic of the subsequent calibration module. Specifically:

[0106] Based on normalized interference value Contribution coefficient of spatiotemporal coupling interference The independent interference of a single environmental parameter on the sensor is calculated using the following expression:

[0107]

[0108] in, For the single-factor interference value of the i-th type parameter; satisfying The larger the value, the stronger the independent interference of the parameter;

[0109] The cooperative interference generated by the interaction of two types of parameters is calculated based on the single-factor interference value, and the expression is as follows:

[0110]

[0111] in, , These represent different soil parameters. ; The coupling interference values ​​of the two types of soil parameters satisfy... ; This represents the single-factor interference value of the i-th type of parameter. This represents the single-factor interference value of the j-th type parameter. For spatial networking coefficients; The parameter coupling coefficient is a symmetric matrix. Based on soil environmental science theory and actual testing experiments, its specific value in this application is:

[0112] This indicates a strong coupling between soil texture and soil temperature;

[0113] This indicates a moderate coupling between soil texture and soil salinity.

[0114] This indicates that soil temperature and soil salinity are highly coupled.

[0115] Therefore, this step in this application realizes the hierarchical quantification of soil environmental disturbances, and integrates the spatiotemporal characteristics of networking into the quantification of coupled interference, which solves the technical pain points of ignoring parameter coupling effects and incomplete interference quantification in traditional technologies, and allows the interference factor matrix to fully characterize the dual characteristics of single interference and coupled interference in the soil environment.

[0116] Based on the above single-factor interference values ​​and coupled interference values, an environmental interference factor matrix is ​​constructed. The expression is:

[0117]

[0118] In this matrix, the diagonal elements are single-factor interference values, and the off-diagonal elements are coupled interference values. All elements of the matrix have spatiotemporal dynamic characteristics and are adapted to the correction dimension of the subsequent interference calibration module. The interference calibration module can directly parse the matrix elements to complete the error correction without additional dimension transformation or parameter mapping.

[0119] Based on the detection time update, the above steps are repeated to generate a new environmental interference factor matrix, thereby achieving real-time dynamic updating of the matrix.

[0120] The interference calibration module calls a preset interference calibration model to dynamically adjust the detection parameters of each sensor unit according to the environmental interference factor matrix, and outputs the interference-free soil data set of each sensor. In this embodiment, based on the structural characteristics of the environmental interference factor matrix, the interference calibration module proposes a matrix element hierarchical mapping calibration model, mapping the single-factor interference elements and coupled interference elements of the matrix to the single error correction layer and coupled error correction layer of the sensor detection parameters, respectively, to achieve hierarchical progressive dynamic calibration. At the same time, a calibration compensation coefficient is introduced to achieve closed-loop optimization of calibration based on the error feedback value detected by the sensor, so that the calibration accuracy continuously improves with the detection process. Moreover, the calibration process is entirely based on the above-mentioned environmental interference factor matrix, and the matrix provides a unique interference quantification input for calibration, forming a closed-loop adaptation of construction-calibration, specifically:

[0121] Based on the structural characteristics of the environmental interference factor matrix, an interference calibration model including a single error correction layer and a coupled error correction layer is constructed.

[0122] Obtain the environmental interference factor matrix The raw soil detection parameter set collected by the multi-parameter sensor unit is input into the interference calibration model, and preset calibration parameters are calibrated based on the hardware of each sensor; the calibration parameters include:

[0123] Calibration base coefficient Dimensionless, with values , which is the calibration correction coefficient for the sensor;

[0124] Parameter coupling coefficient Coupling coefficient with the aforementioned parameters Consistent;

[0125] Error feedback coefficient Dimensionless, with values It is calculated from the calibration error feedback value of the previous detection node.

[0126] The single-factor interference values ​​on the diagonal of the environmental interference factor matrix are mapped to a single error correction layer of the interference calibration model to obtain the single-corrected detection parameters, expressed as:

[0127]

[0128] in, This represents the value of the detection parameter after a single correction. This represents the original sensor reading of the i-th soil parameter. These are the single-factor interference values ​​on the diagonal of the environmental interference factor matrix. To calibrate the baseline coefficients, This is the error feedback coefficient of the previous detection node; The larger the value, the greater the magnitude of a single correction, accurately offsetting the independent interference of a single parameter;

[0129] The off-diagonal coupling interference values ​​of the environmental interference factor matrix are mapped to the coupling error correction layer of the interference calibration model to obtain the coupling-corrected detection parameters, which characterize the final soil data after single-sensor interference cancellation. The expression is as follows:

[0130]

[0131] in, , This represents the detected parameter after coupling correction for the i-th soil parameter. This represents the value of the detection parameter after a single correction. Indicates the coupling interference value. Indicates the parameter coupling coefficient;

[0132] Taking soil texture parameters as an example, their coupled and corrected expression is:

[0133]

[0134] Similarly, the coupled correction formula for soil temperature and soil salinity is:

[0135]

[0136]

[0137] Finally, the interference-free soil data obtained after coupling correction are integrated to obtain an interference-free soil data set.

[0138] The transmission layer is used to transmit the interference-free soil data set generated by the sensing layer to the data processing layer. In this embodiment, the transmission layer adopts a wired + wireless dual-mode networking transmission architecture. The wired transmission unit uses Industrial Ethernet (TCP / IP protocol) to realize short-distance, high-bandwidth fixed node data transmission, while the wireless transmission unit uses LoRa + NB-IoT dual-mode wireless communication protocol to realize long-distance, distributed mobile node data transmission. It is compatible with the distributed data acquisition characteristics of multi-sensor nodes, and realizes low-latency, high-reliability, and anti-interference transmission of data. This ensures that the interference-free soil data set generated by the sensing layer is transmitted to the data processing layer completely, accurately, and in real time, providing a high-quality data foundation for subsequent soil parameter extraction and fusion.

[0139] The data processing layer includes a preprocessing module and a soil parameter extraction module. The preprocessing module preprocesses the received interference-free soil data set to generate preprocessed soil data. The soil parameter extraction module uses a preset soil parameter extraction model to extract spatial and temporal features based on the preprocessed soil data, outputting fused soil parameter values. The preprocessing includes outlier identification, interpolation completion, and normalization. Specifically, outlier identification uses a 3D model... The criteria identify outliers in soil parameters. After identification, time interpolation terms are obtained based on cubic spline interpolation, and spatial interpolation terms are obtained based on kriging interpolation. The outlier data is then interpolated and completed. Finally, normalization is performed to obtain preprocessed soil data. ;

[0140] Acquire preprocessed soil data Based on sensor spatial networking coefficients As the core weight, calculate The global spatial weighted mean of the soil parameters of type i at time i is expressed as:

[0141]

[0142] in, for Spatial characteristic values ​​of the i-th type of soil parameter at time i, For sensor nodes, This represents the total number of sensor nodes.

[0143] With time decay coefficient For time weighting, calculate The time-weighted difference eigenvalue of the i-th type of soil parameter at time i is expressed as:

[0144]

[0145]

[0146] in, , express The global mean of the soil parameter of type i at time i. express The global mean of the soil parameter of type i at time i; express The temporal characteristic value of the i-th type of soil parameter at time i;

[0147] Based on spatial eigenvalues and time eigenvalues Calculate the spatial variation coefficient separately and coefficient of variation over time The expression is:

[0148]

[0149]

[0150] in, Let be the spatial mean of the i-th type of soil parameter. The time characteristic mean of the i-th type of soil parameter;

[0151] Based on spatial variability coefficient and coefficient of variation over time Computational Spatial Fusion Weights Weighting with time The expression is:

[0152]

[0153]

[0154] in, ;

[0155] Based on spatial fusion weights Weighting with time We obtain the weighted sum of the spatial and temporal eigenvalues. The fused value of the soil parameter of type i at time i is expressed as:

[0156]

[0157] The values ​​of soil texture, soil temperature, and soil salinity are integrated to generate a set of soil parameter fusion values, which are then transmitted to the application layer.

[0158] The soil parameter extraction module, as the core module of the data processing layer of this invention, overcomes the technical pain points of spatiotemporal feature separation, fixed weights, and low data utilization in traditional multi-sensor network data processing. Through the innovative design of "spatial differential extraction + temporal dynamic extraction + dynamic weight fusion driven by the coefficient of variation", it achieves in-depth mining and accurate fusion of the spatiotemporal features of soil parameters. Its core innovative effects are reflected in the following four aspects:

[0159] 1. Differentiated spatial feature extraction enhances the representativeness of regional data. It abandons the traditional method of "equal-weighted mean calculation" in soil parameter spatial extraction, and instead uses the spatial networking coefficients calibrated by the sensing layer. By incorporating spatial feature value calculation, the detection data of core monitoring areas and low-interference nodes are given higher weight, while the impact of abnormal data from edge areas and high-interference nodes is weakened. This makes the extracted spatial feature values ​​more consistent with the actual soil conditions of the monitoring area, thereby improving the representativeness and accuracy of spatial dimension data.

[0160] 2. Dynamic extraction of time features, fitting the temporal variation law of parameters through time decay coefficient. A time-weighted difference model is constructed, which combines the mean of parameters at the current time with the difference of parameters at adjacent times. This not only retains the core value of the current detection data, but also incorporates the temporal trend of parameter changes, adapting to the characteristic of soil parameters changing slowly over time. This solves the problem of "only focusing on data at a single time and ignoring temporal correlation" in traditional time feature extraction, making time feature values ​​more reflective of the dynamic change pattern of soil parameters.

[0161] 3. Adaptive dynamic allocation of fusion weights enhances the robustness of fusion results. Based on spatial and temporal coefficients of variation, the fusion weights are dynamically calculated, breaking the limitations of traditional "fixed weight allocation" in data fusion. This allows the weights to be adjusted in real time according to the actual fluctuations of soil parameters. When a parameter in a certain dimension fluctuates significantly, its fusion weight is automatically reduced to prevent abnormal data from dominating the fusion result; when the parameter fluctuation is small, its weight is automatically increased to fully utilize the reference value of that dimension's data, significantly improving the robustness and anti-interference ability of the fused soil parameter values.

[0162] 4. Deep Fusion of Spatiotemporal Features: Fully Leveraging the Multidimensional Value of Networked Data. This module integrates the spatial and temporal characteristics of soil parameters, enabling the dual mining of spatial differences and temporal dynamics in multi-sensor network data. This solves the problems of separate spatiotemporal feature processing and low data utilization in traditional technologies. It combines the spatial coverage advantages of distributed sensor nodes with the temporal series advantages of continuous detection, ensuring that the final output of fused soil parameter values ​​possesses both regional representativeness and temporal continuity. This significantly improves data processing and analysis capabilities, meeting the high-precision monitoring needs of soil parameters in large-scale, complex environments. Overall, the implementation of this module upgrades multi-sensor networked soil data from "simple summarization in a single dimension" to "precise fusion in multiple spatiotemporal dimensions." The output fused values ​​are more closely aligned with actual soil conditions compared to traditional processing methods, providing higher-precision and more valuable soil parameter data support for application-layer scenario generation, visualization rendering, and subsequent scientific decision-making in precision agriculture and ecological restoration.

[0163] The application layer receives the fused soil parameter values ​​and calls a preset soil parameter template to generate and render soil data. The application layer includes a data receiving module, a data parsing module, a soil parameter template management module, and a soil data generation and rendering module.

[0164] The data receiving module receives the set of fused soil parameter values ​​transmitted by the data processing layer based on a standard interface. In this embodiment, the data receiving module receives the set of fused soil parameter values ​​transmitted by the data processing layer through a standardized data interface (such as TCP / IP, MQTT, HTTP / HTTPS). The set includes the fused values ​​of three core parameters: soil texture, soil temperature, and soil salinity, as well as auxiliary information such as the spatial variation coefficient, temporal variation coefficient, fusion weight, detection time, sensor node number, and data reliability of each parameter. It supports the reception of batch data and real-time single data, and has the ability to resume data transmission after interruption to avoid data loss.

[0165] The data parsing module is used to standardize and parse the received set of fused soil parameter values ​​to generate standardized fused soil parameter values. In this embodiment, the data parsing module has a built-in dedicated data parsing engine to standardize and parse the received fused value data, convert the encrypted binary / structured data into readable numerical and character data, extract core parameters and auxiliary information and perform data format verification (such as data range, data type, and integrity verification), mark abnormal data (such as values ​​exceeding a reasonable range or missing information) and trigger an early warning, and simultaneously feed the abnormal data back to the data processing layer for secondary verification.

[0166] The soil parameter template management module presets soil parameter templates based on different application scenarios and needs, and provides operations for modifying, adding, and deleting soil parameter templates. In this embodiment, the soil parameter templates include data display templates, data analysis templates, data output templates, and soil quality evaluation templates, wherein:

[0167] Data display templates are designed according to display dimensions (single parameter / multiple parameters, single time point / time series, single region / spatial distribution) and display carriers (large screen / terminal / mobile), defining attributes such as parameter display location, numerical precision, color coding, and units;

[0168] Data analysis template: integrates basic statistical analysis (mean, variance, trend change) and professional soil analysis (parameter correlation, soil fertility level, soil environmental suitability) algorithms, and defines analysis dimensions and output result formats;

[0169] Data output template: Designed according to output format (report, data file) and output purpose (field guidance, laboratory analysis, platform reporting), defining data output items, layout format, signature position, etc.;

[0170] Soil quality assessment template: Based on national / industry standards, it presets evaluation indicators, evaluation thresholds and grading standards for different soil types (arable land, forest land, grassland, construction land) and different uses (agricultural planting, ecological restoration, geological exploration), and automatically determines the quality grade of the fusion value.

[0171] The soil data generation and rendering module is used to generate and visualize standardized soil parameter fusion values ​​and soil parameter templates in a scenario-based manner. Scenario-based generation is based on a called template, filling the fusion values ​​and associated information into designated positions within the template. Data calculation, analysis, and evaluation are performed according to the rules defined in the template. For example, under the soil quality evaluation template, the system automatically compares the fusion values ​​with evaluation thresholds to determine the soil fertility level / pollution risk level / environmental suitability level and generates corresponding evaluation conclusions. Under the time series analysis template, the system automatically calculates trend change rates, extreme points, and other analytical data for parameters at multiple time points. It also generates structured data results and semi-structured report results: structured data results are used for internal system analysis and integration with third-party platforms; semi-structured report results include numerical values, charts, and textual evaluations for end-user viewing and output.

[0172] The specific rendering formats for visualization are as follows: 1. Numerical rendering: Displaying core parameters according to the precision, unit, and color coding (e.g., black for normal values, yellow for warning values, and red for exceeding standards) defined by the template; 2. Chart rendering: Automatically generating line charts (time trends), bar charts (parameter comparisons), pie charts (soil texture composition), heat maps (spatial distribution), etc., supporting chart scaling, export, and editing; 3. Spatial map rendering: Associating the spatial location of sensor nodes with the corresponding parameter fusion values, and visually marking them on electronic maps / field maps to intuitively display the spatial distribution of soil parameters; 4. Dynamic trend rendering: Dynamically refreshing the parameters collected in real time to generate dynamic trend curves, supporting scaling in the time dimension (hours / days / weeks / months).

[0173] like Figure 2 As shown, in another embodiment of this example, a high-precision soil parameter detection method based on a multi-sensor network is further included, applied to the aforementioned high-precision soil parameter detection device based on a multi-sensor network, comprising:

[0174] S1: Real-time collection of soil environmental parameters to construct an environmental disturbance factor matrix;

[0175] S2: Call the preset interference calibration model to dynamically adjust the detection parameters of each sensor unit according to the environmental interference factor matrix, and output the set of interference-free soil data of each sensor.

[0176] S3: Preprocess the undisturbed soil dataset to generate preprocessed soil data;

[0177] S4: Based on the preprocessed soil data, call the preset soil parameter extraction model to perform spatial feature extraction and temporal feature extraction respectively, and output the fused soil parameter fusion value;

[0178] S5: Receive the fused soil parameter values ​​and call the preset soil parameter template to generate and render soil data.

[0179] The above are merely embodiments of the present invention. Commonly known structures and characteristics are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can, under the guidance of this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A high-precision soil parameter detection device using a multi-sensor network, characterized in that: It includes the perception layer, transmission layer, data processing layer, and application layer, among which: The perception layer includes a multi-parameter integrated sensor module and an interference calibration module. The multi-parameter integrated sensor module is used to construct an environmental interference factor matrix based on real-time acquisition of soil environmental parameters by multiple sensor units. The interference calibration module calls a preset interference calibration model to dynamically adjust the detection parameters of each sensor unit according to the environmental interference factor matrix and outputs a set of interference-free soil data from each sensor. The transport layer is used to transmit the interference-free soil data set generated by the sensing layer to the data processing layer; The data processing layer includes a preprocessing module and a soil parameter extraction module. The preprocessing module is used to preprocess the received set of interference-free soil data to generate preprocessed soil data. The soil parameter extraction module is used to call a preset soil parameter extraction model to perform spatial feature extraction and temporal feature extraction based on the preprocessed soil data, and output the fused soil parameter fusion value. The application layer is used to receive the fused soil parameter values ​​and call the preset soil parameter template to generate and render soil data. In the perception layer, the multi-parameter integrated sensor module includes a multi-parameter sensor unit and an environmental perception unit. The multi-parameter sensor unit is used to collect soil texture parameters, soil temperature, and soil salinity data. The environmental sensing unit is used to acquire soil texture parameters, soil temperature, and soil salinity data, and to construct an environmental disturbance factor matrix, specifically: Obtain any sensor node of the multi-parameter sensor unit During the detection time Soil texture parameters Soil temperature Soil salinity data ; Based on the hardware specifications of the multi-parameter sensor unit, inherent sensor detection parameters and spatiotemporal characteristic parameters of the sensor network are extracted. The inherent sensor detection parameters include the... Sensor detection sensitivity coefficient for soil parameters Sensor basic detection error threshold The spatiotemporal characteristic parameters of sensor networking include the time decay coefficient. Spatial networking coefficient Soil parameter background values ; Based on sensor detection sensitivity coefficient and sensor basic detection error threshold For a single sensor node m in time The original parameters are normalized, and the expression is: in, The normalized interference value of the original sensor parameters. The original values ​​of the sensor for the i-th type of parameter are... , Let be the extreme value of the sensor's detection range for the i-th type of parameter; Based on the real-time detection value of the i-th type of parameter, determine its relationship with the soil parameter background value. The deviation is used to obtain the environmental deviation coefficient. Based on the detection principle of the multi-parameter sensor unit and the types of interference from soil parameters to sensor detection, the foundation contribution coefficient was calibrated. ; Based on the basic contribution coefficient Environmental deviation coefficient Time decay coefficient Spatial networking coefficient Calculate the contribution coefficient of spatiotemporal coupling interference The expression is: in, ; Characterizing soil texture parameters Characterizing soil temperature, Characterizing soil salinity; Based on normalized interference value Contribution coefficient of spatiotemporal coupling interference The independent interference of a single environmental parameter on the sensor is calculated using the following expression: in, For the single-factor interference value of the i-th type parameter; The cooperative interference generated by the interaction of two types of parameters is calculated based on the single-factor interference value, and the expression is as follows: in, , These represent different soil parameters. ; The coupling interference values ​​of the two types of soil parameters; This represents the single-factor interference value of the i-th type of parameter. This represents the single-factor interference value of the j-th type parameter. Indicates the parameter coupling coefficient; For spatial networking coefficients; Based on the above single-factor interference values ​​and coupled interference values, an environmental interference factor matrix is ​​constructed. The expression is: Based on the detection time update, repeat the above steps to generate a new environmental interference factor matrix.

2. The high-precision soil parameter detection device with multi-sensor networking according to claim 1, characterized in that: In the interference calibration module, a preset interference calibration model is invoked to dynamically adjust the detection parameters of each sensor unit according to the environmental interference factor matrix, and the output of the interference-free soil data set of each sensor is as follows: Based on the structural characteristics of the environmental interference factor matrix, an interference calibration model including a single error correction layer and a coupled error correction layer is constructed. Obtain the environmental interference factor matrix The raw soil detection parameter set collected by the multi-parameter sensor unit is input into the interference calibration model, and preset calibration parameters are calibrated based on the hardware calibration of each sensor; the calibration parameters include calibration baseline coefficients. Parameter coupling coefficient Error feedback coefficient ; The single-factor interference values ​​on the diagonal of the environmental interference factor matrix are mapped to a single error correction layer of the interference calibration model to obtain the single-corrected detection parameters, expressed as: in, This represents the value of the detection parameter after a single correction. This represents the original sensor reading of the i-th soil parameter. These are the single-factor interference values ​​on the diagonal of the environmental interference factor matrix. To calibrate the baseline coefficients, This is the error feedback coefficient of the previous detection node; The off-diagonal coupling interference values ​​of the environmental interference factor matrix are mapped to the coupling error correction layer of the interference calibration model to obtain the coupling-corrected detection parameters, which characterize the final soil data after single-sensor interference cancellation. The expression is as follows: in, , This represents the detected parameter after coupling correction for the i-th soil parameter. This represents the value of the detection parameter after a single correction. Indicates the coupling interference value. Indicates the parameter coupling coefficient; The final interference-free soil data obtained after coupling correction are integrated to obtain an interference-free soil data set.

3. The high-precision soil parameter detection device with multi-sensor networking according to claim 2, characterized in that: The soil parameter extraction module is used to call a preset soil parameter extraction model based on the preprocessed soil data to perform spatial feature extraction and temporal feature extraction respectively, and output the fused soil parameter fusion value as follows: Acquire preprocessed soil data Based on sensor spatial networking coefficients As the core weight, calculate The global spatial weighted mean of the soil parameters of type i at time i is expressed as: in, for Spatial characteristic values ​​of the i-th type of soil parameter at time i, For sensor nodes, This represents the total number of sensor nodes. With time decay coefficient For time weighting, calculate The time-weighted difference eigenvalue of the i-th type of soil parameter at time i is expressed as: in, , express The global mean of the soil parameter of type i at time i. express The global mean of the soil parameter of type i at time i; express The temporal characteristic value of the i-th type of soil parameter at time i; Based on spatial eigenvalues and time eigenvalues Calculate the spatial variation coefficient separately and coefficient of variation over time The expression is: in, Let be the spatial mean of the i-th type of soil parameter. The time characteristic mean of the i-th type of soil parameter; Based on spatial variability coefficient and coefficient of variation over time Computational Spatial Fusion Weights Weighting with time The expression is: in, ; Based on spatial fusion weights Weighting with time We obtain the weighted sum of the spatial and temporal eigenvalues. The fused value of the soil parameter of type i at time i is expressed as: The values ​​of soil texture, soil temperature, and soil salinity are integrated to generate a set of soil parameter fusion values, which are then transmitted to the application layer.

4. The high-precision soil parameter detection device with multi-sensor networking according to claim 3, characterized in that: The application layer includes a data receiving module, a data parsing module, a soil parameter template management module, and a soil data generation and rendering module, wherein: The data receiving module receives the set of fused soil parameter values ​​transmitted by the data processing layer based on a standard interface; The data parsing module is used to standardize and parse the received set of fused soil parameter values ​​to generate standardized fused soil parameter values. The soil parameter template management module presets soil parameter templates based on different application scenarios and needs, and provides operations for modifying, adding and deleting soil parameter templates. The soil data generation and rendering module is used to generate and visualize the fused values ​​of the parsed standardized soil parameters and the soil parameter template.

5. A method for high-precision detection of soil parameters using a multi-sensor network, applied to the high-precision soil parameter detection device using a multi-sensor network as described in any one of claims 1-4, characterized in that: include: S1: Real-time collection of soil environmental parameters to construct an environmental disturbance factor matrix; S2: Call the preset interference calibration model to dynamically adjust the detection parameters of each sensor unit according to the environmental interference factor matrix, and output the set of interference-free soil data of each sensor. S3: Preprocess the undisturbed soil dataset to generate preprocessed soil data; S4: Based on the preprocessed soil data, call the preset soil parameter extraction model to perform spatial feature extraction and temporal feature extraction respectively, and output the fused soil parameter fusion value; S5: Receive the fused soil parameter values ​​and call the preset soil parameter template to generate and render soil data.

Images