Agricultural environment monitoring method and system based on Internet of Things

Abstract

The invention provides an agricultural environment monitoring method and system based on the Internet of Things. According to the method, multi-source data weights are dynamically distributed through a soil sensor array connected with Internet of Things nodes, and a global soil parameter set is generated; deploying a micro-electro-mechanical micro-fluidic chip in the coverage area to carry out soil solution selective permeation, converting the target ion concentration into an electric signal, and synchronously transmitting the electric signal and soil parameters; historical time series data are extracted, coherence is established through time correlation analysis, and undeployed area data are filled in combination with a spatial interpolation algorithm to construct a dynamic prediction model; according to the nutrient space change trend output by the model, crop growth requirements are matched to generate a fertilization amount adjustment instruction, and the fertilization amount adjustment instruction is issued to field fertilization equipment through the Internet of Things to execute dynamic regulation. Space-time precise sensing of soil nutrients and self-adaptive fertilization regulation and control are realized, and the utilization efficiency of agricultural resources and crop growth sustainability are improved.

Description

Technical Field

[0001] The present application relates to the technical field of agricultural environment monitoring, and in particular to an agricultural environment monitoring method and system based on the Internet of Things. Background Art

[0002] Modern agricultural integrated water and fertilizer management systems must achieve dynamic and precise control in complex farmland environments. This requires real-time monitoring of multi-dimensional data such as soil moisture, nutrient concentration, and meteorological parameters, and dynamic adjustment of water-fertilizer ratios and irrigation strategies based on crop growth models. Especially in large-scale cropping scenarios, these systems must overcome challenges such as soil heterogeneity, sudden climate change, and widespread equipment distribution. IoT technology enables global perception, coordinated control of edge computing, and execution terminals to ensure efficient use of water and fertilizer resources and optimal alignment with crop growth.

[0003] The current mainstream solution utilizes IoT-based intelligent monitoring and dynamic control technologies. These include obtaining macro-soil soil moisture distribution in farmland through drone remote sensing and satellite imagery; using deep learning models to predict crop water and fertilizer requirements and construct dynamic irrigation decision-making algorithms; and using hybrid networking technology to transmit data to a cloud-based decision-making platform, generating and distributing zoned water and fertilizer ratio instructions to edge nodes, driving intelligent valves and variable-rate fertilizer spreaders for differentiated control. This system supports remote monitoring and multi-site collaborative management, significantly improving water and fertilizer utilization efficiency. Summary of the Invention

[0004] The present application provides an agricultural environment monitoring method and system based on the Internet of Things, which is used to solve the problems of scattered agricultural environment monitoring data, delayed regulation and low utilization rate of fertilization resources in the existing technology.

[0005] In a first aspect, the present application provides an agricultural environment monitoring method based on the Internet of Things, comprising:

[0006] In an agricultural environmental monitoring area, a soil sensor array connected by an IoT node collects multi-source data from the soil, dynamically assigns weights to the measurement values ​​of soil sensors at different locations in the multi-source data, and generates a globally consistent set of soil parameters based on the measurement values ​​after dynamic weight assignment.

[0007] In the coverage area of ​​the soil sensor array, a microfluidic detection chip of a micro-electromechanical system is deployed based on the topological structure of the Internet of Things, and the collected soil solution is selectively infiltrated through the microfluidic detection chip;

[0008] Obtaining a target ion concentration in the infiltrated soil solution, converting the target ion concentration into an electrical signal, and synchronously transmitting the electrical signal and the soil parameter set using a low-power communication protocol of the Internet of Things;

[0009] extracting historical time series data in the set of soil parameters, and performing time correlation analysis on the electrical signal and the historical time series data to establish continuity in the time dimension, and filling data for areas where soil sensors are not deployed by combining a spatial interpolation algorithm, and generating a dynamic prediction model using the continuity and the data filling result;

[0010] outputting a spatial variation trend of soil nutrients according to the dynamic prediction model, and matching the spatial variation trend with preset crop growth demand parameters to generate fertilizer amount adjustment instructions for different soil regions, and issuing the fertilizer amount adjustment instructions to a fertilizer device execution end deployed in the field through an Internet of Things communication link to complete dynamic regulation of soil nutrients.

[0011] Optionally, extracting historical time series data in the set of soil parameters, and performing time correlation analysis on the electrical signal and the historical time series data to establish continuity in the time dimension, and filling data for areas where soil sensors are not deployed by combining a spatial interpolation algorithm, and generating a dynamic prediction model using the continuity and the data filling result, comprising:

[0012] extracting data recorded at different time points from the set of soil parameters to form historical time series data, and matching the electrical signal according to collection time with data at corresponding time points in the historical time series data to calculate fluctuation correlation degree of the electrical signal and the historical time series data at the same time point;

[0013] determining variation rule consistency of the electrical signal and the historical time series data in the time dimension according to the fluctuation correlation degree, and supplementing historical time series data of missing time points based on the variation rule consistency, so that historical time series data of all time points form a continuous time series;

[0014] calculating data difference proportion of areas where soil sensors are not deployed and adjacent deployed areas by a spatial interpolation algorithm, and filling data for areas where soil sensors are not deployed according to the data difference proportion;

[0015] superimposing the continuity result of the time series and the data filling result of the non-deployed area, and constructing a dynamic prediction model according to the superimposed result.

[0016] Optionally, a soil sensor array connected through an Internet of Things node collects multi-source data in soil, and dynamically allocates weights to measurement values of soil sensors at different positions in the multi-source data, and generates a globally consistent set of soil parameters according to the measurement values after dynamic weight allocation, comprising:

[0017] In the agricultural environment detection area, the soil sensor array connected by the Internet of Things nodes continuously collects the measurement values of the positions where each soil sensor is located, and defines the measurement values of multiple soil sensors collected at the same time point as multi-source data;

[0018] According to the position distribution of each soil sensor in the array, the adjacent coverage relationship between soil sensors at different positions is determined, which is characterized by the physical distance between adjacent soil sensors and the consistency of the change trend of the measurement values;

[0019] Based on the adjacent coverage relationship, an initial weight is assigned to the measurement value of each soil sensor, and the measurement value difference proportion of adjacent soil sensors is calculated;

[0020] The initial weight is dynamically adjusted according to the measurement value difference proportion, so that the measurement value difference proportion is negatively correlated with the adjustment amplitude of the corresponding weight of adjacent soil sensors;

[0021] The dynamically adjusted weight is weighted and integrated with the measurement value of the corresponding soil sensor to eliminate the measurement deviation caused by uneven distribution of soil sensors, and a globally consistent soil parameter set covering the entire agricultural environment detection area is generated according to the weighted integration result.

[0022] Optionally, within the coverage area of the soil sensor array, a microfluidic detection chip of a micro-electro-mechanical system is deployed based on the topology of the Internet of Things, and the soil solution is selectively permeated through the microfluidic detection chip, including:

[0023] According to the topology of the Internet of Things nodes in the coverage area of the soil sensor array, the deployment position of the microfluidic detection chip is determined, so that each microfluidic detection chip forms a direct communication link with an Internet of Things node, and the spacing between the chips is adjusted based on the signal strength of the communication link to ensure the signal stability of the communication link;

[0024] A multi-layer permeation membrane is constructed inside the microfluidic detection chip, and the target ions in the collected soil solution are intercepted through the physical structure difference of the different levels of permeation membranes, and the physical structure difference of the permeation membrane refers to the pore size of the permeation membrane;

[0025] According to the matching relationship between the size of the target ion and the pore of the permeation membrane, the target ion is intercepted at the specified level, and the non-target ion is discharged through other levels to realize selective permeation of the soil solution.

[0026] Optionally, the concentration of the target ion in the permeated soil solution is obtained, and the target ion concentration is converted into an electrical signal, and the electrical signal and the soil parameter set are synchronously transmitted using the low-power communication protocol of the Internet of Things, including:

[0027] An electrochemical sensing unit is arranged at the outlet of the penetration channel of the microfluidic detection chip, target ions in the penetrated soil solution are captured by an ion-selective membrane in the electrochemical sensing unit, and a concentration sensing signal is generated based on the accumulation amount of the target ions on the surface of the ion-selective membrane;

[0028] The concentration sensing signal is converted into a corresponding electrical signal based on a conversion circuit built in the electrochemical sensing unit, and the amplitude of the electrical signal changes linearly with the increase of the concentration;

[0029] The electrical signal is time-stamped bound to a set of soil parameters at the same time point through a low-power communication protocol of an Internet of Things node, to realize synchronous transmission of the electrical signal and the set of soil parameters.

[0030] Optionally, the spatial variation trend of soil nutrients is output according to the dynamic prediction model, and the spatial variation trend is matched with a preset crop growth demand parameter to generate a fertilizer application amount adjustment instruction for different soil regions, and the fertilizer application amount adjustment instruction is issued to a fertilizer application device execution end deployed in the field through an Internet of Things communication link to complete dynamic regulation of soil nutrients, including:

[0031] Distribution data of soil nutrients at different geographic locations are extracted from the dynamic prediction model to calculate the spatial variation trend of soil nutrients;

[0032] The data of soil nutrients at each geographic location in the spatial variation trend are compared with the preset crop growth demand parameter point by point to calculate a soil nutrient difference ratio, and a fertilizer priority level of the corresponding region is determined according to the soil nutrient difference ratio;

[0033] Based on the fertilizer priority level, an adjustment instruction containing a fertilizer type and a fertilizer amount is generated for each geographic location, the fertilizer amount in the adjustment instruction is proportional to the soil nutrient difference ratio, and the fertilizer type is determined as supplement or reduction according to the positive or negative direction of the soil nutrient difference ratio;

[0034] The adjustment instruction is distributed to the corresponding fertilizer application device execution end according to the geographic location coordinate information through an Internet of Things communication link to complete dynamic regulation of soil nutrients.

[0035] Optionally, based on the fertilizer priority level, an adjustment instruction containing a fertilizer type and a fertilizer amount is generated for each geographic location, the fertilizer amount in the adjustment instruction is proportional to the soil nutrient difference ratio, and the fertilizer type is determined as supplement or reduction according to the positive or negative direction of the soil nutrient difference ratio, including:

[0036] adjust the amount of fertilization according to the difference proportion of soil nutrients in each geographical location in the fertilization priority, the amount of fertilization being proportional to the difference proportion of soil nutrients;

[0037] determine the positive and negative directions of the difference proportion, compare the positive and negative directions with the nutrient threshold in the preset crop growth demand parameter, and assign a fertilization type to each geographical location, the positive direction in the positive and negative directions indicating that the current nutrient is lower than the nutrient threshold and the corresponding complementary type of fertilizer needs to be applied, and the negative direction indicating that the current nutrient is higher than the nutrient threshold and the corresponding reducing type of fertilizer needs to be applied;

[0038] integrate the fertilization type and the amount of fertilization, and generate an adjustment instruction according to the integration result.

[0039] In a second aspect, the present application provides an agricultural environment monitoring system based on Internet of Things, comprising:

[0040] The distribution module is configured to collect multi-source data in the soil in the agricultural environment detection area through the soil sensor array connected by the Internet of Things nodes, dynamically allocate weights to the measurement values of the soil sensors at different positions in the multi-source data, and generate a globally consistent soil parameter set according to the measurement values after dynamic weight allocation;

[0041] The deployment module is configured to deploy a micro-fluidic detection chip of a micro-electro-mechanical system in the coverage area of the soil sensor array based on the topology of the Internet of Things, and perform selective permeation on the collected soil solution through the micro-fluidic detection chip;

[0042] The conversion module is configured to obtain the target ion concentration in the permeated soil solution, convert the target ion concentration into an electrical signal, and simultaneously transmit the electrical signal and the soil parameter set using a low-power communication protocol of the Internet of Things;

[0043] The generation module is configured to extract historical time series data in the soil parameter set, perform time correlation analysis on the electrical signal and the historical time series data to establish the continuity in the time dimension, perform data filling on the area without deployed soil sensors using a spatial interpolation algorithm, and generate a dynamic prediction model using the continuity and the data filling result;

[0044] The matching module is configured to output the spatial variation trend of soil nutrients according to the dynamic prediction model, match the spatial variation trend with the preset crop growth demand parameter, generate a fertilization amount adjustment instruction for different soil areas, and issue the fertilization amount adjustment instruction to the fertilization equipment execution end deployed in the field through the Internet of Things communication link to complete the dynamic regulation and control of soil nutrients.

[0045] In a third aspect, the embodiments of the present application provide a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component, and implement the method for monitoring an agricultural environment based on Internet of Things as described in the first aspect.

[0046] In a fourth aspect, the embodiments of the present application provide a computer storage medium, which stores a computer program; when the computer program is executed by a computer, the method for monitoring an agricultural environment based on Internet of Things as described in the first aspect is implemented.

[0047] In the technical solution of the present application, a smart agricultural closed-loop system is constructed, and the whole-process intelligentization of soil nutrient management is realized. Through multi-source sensor data fusion and microfluidic chip detection technology, the system obtains high-precision soil multi-parameter information; combined with time-space data modeling and intelligent prediction algorithm, the dynamic change law of nutrients is accurately mastered; finally, a differentiated precision fertilization scheme is generated based on crop demand, and automatic regulation and control is realized through Internet of Things. This multi-technology collaborative innovation mode breaks through the limitations of traditional extensive agricultural management, improves crop yield, significantly reduces resource waste and environmental pollution risk, and provides a complete technical solution for the precision and intelligent development of modern agriculture.

[0048] Further, through the time-space data fusion technology, a high-precision agricultural environment dynamic prediction system is constructed. The system establishes a coherent sequence of historical data based on time correlation analysis, effectively filling in the blanks of the monitoring period; combined with the spatial interpolation algorithm, the discrete monitoring points are expanded to continuous regional data, solving the problem of insufficient sensor coverage. Through the analysis of the time dimension and the intelligent inference of the space dimension, the time-space integrity reconstruction of the monitoring data is realized. This double compensation mechanism not only retains the accuracy of local monitoring, but also has the reliability of global inference, providing a time-space continuous environmental parameter prediction capability for precision agriculture, significantly improving the scientific decision-making level of farmland management.

[0049] These aspects or other aspects of the present application will be more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0051] Figure 1A flow chart of an agricultural environment monitoring method based on Internet of Things is shown;

[0052] Figure 2 A scene diagram of an agricultural environment monitoring system based on Internet of Things is shown;

[0053] Figure 3 A scene diagram of an agricultural environment monitoring system based on Internet of Things is shown;

[0054] Figure 4 A structural schematic diagram of an agricultural environment monitoring system based on Internet of Things is shown;

[0055] Figure 5 A structural schematic diagram of a computing device is shown. DETAILED DESCRIPTION

[0056] In order to enable persons skilled in the art to better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application.

[0057] In some processes described in the specification and claims of the present application and the above-described accompanying drawings, a plurality of operations appearing in a specific order are included, but it should be clearly understood that these operations can be executed or in parallel without the order in which they appear in this text, and the serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes can include more or fewer operations, and the operations can be executed in sequence or in parallel. It should be noted that the "first", "second", etc. in this text are used to distinguish different messages, devices, modules, etc., and do not represent the order of precedence. Also, "first" and "second" are not different types.

[0058] Research has found that although the existing water and fertilizer integrated intelligent regulation and control technology realizes the overall management of farmland through remote sensing monitoring and cloud decision-making, its core defect is that the spatio-temporal matching of macro soil moisture data and micro soil nutrient dynamics is insufficient. On the one hand, unmanned aerial vehicle remote sensing and satellite images are limited by resolution and update frequency, making it difficult to accurately capture the micro-scale nutrient fluctuations in soil heterogeneity areas. On the other hand, the fertilizer demand prediction based on deep learning models relies on historical data fitting and has a lag response to climate mutations or local soil moisture anomalies, resulting in a decrease in the adaptability of water and fertilizer ratio instructions to real-time field conditions. In addition, the cloud centralized decision-making mechanism cannot effectively support multi-node concurrent regulation and control requirements due to communication delays and edge device computing power limitations, and is prone to "regulation and control blind spots" in soil sensor sparse areas, which restricts the dynamic and accurate management efficiency of large-scale farmland.

[0059] To solve the above problems, the application provides a soil nutrient regulation method based on dynamic weight sensing and microfluidic edge computing, which has the following innovative points: a dynamic weight distribution mechanism of a soil sensor array is constructed through an Internet of Things node, local penetration detection of a microfluidic chip and spatiotemporal correlation analysis are combined, and precise regulation of macro-micro coordination is realized. Specifically, the spatial representativeness of multi-source soil data is optimized by using a dynamic weight algorithm, the selective permeability of the microfluidic chip to the soil solution and the real-time conversion technology of ion concentration are combined, and high spatiotemporal resolution dynamic information of nutrients is synchronously obtained; further, through time coherence modeling of historical time series data and a spatial interpolation algorithm, a soil parameter field of the whole field is reconstructed, a dynamic nutrient map with real-time and predictive performance is generated, and precise matching with crop growth demand is achieved. The method breaks through the spatiotemporal lag bottleneck of traditional remote sensing data and cloud decision-making - through the cooperation of edge-side microfluidic detection and dynamic weight distribution of the Internet of Things, the data blind area problem of sparse sensing areas is solved, and the response speed of regulation under sudden climate scenarios is improved; at the same time, the dynamic prediction model based on spatiotemporal correlation significantly enhances the local adaptability of the fertilizer adjustment instruction, realizes the technical leap from "zoning regulation" to "pixel-level precise regulation", and provides reliable support for the optimal allocation of water and fertilizer resources in complex farmland environments.

[0060] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0061] For the above system, the embodiments of the application provide an agricultural environment monitoring method based on the Internet of Things, Figure 1 For the agricultural environment monitoring method based on the Internet of Things, a flowchart is provided, as Figure 1 shown, the method comprises:

[0062] 101. In the agricultural environment detection area, the soil sensor array connected by the Internet of Things node collects multi-source data in the soil, and dynamically distributes the measurement values of the soil sensors at different positions in the multi-source data, and generates a globally consistent soil parameter set according to the measurement values after dynamic weight distribution.

[0063] Optionally, step 101 can specifically include the following steps:

[0064] 1011. In the agricultural environment detection area, the soil sensor array connected by the Internet of Things node continuously collects measurement values at positions of each soil sensor, and defines the measurement values of multiple soil sensors collected at the same time point as multi-source data;

[0065] 1012. Determine the adjacent coverage relationship between different positions of the soil sensors according to the position distribution of each soil sensor in the array, the adjacent coverage relationship being characterized by the physical distance between adjacent soil sensors and the consistency of the change trend of the measured values;

[0066] 1013. Assign an initial weight to the measured value of each soil sensor based on the adjacent coverage relationship, and calculate the measured value difference proportion of adjacent soil sensors;

[0067] 1014. Dynamically adjust the initial weight according to the measured value difference proportion, so that the measured value difference proportion is negatively correlated with the adjustment amplitude of the corresponding weight of adjacent soil sensors;

[0068] 1015. Weight and integrate the dynamically adjusted weight and the measured value of the corresponding soil sensor to eliminate the measurement deviation caused by uneven distribution of soil sensors, and generate a globally consistent soil parameter set covering the entire agricultural environment detection area according to the weighted integration result.

[0069] In the above scheme, the agricultural environment detection area refers to a specific geographical range for monitoring the characteristics of agricultural soil. The Internet of Things node is an intelligent terminal device that connects sensors and realizes data transmission. The soil sensor array is a detection network composed of multiple soil sensors according to a specific layout. The multi-source data refers to the diversified measurement data collected from different sensors at the same time point. The dynamic weight distribution is a calculation method for differentiating and weighting the measured values according to the characteristics of the sensors. The globally consistent soil parameter set is a standardized soil characteristic data set formed after data fusion processing. The measured value is the specific detection result of the soil characteristic parameter by the sensor. The adjacent coverage relationship is a topological structure describing the spatial distribution and measurement correlation between adjacent sensors. The physical distance is the actual distance parameter between adjacent sensors. The measured value difference proportion is a quantitative index of the deviation degree of the detection results of adjacent sensors. The weighted integration is a data fusion process after weight calculation of the measured values of each sensor. The measurement deviation is the data error caused by uneven distribution of sensors.

[0070] In the embodiments of the present application, first, through step 1011, the soil sensor array connected by the Internet of Things node in the agricultural environment detection area collects the measured values (such as humidity, temperature, pH value, etc.) at the position of each sensor in real time. The sensor array synchronously collects data at a fixed frequency, and the measured values of multiple sensors obtained at the same time point are summarized as multi-source data. For example, the soil humidity data at a certain time point contains the measurement results of multiple points, and these data are uploaded to the central processing unit through the Internet of Things communication protocol to provide original input for subsequent analysis.

[0071] Subsequently, in step 1012, the system determines the proximity coverage relationship between different sensors based on the position distribution of the sensors in the array through a spatial clustering algorithm or a proximity relationship model. Specifically, the proximity relationship is determined by two factors: one is the physical distance between the sensors (if the distance is less than the set threshold, it is considered adjacent), and the other is the consistency of the change trend of the historical data (if the fluctuation trend of the adjacent sensor measurement values ​​is highly similar, it is determined to be a strong correlation). For example, if the humidity change curves of two sensors in consecutive time periods are highly overlapped, they may be marked as a proximity relationship even if the physical distance is far. This step provides a basis for the association of space and data for weight allocation.

[0072] Next, based on the proximity and coverage relationships, the system assigns an initial weight to each sensor's measurement. The initial weights are set as follows: nodes associated with more neighboring sensors receive higher weights, while isolated nodes receive lower weights. For example, sensors located in densely populated areas, where their coverage overlaps significantly, receive an initial weight of 0.8; sensors in sparsely populated areas receive an initial weight of 0.5. Furthermore, the difference ratio between the measurements of adjacent sensors (e.g., the percentage difference in humidity between two adjacent sensors) is calculated to measure the consistency of local data.

[0073] The initial weights are then dynamically adjusted based on the difference ratio of the measured values. The adjustment logic is as follows: if the difference ratio between the measurements of adjacent sensors is small (e.g., humidity differences can be less than 5%), the weights of both are increased (e.g., from 0.8 to 0.9) to enhance their data contribution; if the difference ratio is large (e.g., humidity differences can exceed 15%), the weights are reduced (e.g., from 0.8 to 0.6) to suppress interference from abnormal data. During the adjustment process, the system uses a negative feedback mechanism to ensure that the weight change is inversely proportional to the difference ratio, thereby prioritizing the use of regional data with high consistency.

[0074] Finally, in step 1015, the dynamically adjusted weights are weighted and integrated with the corresponding sensor measurements. For example, all humidity values ​​at the same time point are weighted and summed to obtain a global humidity parameter. This process eliminates measurement bias caused by uneven sensor distribution or local anomalies, generating a globally consistent set of soil parameters (such as average humidity and temperature distribution maps) covering the entire detection area. The integrated parameters are output through a visual interface or decision-making system, providing a unified data benchmark for precision agriculture management (such as irrigation scheduling).

[0075] In a practical application, in a greenhouse soil moisture monitoring scenario, a soil sensor array deployed in the crop planting area collects multi-source data in real time through IoT nodes (step 1011). For example, the humidity measurements of sensors located on the east side of the greenhouse fluctuate periodically due to their proximity to irrigation pipes, while those on the west side show a continuous downward trend due to solar evaporation. The system determines proximity based on the sensor array's honeycomb layout (step 1012): nodes with adjacent sensors within two meters of each other and synchronized historical measurement changes are marked as strongly correlated, while nodes with a distance exceeding three meters or significant data fluctuations are classified as weakly correlated. Based on this, sensors in the central area are assigned higher initial weights, while edge nodes are assigned lower initial weights (step 1013). If, during a particular acquisition, the humidity difference ratio between two adjacent sensor groups on the east side increases abnormally (step 1013), the system dynamically adjusts the weights downward (step 1014), mitigating the impact of local irrigation disturbances on global data. Finally, the measurement values ​​of all sensors are integrated through dynamic weighting (step 1015) to generate a global parameter set reflecting the uniformity of soil moisture in the entire greenhouse, accurately guiding the water distribution of the zoned drip irrigation system and avoiding the problems of over-irrigation or under-irrigation caused by traditional single-point monitoring.

[0076] The overall solution of step 101 above effectively solves the problem of data bias caused by uneven sensor distribution or soil spatial heterogeneity in agricultural detection through dynamic weight allocation and multi-source data fusion. Based on the neighboring coverage relationship, the measurement trend correlation between sensors is established, and the weight is dynamically adjusted using the difference ratio to form a local measurement value self-correction mechanism, which significantly improves data reliability. Through weight negative feedback adjustment, the system can automatically suppress the impact of abnormal sensors or local environmental mutations on global data and enhance anti-interference capabilities. The weighted integration process takes into account spatial distribution characteristics and real-time data changes, eliminates measurement blind spots caused by differences in sensor density, and generates a highly consistent global soil parameter set. This adaptive dynamic compensation mechanism breaks through the limitations of traditional uniformly weighted data integration, realizes high-precision spatial mapping of soil conditions in complex farmland environments, and provides a reliable data foundation for precision agricultural decision-making.

[0077] 102. Deploy a microfluidic detection chip of a micro-electromechanical system based on the topological structure of the Internet of Things within the coverage area of ​​the soil sensor array, and selectively infiltrate the collected soil solution through the microfluidic detection chip.

[0078] Optionally, step 102 includes:

[0079] 1021. Determine the deployment position of the microfluidic detection chip according to the topology of the Internet of Things nodes within the coverage area of the soil sensor array, so that each microfluidic detection chip forms a direct communication link with one Internet of Things node, and adjust the spacing between the chips based on the signal strength of the communication link to ensure signal stability of the communication link;

[0080] 1024. Construct a multi-layer permeable membrane inside the microfluidic detection chip, and intercept the target ions in the collected soil solution through the physical structure difference of the different levels of permeable membrane, wherein the physical structure difference of the permeable membrane refers to the pore size of the permeable membrane;

[0081] 1023. According to the matching relationship between the size of the target ion and the pore of the permeable membrane, the target ion is intercepted at the specified level, and the non-target ion is discharged through other levels, so as to realize selective permeation of the soil solution.

[0082] In the above scheme, the topology is a spatial model describing the connection relationship between the Internet of Things nodes and the devices. The micro-electro-mechanical system is a miniaturized system integrating micro-mechanical structures and electronic components. The microfluidic detection chip is a miniaturized detection device based on microfluidic analysis. Selective permeation is a technology that separates specific components using physical or chemical properties. Soil solution is a liquid component containing dissolved substances in soil. The deployment position is the installation coordinates of the microfluidic detection chip in the monitoring area. The direct communication link is a point-to-point data transmission channel between devices. The signal strength is the energy strength index of electromagnetic waves in wireless communication. The multi-layer permeable membrane is a filtering structure composed of different pore membranes. The physical structure difference refers to the difference in physical properties such as pore size and thickness of the permeable membrane. The target ion is a specific charged particle that needs to be separated and extracted in the detection process. The matching relationship is the geometric correspondence between the diameter of the target ion and the pore diameter of the membrane. The non-target ion is a dissolved charged particle that does not need to be retained in the detection process.

[0083] In the embodiments of the present application, first, according to the topology of the Internet of Things nodes within the coverage area of the soil sensor array, the system determines the installation point of the microfluidic detection chip through a position deployment algorithm. Specifically, the deployment position of each chip needs to form a direct communication link with one Internet of Things node to ensure the real-time data transmission. During the deployment process, the system monitors the signal strength of the communication link in real time (such as through a radio frequency signal detection module), and if the signal strength is lower than the threshold, the physical spacing between the chips is adjusted dynamically (such as moving the chip position or adding a relay node) until the signal is stable. This process combines topology optimization and adaptive communication technology to ensure the collaborative working ability of the chip and the Internet of Things node.

[0084] Subsequently, a structure with multiple layers of permeable membranes is constructed inside the deployed microfluidic detection chip through micro-nano processing technology through step 1022. The different levels of permeable membranes form differentiated pore sizes (e.g., larger pores on the top layer, and smaller pores on the lower layers) by controlling the material deposition or etching process. For example, a multi-layer membrane structure is prepared using silicon-based micro-processing technology, and the pore size of each layer is pre-designed according to the physical characteristics (such as ion radius) of the target ions, ensuring that different levels can be targeted to intercept ions of specific sizes.

[0085] Finally, through step 1023, when the soil solution flows through the microfluidic detection chip, based on the matching relationship between the size of the target ion and the pore size of the permeable membrane, the system realizes selective permeation through hierarchical filtration. For example, larger target ions (such as heavy metal ions) are initially intercepted by the top layer of large-pore membranes, while smaller target ions (such as specific nutrient elements) are intercepted by the lower layer of small-pore membranes; non-target ions can directly pass through each layer due to their small size. The intercepted target ions are analyzed for concentration by the micro-sensor or detection unit inside the chip, and the results are uploaded to the central system through the Internet of Things node, ultimately achieving accurate detection and dynamic monitoring of the composition of the soil solution.

[0086] In practical applications, in the context of orchard soil nutrient monitoring, Internet of Things nodes are deployed in a hexagonal topology in areas with dense root systems of fruit trees, and the installation position of the microfluidic detection chip is dynamically optimized according to the signal strength of the node communication link (step 1021). For example, when the signal attenuation between nodes is caused by terrain undulations in a certain area, the system automatically shortens the distance between adjacent chips to within the communication stability threshold, ensuring real-time data transmission between the chip and the node. Each microfluidic detection chip has three layers of permeable membranes (step 1024): the first layer of pores intercepts humic acid macromolecules in the soil solution, the second layer of pores matches the size of potassium ions to achieve selective interception, and the third layer of ultra-small pores discharges water and dissolved impurities. When the soil solution flows through the chip (step 1023), potassium ions are captured due to their size matching the second layer of pores, while sodium ions and nitrate small molecules penetrate to the third layer and are discharged, thereby accurately extracting the potassium concentration characteristics. The captured potassium ion data is uploaded to the control center through the Internet of Things node, and is combined with the global parameter set of the soil sensor array to dynamically generate a precise fertilization plan for the fruit tree subarea, avoiding the nutrient imbalance caused by traditional extensive fertilization.

[0087] The overall scheme of step 102 above builds a high-precision soil solution in-situ detection network through the cooperative deployment of microfluidic detection chips and Internet of Things nodes. Based on the dynamic optimization of the chip spacing of the Internet of Things topology, the system ensures stable transmission of detection data while achieving seamless coverage of the monitoring area. By utilizing the pore differences of the multi-layer permeable membrane, the system can selectively retain target ions in the soil solution, effectively excluding non-target component interference and improving detection specificity. This physical screening mechanism at the micro-nano scale breaks through the limitations of complex pretreatment in traditional chemical detection and achieves efficient separation through pore-ion size matching. The direct connection architecture of the chip and the Internet of Things node further enhances the real-time and reliability of the detection data, providing an in-situ and efficient means of soil solution component analysis for precision agriculture, significantly improving ion monitoring capability and data quality in complex farmland environments.

[0088] 103. Obtain the target ion concentration in the permeated soil solution and convert the target ion concentration into an electrical signal, while using the low-power communication protocol of the Internet of Things to synchronize the transmission of the electrical signal and the soil parameter set.

[0089] Optionally, step 103 includes:

[0090] 1031. An electrochemical sensing unit is arranged at the outlet of the permeation channel of the microfluidic detection chip. The target ions in the permeated soil solution are captured by the ion-selective membrane in the electrochemical sensing unit, and a concentration sensing signal is generated based on the accumulation of the target ions on the surface of the ion-selective membrane.

[0091] 1032. The concentration sensing signal is converted into a corresponding electrical signal based on the conversion circuit built-in the electrochemical sensing unit. The amplitude of the electrical signal changes linearly with the increase of the concentration.

[0092] 1033. The electrical signal is timestamped bound with the soil parameter set at the same time point through the low-power communication protocol of the Internet of Things node, to realize the synchronous transmission of the electrical signal and the soil parameter set.

[0093] In the above scheme, the soil solution after permeation is a liquid soil sample processed by the microfluidic detection chip after selective permeation. The target ion concentration is the content level of specific charged particles in the soil solution. The electrical signal is the physical quantity converted from the chemical concentration to voltage or current form by the sensor. The low-power communication protocol of the Internet of Things is the rule of optimizing energy consumption for wireless data transmission. The set of soil parameters is a standardized soil feature dataset formed by integrating multi-source data. Synchronous transmission is a technology to ensure the simultaneous transmission of information from different data sources. The microfluidic detection chip is a miniature detection device based on microfluidic channels for fluid analysis. The outlet of the permeation channel is the terminal position of the solution outflow in the microfluidic chip. The electrochemical sensing unit is a detection module that converts ion concentration into electrical signals. The ion-selective membrane is a thin film material that selectively permeates specific ions. The target ion is the charged particle that needs to be separated and extracted in the detection process. The cumulative amount is the physical quantity of the target ion accumulated on the membrane surface. The concentration sensing signal is the original detection signal reflecting the concentration change of the target ion. The conversion circuit is a combination of electronic elements that converts analog signals into standard electrical signals. The amplitude is the voltage or current intensity value of the electrical signal. The linear change is the response characteristic that the amplitude of the electrical signal is directly proportional to the concentration of the target ion. The Internet of Things node is an intelligent terminal device with data acquisition and communication functions. The same time point is the time marker that is completely consistent in the data acquisition process. Time stamp binding is the operation of adding a unified time identifier to the electrical signal and the set of soil parameters.

[0094] In the embodiments of the present application, first, an electrochemical sensing unit is arranged at the outlet of the permeation channel of the microfluidic detection chip by step 1031, and an ion-selective membrane is built-in the unit. When the soil solution after selective permeation flows through the outlet, the ion-selective membrane captures specific target ions (such as potassium ions or nitrate ions) in the solution through physical adsorption or chemical bonding. As the target ions continue to accumulate on the membrane surface, the change in ion concentration causes the electrochemical characteristics (such as potential or current) of the membrane surface to change, thereby generating a concentration sensing signal directly related to the concentration. This process relies on the specific recognition ability of the material properties (such as polymer or biomolecule modified membrane) of the ion-selective membrane to the target ions.

[0095] Subsequently, the conversion circuit inside the electrochemical sensing unit converts the concentration sensing signal generated in step 1031 into a standardized electrical signal by step 1032. The conversion circuit converts the electrochemical change on the membrane surface (such as microampere current) into a voltage signal through amplification, filtering, and linearization, and the voltage amplitude is directly and linearly related to the concentration of the target ion. For example, when the ion concentration increases, the output voltage can gradually increase from 0.1V to 5V, ensuring that the signal amplitude can be directly used for digital processing and transmission.

[0096] Finally, through step 1033, the system binds the electrical signal generated in step 1032 with the set of soil parameters (such as humidity, temperature) obtained from the soil sensor array at the same time point through the low-power communication protocol of the Internet of Things node. The bound data packet is uploaded to the cloud or local server through the wireless network, ensuring that the collection time of the electrical signal and the soil parameters is strictly synchronized, providing a time-consistent multi-source data basis for subsequent data analysis (such as ion concentration and soil environment correlation modeling).

[0097] In practical application, in the soil nutrient monitoring scene of the vegetable greenhouse, after the microfluidic detection chip deployed in the crop root zone completes the selective permeation of the soil solution (step 102), the electrochemical sensing unit at the outlet of the permeation channel starts the ion capture process (step 1031). For example, for the monitoring of excess nitrate in the soil after fertilization, the ion-selective membrane in the chip preferentially adsorbs nitrate ions, and the surface accumulation increases with the permeation time, triggering the increase in the intensity of the concentration-induced signal. The built-in conversion circuit converts the induced signal into an electrical signal in real time (step 1032), and the higher the nitrate concentration, the greater the voltage amplitude of the output electrical signal. At the same time, the humidity, pH value and other parameters collected by the soil sensor array are integrated into a global set of soil parameters (step 101). The Internet of Things node binds the electrical signal and the set of soil parameters to the same timestamp through the low-power communication protocol (step 1033), ensuring that the nitrate concentration data and the soil environment state are transmitted synchronously to the cloud analysis platform. When the system detects that the electrical signal amplitude in a certain area is continuously over-limit, it automatically triggers an alarm and generates a nitrogen reduction irrigation instruction, accurately controls the valve opening degree of the water and fertilizer integrated machine, and realizes the dynamic balance of the nutrient excess area.

[0098] The overall scheme of step 103 above realizes high-precision in-situ detection and synchronous data transmission of soil ion concentration through the deep integration of electrochemical sensing and Internet of Things technology. The electrochemical sensing unit based on ion-selective membrane can accurately capture target ions and convert concentration information into reliable electrical signals through linear conversion circuit, ensuring high sensitivity and accuracy of the detection results. The use of low-power communication protocol for timestamp binding transmission not only ensures the spatiotemporal consistency of the electrical signal and the set of soil parameters, but also significantly reduces the system energy consumption. This integrated design of "sensing-conversion-synchronization" effectively solves the problem of data fragmentation and transmission delay in traditional soil detection, providing real-time and collaborative multi-parameter monitoring capability for precision agriculture, and greatly improving the timeliness and reliability of soil nutrient dynamic analysis.

[0099] 104、extracting historical time series data from the set of soil parameters, and performing time correlation analysis on the electrical signal and the historical time series data to establish the continuity of the time dimension, and filling data for areas where soil sensors are not deployed by combining a spatial interpolation algorithm, and generating a dynamic prediction model using the continuity and the data filling result.

[0100] Optionally, step 104 comprises:

[0101] 1041、extracting data recorded at different time points from the set of soil parameters to form historical time series data, and matching the electrical signal with data at corresponding time points in the historical time series data according to collection time to calculate the fluctuation correlation degree of the electrical signal and the historical time series data at the same time point;

[0102] 1042、determining the consistency of the change rule of the electrical signal and the historical time series data in the time dimension according to the fluctuation correlation degree, and supplementing historical time series data for missing time points based on the consistency of the change rule, so that the historical time series data at all time points form a continuous time series;

[0103] 1043、calculating the data difference ratio of areas where soil sensors are not deployed and adjacent deployed areas by a spatial interpolation algorithm, and filling data for areas where soil sensors are not deployed according to the data difference ratio;

[0104] 1044、superimposing the continuity result of the time series and the data filling result of the non-deployed area, and constructing a dynamic prediction model according to the superimposed result.

[0105] In the above scheme, the time correlation analysis is an analysis method for studying the correlation of the electrical signal and the historical time series data in the time axis. The continuity of the time dimension is the feature that the data maintains continuity and consistency in the time series. The spatial interpolation algorithm is a mathematical method for calculating the data of the unmonitored area according to the known monitoring point data. The area where soil sensors are not deployed is a geographical area where sensors are not installed in the monitoring network. Data filling is the process of supplementing soil parameter data for the non-deployed area using the interpolation algorithm. The dynamic prediction model is a mathematical model for predicting future changes based on spatio-temporal data. The fluctuation correlation degree is a quantitative index of the similarity of changes of the electrical signal and the historical time series data at the same time point. The consistency of the change rule is the same change trend feature presented by the electrical signal and the historical time series data. The missing time point is a time node with a record gap in the historical time series data. The continuity result of the time series is the complete time series after the missing data is supplemented. The data filling result of the non-deployed area is the data of the non-deployed area obtained by using the interpolation algorithm. The superimposed result is the integrated data obtained by integrating the time series continuity result and the spatial filling data. The construction of the dynamic prediction model is the process of establishing a mathematical model for predicting future changes according to the superimposed result.

[0106] In the embodiments of the present application, first, the system extracts historical time series data (such as daily soil humidity, temperature, etc.) recorded at different time points from the soil parameter set through step 1041, and matches the electrical signal (corresponding to the target ion concentration) with the data at the corresponding time point in the historical time series data according to the collection time. The matching process is realized by a timestamp alignment algorithm, for example, the electrical signal at the same time is bound with the soil humidity data, and then the correlation analysis (such as covariance calculation in the sliding window) is used to quantify the fluctuation correlation degree of the two at the same time point. For example, if the electrical signal (ion concentration) and soil humidity rise synchronously in a certain time period, it is determined that there is a strong correlation between the two, otherwise the correlation is weak.

[0107] Subsequently, through step 1042, the system analyzes the consistency of the change rule of the electrical signal and the historical time series data in the time dimension according to the fluctuation correlation degree obtained in step 1041. For example, if the fluctuation trend of the electrical signal and the soil humidity is highly consistent for three consecutive days, it is considered that the two follow the same environmental change rule. Based on this rule, the system supplements the historical time series data at the missing time point by linear interpolation or time series prediction model, for example, the missing humidity data of a certain day is completed according to the trend of the electrical signal, and finally a coherent time series data is formed.

[0108] Secondly, through step 1043, for the area where the soil sensor has not been deployed, the system uses a spatial interpolation algorithm (such as inverse distance weighting or Kriging interpolation) to calculate the data difference ratio of the area and the surrounding area according to the sensor data of the adjacent deployed area. The calculation of the data difference ratio is based on the geographical spatial relationship (such as distance, terrain), which ensures that the filling result conforms to the actual environmental distribution characteristics.

[0109] Finally, through step 1044, the time series coherence result (completed time series data) generated in step 1042 and the data filling result of the non-deployed area in step 1043 are superimposed to integrate into a complete data set covering the entire detection area and being time continuous. Based on the data set, the system constructs a dynamic prediction model through a machine learning algorithm (such as random forest or LSTM neural network), which can predict the change trend of soil parameters (such as humidity decline rate) or target ion concentration abnormal events at future time points, providing support for precision agriculture decision-making.

[0110] In practical applications, in the tea garden soil nutrient monitoring scene, the system extracts the historical time series data of the soil parameter set in the past three months (step 1041), including the periodic fluctuation of the pH value, humidity and potassium ion electric signal. For example, the potassium ion electric signal collected after spring rainfall has high correlation with the rising curve of the humidity parameter in the historical time series data, indicating that the two have consistent change regularity in the time dimension. When a certain collection is missing the electric signal in the early morning due to communication interruption, the system automatically supplements the data in this period according to the established correlation (step 1042), ensuring that the historical time series data is coherent and complete on the time axis. For the edge area of the terrace where the soil sensor is not deployed, the system calculates the difference ratio of potassium ion concentration between it and the adjacent deployed sensor area through a spatial interpolation algorithm (step 1043), and combines the terrain slope and irrigation path characteristics to generate a virtual data filling result for the edge area. For example, the upper area of the terrace is caused by rainwater erosion, and the interpolation data and the measured area are dynamically corrected. Finally, the coherence result of the time series is integrated with the spatial filling data (step 1044) to construct a dynamic prediction model of the potassium ion concentration in the tea garden. When the model predicts that the potassium ion concentration in a certain area will be below the threshold in the next three days, it automatically triggers the precise fertilization instruction to guide the unmanned aerial vehicle to broadcast potassium fertilizer at the designated point, realizing the full-area coverage and forward-looking regulation of tea garden nutrient management.

[0111] The overall scheme of step 104 realizes the full-area coverage and dynamic prediction of agricultural environment monitoring through spatio-temporal data fusion and intelligent prediction modeling. Based on time correlation analysis, the system establishes the time series coherence of the electric signal and the historical data, effectively filling the time breakpoints of the monitoring data. Combined with the spatial interpolation algorithm, the discrete point-shaped monitoring is expanded to area-shaped coverage, solving the problem of sensor deployment blind area. Through the double data compensation of time and space dimensions, the dynamic prediction model not only retains the high precision characteristics of local monitoring, but also has global deduction ability, significantly improving the spatio-temporal continuity of soil parameter prediction. This "time correlation + spatial interpolation" collaborative modeling method breaks through the limitations of the traditional monitoring of spatio-temporal data fragmentation, and provides high-resolution and high-reliability full-area dynamic prediction support for precision agriculture decision-making.

[0112] 105. Output the spatial variation trend of soil nutrients according to the dynamic prediction model, and match the spatial variation trend with the preset crop growth demand parameter to generate a fertilizer amount adjustment instruction for different soil regions, and issue the fertilizer amount adjustment instruction to the fertilizer equipment execution end deployed in the field through the Internet of Things communication link to complete the dynamic regulation of soil nutrients.

[0113] Optionally, step 105 includes:

[0114] 1051、extracting the distribution data of soil nutrients in different geographical locations from the dynamic prediction model to calculate the spatial variation trend of soil nutrients;

[0115] 1052、point-by-point comparing the data of soil nutrients in each geographical location in the spatial variation trend with the preset crop growth demand parameters to calculate the soil nutrient difference proportion, and determining the fertilization priority level of the corresponding area according to the soil nutrient difference proportion;

[0116] 1053、generating adjustment instructions containing fertilization type and fertilization amount for each geographical location based on the fertilization priority level, the fertilization amount in the adjustment instructions being proportional to the soil nutrient difference proportion, and the fertilization type being determined as supplement or reduction according to the positive or negative direction of the soil nutrient difference proportion;

[0117] Wherein, step 1053 can specifically include the following process:

[0118] According to the soil nutrient difference proportion of each geographical location in the fertilization priority level, adjusting the fertilization amount, which is proportional to the soil nutrient difference proportion; determining the positive or negative direction of the difference proportion, comparing the positive or negative direction with the nutrient threshold in the preset crop growth demand parameters, assigning the fertilization type for each geographical location, the positive direction in the positive or negative direction indicating that the current nutrient is lower than the nutrient threshold, and the corresponding supplement type of fertilizer needs to be applied, and the negative direction indicating that the current nutrient is higher than the nutrient threshold, and the corresponding reduction type of fertilizer needs to be applied; integrating the fertilization type and the fertilization amount, and generating adjustment instructions according to the integration result.

[0119] 1054、distributing the adjustment instructions to the corresponding fertilization equipment execution end according to the geographical location coordinate information through the Internet of Things communication link to complete the dynamic regulation and control of soil nutrients.

[0120] In the above scheme, the dynamic prediction model is a mathematical model for predicting the change of soil nutrients based on spatio-temporal data. The spatial variation trend of soil nutrients is the distribution change rule of nutrient content in soil at different geographical locations. The preset crop growth requirement parameter is the standard requirement threshold of nutrients for crops at different growth stages. The fertilizer amount adjustment instruction is a control command for the type and amount of fertilizer generated according to the difference in soil nutrients. The Internet of Things communication link is a data transmission channel between devices established through the Internet of Things protocol. The fertilizer equipment execution end is a field fertilization machine that receives and executes the adjustment instruction. Dynamic regulation is a closed-loop control process of adjusting the fertilization operation according to real-time data. Soil nutrients are nutrient components in soil that can be absorbed by plants. The type of fertilizer is a classification identifier of fertilizer such as nitrogen fertilizer, phosphorus fertilizer, etc. The amount of fertilizer is the amount of fertilizer per unit area. The geographical location coordinate information is the latitude and longitude or grid coding data used to identify the location of the field. The point-by-point comparison is an operation of comparing soil nutrient data at each location with crop requirement parameters one by one. The difference ratio of soil nutrients is a quantitative value of the deviation degree of the actual soil nutrients from the standard requirement. The priority of fertilization is the level of fertilization urgency divided according to the difference degree. The type of supplementary fertilizer is a fertilizer category used to increase soil nutrients. The type of reduced fertilizer is a fertilizer category used to reduce excess soil nutrients. The nutrient threshold is the minimum or maximum nutrient standard value required for crop growth. The positive direction is the state identifier of the soil nutrients below the requirement threshold. The closed-loop control is a cyclic control system formed by real-time monitoring and adjustment.

[0121] In the embodiments of the present application, first, the soil nutrient distribution data of different geographical locations is extracted from the dynamic prediction model through step 1051, and these data are processed through geographical space analysis techniques (such as gradient calculation or heat map generation) to generate a spatial variation trend of soil nutrients reflecting the distribution difference of nutrient content in the farmland. For example, the system identifies that the nutrients in a certain area present a feature of decreasing from the center to the periphery, providing a spatial reference for subsequent decision-making.

[0122] Subsequently, the soil nutrient data at each location in the spatial variation trend is matched with the preset crop growth requirement parameter through step 1052, and the difference ratio of soil nutrients in each region is obtained through difference calculation. The positive and negative directions of the difference ratio represent that the current nutrient level is lower or higher than the crop requirement threshold, and the system divides the priority of fertilization according to the size of the difference ratio - the larger the difference ratio, the higher the priority, which needs to be adjusted first.

[0123] Then, the system generates fertilization amount adjustment instructions for each geographic location based on the fertilization priority level through step 1053. The fertilization amount of the adjustment instruction is dynamically determined according to the difference ratio, and the larger the ratio, the larger the adjustment amount; the fertilization type is determined by the difference direction: the positive direction (nutrient deficiency) triggers the supplementary fertilizer instruction, and the negative direction (excess nutrients) triggers the reduction or neutralization measure instruction. For example, a certain area needs to increase nitrogen fertilizer due to insufficient nitrogen content, while another area needs to reduce phosphorus fertilizer due to excess phosphorus.

[0124] Finally, the adjustment instructions are distributed to the corresponding field fertilization equipment execution end according to the geographic location coordinate information through the Internet of Things communication link. After receiving the instructions, the equipment automatically adjusts the fertilizer type and the amount of fertilizer, such as precisely supplementing fertilizer to areas with insufficient nutrients or starting a neutralization program for areas with excess nutrients, thereby achieving dynamic balance regulation and control of farmland nutrients. The whole process is through the closed-loop logic of "prediction-matching-instruction-execution", ensuring that the soil nutrients and crop demand are real-time adapted.

[0125] In actual application, in the scenario of nitrogen management in rice field soil, the dynamic prediction model outputs the spatial variation trend of nitrogen concentration in the rice field based on historical time series data and spatial interpolation results (step 1051), showing that the nitrogen concentration gradient on the south side of the field is lower than that on the north side due to irrigation water flow erosion. The system matches the nitrogen data on the south side with the growth demand parameters of rice tillering stage point by point (step 1052), and calculates that the nitrogen difference ratio on the south side is positive, and nitrogen fertilizer needs to be supplemented; while the nitrogen difference ratio on the north side is negative, and the amount of fertilizer needs to be reduced. Based on the fertilization priority level (step 1053), the south generates a "liquid urea supplement" instruction, and the amount of fertilizer is proportional to the difference ratio, while the north triggers a "slow-release fertilizer reduction" instruction. The Internet of Things communication link distributes the instructions to the intelligent fertilizer distribution machine execution end deployed on the field ridge according to the field grid coordinates (step 1054): the south side fertilizer distribution machine nozzle expands the opening degree to increase the urea spraying intensity, and the north side fertilizer distribution machine switches to low flow mode and mixes straw degradation agent to inhibit nitrogen release. Through dynamic regulation, the nitrogen concentration in the whole rice field is adapted to the growth curve of rice within 48 hours, avoiding insufficient panicle number caused by fertilizer deficiency on the south side, and preventing nitrogen excess on the north side from causing greedy green lodging, achieving closed-loop control of precision agriculture.

[0126] The overall scheme of step 105 above realizes the dynamic optimization of farmland nutrients through intelligent prediction and precise regulation of closed-loop management. Based on the intelligent matching of soil nutrient spatial variation trend and crop demand, the system can automatically identify the nutrient surplus and deficit status of different regions and generate differentiated fertilization strategies. Through positive and negative direction discrimination, it can not only supplement the missing nutrients, but also reduce the excess components, effectively avoiding the waste of resources or environmental pollution caused by traditional uniform fertilization. The precise issuance of Internet of Things instructions ensures the spatio-temporal consistency of the control measures with the actual situation in the field, forming a complete closed loop of "monitoring-prediction-decision-execution". This data-driven precision fertilization mode significantly improves the scientificity and timeliness of farmland management, ensuring crop growth while optimizing resource utilization.

[0127] The following is a complete embodiment for steps 101-105:

[0128] In the soil nutrient monitoring scene of a grape vineyard, an array of soil sensors connected by Internet of Things nodes is deployed in a hexagonal topology in the grapevine root zone, continuously collecting multi-source data such as humidity, pH value, and conductivity at each node location. According to the adjacent coverage relationship of the sensors, nodes with a distance of less than 1.5 meters and synchronous historical data fluctuations are marked as strongly correlated groups, and the system assigns them a higher initial weight. When the humidity difference ratio of adjacent sensors suddenly increases in a certain sampling, the system dynamically reduces its weight adjustment range, and the weighted integrated global soil parameter set is generated. At the same time, the microfluidic detection chip adjusts the deployment distance according to the signal strength of the Internet of Things nodes to ensure stable communication links with the nodes. The three-layer permeable membrane in the chip selectively retains potassium ions and discharges impurities such as sodium ions from the soil solution through pore differences. The electrochemical sensing unit at the outlet of the permeation channel captures potassium ions and generates a concentration sensing signal, which is converted into a linear electrical signal by the conversion circuit and transmitted synchronously with the soil parameter set through a low-power protocol.

[0129] The system extracts historical time series data, analyzes the correlation between potassium ion electrical signals and humidity parameters (claim 2), fills in the missing data in the early morning period due to communication interruption, and forms a coherent time series. For areas on the slope where sensors are not deployed, the system calculates the data difference ratio with adjacent areas based on a spatial interpolation algorithm to generate virtual fill values. The dynamic prediction model constructed after superposition shows that the potassium concentration on the southeast slope will be below the threshold, matching the preset grape berry expansion period demand parameters. According to the positive direction of the difference ratio, the "liquid potassium fertilizer increase" instruction is generated for the southeast slope, with the fertilizer amount being proportional to the difference value; the "water-soluble irrigation dilution" instruction is triggered for the northwest slope due to potassium excess. The adjusted instructions are distributed to the intelligent fertilizer machines at the corresponding coordinates through the Internet of Things, and the southeast slope sprinklers increase the potassium fertilizer flow, while the northwest slope drip irrigation system mixes water to reduce the local concentration, achieving precise dynamic regulation of the nutrients in the grape vineyard.

[0130] As Figure 2 andFigure 3 As shown in the figure, the three-dimensional architecture and operation logic of the Internet of Things-based agricultural environment monitoring system are demonstrated. The first figure presents the core components in a vertical layered structure: the top layer of the Internet of Things gateway connects the soil sensor array through low-power protocols, these sensors collect multi-source data through dynamic weight distribution, and work cooperatively with three microfluidic detection chips - the latter converts soil ion concentration into electrical signals through selective permeation technology. The middle layer of the data processing system uses time correlation analysis and spatial interpolation algorithms to build a dynamic prediction model, and finally outputs precise instructions matching crop requirements by the fertilizer application device. Figure 3 Then the planar perspective supplements the details of the grid layout of the monitoring nodes, through the heterogeneous node network composed of S1-S8 numbered sensors and IM / M microfluidic chips, forming a multi-dimensional data acquisition matrix. The two figures together emphasize the "perception-transmission-analysis-decision" closed-loop process, through the integration of physical sensing and chemical detection technology, realizing the spatio-temporal dynamic modeling and intelligent fertilization control of soil nutrients, embodying the systematic application value of Internet of Things technology in modern precision agriculture.

[0131] Figure 4 A structure diagram of an agricultural environment monitoring system based on the Internet of Things is provided for the embodiments of the present application, as shown in the figure, the system comprises: Figure 2

[0132] The distribution module 41 is configured to collect multi-source data in the soil in the agricultural environment detection area through the soil sensor array connected by the Internet of Things nodes, and to perform dynamic weight distribution on the measurement values of the soil sensors at different positions in the multi-source data, and to generate a globally consistent soil parameter set according to the measurement values after dynamic weight distribution.

[0133] The deployment module 42 is configured to deploy microfluidic detection chips of micro-electro-mechanical systems based on the topological structure of the Internet of Things within the coverage area of the soil sensor array, and to perform selective permeation on the collected soil solution through the microfluidic detection chips.

[0134] The conversion module 43 is configured to obtain the target ion concentration in the permeated soil solution, and to convert the target ion concentration into an electrical signal, and to simultaneously transmit the electrical signal and the soil parameter set using the low-power communication protocol of the Internet of Things.

[0135] The generation module 44 is configured to extract historical time series data in the soil parameter set, and to perform time correlation analysis on the electrical signal and the historical time series data to establish the continuity in the time dimension, and to combine a spatial interpolation algorithm to fill data in the area where no soil sensor is deployed, and to generate a dynamic prediction model using the continuity and the data filling result.

[0136] ​The matching module 45 is configured to output a spatial variation trend of soil nutrients according to the dynamic prediction model, match the spatial variation trend with preset crop growth demand parameters, generate fertilization amount adjustment instructions for different soil regions, and issue the fertilization amount adjustment instructions to a fertilization device execution end deployed in a field through an Internet of Things communication link to complete dynamic regulation of soil nutrients.

[0137] Figure 4 The agricultural environment monitoring system based on the Internet of Things can perform Figure 1 The implementation principle and technical effects of the agricultural environment monitoring method based on the Internet of Things are not repeated. The specific manner in which each module, unit of the agricultural environment monitoring system based on the Internet of Things in the above embodiments performs operations has been described in detail in the embodiments related to the method, and will not be described in detail here.

[0138] In one possible design, Figure 5 The automatic wheel flaw detection and wheelset parallelism correction system of the embodiment can be implemented as a computing device, such as a computer. Figure 5 As shown, the computing device can include a storage component 51 and a processing component 52.

[0139] The storage component 51 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 52.

[0140] The processing component 52 is configured to perform the above Figure 1 The automatic wheel flaw detection and wheelset parallelism correction method of the embodiment.

[0141] The processing component 52 can include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic elements for executing the above method.

[0142] The storage component 51 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0143] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0144] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0145] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0146] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0147] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is an automated wheel flaw detection and wheelset parallelism correction method.

[0148] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0149] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0150] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0151] Finally, it should be noted that the above examples are only used to illustrate the technical solutions of the present application, and are not intended to limit the same; although the present application has been described in detail with reference to the foregoing examples, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An agricultural environment monitoring method based on the Internet of Things, characterized in that: include: In an agricultural environmental monitoring area, a soil sensor array connected by an IoT node collects multi-source data from the soil, dynamically assigns weights to the measurement values ​​of soil sensors at different locations in the multi-source data, and generates a globally consistent set of soil parameters based on the measurement values ​​after dynamic weight assignment. In the coverage area of ​​the soil sensor array, a microfluidic detection chip of a micro-electromechanical system is deployed based on the topological structure of the Internet of Things, and the collected soil solution is selectively infiltrated through the microfluidic detection chip; Obtaining a target ion concentration in the infiltrated soil solution, converting the target ion concentration into an electrical signal, and synchronously transmitting the electrical signal and the soil parameter set using a low-power communication protocol of the Internet of Things; Extracting historical time series data from the soil parameter set, performing a temporal correlation analysis between the electrical signal and the historical time series data to establish temporal continuity, and using a spatial interpolation algorithm to perform data filling in areas where soil sensors are not deployed, generating a dynamic prediction model using the results of the continuity and data filling; The spatial variation trend of soil nutrients is output according to the dynamic prediction model, and the spatial variation trend is matched with the preset crop growth requirement parameters to generate fertilizer adjustment instructions for different soil areas. The fertilizer adjustment instructions are then sent to the execution end of the fertilizer equipment deployed in the field through the Internet of Things communication link to complete the dynamic regulation of soil nutrients.

2. The method according to claim 1, characterized in that Extracting historical time series data from the soil parameter set, performing temporal correlation analysis on the electrical signal and the historical time series data to establish temporal continuity, and using a spatial interpolation algorithm to perform data filling on areas where soil sensors are not deployed. Generating a dynamic prediction model using the results of the continuity and data filling includes: Extracting data recorded at different time points from the soil parameter set to form historical time series data, and matching the electrical signal with data at corresponding time points in the historical time series data according to the acquisition time to calculate the degree of correlation between the electrical signal and the historical time series data at the same time point; Determining the consistency of the change rules between the electrical signal and the historical time series data in the time dimension based on the degree of fluctuation correlation, and supplementing the historical time series data of missing time points based on the consistency of the change rules, so that the historical time series data of all time points form a coherent time series; Calculating the data difference ratio between the area where soil sensors are not deployed and the adjacent area where soil sensors are deployed by a spatial interpolation algorithm, and filling the area where soil sensors are not deployed with data according to the data difference ratio; The consistency results of the time series are superimposed with the data filling results of the undeployed areas, and a dynamic prediction model is constructed based on the superimposed results.

3. The method according to claim 1, characterized in that In the agricultural environment detection area, a soil sensor array connected by an Internet of Things node collects multi-source data in the soil, dynamically weights the measurement values ​​of soil sensors at different locations in the multi-source data, and generates a globally consistent set of soil parameters based on the measurement values ​​after dynamic weight allocation, including: In the agricultural environment detection area, a soil sensor array connected by an Internet of Things node continuously collects measurement values ​​at the location of each soil sensor, and the measurement values ​​of multiple soil sensors collected at the same time point are defined as multi-source data; Determine, based on the position distribution of each soil sensor in the array, a proximity coverage relationship between soil sensors at different positions, wherein the proximity coverage relationship is characterized by the consistency of the physical spacing between adjacent soil sensors and the change trend of the measured values; Based on the proximity coverage relationship, an initial weight is assigned to the measurement value of each soil sensor, and a difference ratio of the measurement values ​​of adjacent soil sensors is calculated; Dynamically adjusting the initial weights according to the measured value difference ratios, such that the measured value difference ratios are negatively correlated with the adjustment amplitudes of the corresponding weights of adjacent soil sensors; The dynamically adjusted weights are weightedly integrated with the measurement values ​​of the corresponding soil sensors to eliminate the measurement bias caused by the uneven distribution of soil sensor locations, and a globally consistent set of soil parameters covering the entire agricultural environment monitoring area is generated based on the weighted integration results.

4. The method according to claim 1, wherein In the coverage area of ​​the soil sensor array, a microfluidic detection chip of a micro-electromechanical system is deployed based on the topological structure of the Internet of Things, and the soil solution is selectively infiltrated by the microfluidic detection chip, including: Determine the deployment positions of the microfluidic detection chips based on the topological structure of the IoT nodes within the coverage area of ​​the soil sensor array, so that each microfluidic detection chip forms a direct communication link with an IoT node, and adjust the spacing between the chips based on the signal strength of the communication link to ensure signal stability of the communication link; A multi-layer permeable membrane is constructed inside the microfluidic detection chip, and the target ions in the collected soil solution are intercepted by the physical structural differences of the permeable membranes at different levels, wherein the physical structural differences of the permeable membranes refer to the pore size of the permeable membranes; According to the matching relationship between the size of the target ions and the pores of the permeable membrane, the target ions are retained in a designated layer, and non-target ions are discharged through other layers to achieve selective permeation of the soil solution.

5. The method according to claim 1, wherein Obtaining a target ion concentration in the infiltrated soil solution, converting the target ion concentration into an electrical signal, and synchronously transmitting the electrical signal and the soil parameter set using a low-power communication protocol of the Internet of Things, including: An electrochemical sensing unit is provided at the outlet of the permeation channel of the microfluidic detection chip, the target ions in the permeated soil solution are captured by the ion selective membrane in the electrochemical sensing unit, and a concentration sensing signal is generated based on the accumulation of the target ions on the surface of the ion selective membrane; Based on the conversion circuit built into the electrochemical sensing unit, the concentration sensing signal is converted into a corresponding electrical signal, and the amplitude of the electrical signal changes linearly with the increase of concentration; Through the low-power communication protocol of the Internet of Things node, the electrical signal is timestamped and bound to the soil parameter set at the same time point to achieve synchronous transmission of the electrical signal and the soil parameter set.

6. The method according to claim 1, wherein Outputting the spatial variation trend of soil nutrients based on the dynamic prediction model and matching the spatial variation trend with preset crop growth requirement parameters to generate fertilizer adjustment instructions for different soil areas. The fertilizer adjustment instructions are then sent to the execution end of fertilizer equipment deployed in the field via an IoT communication link to complete the dynamic regulation of soil nutrients, including: Extracting the distribution data of soil nutrients in different geographical locations from the dynamic prediction model to calculate the spatial variation trend of soil nutrients; Comparing the soil nutrient data of each geographical location in the spatial variation trend with the preset crop growth requirement parameters point by point to calculate the soil nutrient difference ratio, and determining the fertilization priority level of the corresponding area based on the soil nutrient difference ratio; Based on the fertilization priority level, generating an adjustment instruction including a fertilization type and a fertilization amount for each geographical location, wherein the fertilization amount in the adjustment instruction is proportional to the soil nutrient difference ratio, and the fertilization type is determined as supplementation or reduction based on the positive or negative direction of the soil nutrient difference ratio; The adjustment instructions are distributed to the corresponding fertilizing equipment execution end according to the geographic location coordinate information through the Internet of Things communication link to complete the dynamic regulation of soil nutrients.

7. The method according to claim 6, characterized in that Based on the fertilization priority level, an adjustment instruction including a fertilization type and an amount is generated for each geographical location, wherein the amount of fertilization in the adjustment instruction is proportional to the soil nutrient difference ratio, and the fertilization type is determined as supplementation or reduction based on the positive or negative direction of the soil nutrient difference ratio, including: adjusting the amount of fertilizer applied according to the soil nutrient difference ratio at each geographical location in the fertilization priority level, wherein the amount of fertilizer applied is proportional to the soil nutrient difference ratio; Determine the positive and negative directions of the difference ratio, compare the positive and negative directions with the nutrient thresholds in the preset crop growth requirement parameters, and assign a fertilization type to each geographical location, where a positive direction indicates that the current nutrient is lower than the nutrient threshold and a corresponding supplementary type of fertilizer needs to be applied, and a negative direction indicates that the current nutrient is higher than the nutrient threshold and a corresponding reduced type of fertilizer needs to be applied; The fertilization type and fertilization amount are integrated, and an adjustment instruction is generated according to the integration result.

8. An agricultural environment monitoring system based on the Internet of Things, characterized in that: include: An allocation module is configured to collect multi-source data from the soil in an agricultural environment monitoring area through a soil sensor array connected to an Internet of Things node, dynamically weight the measurement values ​​of soil sensors at different locations in the multi-source data, and generate a globally consistent set of soil parameters based on the measurement values ​​after dynamic weight allocation; A deployment module is used to deploy a microfluidic detection chip of a micro-electromechanical system based on the topology of the Internet of Things within the coverage area of ​​the soil sensor array, and selectively penetrate the collected soil solution through the microfluidic detection chip; a conversion module for obtaining a target ion concentration in the soil solution after infiltration, converting the target ion concentration into an electrical signal, and synchronously transmitting the electrical signal and the soil parameter set using a low-power communication protocol of the Internet of Things; a generation module for extracting historical time series data from the soil parameter set, performing a temporal correlation analysis between the electrical signal and the historical time series data to establish temporal continuity, performing data filling for areas where soil sensors are not deployed using a spatial interpolation algorithm, and generating a dynamic prediction model using the results of the continuity and data filling; A matching module is used to output the spatial variation trend of soil nutrients based on the dynamic prediction model, and match the spatial variation trend with preset crop growth requirement parameters to generate fertilizer adjustment instructions for different soil areas, and send the fertilizer adjustment instructions to the execution end of the fertilizer equipment deployed in the field through the Internet of Things communication link to complete the dynamic regulation of soil nutrients.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an agricultural environment monitoring method based on the Internet of Things as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, an agricultural environment monitoring method based on the Internet of Things as claimed in any one of claims 1 to 7 is implemented.

Images