Cultivated land quality monitoring system based on cultivated land protection

Abstract

The invention provides a cultivated land quality monitoring system based on cultivated land protection, and relates to the technical field of agricultural management monitoring, and the system comprises the following modules: a data collection module which collects biological activity data including enzyme activity, earthworm biomass, microbial respiration rate and earth surface hardening index; the biological activity analysis module is used for constructing a biological activity digital model and outputting a soil biological activity index of the cultivated land to be detected; the degradation correction module is used for establishing degradation grades; and the quality evaluation module is used for correcting and adjusting the soil biological activity index based on the obtained degradation grading index, and outputting an evaluation result through the soil biological activity correction index. According to the method, multi-dimensional biological activity data such as enzyme activity, earthworm biomass, microbial respiration rate and earth surface hardening index are collected, and data consistency is ensured by using a standardization method, so that the limitation that a conventional technology depends on a single index is avoided; the defect that a traditional method neglects the influence of climate change accumulation is overcome.

Description

Technical Field

[0001] This invention relates to the field of agricultural management and monitoring technology, specifically to a farmland quality monitoring system based on farmland protection. Background Technology

[0002] Soil biological activity is a key indicator for measuring arable land quality. It directly participates in processes such as organic matter decomposition, nutrient cycling, and structural improvement through the life activities of soil microorganisms, animals, and enzyme systems. On the one hand, microorganisms decompose organic matter, releasing nutrients such as nitrogen and phosphorus, promoting soil aggregate formation, and enhancing water and fertilizer retention capacity. On the other hand, soil animals such as earthworms can open pores and improve aeration, while enzyme activity drives nutrient transformation and inhibits pathogen reproduction. Decreased biological activity leads to soil compaction, fertility decline, and ecological imbalance. Conversely, high biological activity enhances arable land productivity, resilience, and sustainability, serving as a crucial foundation for achieving high-quality, high-yield agriculture and ecological security. Assessing the intensity of soil biological activity can be directly applied to arable land protection and to evaluate the soil quality within a given area.

[0003] In the prior art, CN117036087A discloses a farmland quality monitoring system based on farmland protection. This technology includes a front-end, a terminal, and a cloud platform. The front-end, terminal, and cloud platform are connected via internet communication. The cloud platform includes a data acquisition module, a main control module, a data processing module, a data storage module, a data output module, an evaluation model, a suitable crop recommendation system, and a land improvement recommendation module. The data acquisition module, main control module, data processing module, data storage module, data output module, and evaluation model are sequentially connected in communication. The evaluation model, suitable crop recommendation system, and land improvement recommendation module are also sequentially connected in communication. This system enables rapid assessment of farmland quality, facilitating understanding of the soil during the planting process, and also recommends suitable crops based on farmland quality.

[0004] However, the existing technologies mentioned above rely on single or limited bioactivity indicators for monitoring arable land quality, resulting in incomplete monitoring results that fail to fully reflect the health status of arable land. Furthermore, conventional technologies generally ignore the cumulative effects of environmental stress factors and fail to integrate climate stress into the assessment, making the monitoring results unsuitable for adapting to climate change. In addition, conventional methods do not support flexible handling of missing data, leading to assessment results that are either too high or too low. Their evaluation systems are mostly static thresholds, based on fixed standards rather than dynamic calibration, and cannot be updated and warned in real time using historical data. Ultimately, this results in arable land quality assessments lacking scientific rigor, adaptability, and practicality, making it difficult to provide effective protection recommendations.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a farmland quality monitoring system based on farmland protection, to address the problems mentioned in the background section. This invention collects multidimensional bioactivity data, including enzyme activity, earthworm biomass, microbial respiration rate, and soil compaction index, and uses standardized methods to ensure data consistency, avoiding the limitations of conventional techniques that rely on single indicators. Simultaneously, the system integrates environmental stress factors through a degradation correction module, calculates the degradation index, and dynamically corrects the soil bioactivity index, overcoming the shortcomings of traditional methods that ignore the cumulative effects of climate change. This provides a more scientific, adaptable, and practical monitoring solution.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A farmland quality monitoring system based on farmland protection includes the following modules:

[0009] Data acquisition module: Divide the cultivated land area into a square with a side length of M, set up representative sampling points in the cultivated land area, and collect bioactivity data from each representative sampling point. The bioactivity data includes enzyme activity, earthworm biomass, microbial respiration rate and surface compaction index. All data are collected in a unified time period.

[0010] Bioactivity Analysis Module: Constructs a digital model of bioactivity using a neural network convolutional structure, including an input layer, a hidden layer, and an output layer. The input layer receives bioactivity data, the hidden layer processes the bioactivity data, and the output layer outputs the soil bioactivity index of the cultivated land to be tested.

[0011] Deterioration correction module: Obtain the difference between the daily maximum and minimum temperatures and the drought index of the farmland area to be monitored during the Y-day period before monitoring, calculate the daily average temperature difference fluctuation range and the average drought index during the Y-day period, calculate the deterioration index of farmland soil biological activity through the daily average temperature difference fluctuation range and the average drought index, establish a deterioration grading system, and obtain the corresponding deterioration grading index by comparing the calculated deterioration index of the farmland area to be monitored with the deterioration grading system.

[0012] Quality assessment module: Based on the obtained deterioration grading index, the soil biological activity index is corrected and adjusted to generate a soil biological activity correction index adapted to environmental changes. Finally, the quality assessment results of the cultivated land under test are output through the soil biological activity correction index.

[0013] Furthermore, in the data acquisition module, the cultivated land area to be tested is divided by grid division method, soil samples are collected at each representative sampling point, and the activity of key enzymes is determined by fluorescence method. The key enzymes include urease and phosphatase. Then, earthworms are manually sorted by excavating a quantitative soil volume, dried, weighed, and the earthworm biomass is calculated.

[0014] The CO2 release of soil samples was measured to calculate the microbial respiration rate. Finally, the soil samples were sieved, and the weight percentage of compaction after sieving was calculated and normalized. Then, the samples were graded into ten levels from 0 to 9 as the final surface compaction index. The collection of each biological activity data was completed within the same time period.

[0015] Furthermore, the microbial respiration rate is calculated through the following process: after removing visible plant debris and stones from the collected fresh soil sample, it is quickly sieved and the humidity is adjusted to 50%±5%. The treated soil is weighed and placed into a sealed culture bottle. At the same time, a small container containing NaOH standard solution is placed in the bottle to absorb CO2. The sealed culture bottle is placed in a constant temperature and dark environment for 24 hours.

[0016] After the culture was completed, the small container was removed, and excess BaCl2 solution was added to precipitate carbonate ions. The remaining NaOH was titrated with standard hydrochloric acid solution using phenolphthalein as an indicator. The mass of CO2 absorbed was calculated based on the amount of NaOH consumed. Combined with the soil dry weight and culture time, the CO2 release rate per unit mass of soil per unit time was calculated to obtain the final microbial respiration rate.

[0017] Furthermore, in the bioactivity analysis module, the soil bioactivity index is calculated using the following formula:

[0018]

[0019] in:

[0020] Soil biological activity index;

[0021] Enzyme activity is used to reflect the efficiency of soil organic matter decomposition and nutrient cycling.

[0022] Earthworm biomass is used to indicate soil structural stability and the intensity of biological disturbance.

[0023] Microbial respiration rate characterizes microbial metabolic activity and organic matter mineralization capacity;

[0024] The soil surface compaction index is used to describe the degree of degradation of soil physical structure. The formula contains... The attenuation factor is based on the surface compaction index after conversion. The value range is [0, 9];

[0025] This is the basic attenuation value, used to adjust the degree of influence after the surface compaction index is converted into an attenuation factor;

[0026] If no earthworm biomass is collected at the representative sampling point to be tested, the following formula is used for calculation instead:

[0027]

[0028] in: The correction coefficient is used to adjust the basic attenuation value. When earthworm biomass loss causes bioturbation, it mitigates the surface compaction index and leads to an overestimation of the soil biological activity index. This correction coefficient is used to weaken the overestimation and improve the accuracy of the final calculated soil biological activity index. The range of values ​​for is: .

[0029] Furthermore, in the bioactivity analysis module, the input layer receives four indicators from representative sampling points in step S1: enzyme activity, earthworm biomass, microbial respiration rate, and surface compaction index. The data is input in vector form. The hidden layer uses a convolutional neural network structure for processing, including convolutional layers, activation function layers, and pooling layers. The convolutional layers use multiple 5x1 convolutional kernels to adapt to one-dimensional data features. The input vector is weighted and summed through a sliding window, and then local features are extracted.

[0030] The feature dimensionality is reduced by using a max pooling layer to retain key information and reduce the risk of overfitting. Then, the processed feature data is passed to a fully connected layer for further integration, and latent patterns are calculated using weight matrices and bias terms. Finally, the output layer generates a comprehensive soil bioactivity index through a linear activation function.

[0031] The soil bioactivity index generated by the output layer is normalized and output as a standard value between 0 and 1. The convolutional neural network structure is trained by model training, supervised learning based on historical farmland datasets, and iteratively optimized using the mean squared error loss function.

[0032] Furthermore, in the degradation correction module, the degradation index of arable land soil biological activity is calculated using the following formula:

[0033]

[0034] in:

[0035] The degradation index of soil biological activity in the cultivated land to be tested;

[0036] The number of consecutive days prior to monitoring, with a value range of [31, 60]; The consecutive days before monitoring are the sequence number, with a value range of [1, 30], and the value of the consecutive days sequence number is determined by the number of consecutive days before monitoring.

[0037] For the first The highest temperature of the day; For the first The lowest temperature of the day;

[0038] The average drought index;

[0039] The average drought index After continuously collecting precipitation, temperature and soil moisture data, a standardized drought model is used to calculate the daily drought index value. The obtained daily drought index value is normalized to a 1-10 scale. The sum of all drought indices and the result is divided by the total number of days Y to quantify the cumulative impact of environmental stress on soil biological activity.

[0040] Furthermore, the average drought index is calculated using the following formula:

[0041]

[0042] in:

[0043] For the first The drought index value for the day;

[0044] The minimum drought index value within the total number of days Y; The maximum drought index value within the total number of days Y;

[0045] After normalizing the daily drought index values, they are mapped to a 1-10 scale. If the calculated result exceeds the boundary, it is truncated to the [1,10] interval. This ensures that all data are converted to a uniform and comparable scale to quantify the intensity of drought stress, and when:

[0046] When the time period is Y days prior to monitoring, it means that there was no drought in the cultivated land area to be monitored, and the degradation index obtained at this time is least affected by drought.

[0047] when The time period represents the period Y days prior to monitoring when the cultivated land area under test was in a state of extreme drought. The degradation index obtained at this time is most affected by drought.

[0048] Furthermore, in the quality evaluation module, the soil biological activity correction index is calculated using the following formula: ;in: Soil biological activity correction index; To match the coefficient values ​​of the degradation index;

[0049] pass The degradation index is converted into a degradation factor, and then the coefficient value of the degradation index is used. Inverse relationship with soil biological activity index Corrective treatment is performed to generate a soil biological activity correction index. .

[0050] Furthermore, the soil biological activity correction index Mapped to a five-level farmland quality evaluation system, the threshold boundaries are dynamically calibrated by combining historical data, and the final evaluation results with deterioration markers are output. Simultaneously, a farmland protection measure suggestion library is generated with linked prompts. The evaluation process ensures that the evaluation results reflect the superposition of biological activity background value and climate stress through quantitative compensation of environmental stress factors. Moreover, the correction coefficient library supports seasonal updates by crop type.

[0051] Furthermore, the five-level farmland quality evaluation system integrates farmland quality monitoring data from the entire process, determines initial boundaries through cluster analysis, adjusts the boundary points by introducing environmental stress factor weights, constructs a verification mechanism, and uses field sampling feedback and model output results for regression testing to form a standardized evaluation system with correction rules. This system is then embedded into the farmland protection and management platform to achieve tiered early warning.

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] This farmland quality monitoring system significantly improves the accuracy and comprehensiveness of farmland protection. By collecting multi-dimensional bioactivity data such as enzyme activity, earthworm biomass, microbial respiration rate, and surface compaction index, and using standardized methods to ensure data consistency, it avoids the limitations of conventional technologies that rely on single indicators. Simultaneously, the system integrates environmental stress factors through a degradation correction module, calculates a degradation index, and dynamically corrects the soil bioactivity index, overcoming the shortcomings of traditional methods that ignore the cumulative impact of climate change. The system also supports data missing handling and dynamic evaluation mechanisms, enabling farmland protection recommendations with early warnings. Finally, through quantitative compensation of environmental stress factors, it ensures that the evaluation results simultaneously reflect the superposition of biological background values ​​and climate stress, providing a more scientific, adaptable, and practical monitoring solution. Attached Figure Description

[0054] Figure 1 This is a system block diagram of a farmland quality monitoring system based on farmland protection according to the present invention.

[0055] Figure 2 This is a flowchart of the operation of a farmland quality monitoring system based on farmland protection. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0057] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0058] Example:

[0059] Please see Figures 1-2 The present invention provides the following technical solutions:

[0060] A farmland quality monitoring system based on farmland protection includes the following modules:

[0061] Data Acquisition Module: A square area of ​​cultivated land to be tested, with sides of 20m × 20m, is divided. Multiple representative sampling points are scientifically established within this square area to ensure reasonable distribution and comprehensive coverage. Next, the required bioactivity data are systematically collected from each designated representative sampling point. These data specifically include enzyme activity reflecting soil biochemical processes, earthworm biomass indicating biodiversity and structural stability, microbial respiration rate characterizing microbial metabolic intensity, and surface compaction index describing soil physical state. Finally, all the above data collection work must be completed synchronously within a strictly set unified time period to ensure timeliness and eliminate the potential impact of time factors on data consistency.

[0062] In the data acquisition module, a grid division method is used to divide the cultivated land area to be tested, ensuring uniform spatial distribution and comprehensive coverage of representative sampling points. Next, standardized soil samples are accurately collected at each selected representative sampling point, and sterile tools are used to control the depth and volume to maintain sample consistency. Subsequently, the activity of key enzymes is determined using fluorescence methods, specifically involving fluorescent substrate reactions and optical detection equipment. The key enzymes measured include urease and phosphatase. Then, a soil sample of a preset volume is excavated at the sampling point, all visible earthworm individuals are manually sorted and cleaned to remove impurities, and the earthworm samples are dried in a constant temperature oven to constant weight before being accurately weighed. Finally, the earthworm biomass is calculated.

[0063] The amount of CO2 gas released from soil samples under specific conditions was accurately measured using a closed-system laboratory incubation method, and the microbial respiration rate was calculated accordingly. Then, the same batch of soil samples was sieved through a standard sieve device to separate the compacted material. Next, the remaining compacted particles on the sieve were collected, dried, and accurately weighed to calculate the percentage of their weight relative to the total weight of the original soil sample. This percentage data was then normalized to convert it into a standard numerical range. Based on this, the normalization results were divided into ten equally spaced levels, forming a continuous rating system from 0 (representing no compaction) to 9 (representing severe compaction), serving as the final surface compaction index. Simultaneously, all the above-mentioned biological activity data collection operations were completed synchronously within a strictly unified preset time period to ensure the timeliness and consistency of data collection.

[0064] The microbial respiration rate was calculated through the following process: First, the freshly collected soil sample was preliminarily cleaned to thoroughly remove all visible plant debris, root fragments, and other stone impurities, ensuring the sample was pure and undisturbed. Next, the cleaned sample was quickly passed through a standard aperture sieve (in this embodiment, a 2mm sieve was used) to separate fine particles. Subsequently, the soil moisture was adjusted to the target level (50% ± 5%) using a precision hygrometer, achieving humidity equilibrium by adding sterile water or controlling the drying environment. Then, 200 grams of the treated soil sample was accurately weighed and carefully placed into a pre-prepared sealed glass culture bottle. Simultaneously, a glass beaker containing a known concentration of NaOH standard solution was placed inside the bottle to absorb CO2 gas released during subsequent culture. Finally, the sealed culture bottle was transferred to a constant temperature, in this embodiment, 25°C, dark incubator, and cultured continuously for 24 hours to ensure that microbial activity occurred under stable, light-free conditions.

[0065] After the 24-hour incubation period, the small container containing the NaOH standard solution in the sealed culture bottle was manually removed. Then, an excess of BaCl2 solution was added to this small container to induce a reaction between carbonate ions and barium ions, forming an insoluble barium carbonate precipitate, ensuring the precipitation process was complete. Subsequently, phenolphthalein indicator was added, and the remaining liquid in the container was slowly titrated with a standard hydrochloric acid solution of known concentration. The color change was closely observed, and the titration endpoint was determined when the solution faded from pink to colorless. Based on the volume and concentration of standard hydrochloric acid consumed at the titration endpoint, the amount of remaining NaOH was accurately calculated, and the mass of absorbed CO2 was then deduced. Next, combining the pre-measured dry weight of the soil sample and the 24-hour incubation time, obtained by drying and weighing, the CO2 release rate per unit mass of soil per unit time was comprehensively calculated. Finally, through this series of calculations, the accurate microbial respiration rate value was obtained.

[0066] In the bioactivity analysis module, the input layer first receives a standardized data stream from step S1. This data stream contains four core indicators collected from each representative sampling point: enzyme activity, earthworm biomass, microbial respiration rate, and surface compaction index. These indicator data are preprocessed and input into the system as fixed-dimensional four-dimensional feature vectors. The hidden layer uses a convolutional neural network (CNN) architecture for deep feature processing, and its processing flow includes three core operations:

[0067] Convolutional layer processing: Deploy multiple 5×1 convolutional kernels to perform local feature scanning based on the characteristics of one-dimensional vector data; Each convolutional kernel moves step by step along the input vector through a sliding window mechanism to perform weighted sum calculations to extract local pattern features;

[0068] Nonlinear activation: A ReLU activation function layer is introduced after the output of the convolutional layer to apply a nonlinear transformation to the feature map to enhance the model's expressive power;

[0069] Feature dimensionality reduction: The activated features are downsampled by a max pooling layer, which preserves key feature responses while compressing data dimensionality, significantly reducing subsequent computational complexity and the risk of overfitting.

[0070] The entire process leverages the local perceptual properties of convolutional kernels to efficiently extract hidden bioactivity association patterns from the input vector.

[0071] The feature map is downsampled using a max-pooling layer, selecting the maximum value within the local receptive field as the representative output. This effectively reduces the feature dimensionality while retaining the most significant feature information, suppressing noise interference and reducing the risk of model overfitting. Subsequently, the sparsed feature data after dimensionality reduction is passed to a fully connected layer. In this layer, all neurons establish global connections with the output of the previous layer. Through linear transformation of the weight matrix and adjustment of the bias terms, the nonlinear relationships between different features are deeply integrated, thereby calculating and mining the complex latent patterns hidden in the data. Finally, a linear activation function, i.e., identity transformation, is applied to the integrated high-dimensional features in the output layer, directly mapping and generating a continuous numerical index that comprehensively represents soil biological activity. This index serves as the final output of the end-to-end processing.

[0072] After generating the original soil bioactivity index, the output layer performs normalization on the index, mapping the values ​​to a uniform continuous range of 0-1 using a linear scaling method to form standardized and comparable output values. This aims to eliminate dimensional differences and improve the generality of the results. Secondly, the optimization of the convolutional neural network structure depends on the model training process, which implements a supervised learning strategy based on historical farmland datasets, using a large number of labeled samples as input-output pairs to adjust parameters. Then, during training, the mean squared error loss function is applied to accurately quantify the deviation between the predicted and actual values. The network weights are repeatedly adjusted through an iterative optimization algorithm of gradient descent, gradually converging to the minimum loss state, ultimately achieving a stable improvement in model performance.

[0073] Bioactivity Analysis Module: Constructs a digital model of bioactivity using a neural network convolutional structure, including an input layer, a hidden layer, and an output layer. The input layer receives bioactivity data, the hidden layer processes the bioactivity data, and the output layer outputs the soil bioactivity index of the cultivated land to be tested.

[0074] The soil biological activity index is calculated using the following formula:

[0075]

[0076] in:

[0077] Soil biological activity index;

[0078] Enzyme activity is used to reflect the efficiency of soil organic matter decomposition and nutrient cycling; that is, the higher the enzyme activity value, the higher the soil biological activity index. The higher the level, the more it can directly improve. The overall value. Enzyme activity reflects the efficiency of key biochemical processes in the soil (such as the activity of urease and phosphatase). Higher activity indicates more efficient organic matter decomposition and nutrient cycling;

[0079] Earthworm biomass is used to indicate soil structure stability and the intensity of biological disturbance. Earthworm biomass indicates the stability of soil structure and the intensity of biological disturbance, and has a positive feedback effect on soil structure.

[0080] Microbial respiration rate is used to characterize microbial metabolic activity and organic matter mineralization capacity. A high CO2 release rate indicates vigorous microbial activity. When soil degrades, the positive contribution of microbial respiration is partially offset to prevent the index from being overly optimistic.

[0081] The soil surface compaction index is used to describe the degree of degradation of soil physical structure. The formula contains... The attenuation factor is based on the surface compaction index after conversion. The value range is [0,9]; the effect is to reverse the impact of physical degradation on biological activity. Environmental stress is reflected by reducing the weight of microbial respiration rate.

[0082] The basic attenuation value is used to adjust the degree of influence after the surface compaction index is converted into an attenuation factor. The basic attenuation value is used to adjust the magnitude of the influence.

[0083] If no earthworm biomass is collected at the representative sampling point to be tested, the following formula is used for calculation instead:

[0084]

[0085] in: The correction coefficient is used to adjust the basic attenuation value. When earthworm biomass loss causes bioturbation, it mitigates the surface compaction index and leads to an overestimation of the soil biological activity index. This correction coefficient is used to weaken the overestimation and improve the accuracy of the final calculated soil biological activity index. The range of values ​​for is: .

[0086] Deterioration correction module: Obtain the difference between the daily maximum and minimum temperatures and the drought index of the farmland area to be monitored during the Y-day period before monitoring, calculate the daily average temperature difference fluctuation range and the average drought index during the Y-day period, calculate the deterioration index of farmland soil biological activity through the daily average temperature difference fluctuation range and the average drought index, establish a deterioration grading system, and obtain the corresponding deterioration grading index by comparing the calculated deterioration index of the farmland area to be monitored with the deterioration grading system.

[0087] In the degradation correction module, the degradation index of arable land soil biological activity is calculated using the following formula:

[0088]

[0089] in:

[0090] The degradation index of soil biological activity in the cultivated land to be tested;

[0091] The number of consecutive days prior to monitoring, with a value range of [31, 60]; This represents the sequence number of consecutive days prior to monitoring, with a value ranging from [1, 30]. The value of the consecutive day sequence number is determined by the number of consecutive days prior to monitoring. The negative correlation design, which represents the duration of environmental stress, is derived from the dilution effect of stress and converts the total stress into the daily average impact intensity, making the monitoring results of different durations comparable.

[0092] For the first The highest temperature of the day; For the first The lowest temperature of the day, temperature difference It is positively correlated with the degradation index. The greater the daily temperature difference, the higher the contribution of stress on that day.

[0093] The average drought index increases the "efficiency" of a unit temperature difference, for example, when... At that time, the contribution of temperature difference is reduced to 1 / 10, but because the actual destructive power of temperature difference is doubled, a higher value compensation is needed to maintain the same amount of soil biological activity degradation index of arable land.

[0094] The average drought index After continuously collecting precipitation, temperature and soil moisture data, a standardized drought model is used to calculate the daily drought index value. The obtained daily drought index value is normalized to a 1-10 scale. The sum of all drought indices and the result is divided by the total number of days Y to quantify the cumulative impact of environmental stress on soil biological activity.

[0095] The average drought index is calculated using the following formula:

[0096]

[0097] in:

[0098] For the first The drought index value of a day reflects the balance between water supply and demand on that day;

[0099] The minimum drought index value within the total number of days Y is used as a dynamic benchmark: it varies with the drought extreme value within the monitoring period; The maximum drought index value within the total number of days Y reflects the optimal water conditions within the cycle.

[0100] After normalizing the daily drought index values, they are mapped to a 1-10 scale. If the calculated result exceeds the boundary, it is truncated to the [1,10] interval. This ensures that all data are converted to a uniform and comparable scale to quantify the intensity of drought stress, and when:

[0101] When the time period is Y days prior to monitoring, it means that there was no drought in the cultivated land area to be monitored, and the degradation index obtained at this time is least affected by drought.

[0102] when The time period represents the period Y days prior to monitoring when the cultivated land area under test was in a state of extreme drought. The degradation index obtained at this time is most affected by drought.

[0103] Quality assessment module: Based on the obtained deterioration grading index, the soil biological activity index is corrected and adjusted to generate a soil biological activity correction index adapted to environmental changes. Finally, the quality assessment results of the cultivated land under test are output through the soil biological activity correction index.

[0104] In the quality assessment module, the soil biological activity correction index is calculated using the following formula: ;in: Soil biological activity correction index; To match the coefficient values ​​of the degradation index;

[0105] pass The degradation index is converted into a degradation factor, and then the coefficient value of the degradation index is used. Inverse relationship with soil biological activity index Corrective treatment is performed to generate a soil biological activity correction index. .

[0106] Calculated soil biological activity correction index Mapped to a five-level farmland quality evaluation system, based on preset threshold rules, the system categorizes farmland quality from best to worst by comparing index values ​​with the boundaries of each level. Simultaneously, it performs dynamic calibration by incorporating historical farmland quality monitoring datasets, periodically and automatically adjusting threshold boundaries to adapt to long-term trends. Finally, it outputs a final evaluation result with a degradation marker, which is directly appended to the evaluation level based on degradation grading indicators to highlight the risk.

[0107] A synchronized triggering mechanism for the farmland protection measures recommendation database is implemented. This database contains customized protection plans for different quality levels, such as crop rotation or fertilization recommendations, and provides real-time push notifications. The entire evaluation process utilizes a quantitative compensation mechanism for environmental stress factors, such as drought or temperature fluctuations, to ensure that the final output not only reflects the baseline level of soil biological activity but also accurately covers the superposition of climate stresses. Furthermore, the system's built-in correction coefficient database supports regular updates and maintenance based on major crop types and seasonal cycles. In constructing the five-level farmland quality evaluation system, comprehensive integration of monitoring data from data collection to quality evaluation is achieved. Cluster analysis algorithms are used to process historical samples to scientifically determine the initial level division boundaries.

[0108] Further, environmental stress factor weighting coefficients are introduced, and the boundary points are dynamically adjusted to enhance the sensitivity of stress impacts. A comprehensive verification mechanism is constructed, using feedback data from field sampling points and model output results to perform regression analysis, identify biases, and iteratively optimize. Ultimately, a standardized evaluation system with periodic correction rules is formed, such as annual parameter review, which is seamlessly integrated into the farmland protection management platform, enabling tiered early warning functions through a visual interface.

[0109] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0110] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0111] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0112] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A farmland quality monitoring system based on farmland protection, characterized in that, Includes the following modules: Data acquisition module: Divide the cultivated land area into a square with a side length of M, set up representative sampling points in the cultivated land area, and collect bioactivity data from each representative sampling point. The bioactivity data includes enzyme activity, earthworm biomass, microbial respiration rate and surface compaction index. All data are collected in a unified time period. Bioactivity Analysis Module: Constructs a digital model of bioactivity using a neural network convolutional structure, including an input layer, a hidden layer, and an output layer. The input layer receives bioactivity data, the hidden layer processes the bioactivity data, and the output layer outputs the soil bioactivity index of the cultivated land to be tested. Deterioration correction module: Obtain the difference between the daily maximum and minimum temperatures and the drought index of the farmland area to be monitored during the Y-day period before monitoring, calculate the daily average temperature difference fluctuation range and the average drought index during the Y-day period, calculate the deterioration index of farmland soil biological activity through the daily average temperature difference fluctuation range and the average drought index, establish a deterioration grading system, and obtain the corresponding deterioration grading index by comparing the calculated deterioration index of the farmland area to be monitored with the deterioration grading system. Quality assessment module: Based on the obtained deterioration grading index, the soil biological activity index is corrected and adjusted to generate a soil biological activity correction index adapted to environmental changes. Finally, the quality assessment results of the cultivated land under test are output through the soil biological activity correction index.

2. The farmland quality monitoring system based on farmland protection according to claim 1, characterized in that: In the data acquisition module, the cultivated land area to be tested is divided by grid division method, soil samples are collected at each representative sampling point, and the activity of key enzymes is determined by fluorescence method. The key enzymes include urease and phosphatase. Then, earthworms are manually sorted by excavating a quantitative soil volume, dried, weighed and the earthworm biomass is calculated. The CO2 release of soil samples was measured to calculate the microbial respiration rate. Finally, the soil samples were sieved, and the weight percentage of compaction after sieving was calculated and normalized. Then, the samples were graded into ten levels from 0 to 9 as the final surface compaction index. The collection of each biological activity data was completed within the same time period.

3. The farmland quality monitoring system based on farmland protection according to claim 2, characterized in that, The microbial respiration rate was calculated through the following process: After removing visible plant debris and stones from the collected fresh soil sample, it was quickly sieved and the humidity was adjusted to 50%±5%. The treated soil was weighed and placed into a sealed culture bottle. At the same time, a small container containing NaOH standard solution was placed in the bottle to absorb CO2. The sealed culture bottle was placed in a constant temperature and dark environment for 24 hours. After the culture was completed, the small container was removed, and excess BaCl2 solution was added to precipitate carbonate ions. The remaining NaOH was titrated with standard hydrochloric acid solution using phenolphthalein as an indicator. The mass of CO2 absorbed was calculated based on the amount of NaOH consumed. Combined with the soil dry weight and culture time, the CO2 release rate per unit mass of soil per unit time was calculated to obtain the final microbial respiration rate.

4. A farmland quality monitoring system based on farmland protection according to claim 2, characterized in that: In the bioactivity analysis module, the soil bioactivity index is calculated using the following formula: in: Soil biological activity index; Enzyme activity is used to reflect the efficiency of soil organic matter decomposition and nutrient cycling. Earthworm biomass is used to indicate soil structural stability and the intensity of biological disturbance. Microbial respiration rate characterizes microbial metabolic activity and organic matter mineralization capacity; The soil surface compaction index is used to describe the degree of degradation of soil physical structure. The formula contains... The attenuation factor is based on the surface compaction index after conversion. The value range is [0, 9]; This is the basic attenuation value, used to adjust the degree of influence after the surface compaction index is converted into an attenuation factor; If no earthworm biomass is collected at the representative sampling point to be tested, the following formula is used for calculation instead: in: The correction coefficient is used to adjust the basic attenuation value. When earthworm biomass loss causes bioturbation, it mitigates the surface compaction index and leads to an overestimation of the soil biological activity index. This correction coefficient is used to weaken the overestimation and improve the accuracy of the final calculated soil biological activity index. The range of values ​​for is: .

5. A farmland quality monitoring system based on farmland protection according to claim 4, characterized in that: In the bioactivity analysis module, the input layer receives four indicators from representative sampling points in step S1: enzyme activity, earthworm biomass, microbial respiration rate, and surface compaction index. The data is input in vector form. The hidden layer uses a convolutional neural network structure for processing, including convolutional layers, activation function layers, and pooling layers. The convolutional layers use multiple 5x1 convolutional kernels to adapt to one-dimensional data features. The input vector is weighted and summed through a sliding window, and then local features are extracted. The feature dimensionality is reduced by using a max pooling layer to retain key information and reduce the risk of overfitting. Then, the processed feature data is passed to a fully connected layer for further integration, and latent patterns are calculated using weight matrices and bias terms. Finally, the output layer generates a comprehensive soil bioactivity index through a linear activation function. The soil bioactivity index generated by the output layer is normalized and output as a standard value between 0 and 1. The convolutional neural network structure is trained by model training, supervised learning based on historical farmland datasets, and iteratively optimized using the mean squared error loss function.

6. A farmland quality monitoring system based on farmland protection according to claim 2, characterized in that, In the degradation correction module, the degradation index of arable land soil biological activity is calculated using the following formula: in: The degradation index of soil biological activity in the cultivated land to be tested; The number of consecutive days prior to monitoring, with a value range of [31, 60]; The consecutive days before monitoring are the sequence number, with a value range of [1, 30], and the value of the consecutive days sequence number is determined by the number of consecutive days before monitoring. For the first The highest temperature of the day; For the first The lowest temperature of the day; The average drought index; The average drought index After continuously collecting precipitation, temperature and soil moisture data, a standardized drought model is used to calculate the daily drought index value. The obtained daily drought index value is normalized to a 1-10 scale. The sum of all drought indices and the result is divided by the total number of days Y to quantify the cumulative impact of environmental stress on soil biological activity.

7. A farmland quality monitoring system based on farmland protection according to claim 6, characterized in that, The average drought index is calculated using the following formula: in: For the first The drought index value for the day; The minimum drought index value within the total number of days Y; The maximum drought index value within the total number of days Y; After normalizing the daily drought index values, they are mapped to a 1-10 scale. If the calculated result exceeds the boundary, it is truncated to the [1,10] interval. This ensures that all data are converted to a uniform and comparable scale to quantify the intensity of drought stress, and when: When the time period is Y days prior to monitoring, it means that there was no drought in the cultivated land area to be monitored, and the degradation index obtained at this time is least affected by drought. when The time period represents the period Y days prior to monitoring when the cultivated land area under test was in a state of extreme drought. The degradation index obtained at this time is most affected by drought.

8. A farmland quality monitoring system based on farmland protection according to claim 1, characterized in that: In the quality assessment module, the soil biological activity correction index is calculated using the following formula: ;in: Soil biological activity correction index; To match the coefficient values ​​of the degradation index; pass The degradation index is converted into a degradation factor, and then the coefficient value of the degradation index is used. Inverse relationship with soil biological activity index Corrective treatment is performed to generate a soil biological activity correction index. .

9. A farmland quality monitoring system based on farmland protection according to claim 8, characterized in that: Soil biological activity correction index Mapped to a five-level farmland quality evaluation system, the threshold boundaries are dynamically calibrated by combining historical data, and the final evaluation results with deterioration markers are output. At the same time, a farmland protection measure suggestion library is generated to provide linkage prompts. The evaluation process ensures that the evaluation results reflect the superposition of biological activity background value and climate stress through quantitative compensation of environmental stress factors.

10. A farmland quality monitoring system based on farmland protection according to claim 9, characterized in that: The five-level farmland quality evaluation system integrates data from the entire farmland quality monitoring process, determines the initial boundary through cluster analysis, adjusts the boundary point by introducing environmental stress factor weights, constructs a verification mechanism, and uses field sampling feedback and model output results for regression testing to form a standardized evaluation system with correction rules. This system is then embedded into the farmland protection and management platform to achieve tiered early warning.

Images