Ploughing system toughness evaluation and optimization method facing composite interference

Abstract

The invention discloses a composite interference-oriented cultivated land system toughness evaluation and optimization method, and relates to the technical field of cultivated land toughness optimization, and the method comprises the steps: collecting and preprocessing multi-source cultivated land data, constructing a toughness evaluation index system, and calculating the weight of each index; and calculating a comprehensive toughness score based on weight distribution, drawing a time sequence change curve based on the score, and performing region division to generate a toughness distribution diagram. According to the method, the single-factor pollution index, the Nemo comprehensive index and the interaction effect index of heavy metal pollution are combined, the soil pollution condition is accurately quantified through a weighted summation method, a scientific and reliable basis is provided for toughness evaluation, the dynamic toughness distribution diagram is generated in combination with the LSTM model and the GIS technology, the spatio-temporal dynamic characteristics of a cultivated land system under composite interference are revealed, and the method has the advantages of being high in reliability and high in reliability. According to the method, the defects of an existing model in space-time analysis are overcome, differentiated optimization strategies are formulated based on region division and weak region labeling, and the resource allocation efficiency and the pertinence of an optimization scheme are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of farmland resilience optimization, and in particular to a farmland system resilience assessment and optimization method facing composite disturbances. Background Art

[0002] In recent years, with the intensification of global climate change and the increase in the intensity of human activities, the risk of compound disturbance faced by agricultural cultivated land systems has increased significantly. The assessment and optimization of cultivated land productivity and ecological resilience have gradually become an important research field for sustainable agricultural development. The assessment of cultivated land system resilience aims to measure the ability of cultivated land to maintain production functions and restore ecological balance after being disturbed. Its optimization methods usually involve multi-source data collection and analysis, model construction, and regional management strategy design. Traditional methods mainly rely on a single indicator to evaluate the health of cultivated land, while technological advances in recent years have made multi-dimensional assessment possible. For example, the development of GIS and remote sensing technology supports the precise monitoring of the spatial distribution and dynamic changes of cultivated land systems, and the introduction of machine learning algorithms has significantly improved the accuracy and robustness of resilience prediction. At the same time, with the popularization of database technology and visualization tools, cultivated land optimization strategies have gradually realized real-time data-driven and interactive decision-making. Despite this, existing technologies still face many challenges in the implementation of comprehensive assessment and optimization strategies. For example, in complex ecosystems, the interactive effects of different indicators are difficult to quantify, and existing models do not adequately describe the dynamic response mechanism of compound disturbances. In addition, the optimization strategy after regional division lacks pertinence and fails to formulate differentiated management plans based on regional characteristics. Especially under the combined effects of complex disturbances (such as climate change, soil pollution and social intervention), the existing resilience assessment framework is difficult to fully describe the spatiotemporal dynamic characteristics of cultivated land systems, resulting in the limitation of the practicality and operability of optimization strategies. Summary of the invention

[0003] In view of the above existing problems, the present invention is proposed.

[0004] Therefore, the present invention provides a method for evaluating and optimizing the resilience of cultivated land systems oriented to complex disturbances, which solves the problem that the existing resilience assessment framework is difficult to fully describe the spatiotemporal dynamic characteristics of the cultivated land system, resulting in the limitation of the practicability and operability of the optimization strategy.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides a method for evaluating and optimizing the resilience of a farmland system under complex disturbances, which comprises:

[0007] Collect multi-source cultivated land data and preprocess them, build a resilience assessment index system and calculate the weight of each index; calculate the comprehensive resilience score based on the weight distribution, draw the time series change curve based on the score and generate a resilience distribution map by regional division; optimize according to the regional division results, visualize the data and store the data generated during the optimization process.

[0008] As a preferred solution of the method for evaluating and optimizing the resilience of cultivated land systems facing complex disturbances described in the present invention, the method comprises the following steps: collecting multi-source cultivated land data and performing preprocessing thereof refers to using a stratified random sampling method to divide sampling units according to regional soil types and geomorphological characteristics, collecting surface soil at each sampling point, and removing coarse particles after naturally air-drying the soil samples, analyzing the soil organic matter content by potassium dichromate method, detecting the heavy metal content by ICP-MS, and taking the average value after detecting each sample a times as the final concentration, obtaining the average annual precipitation in the past A years from the meteorological monitoring station in the region, and performing time series analysis on the collected data, extracting annual irrigation records from the agricultural irrigation management platform, and calculating the total annual irrigation times based on the irrigation times data for each piece of cultivated land;

[0009] Preprocess the collected data.

[0010] As a preferred solution of the method for evaluating and optimizing the resilience of cultivated land systems facing complex disturbances described in the present invention, wherein: the construction of a resilience evaluation index system and calculation of the weight of each index refers to calculating the dimensions of resilience evaluation indexes, including productivity, health status and site conditions;

[0011] Said productivity includes soil organic matter content, precipitation, and irrigation probability;

[0012] The health conditions include heavy metal pollution index and vegetation coverage;

[0013] The heavy metal pollution index includes: calculating the single factor pollution index based on the pre-processed data ;

[0014] Combine the single factor pollution index and normalized data to calculate the pollution risk value ;

[0015] Generate a pollution risk matrix of the sample based on the risk value, and generate a normalized soil quality standard matrix based on the preprocessed soil data;

[0016] Calculate the difference based on the contamination risk value and calculate the minimum difference between each sample and the standard and the maximum difference ;

[0017] Calculate the correlation coefficients of factors and quality criterion levels based on difference series and extreme values;

[0018] Calculate the comprehensive correlation of each sample based on the correlation coefficient ;

[0019] The Nemero comprehensive index B of each sample was calculated based on the comprehensive correlation and single factor contamination index;

[0020] Calculate the interaction effect index D of each heavy metal factor in the sample;

[0021] Combining the Nemero comprehensive index and the interaction effect index, the final soil comprehensive pollution index G of the sample was calculated by weighted summation method;

[0022] The vegetation coverage rate refers to the NDVI value of the area obtained through remote sensing images, and the vegetation coverage rate FVC is calculated based on the NDVI value;

[0023] The site conditions include extracting the total length of regional roads through GIS software, allocating the total length of roads to grids per square kilometer according to the distribution of cultivated land, calculating the road grid density of each grid, calculating the shortest distance from cultivated land to the nearest market based on GIS buffer analysis, and calculating the slope distribution through DEM model data;

[0024] After standardizing the indicator data, a standardized evaluation matrix is ​​constructed. The rows in the matrix represent samples, and the columns represent indicators. The contribution value of each sample to the indicator is calculated. , information entropy is calculated based on contribution value , calculate the weight according to the entropy value ;

[0025] Organize experts in the field to compare each indicator pairwise, build a comparison matrix, and calculate the weight vector;

[0026] The final indicator weight W is obtained by combining the entropy weight method weight and AHP weight.

[0027] As a preferred solution of the method for evaluating and optimizing the resilience of cultivated land systems facing complex disturbances described in the present invention, the comprehensive resilience score calculated based on weight distribution refers to calculating the productivity score of each sample by selecting soil organic matter content, precipitation and irrigation probability indicators in the productivity dimension based on the weight calculation result. ;

[0028] Calculate health status score based on metal pollution index and vegetation coverage ;

[0029] Calculates site condition score based on road network density, distance to market and slope ;

[0030] The comprehensive resilience score is calculated by weighted summation based on the productivity score, health score and site condition score. .

[0031] As a preferred solution of the method for evaluating and optimizing the resilience of cultivated land systems facing complex disturbances described in the present invention, the method of drawing a time series change curve based on the score and performing regional division to generate a resilience distribution map includes:

[0032] After collecting and preprocessing historical meteorological data and socioeconomic data, the preprocessed data were combined with the comprehensive resilience score to construct a dataset;

[0033] Use the LSTM model to predict changes in resilience scores. Input the data set as a training set into the LSTM model for model training. Define the loss function and Adam optimizer to iteratively optimize the model parameters. When the loss of the LSTM model no longer decreases significantly during continuous iterations, stop iterating, output the model parameters, and update the LSTM model. Input the real-time resilience score calculation data into the LSTM model to obtain future resilience score changes.

[0034] Collect historical resilience scores, combine historical resilience scores with forecast scores, use dotted lines to represent future resilience scores, and use solid lines to represent historical resilience scores;

[0035] Calculate the annual change in resilience score using time series curve data ;

[0036] Calculate the standard deviation of the change as the change threshold H. If the change is greater than the threshold H, mark the year as a key time node.

[0037] If the change is less than 0, it means that the toughness decreases rapidly, and the node is marked as a decreasing node;

[0038] If the change is greater than 0, it means that the resilience is quickly recovered, and the node is marked as a recovery node;

[0039] Obtain the resilience score and area of ​​each plot, perform weighted average by spatial unit, and calculate the annual comprehensive resilience score for each spatial unit;

[0040] Thresholds U and O are set, and threshold U>O. If the comprehensive toughness score is greater than threshold U, it is a high toughness area. If the comprehensive toughness score is greater than or equal to threshold O and less than or equal to threshold U, it is a medium toughness area. If the comprehensive toughness score is less than threshold O, it is a low toughness area.

[0041] Using GIS software, the resilience score of each geographic unit is mapped according to a color gradient, and a pollution source distribution layer is added to mark the pollution points and areas. The corresponding resilience distribution map is generated for each year, and the annual resilience distribution map is synthesized into a dynamic interactive map using GIS software.

[0042] As a preferred solution of the method for evaluating and optimizing the resilience of cultivated land systems facing complex disturbances described in the present invention, the optimization according to the regional division results includes:

[0043] If the area is a high-resilience area, maintain the existing high-resilience state, promote ecological planting models, reduce the use of pesticides and fertilizers, strengthen ecological protection, and strengthen long-term monitoring;

[0044] If the region is a medium-resilience area, priority will be given to restoring the ecosystem, carrying out vegetation restoration, returning farmland to forest, concentrating resources on cleaning up pollution sources, and reducing heavy metal pollution concentrations;

[0045] If the area is a low-resilience area, the ecological barrier function will be enhanced, key restoration projects will be implemented, a dynamic monitoring mechanism will be established, and real-time monitoring of disturbance sources will be enhanced.

[0046] As a preferred scheme of the method for assessing and optimizing the resilience of cultivated land systems facing complex disturbances described in the present invention, the visual display of data refers to using the front-end framework React.js to build a visual interface, visually displaying the regional division results and dynamic interactive maps, and superimposing optimization strategy points on the map, marking high-resilience areas in green to display protection strategies, marking medium-resilience areas in yellow to display restoration strategies, and marking low-resilience areas in red to display priority restoration strategies.

[0047] As a preferred solution of the method for evaluating and optimizing the resilience of cultivated land systems facing complex disturbances described in the present invention, the data generated during the storage optimization process refers to storing the resilience score data, optimization strategy data and resilience area annotation data in a database in order according to timestamps, and automatically updating the database after each new resilience score data is added;

[0048] Synchronize and regularly back up the stored data, regularly perform security and integrity checks on the stored data and backup data, and generate test reports.

[0049] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for assessing and optimizing the resilience of a cultivated land system facing complex disturbances as described in the first aspect of the present invention is implemented.

[0050] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the method for assessing and optimizing the resilience of a cultivated land system oriented to complex disturbances as described in the first aspect of the present invention.

[0051] The beneficial effects of the present invention are as follows: the present invention combines the single factor pollution index of heavy metal pollution, the Nemero comprehensive index and the interaction effect index, and uses the weighted summation method to accurately quantify the soil pollution status, thereby providing a scientific and reliable basis for resilience assessment. The LSTM model and GIS technology are combined to generate a dynamic resilience distribution map, which reveals the spatiotemporal dynamic characteristics of the cultivated land system under complex interference, and makes up for the shortcomings of the existing models in spatiotemporal analysis. Based on regional division and weak area marking, differentiated optimization strategies are formulated, which significantly improves the resource allocation efficiency and the pertinence of the optimization scheme, which not only significantly improves the comprehensiveness and accuracy of the resilience assessment of the cultivated land system, but also realizes the practicality and dynamic adjustment capability of the optimization scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0053] Figure 1 This is a flow chart of the method for assessing and optimizing the resilience of cultivated land systems facing complex disturbances in Example 1. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0056] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0057] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides a method for evaluating and optimizing the resilience of a farmland system facing complex disturbances, comprising the following steps:

[0058] S1. Collect and preprocess multi-source farmland data, build a resilience assessment index system and calculate the weight of each index;

[0059] Specifically, collecting multi-source cultivated land data and preprocessing them means using stratified random sampling to divide sampling units according to regional soil types and geomorphological features, collecting surface soil at each sampling point, and removing coarse particles after naturally air-drying the soil samples. The stratified random sampling method fully considers the diversity of soil types and geomorphological features in the region by scientifically dividing the sampling units, significantly improving the comprehensiveness and representativeness of the data. This sampling method avoids the regional bias problem that may occur in traditional random sampling, and provides a high-quality data basis for subsequent analysis. The soil organic matter content is obtained by potassium dichromate analysis, and the heavy metal content is detected by ICP-MS. ICP-MS can sensitively detect the content of heavy metals such as cadmium, lead, chromium, and mercury in soil samples. The test results are accurate and reliable. Compared with traditional methods, ICP-MS It has high throughput, low detection limit and good reproducibility, and is suitable for the analysis of complex soil samples. After testing each sample a times, the average value is taken as the final concentration. The average annual precipitation in the past A years is obtained from the meteorological monitoring station in the region, and the collected data is analyzed in time series. The annual irrigation records are extracted from the agricultural irrigation management platform, and the total annual irrigation times are calculated based on the irrigation times data of each cultivated land.

[0060] Preprocess the collected data;

[0061] The preprocessing includes cleaning and standardizing the data;

[0062] The heavy metal content includes cadmium ( ),lead( ),chromium( ),mercury( ).

[0063] Through the comprehensive collection and preprocessing of multi-source data of cultivated land systems, the problems of narrow coverage, low detection accuracy and insufficient time dynamic response of traditional data collection have been solved. The stratified random sampling method effectively improves the representativeness of the samples, the potassium dichromate method and ICP-MS and other technologies ensure the accuracy of key data, and time series analysis provides long-term trend support for dynamic resilience assessment. These innovative technical means not only significantly improve the scientificity and comprehensiveness of cultivated land resilience assessment, but also lay a solid data foundation for optimizing cultivated land management strategies.

[0064] Furthermore, constructing a resilience assessment index system and calculating the weight of each index refers to calculating the dimensions of resilience assessment indicators, including productivity, health status, and site conditions;

[0065] Said productivity includes soil organic matter content, precipitation, and irrigation probability;

[0066] The health conditions include heavy metal pollution index and vegetation coverage;

[0067] The resilience of cultivated land systems is comprehensively quantified through multiple dimensions (productivity, health status, site conditions), which is significantly better than the traditional single indicator evaluation method. Taking the productivity dimension as an example, the combination of soil organic matter content, precipitation and irrigation probability reflects the potential production capacity of cultivated land. The health status dimension combines heavy metal pollution and vegetation coverage, achieving a systematic analysis from pollution control to ecological protection.

[0068] The heavy metal pollution index includes: calculating the single factor pollution index based on the pre-processed data :

[0069] ,

[0070] In the formula, is the concentration of the first heavy metal factor measured by experiment, It is the safe concentration of a heavy metal given by national or international environmental standards;

[0071] Combine the single factor pollution index and normalized data to calculate the pollution risk value :

[0072] ,

[0073] In the formula, It is a sample No. Normalized measurements of factors;

[0074] Generate a pollution risk matrix of the sample based on the risk value, and generate a normalized soil quality standard matrix based on the preprocessed soil data;

[0075] Calculate the difference based on the contamination risk value:

[0076]

[0077] In the formula, It is a sample No. Factors (pollution risk values) and quality standard levels The difference between the normalized standard values, quality standard level Obtained through the soil safety threshold specified in the Soil Environmental Quality Standards, It is Level The normalized standard value of the factor, is the total number of heavy metal factors;

[0078] Calculate the minimum difference between each sample and the standard and the maximum difference :

[0079] ,

[0080] ;

[0081] In the formula, It is a calculation sample No. Level, all The minimum value of the difference between the item factors, It is a calculation sample No. Level, all The maximum value of the difference between the item factors;

[0082] Calculate the correlation coefficient of the factor and the quality criterion level based on the difference series and extreme values:

[0083] ,

[0084] In the formula, It is a sample Middle Factors and quality standards Correlation coefficient of levels;

[0085] Calculate the comprehensive correlation of each sample based on the correlation coefficient :

[0086] ,

[0087] The Nemero comprehensive index B of each sample was calculated based on the comprehensive correlation and single factor contamination index:

[0088] ,

[0089] In the formula, is the maximum value of the single factor pollution index. is the total number of heavy metal factors, It is a sample Soil quality standard grade The comprehensive relevance of

[0090] Calculate the interaction effect index D of each heavy metal factor in the sample:

[0091] ,

[0092] In the formula, represents the interaction strength between heavy metal factors l and f, obtained through experiments, is the pollution index of heavy metal factor f, and p is the total number of heavy metal factors;

[0093] Combining the Nemero comprehensive index and the interaction effect index, the final soil comprehensive pollution index G of the sample was calculated by weighted summation method;

[0094] By combining the single factor pollution index with the comprehensive correlation, the quantification process of the pollution degree was refined, and based on the Nemero comprehensive index and the interaction effect index, an overall assessment framework for heavy metal pollution was constructed. The interaction effect index considers the synergy and antagonism between heavy metal factors by weighting, which effectively makes up for the deficiency of existing technologies that ignore the complex relationship between factors, thus providing a comprehensive basis for accurately identifying the risk of cultivated land pollution.

[0095] The vegetation coverage rate refers to the NDVI value of the area obtained through remote sensing images, and the vegetation coverage rate is calculated based on the NDVI value;

[0096] The site conditions include extracting the total length of regional roads through GIS software, allocating the total length of roads to grids per square kilometer according to the distribution of cultivated land, calculating the road grid density of each grid, calculating the shortest distance from cultivated land to the nearest market based on GIS buffer analysis, and calculating the slope distribution through DEM (digital elevation model) data;

[0097] The site conditions are reflected in the distribution of road density, market distance and slope extracted by GIS, which reflects the socio-economic accessibility of cultivated land. By automatically extracting road density, market distance and slope data through GIS technology and combining with the DEM model, the present invention achieves accurate quantification of site conditions, helps to identify key areas with inconvenient transportation or insufficient infrastructure, and provides data support for optimizing cultivated land utilization;

[0098] After standardizing the indicator data, a standardized evaluation matrix is ​​constructed. The rows in the matrix represent samples, and the columns represent indicators. The contribution value of each sample to the indicator is calculated. :

[0099] ,

[0100] In the formula, is the standardized value of sample q on index w, and m is the total number of samples;

[0101] Calculating information entropy based on contribution value :

[0102] ,

[0103] Where K is a constant used to normalize information entropy. ;

[0104] Calculate weights based on entropy :

[0105] ,

[0106] In the formula, is the total number of indicators;

[0107] Experts in the field of organization compare each indicator pairwise, construct a comparison matrix, and calculate the weight vector:

[0108] ,

[0109] In the formula, is the maximum eigenvalue of the comparison matrix, v is the dimension of the comparison matrix;

[0110] The final indicator weight W is obtained by combining the entropy weight method weight and AHP weight:

[0111] ,

[0112] In the formula, is the adjustment coefficient, which is set by genetic algorithm.

[0113] By calculating the information contribution value of each indicator through the entropy weight method and combining the subjective weight distribution provided by AHP, this paper realizes the dynamic fusion of objective and subjective evaluation methods for the first time, combining the Nemero comprehensive index with the interaction effect between heavy metal factors, and calculating the comprehensive soil pollution index through weighted summation, providing a complete solution for the quantification of health status dimensions, and calculating the vegetation coverage rate (FVC) through NDVI data, which intuitively reflects the dynamic changes of cultivated land vegetation under complex interference. Combined with GIS map analysis, this step can help quickly identify areas with insufficient vegetation coverage and provide a basis for precise restoration.

[0114] By building a resilience assessment index system and dynamically calculating weights, combined with advanced data processing and model calculation methods, the scientificity and operability of the resilience assessment of cultivated land systems have been comprehensively improved. From multi-dimensional quantitative resilience characteristics to comprehensive pollution assessment and site condition analysis, to the final comprehensive scoring and dynamic optimization strategy design, all key steps are closely linked, ultimately achieving accurate assessment and optimization support for cultivated land systems under complex disturbances.

[0115] S2. Calculate the comprehensive resilience score based on the weight distribution, draw the time series change curve based on the score and generate a resilience distribution map by regional division;

[0116] Specifically, the calculation of the comprehensive resilience score based on weight distribution means that based on the weight calculation results, the soil organic matter content, precipitation and irrigation probability indicators in the productivity dimension are selected to calculate the productivity score of each sample. :

[0117] ,

[0118] In the formula, is the weight of the bth indicator in the productivity dimension, is the Ith productivity index value of the Lth sample, including soil organic matter content, precipitation, and irrigation probability, obtained by standardizing the data;

[0119] Soil organic matter content, precipitation and irrigation probability are selected as core indicators of productivity, and the role of natural conditions and human intervention are comprehensively considered. Through the weight distribution method, the contribution of each indicator to productivity is accurately quantified, which overcomes the problem of insufficient adaptability caused by the fixed indicator weights in the existing model, provides a comprehensive productivity evaluation framework, and provides scientific guidance for regional productivity optimization.

[0120] Calculate health status score based on metal pollution index and vegetation coverage :

[0121] ,

[0122] In the formula, and It is the weight of the soil comprehensive pollution index G and vegetation coverage FVC, which is determined by the expert evaluation method;

[0123] The health status dimension combines soil health (reflected by the heavy metal pollution index) with ecological status (reflected by vegetation coverage) to comprehensively measure the sustainability of cultivated land ecosystems. It uses expert evaluation to dynamically adjust weights, which can flexibly optimize the importance of indicators according to specific regional characteristics and improve the model's adaptability to different disturbance scenarios. The introduction of vegetation coverage makes up for the lack of description of ecological recovery capacity in traditional health status assessments.

[0124] Calculates site condition score based on road network density, distance to market and slope :

[0125] ,

[0126] In the formula, is the weight of the bth indicator in the site condition dimension, is the jth site condition index value of the i-th sample, including road density, market distance and slope, obtained by standardizing the data;

[0127] The infrastructure conditions and location advantages of cultivated land are comprehensively measured through three core indicators: road network density, market distance and slope. The high-precision data extraction method supported by GIS technology significantly improves the accuracy and reliability of site condition scores.

[0128] Based on the weighted summation calculation of productivity score, health status score and site condition score, a comprehensive resilience score is obtained. Through the weighted summation method, the present invention effectively integrates the three dimensions of productivity, health status and site condition, and constructs a complete resilience assessment framework. The weight of each dimension can be flexibly adjusted according to regional characteristics and research needs, which improves the versatility and adaptability of the model.

[0129] By calculating productivity scores, health scores and site condition scores based on weight distribution, a comprehensive resilience scoring model was constructed to comprehensively measure the performance of cultivated land systems under complex disturbances. Its multi-source data integration, dynamic weight distribution and multi-dimensional indicator calculation methods not only overcome the shortcomings of existing technologies such as poor model adaptability and incomplete description, but also provide strong technical support for resilience assessment and optimization strategies.

[0130] Furthermore, based on the score, a time series change curve is drawn and regional division is performed to generate a resilience distribution map, including:

[0131] After collecting and preprocessing historical meteorological data and socioeconomic data, the preprocessed data were combined with the comprehensive resilience score to construct a dataset;

[0132] Use the LSTM model to predict changes in resilience scores. Input the data set as a training set into the LSTM model for model training. Define the loss function and Adam optimizer to iteratively optimize the model parameters. When the loss of the LSTM model no longer decreases significantly during continuous iterations, stop iterating, output the model parameters, and update the LSTM model. Input the real-time resilience score calculation data into the LSTM model to obtain future resilience score changes.

[0133] Collect historical resilience scores, combine historical resilience scores with forecast scores, use dotted lines to represent future resilience scores, and use solid lines to represent historical resilience scores;

[0134] Calculate the annual change in resilience score using time series curve data :

[0135] ,

[0136] In the formula, is the time step, t is the time point;

[0137] Calculate the standard deviation of the change as the change threshold H. If the change is greater than the threshold H, mark the year as a key time node.

[0138] If the change is less than 0, it means that the toughness decreases rapidly, and the node is marked as a decreasing node;

[0139] If the change is greater than 0, it means that the resilience is quickly recovered, and the node is marked as a recovery node;

[0140] Obtain the resilience score and area of ​​each plot, perform weighted average by spatial unit, calculate the annual comprehensive resilience score of each spatial unit, set thresholds U and O by statistical analysis of historical data of the plots, and threshold U>O. If the comprehensive resilience score is greater than threshold U, it is a high-resilience area. If the comprehensive resilience score is greater than or equal to threshold O and less than or equal to threshold U, it is a medium-resilience area. If the comprehensive resilience score is less than threshold O, it is a low-resilience area.

[0141] Using GIS software, the resilience score of each geographic unit is mapped according to a color gradient, and a pollution source distribution layer is added to mark the pollution points and areas. The corresponding resilience distribution map is generated for each year, and the annual resilience distribution map is synthesized into a dynamic interactive map using GIS software.

[0142] Based on the dynamic prediction of multi-source data fusion and LSTM model, a time series change curve was constructed and regional division was performed to generate a dynamic resilience distribution map, which significantly improved the accuracy and operability of the resilience assessment and optimization of the cultivated land system. Its main beneficial effects include: enhancing the scientific nature of the assessment through dynamic prediction and identification of key time nodes, achieving the accuracy of regional division by weighted average and spatial mapping; and providing efficient visualization support for optimization decisions through dynamic interactive map display. Based on academic rigor and innovation, this solution comprehensively solves the problems of insufficient data utilization, inaccurate prediction, and lack of targeted regional optimization in traditional methods, and provides a reliable technical means for the sustainable management of cultivated land systems under complex interference conditions.

[0143] S3. Optimize according to the regional division results, visualize the data and store the data generated during the optimization process;

[0144] Specifically, the optimization according to the regional division results includes:

[0145] If the area is a high-resilience area, then maintain the existing high-resilience state and promote ecological planting models, which can reduce the use of pesticides and fertilizers, effectively prevent chemical pollution of soil and groundwater, further enhance soil fertility and vegetation coverage, reduce the use of pesticides and fertilizers, strengthen ecological protection, and strengthen long-term monitoring;

[0146] If the area is a medium-resilience area, priority should be given to restoring the ecosystem. Vegetation restoration and returning farmland to forest can improve soil structure and fertility, increase vegetation coverage, reduce the risk of soil erosion, and concentrate resources to clean up pollution sources and reduce heavy metal pollution concentrations. Cleaning up pollution sources can effectively reduce heavy metal concentrations and improve soil health. Multi-dimensional ecological restoration measures work synergistically to gradually upgrade medium-resilience areas to high-resilience areas.

[0147] If the region is a low-resilience area, enhancing the ecological barrier function can effectively prevent the spread of pollutants and improve the microenvironment. For example, pollutants can be fixed or adsorbed by planting heavy metal-resistant plants. The implementation of key restoration projects (such as soil improvement and pollution control) can fundamentally improve the health status and productivity of low-resilience areas. The establishment of a dynamic monitoring mechanism can quickly identify disturbance sources and take countermeasures through real-time data feedback, thereby reducing the degree of damage to the ecosystem and enhancing real-time monitoring of disturbance sources (such as abnormal precipitation and pollutant emissions).

[0148] Through differentiated optimization strategies for high-resilience areas, medium-resilience areas, and low-resilience areas, scientific allocation of resources and maximization of ecological benefits are achieved. The protection measures in high-resilience areas continue their existing advantages and provide a demonstration for farmland management. The restoration measures in medium-resilience areas effectively improve the quality of the ecological environment and provide an important path for achieving regional resilience improvement. The key governance measures in low-resilience areas significantly improve the resilience of the ecosystem and lay the foundation for ensuring agricultural production safety and sustainable development. By introducing a real-time monitoring mechanism for disturbance sources, the present invention can quickly adjust the optimization strategy in a dynamic environment to ensure the long-term stability of the optimization effect.

[0149] Further, visualizing data refers to using a front-end framework Build a visualization interface to visualize the regional division results and dynamic interactive map, and overlay the optimization strategy points on the map, mark the high-resilience areas in green to show the protection strategy, mark the medium-resilience areas in yellow to show the repair strategy, and mark the low-resilience areas in red to show the priority repair strategy.

[0150] By using React.js to build a visualization interface, combined with dynamic interactive maps and optimization strategy point marking technology, the intuitive presentation of cultivated land resilience distribution and optimization schemes is achieved. The dynamic interactive map supports dynamic monitoring and comprehensive analysis of spatiotemporal distribution, and the optimization strategy points provide accurate strategy recommendations for different regions. Compared with traditional static data display methods, the visualization interface of the present invention not only improves the intuitiveness of data expression and the convenience of operation, but also enhances the pertinence and operability of the optimization scheme. At the same time, the efficient development model and flexible expansion capabilities of React.js ensure the practicality and long-term applicability of the platform, providing a scientific and efficient solution for agricultural resource management.

[0151] Furthermore, storing the data generated during the optimization process means storing the resilience score data, optimization strategy data, and resilience region annotation data in a database in order according to timestamps, and automatically updating the database after each new resilience score data is added;

[0152] Synchronize and regularly back up the stored data, regularly perform security and integrity checks on the stored data and backup data, and generate test reports.

[0153] By optimizing the storage and management process of resilience data, efficient data management, dynamic updating and long-term preservation are achieved, which can be applied to a wide range of scenarios for resilience assessment and optimization of cultivated land systems, especially in the face of complex compound disturbances, and can provide accurate real-time data support for decision makers. The automatic update and regular backup mechanism of the database not only reduces the risk of data loss, but also enhances the credibility and scientific nature of the data, while the generation of visual display and test reports further improves the transparency and operability of data analysis.

[0154] This embodiment also provides a computer device, which is suitable for the method of assessing and optimizing the resilience of a cultivated land system facing complex disturbances, and includes: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the method of assessing and optimizing the resilience of a cultivated land system facing complex disturbances as proposed in the above embodiment.

[0155] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0156] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing the resilience assessment and optimization of cultivated land systems oriented to complex interference as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0157] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for evaluating and optimizing the resilience of cultivated land systems under complex disturbances, characterized by: include, Collect and preprocess multi-source farmland data, build a resilience assessment index system and calculate the weight of each index; Calculate the comprehensive resilience score based on weight allocation, draw a time series change curve based on the score and generate a resilience distribution map by dividing the regions; Optimize based on the regional division results, visualize the data and store the data generated during the optimization process.

2. The method for evaluating and optimizing farmland system resilience against combined disturbances according to claim 1, characterized in that: The collecting of multi-source cultivated land data and preprocessing thereof refers to using a stratified random sampling method to divide sampling units according to regional soil types and geomorphological features, collecting surface soil at each sampling point, and removing coarse particles after the soil samples are naturally air-dried, analyzing the soil organic matter content by potassium dichromate method, using ICP-MS to detect the heavy metal content, taking the average value after testing each sample a times as the final concentration, obtaining the average annual precipitation in the past A years from the meteorological monitoring station in the region, and performing time series analysis on the collected data, extracting annual irrigation records from the agricultural irrigation management platform, and calculating the total annual irrigation times based on the irrigation times data for each piece of cultivated land; Preprocess the collected data.

3. The method for evaluating and optimizing farmland system resilience against combined disturbances according to claim 2, characterized in that: The construction of the resilience assessment index system and calculation of the weight of each index refers to the calculation of the resilience assessment index dimensions, including productivity, health status and site conditions; Said productivity includes soil organic matter content, precipitation, and irrigation probability; The health conditions include heavy metal pollution index and vegetation coverage; The heavy metal pollution index includes: calculating the single factor pollution index based on the pre-processed data ; Combine the single factor pollution index and normalized data to calculate the pollution risk value ; Generate a pollution risk matrix of the sample based on the risk value, and generate a normalized soil quality standard matrix based on the preprocessed soil data; Calculate the difference based on the contamination risk value and calculate the minimum difference between each sample and the standard and the maximum difference ; Calculate the correlation coefficients of factors and quality criterion levels based on difference series and extreme values; Calculate the comprehensive correlation of each sample based on the correlation coefficient ; The Nemero comprehensive index B of each sample was calculated based on the comprehensive correlation and single factor contamination index; Calculate the interaction effect index D of each heavy metal factor in the sample; Combining the Nemero comprehensive index and the interaction effect index, the final soil comprehensive pollution index G of the sample was calculated by weighted summation method; The vegetation coverage rate refers to the NDVI value of the area obtained through remote sensing images, and the vegetation coverage rate FVC is calculated based on the NDVI value; The site conditions include extracting the total length of regional roads through GIS software, allocating the total length of roads to grids per square kilometer according to the distribution of cultivated land, calculating the road grid density of each grid, calculating the shortest distance from cultivated land to the nearest market based on GIS buffer analysis, and calculating the slope distribution through DEM model data; After standardizing the indicator data, a standardized evaluation matrix is ​​constructed. The rows in the matrix represent samples, and the columns represent indicators. The contribution value of each sample to the indicator is calculated. , information entropy is calculated based on contribution value , calculate the weight according to the entropy value ; Organize experts in the field to compare each indicator pairwise, build a comparison matrix, and calculate the weight vector; The final indicator weight W is obtained by combining the entropy weight method weight and AHP weight.

4. The method for evaluating and optimizing farmland system resilience against combined disturbances according to claim 3, characterized in that: The calculation of comprehensive resilience score based on weight distribution refers to the calculation of productivity score of each sample by selecting soil organic matter content, precipitation and irrigation probability indicators in productivity dimension based on weight calculation results. ; Calculate health status score based on metal pollution index and vegetation coverage ; Calculates site condition score based on road network density, distance to market and slope ; The comprehensive resilience score is calculated by weighted summation based on the productivity score, health score and site condition score. .

5. The method for evaluating and optimizing farmland system resilience against combined disturbances according to claim 4, characterized in that: Drawing a time series change curve based on the score and performing regional division to generate a resilience distribution map includes: After collecting and preprocessing historical meteorological data and socioeconomic data, the preprocessed data were combined with the comprehensive resilience score to construct a dataset; Use the LSTM model to predict changes in resilience scores. Input the data set as a training set into the LSTM model for model training. Define the loss function and Adam optimizer to iteratively optimize the model parameters. When the loss of the LSTM model no longer decreases significantly during continuous iterations, stop iterating, output the model parameters, and update the LSTM model. Input the real-time resilience score calculation data into the LSTM model to obtain future resilience score changes. Collect historical resilience scores, combine historical resilience scores with forecast scores, use dotted lines to represent future resilience scores, and use solid lines to represent historical resilience scores; Calculate the annual change in resilience score using time series curve data ; Calculate the standard deviation of the change as the change threshold H. If the change is greater than the threshold H, mark the year as a key time node. If the change is less than 0, it means that the toughness decreases rapidly, and the node is marked as a decreasing node; If the change is greater than 0, it means that the resilience is quickly recovered, and the node is marked as a recovery node; Obtain the resilience score and area of ​​each plot, perform weighted average by spatial unit, and calculate the annual comprehensive resilience score for each spatial unit; Thresholds U and O are set, and threshold U>O. If the comprehensive toughness score is greater than threshold U, it is a high toughness area. If the comprehensive toughness score is greater than or equal to threshold O and less than or equal to threshold U, it is a medium toughness area. If the comprehensive toughness score is less than threshold O, it is a low toughness area. Using GIS software, the resilience score of each geographic unit is mapped according to a color gradient, and a pollution source distribution layer is added to mark the pollution points and areas. The corresponding resilience distribution map is generated for each year, and the annual resilience distribution map is synthesized into a dynamic interactive map using GIS software.

6. The method for evaluating and optimizing farmland system resilience against combined disturbances according to claim 5, characterized in that: The optimization according to the regional division result includes: If the area is a high-resilience area, maintain the existing high-resilience state, promote ecological planting models, reduce the use of pesticides and fertilizers, strengthen ecological protection, and strengthen long-term monitoring; If the region is a medium-resilience area, priority will be given to restoring the ecosystem, carrying out vegetation restoration, returning farmland to forest, concentrating resources on cleaning up pollution sources, and reducing heavy metal pollution concentrations; If the area is a low-resilience area, the ecological barrier function will be enhanced, key restoration projects will be implemented, a dynamic monitoring mechanism will be established, and real-time monitoring of disturbance sources will be enhanced.

7. The method for evaluating and optimizing farmland system resilience against complex disturbances according to claim 6, characterized in that: The data visualization refers to using the front-end framework React.js to build a visualization interface, visualizing the regional division results and the dynamic interactive map, and superimposing optimization strategy points on the map, marking the high-resilience areas in green to show the protection strategy, marking the medium-resilience areas in yellow to show the repair strategy, and marking the low-resilience areas in red to show the priority repair strategy.

8. The method for evaluating and optimizing farmland system resilience against complex disturbances according to claim 7, characterized in that: The data generated during the storage optimization process refers to storing the resilience score data, optimization strategy data and resilience area annotation data in a database in order according to timestamps, and automatically updating the database after each new resilience score data is added; Synchronize and regularly back up the stored data, regularly perform security and integrity checks on the stored data and backup data, and generate test reports.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for assessing and optimizing the resilience of a cultivated land system facing complex disturbances as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for assessing and optimizing the resilience of a cultivated land system facing complex disturbances as described in any one of claims 1 to 8 are implemented.

Images