Artificial Intelligence-Based Soil Carbon Sequestration and Carbon Increase Optimization Method and System

Abstract

The present invention provides an optimization method and system for soil carbon sequestration and carbon increase based on artificial intelligence, belonging to the technical field of agricultural carbon sequestration and carbon increase optimization, including the following steps: S101: Obtain the area of the agricultural planting soil area and the boundary of the agricultural planting soil area, and obtain weather data; S102: Obtain different plant species planted, perform area division operations, and number the soil areas; S103: Obtain a set of data change fluctuation curves and a set of weather factors, and obtain a carbon sequestration amount-time change fluctuation curve set; S104: Establish an agricultural soil carbon sequestration amount model, and obtain carbon sequestration amount prediction data matching the soil area number; S105: Design and establish a carbon sequestration verification model, obtain carbon sequestration amount verification data matching the soil area number, and perform data verification and calibration operations; S106: Based on the verified carbon sequestration amount prediction data, perform carbon increase optimization operations on different sub-areas with soil area numbers, improving the accuracy of the prediction data.

Description

Technical Field

[0001] The present invention belongs to the technical field of agricultural carbon sequestration and carbon increase optimization, and more specifically, particularly relates to a method and system for optimizing soil carbon sequestration and carbon increase based on artificial intelligence. Background Art

[0002] In the agricultural field, soil quality is the cornerstone of crop growth and yield, and soil carbon content is one of the core indicators for measuring soil quality. Carbon in the soil exists in the form of organic carbon and inorganic carbon. Among them, soil organic carbon not only provides a long-term nutrient source for crops, but also improves soil structure and enhances the water and fertilizer retention capacity of the soil, playing a key role in improving the stress resistance and quality of crops. However, when traditional carbon sequestration and carbon increase optimization methods predict and obtain data on future soil carbon sequestration, the data accuracy is relatively low, and when increasing carbon in the soil, there is often a time lag. Summary of the Invention

[0003] In order to solve the above technical problems, the present invention provides a method and system for optimizing soil carbon sequestration and carbon increase based on artificial intelligence, so as to solve the technical problems that in the prior art, when traditional carbon sequestration and carbon increase optimization methods predict and obtain data on future soil carbon sequestration, the data accuracy is relatively low, and when increasing carbon in the soil, there is often a time lag.

[0004] The objectives and effects of the method and system for optimizing soil carbon sequestration and carbon increase based on artificial intelligence of the present invention are achieved by the following specific technical means:

[0005] The method for optimizing soil carbon sequestration and carbon increase based on artificial intelligence includes the following steps:

[0006] S101: Conduct an investigation on the agricultural planting soil area to obtain the area and boundary of the agricultural planting soil area, establish an agricultural weather detection station and an agricultural weather detection database, detect the weather of the agricultural planting soil area based on the agricultural weather detection station, and obtain weather data. The weather data includes the past dates of the agricultural planting soil area and the weather types matching the past dates, and import the weather data into the agricultural weather detection database for storage;

[0007] S102: Obtain different plant species planted in the agricultural planting soil area, assign character numbers to different plant species, perform area division operations based on the plant species in the agricultural planting soil area, divide the agricultural planting soil area into multiple groups of different agricultural planting soil sub-areas according to the planted plant species based on the area division operations, assign digital numbers to different sub-areas, and match and combine the digital numbers with the character numbers to form soil area numbers;

[0008] S103: Obtain historical carbon sequestration data and historical dates for different sub-regions, perform timestamp alignment operations to match the historical carbon sequestration data with the historical dates, obtain a data change fluctuation curve group based on the historical carbon sequestration data and historical dates, import the soil area numbers into the data change fluctuation curve group, and at the same time, call the agricultural weather detection database, obtain a weather factor group based on the weather data in the agricultural weather detection database, and obtain a carbon sequestration - time change fluctuation curve group based on the data change fluctuation curve group, weather factor group, and soil area numbers;

[0009] S104: Establish an agricultural soil carbon sequestration model, copy the carbon sequestration - time change fluctuation curve group, import the copy as a training sample into the agricultural soil carbon sequestration model, and export an agricultural soil carbon sequestration prediction model based on iterative training. The agricultural soil carbon sequestration prediction model predicts the carbon sequestration of different sub-regions in the next year based on the monthly change rate of the carbon sequestration - time change fluctuation curve group, so as to obtain carbon sequestration prediction data matching the soil area numbers;

[0010] S105: Design and establish a carbon sequestration verification model, obtain carbon sequestration verification data matching the soil area numbers based on the carbon sequestration verification model, perform data verification and calibration operations based on the carbon sequestration verification data and carbon sequestration prediction data, and verify the carbon sequestration prediction data based on the data verification and calibration operations;

[0011] S106: Perform carbon sequestration optimization operations on different sub-regions with verified carbon sequestration prediction data according to the soil area numbers.

[0012] As a further solution of the present invention, design and establish a carbon sequestration verification model, and obtain carbon sequestration verification data matching the soil area numbers based on the carbon sequestration verification model, including:

[0013] A1: Construct a small-scale regional digital map, which is constructed based on the actual geographical data of the actual soil area; construct a small-scale agricultural planting soil area digital sand table based on the small-scale regional digital map. The small-scale agricultural planting soil area digital sand table contains planted plants and is used to simulate the environment of the agricultural planting soil area. Perform area division operations on the small-scale agricultural planting soil area digital sand table, and at the same time perform soil area numbering;

[0014] A2: Establish a virtual weather observation station and a virtual weather generation station in the small-scale agricultural planting soil area digital sand table. The virtual weather generation station is used to dynamically simulate the actual weather, and the virtual weather observation station is used to detect the weather of the small-scale agricultural planting soil area sand table and obtain virtual weather data;

[0015] A3: Build a carbon sequestration verification model, and the DAYCENT model is selected to build the carbon sequestration verification model;

[0016] A4: Import virtual weather data and relevant parameters of the agricultural planting soil area into the carbon sequestration verification model, perform verification simulation through the carbon sequestration verification model, and generate carbon sequestration verification data matching the soil area number.

[0017] As a further solution of the present invention, the carbon sequestration verification model is simulated based on the carbon cycle function, and the carbon cycle function is expressed as:

[0018] SOC T = SOC T-1 + GPP T - R T - M T ;

[0019] Among them, SOC T represents the soil carbon content at time T, SOC T-1 represents the soil carbon content at the previous time T, GPP T represents the total primary productivity at time T, R T represents the carbon released by the respiration of plants and soil at time T, M T represents the carbon released by soil organic matter;

[0020] When performing simulation, the virtual weather data is imported into the carbon sequestration verification model as a virtual weather factor. After the virtual weather factor is imported, the carbon sequestration verification model performs simulation in a dynamic calculation simulation manner.

[0021] As a further solution of the present invention, the regional division operation specifically includes:

[0022] B1: Obtain the rectangular NDVI image of the agricultural planting soil area based on remote sensing technology;

[0023] B2: Denoise the rectangular NDVI image based on wavelet transform technology. After denoising, perform sharpening and contrast enhancement operations on the rectangular NDVI image;

[0024] B3: Use the Canny edge detection algorithm to detect the boundary of the soil area based on the brightness gradient of the rectangular NDVI image. At the same time, remove the useless image in the rectangular NDVI image based on the Canny edge detection algorithm to obtain a curved NDVI image. The useless image is the image part in the rectangular image that does not contain the agricultural planting soil area;

[0025] B4: Perform image segmentation on the curved NDVI image based on the region growing algorithm, and divide the regions with similar spectral characteristics in the NDVI image into multiple sub-regions;

[0026] B5: Establish a color gradient table, which is matched with the character numbers of different plant species. Among them, the color of the bare soil road area is defined as white. Based on color mapping technology, the NDVI values in the curved NDVI image sub-region are mapped to different colors, and the color mapping is performed based on the color gradient table.

[0027] As a further solution of the present invention, based on the Canny edge detection algorithm, useless images in the rectangular NDVI image are removed to obtain a curved NDVI image, including:

[0028] After obtaining the curved NDVI image, a correction operation is performed on the curved NDVI image based on the area of the agricultural planting soil area and the boundary of the agricultural planting soil area.

[0029] As a further solution of the present invention, a rectangular image of the agricultural planting soil area is obtained based on remote sensing technology, including:

[0030] The obtained rectangular image of the agricultural planting soil area is a high-resolution image.

[0031] As a further solution of the present invention, a matching combination is performed based on the digital number and the character number, and combined into a soil area number, including:

[0032] When performing the digital numbering, numbers 1, 2, 3, 4 to n are used for numbering, where n represents the number of sub-regions of the agricultural planting soil;

[0033] When performing the character numbering, the names of different plant species are obtained, and the first letters are extracted based on the pinyin of different plant species. Character coding is performed based on the combination of the first letters of the pinyin of different plant species. If the character coding is the same, a number is added at the end of the combination of the first letters of the pinyin. At the same time, the character coding is assigned to different plant species for character numbering;

[0034] Based on the matching combination of the digital number and the character number, it is combined into a soil area number, and the soil area number is expressed as (count, name), where count represents the digital number and name represents the character number.

[0035] As a further solution of the present invention, the carbon increase optimization operation specifically includes:

[0036] Carbon is added to the soil, and when adding carbon, a combined carbon addition method is used for carbon addition;

[0037] A large amount of the first carbon addition material is used to add carbon to the soil. The first carbon addition material includes decomposed straw and feces treated by harmless treatment;

[0038] Appropriately utilize the second carbon-increasing material to increase carbon in the soil. The second carbon-increasing material is represented by strains that meet the standards of food, feed, and fertilizer.

[0039] Precisely supplement a small amount of the third carbon-increasing material to increase carbon in the soil. The third carbon-increasing material is represented by mineral elements that meet the standards of organic farming.

[0040] Users can adjust the ratio and total usage amount of the first, second, and third carbon-increasing materials based on the carbon sequestration prediction data.

[0041] As a further solution of the present invention, the data verification and calibration operation specifically includes:

[0042] Obtain a verification threshold, which is used to perform a calibration operation on the carbon sequestration prediction data and the carbon sequestration verification data. Take the absolute value of the difference between the carbon sequestration prediction data and the carbon sequestration verification data to obtain a carbon sequestration comparison value, and compare the carbon sequestration comparison value with the verification threshold.

[0043] If the carbon sequestration comparison value is less than or equal to the verification threshold, then export the carbon sequestration prediction data as the verified carbon sequestration prediction data.

[0044] If the carbon sequestration comparison value is greater than the verification threshold, re-predict based on the agricultural soil carbon sequestration prediction model.

[0045] An artificial intelligence-based soil carbon sequestration and carbon-increasing optimization system includes:

[0046] An acquisition module, which is used to acquire the area and boundary of the agricultural planting soil area, acquire the different plant species planted in the agricultural planting soil area, and acquire the historical carbon sequestration data and historical dates of different sub-areas.

[0047] A weather module, which includes an agricultural weather detection station and an agricultural weather detection database.

[0048] The agricultural weather detection station is used to detect the weather in the agricultural planting soil area and acquire weather data.

[0049] The agricultural weather detection database is used to store the weather data detected by the agricultural weather detection station.

[0050] A region module, which is used to perform region division operations.

[0051] A numbering module, which is used to assign character numbers to different plant species, assign digital numbers to different sub-areas, and generate soil area numbers based on the character numbers and digital numbers.

[0052] A processing module, configured to obtain a data change fluctuation curve group based on historical carbon sequestration data and historical dates, import the soil area number into the data change fluctuation curve group, and call an agricultural weather detection database to obtain a weather factor group, and obtain a carbon sequestration - time change fluctuation curve group based on the weather factor group, the data change fluctuation curve group, and the soil area number;

[0053] A model module, configured to establish an agricultural soil carbon sequestration model, and generate an agricultural soil carbon sequestration prediction model based on the carbon sequestration - time change fluctuation curve group, so as to obtain carbon sequestration prediction data matching the soil area number based on the agricultural soil carbon sequestration prediction model;

[0054] A verification and simulation module, configured to design and establish a carbon sequestration verification model, and obtain carbon sequestration verification data matching the soil area number based on the carbon sequestration verification model;

[0055] A verification module, configured to perform a data verification and check operation on the carbon sequestration verification data and the carbon sequestration prediction data.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] First, conduct an investigation on the agricultural planting soil area to obtain the area of the agricultural planting soil area and the boundary of the agricultural planting soil area. Then, establish an agricultural weather detection station and an agricultural weather detection database. Import the weather data obtained from the agricultural weather detection station into the agricultural weather detection database for storage. Next, obtain the different plant species planted in the agricultural planting soil area, and perform regional division operations based on the plant species in the agricultural planting soil area. Thus, divide the agricultural planting soil area into multiple different agricultural planting soil sub-areas according to the planted plant species, and number the soil areas of the sub-areas. After that, obtain the historical carbon sequestration data and historical dates of different sub-areas, obtain a group of data change fluctuation curves based on the historical carbon sequestration data and historical dates, obtain a group of weather factors, obtain a carbon sequestration-time change fluctuation curve group based on the data change fluctuation curve group, the weather factor group, and the soil area number, establish an agricultural soil carbon sequestration model, and derive an agricultural soil carbon sequestration prediction model based on iterative training. Thus, obtain carbon sequestration prediction data matching the soil area number. Subsequently, design and establish a carbon sequestration verification model, obtain carbon sequestration verification data matching the soil area number, perform data verification and calibration operations, and perform carbon sequestration optimization operations on the sub-areas with different soil area numbers based on the verified carbon sequestration prediction data. This method can predict the carbon sequestration of soil sub-areas. If the predicted carbon sequestration is too low, the user can promptly perform carbon sequestration operations through manual intervention, and the manual intervention can be carried out by sub-area, enabling more accurate carbon sequestration operations. Moreover, when obtaining the carbon sequestration prediction data of each sub-area, the data can be verified and calibrated based on the carbon sequestration verification model, ensuring the accuracy of the prediction data. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 FIG. is a flowchart of the steps of the soil carbon sequestration and carbon increment optimization method based on artificial intelligence of the present invention;

[0059] Figure 2 FIG. is a flowchart of the steps of designing and establishing a carbon sequestration verification model in the soil carbon sequestration and carbon increment optimization method based on artificial intelligence of the present invention;

[0060] Figure 3 FIG. is a flowchart of the steps of the regional division operation in the soil carbon sequestration and carbon increment optimization method based on artificial intelligence of the present invention;

[0061] Figure 4 FIG. is an example schematic diagram of the agricultural planting soil area after the regional division operation in the soil carbon sequestration and carbon increment optimization method based on artificial intelligence of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] The following further describes in detail the embodiments of the present invention in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the technical solutions of the present invention, but cannot be used to limit the protection scope of the present invention.

[0063] Example 1:

[0064] As shown in the attached Figure 1 , Figure 2 , Figure 3 , Figure 4 figures:

[0065] The present invention provides an optimization method for soil carbon sequestration and carbon increase based on artificial intelligence, which is applicable to soil carbon sequestration and carbon increase, and includes the following steps:

[0066] S101: Investigate the agricultural planting soil area, obtain the area and boundary of the agricultural planting soil area, establish an agricultural weather detection station and an agricultural weather detection database, detect the weather of the agricultural planting soil area based on the agricultural weather detection station, and obtain weather data. The weather data includes the past dates of the agricultural planting soil area and the weather types matching the past dates, and import the weather data into the agricultural weather detection database for storage.

[0067] Specifically, when investigating the agricultural planting soil area, the user can use a drone or other means to investigate the agricultural planting soil area, and the weather conditions of the agricultural planting soil area are important data in evaluating the carbon sequestration amount. Establishing an agricultural weather detection station can detect and record the weather of the agricultural planting soil area.

[0068] S102: Obtain different plant species planted in the agricultural planting soil area, assign character numbers to different plant species, perform regional division operations based on the plant species in the agricultural planting soil area, divide the agricultural planting soil area into multiple different agricultural planting soil sub-areas according to the planted plant species based on the regional division operations, assign digital numbers to different sub-areas, and match and combine the digital numbers with the character numbers to form soil area numbers.

[0069] Specifically, the regional division operations include:

[0070] B1: Obtain the rectangular NDVI image of the agricultural planting soil area based on remote sensing technology.

[0071] When obtaining the image of the agricultural planting soil area through remote sensing technology, the obtained rectangular image of the agricultural planting soil area is a high-resolution image. It can be understood that finer-grained soil area division requires higher-resolution image data to ensure the accuracy of the division.

[0072] B2: Denoise the rectangular NDVI image based on wavelet transform technology, and perform sharpening and contrast enhancement operations on the rectangular NDVI image after denoising.

[0073] It is understandable that after obtaining remote sensing image data, preprocessing and image enhancement are usually required to improve the effect of edge detection. There may be noise in the remote sensing image, and denoising can reduce the impact of noise on subsequent processing. After denoising, the edges in the enhanced image can be sharpened through sharpening operations, making the boundaries between different regions clearer. By sharpening the image, details such as soil areas and vegetation boundaries can be highlighted, which helps with regional division and analysis. Through contrast enhancement operations, the contrast within local regions of the image can be enhanced, especially for regions with less plant cover, enabling better display of the differences between soil and other regions.

[0074] B3: The Canny edge detection algorithm is used to detect the boundaries of the soil area based on the brightness gradient of the rectangular NDVI image. At the same time, based on the Canny edge detection algorithm, useless images in the rectangular NDVI image are removed to obtain a curved NDVI image. Useless images are the parts of the rectangular image that do not contain the agricultural planting soil area.

[0075] Furthermore, after obtaining the curved NDVI image, a correction operation is performed on the curved NDVI image based on the area of the agricultural planting soil area and the boundary of the agricultural planting soil area. Through the correction operation, the curved NDVI image can be matched and corrected with the agricultural planting soil area under investigation, preventing errors in image division during subsequent steps.

[0076] B4: Image segmentation of the curved NDVI image is performed based on the region growing algorithm, and regions with similar spectral characteristics in the NDVI image are divided into multiple sub-regions.

[0077] Specifically, the region growing algorithm can select multiple "seed points" in the curved NDVI image, and then expand the region based on the "seed points" and according to relevant thresholds, enabling accurate identification of the boundaries and merging of similar regions.

[0078] B5: A color gradient table is established, which is matched with the character numbers of different plant species. Among them, the color of the bare soil road area is defined as white, and the NDVI values in the sub-regions of the curved NDVI image are mapped to different colors based on color mapping technology. The color mapping is performed based on the color gradient table.

[0079] It is understandable that using color mapping technology to map NDVI values to different colors helps enhance the visual effect of different planted plants in the image, making the differences between different planted plant types more obvious, and can enhance the contrast between different planted plant types or soil areas, making the visual effect of the curved NDVI image clearer and facilitating subsequent analysis of the soil area by users.

[0080] Specifically, based on the matching combination of digital numbers and character numbers, they are combined into soil area numbers, including: when performing digital numbering, numbers 1, 2, 3, 4 to n are used for numbering, where n represents the number of sub-areas of agricultural planting soil; when performing character numbering, the names of different plant species are obtained, and the first letters are extracted based on the pinyin of different plant species. Character coding is performed based on the combination of the first letters of the pinyin of different plant species. If the character coding is the same, a number is added at the end of the combination of the first letters of the pinyin. At the same time, the character coding is assigned to different plant species for character numbering.

[0081] Furthermore, when performing character numbering, if the first pinyin letter is the same, a character is added to the second pinyin letter. If the second pinyin letter is the same, a character is added to the third pinyin letter, and so on. If all the pinyin letters are used up and the characters are the same, a number is added at the end of the character.

[0082] After the digital numbering and character numbering are completed, based on the digital numbering and character numbering, they are matched and combined into a soil area number. The soil area number is expressed as (count, name), where count represents the digital number and name represents the character number. For example, (1, ng2), the digital number is 1, indicating that it is within sub-area 1, and the character number is ng2, indicating that the plant species is pumpkin. Then this soil area number represents sub-area 1 where the planted plant is pumpkin.

[0083] For example, as Figure 4 shown, (1, bc) represents sub-area 1 where the planted plant is spinach, (2, qc) represents sub-area 2 where the planted plant is green vegetables, (3, jc) represents sub-area 3 where the planted plant is Chinese chives, and (4, pg) represents sub-area 4 where the planted plant is apples.

[0084] S103: Obtain the historical carbon sequestration data and historical dates of different sub-areas, perform an alignment timestamp operation, so that the historical carbon sequestration data and historical dates are matched. Based on the historical carbon sequestration data and historical dates, obtain a data change fluctuation curve group, and import the soil area number into the data change fluctuation curve group. At the same time, call the agricultural weather detection database, obtain a weather factor group based on the weather data in the agricultural weather detection database, and obtain a carbon sequestration-time change fluctuation curve group based on the data change fluctuation curve group, the weather factor group, and the soil area number.

[0085] For example, the carbon sequestration-time change fluctuation curve group can be expressed as:

[0086]

[0087] Among them, F represents the carbon sequestration-time change fluctuation curve group, Δr Denoted as the weather factor group, a 1 、a 2 、a 3 and a 4 Denoted as the weight, l denoted as the constant. When performing calculations, calculations are carried out based on the appropriate weather factors in the weather factor group.

[0088] S104: Establish an agricultural soil carbon sequestration model, make a copy of the carbon sequestration - time variation fluctuation curve group, import the copy as a training sample into the agricultural soil carbon sequestration model, and derive an agricultural soil carbon sequestration prediction model based on iterative training. The agricultural soil carbon sequestration prediction model predicts the carbon sequestration of different sub - regions in the next year based on the monthly change rate of the carbon sequestration - time variation fluctuation curve group, so as to obtain carbon sequestration prediction data matching the soil area number.

[0089] Furthermore, when establishing the agricultural soil carbon sequestration model, a hybrid model of the CENTURY model and the random forest model can be selected. By combining the process model and the data - driven model, it not only retains the physical interpretability of the process model but also utilizes the strong non - linear fitting ability of the data - driven model to achieve higher prediction accuracy, and the over - fitting risk of a single model can be reduced through the hybrid model, improving the robustness of the overall model.

[0090] In this method, soil carbon sequestration is affected by multiple factors, and there are complex non - linear relationships among these factors. The hybrid model can capture these complex relationships through the data - driven part (such as neural networks, random forests, etc.), and at the same time use the process - based model to explain the basic mechanism of the soil carbon sequestration process. The influence of weather and climate reasons on carbon sequestration is dynamic and non - linear. The hybrid model can combine the weather factor group and adjust the prediction model through machine learning algorithms, thereby providing a more accurate carbon sequestration prediction.

[0091] S105: Design and establish a carbon sequestration verification model, obtain carbon sequestration verification data matching the soil area number based on the carbon sequestration verification model, perform a data verification check operation based on the carbon sequestration verification data and the carbon sequestration prediction data, and verify the carbon sequestration prediction data based on the data verification check operation.

[0092] Specifically, the data verification check operation includes:

[0093] Obtain a verification threshold, which is used to perform a check operation on the carbon sequestration prediction data and the carbon sequestration verification data. Take the absolute value of the difference between the carbon sequestration prediction data and the carbon sequestration verification data to obtain a carbon comparison value, and compare the carbon comparison value with the verification threshold;

[0094] If the carbon comparison value is less than or equal to the verification threshold, then export the carbon sequestration prediction data as the verified carbon sequestration prediction data;

[0095] If the carbon sequestration comparison value is greater than the verification threshold, re-prediction is performed based on the agricultural soil carbon sequestration prediction model.

[0096] Specifically, a carbon sequestration verification model is designed and established, and carbon sequestration verification data matching the soil area number is obtained based on the carbon sequestration verification model, including:

[0097] A1: Construct a small-scale regional digital map, which is constructed based on the actual geographical data of the actual soil area; construct a small-scale agricultural planting soil area digital sand table based on the small-scale regional digital map. The small-scale agricultural planting soil area digital sand table contains planted plants and is used to simulate the environment of the agricultural planting soil area. Perform area division operations on the small-scale agricultural planting soil area digital sand table and assign soil area numbers at the same time.

[0098] A2: Establish a virtual weather observation station and a virtual weather generation station in the small-scale agricultural planting soil area digital sand table. The virtual weather generation station is used to dynamically simulate the actual weather, and the virtual weather observation station is used to detect the weather of the small-scale agricultural planting soil area sand table and obtain virtual weather data.

[0099] Combine the simulated soil area information with the data collected by the weather observation station to form a complete virtual environment. Based on these data, simulate the carbon sequestration process of the soil.

[0100] A3: Construct a carbon sequestration verification model, and the DAYCENT model is selected to construct the carbon sequestration verification model.

[0101] A4: Import virtual weather data and parameters related to the agricultural planting soil area into the carbon sequestration verification model, perform verification simulation through the carbon sequestration verification model, and generate carbon sequestration verification data matching the soil area number.

[0102] Furthermore, the carbon sequestration verification model is simulated based on the carbon cycle function, and the carbon cycle function is expressed as:

[0103] SOC T = SOC T-1 + GPP T - R T - M T ;

[0104] Among them, SOC T represents the soil carbon content at time T, SOC T-1 represents the soil carbon content at the previous time T, GPP T represents the total primary productivity at time T, R T represents the carbon released by the respiration of plants and soil at time T, and M TIndicates the amount of carbon released from soil organic matter.

[0105] When conducting the simulation, virtual weather data is imported into the carbon sequestration verification model as virtual weather factors. After the virtual weather factors are imported, the carbon sequestration verification model conducts the simulation in a dynamic calculation and simulation manner.

[0106] Through the carbon sequestration verification model, the carbon sequestration prediction data and the carbon sequestration verification data can be tested to ensure the accuracy of the data. In actual applications, through the carbon sequestration verification model, the carbon sequestration prediction model can also be calibrated. If the verification results show that the carbon sequestration prediction deviation is large in certain regions or scenarios, the model can be optimized by adjusting the parameters of the prediction model or improving the input variables to make it more adaptable to the actual situation.

[0107] S106: Perform carbon addition optimization operations on sub-regions with different soil area numbers based on the verified carbon sequestration prediction data.

[0108] It can be understood that through this method, the carbon sequestration amount of sub-regions can be predicted. If the predicted carbon sequestration amount is too small, carbon addition operations can be carried out in a timely manner through artificial intervention, and artificial intervention can be carried out in sub-regions to make the carbon addition operation more accurate.

[0109] Specifically, the carbon addition optimization operation specifically includes:

[0110] Add carbon to the soil. When adding carbon, use a combined carbon addition method;

[0111] Use a large amount of the first carbon addition material to add carbon to the soil. The first carbon addition material includes decomposed straw and manure treated harmlessly;

[0112] Use an appropriate amount of the second carbon addition material to add carbon to the soil. The second carbon addition material is expressed as strains that meet the standards of food, feed, and fertilizer;

[0113] Precisely supplement a small amount of the third carbon addition material to add carbon to the soil. The third carbon addition material is expressed as mineral elements that meet the standards of organic farming;

[0114] Users can adjust the ratio of the first carbon addition material, the second carbon addition material, and the third carbon addition material based on the carbon sequestration prediction data, and can also adjust the total usage amount of the first carbon addition material, the second carbon addition material, and the third carbon addition material based on the carbon sequestration prediction data.

[0115] The soil carbon sequestration and carbon addition optimization system based on artificial intelligence includes:

[0116] An acquisition module, used to acquire the area of the agricultural planting soil area and the boundary of the agricultural planting soil area, acquire different plant species planted in the agricultural planting soil area, and acquire the historical carbon sequestration data and historical dates of different sub-regions;

[0117] Weather module, the weather module includes an agricultural weather detection station and an agricultural weather detection database;

[0118] Agricultural weather detection station, used to detect the weather in the agricultural planting soil area and obtain weather data;

[0119] Agricultural weather detection database, used to store the weather data detected by the agricultural weather detection station;

[0120] Area module, used to perform area division operations;

[0121] Numbering module, used to assign character numbers to different plant species, assign digital numbers to different sub-areas, and generate soil area numbers based on the character numbers and digital numbers;

[0122] Processing module, used to obtain a data change fluctuation curve group based on historical carbon sequestration data and historical dates, import the soil area number into the data change fluctuation curve group, and call the agricultural weather detection database to obtain a weather factor group, and obtain a carbon sequestration - time change fluctuation curve group based on the weather factor group, the data change fluctuation curve group and the soil area number;

[0123] Model module, used to establish an agricultural soil carbon sequestration model, generate an agricultural soil carbon sequestration prediction model based on the carbon sequestration - time change fluctuation curve group, and obtain carbon sequestration prediction data matching the soil area number based on the agricultural soil carbon sequestration prediction model;

[0124] Verification simulation module, used to design and establish a carbon sequestration verification model, and obtain carbon sequestration verification data matching the soil area number based on the carbon sequestration verification model;

[0125] Verification module, used to perform data verification and checking operations on the carbon sequestration verification data and the carbon sequestration prediction data.

[0126] Specific usage and functions of the first embodiment:

[0127] First, conduct an investigation on the agricultural planting soil area to obtain the area and boundary of the agricultural planting soil area, establish an agricultural weather detection station and an agricultural weather detection database, import the weather data obtained from the agricultural weather detection station into the agricultural weather detection database for storage. Then, obtain the different plant species planted in the agricultural planting soil area, and perform regional division operations based on the plant species in the agricultural planting soil area, so as to divide the agricultural planting soil area into multiple different agricultural planting soil sub-areas according to the planted plant species, number the soil areas of the sub-areas. After that, obtain the historical carbon sequestration data and historical dates of different sub-areas, obtain a data change fluctuation curve group based on the historical carbon sequestration data and historical dates, obtain a weather factor group, obtain a carbon sequestration-time change fluctuation curve group based on the data change fluctuation curve group, the weather factor group and the soil area number, establish an agricultural soil carbon sequestration model, and derive an agricultural soil carbon sequestration prediction model based on iterative training, so as to obtain carbon sequestration prediction data matching the soil area number. Subsequently, design and establish a carbon sequestration verification model, obtain carbon sequestration verification data matching the soil area number, perform data verification and calibration operations, and perform carbon sequestration optimization operations on the sub-areas with different soil area numbers based on the verified carbon sequestration prediction data. This method can predict the carbon sequestration of soil sub-areas. If the predicted carbon sequestration is too low, the user can timely perform carbon sequestration operations through manual intervention, and can perform manual intervention in sub-areas, which can perform carbon sequestration operations more accurately. Moreover, when obtaining the carbon sequestration prediction data of each sub-area, the data can be verified and calibrated based on the carbon sequestration verification model, ensuring the accuracy of the prediction data.

[0128] An electronic device, comprising:

[0129] At least one processor; and a memory communicatively connected to at least one processor; wherein, the memory stores instructions executable by at least one processor, and the instructions are executed by at least one processor so that at least one processor can execute the method proposed in Embodiment 1 of the present invention.

[0130] The following specifically introduces each component of the electronic device:

[0131] Among them, the processor is the control center of the electronic device, which can be a single processor or a collective term for multiple processing elements. For example, the processor can be one or more central processing units (CPUs), or can be an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement Embodiment 1 of the present invention. For example: one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs).

[0132] Among them, the processor can execute various functions of the electronic device by running or executing software programs stored in the memory and calling data stored in the memory.

[0133] The memory is used to store the software program for implementing the solution of the present invention and is controlled by the processor for execution. The specific implementation manner can refer to the above method embodiments and will not be elaborated here.

[0134] The memory can be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or can also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only (CD-ROM), or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory can be integrated with the processor or exist independently and be coupled to the processor through the interface circuit of the electronic device. The embodiments of the present invention do not make specific limitations on this.

[0135] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0136] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can mean that A exists alone, A and B exist simultaneously, or B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this document generally indicates an "or" relationship between the associated objects before and after, but it may also indicate an "and / or" relationship, which can be understood specifically with reference to the context before and after.

[0137] It should be understood that in the embodiments of the present invention, the order numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0138] The above embodiments are only used to illustrate the technical solutions of the present invention, not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. An artificial intelligence-based soil carbon sequestration and carbon enhancement optimization method, characterized in that: The following steps are included: S101: Conduct a survey on the agricultural planting soil area, obtain the area and boundary of the agricultural planting soil area, establish an agricultural weather detection station and an agricultural weather detection database, detect the weather in the agricultural planting soil area based on the agricultural weather detection station, and obtain weather data, the weather data includes past dates in the agricultural planting soil area and weather types matching the past dates, and import the weather data into the agricultural weather detection database for storage; S102: different plant species planted in the agricultural planting soil area are obtained, different plant species are numbered by characters, and a region division operation is performed based on the plant species in the agricultural planting soil area. Based on the region division operation, the agricultural planting soil area is divided into a plurality of different agricultural planting soil sub-regions according to the planted plant species, different sub-regions are numbered, and the number numbers are matched and combined with the character numbers to form soil region numbers; S103: Obtain historical carbon fixation data and historical dates of different sub-regions, perform a timestamp alignment operation to match the historical carbon fixation data with the historical dates, obtain a data change fluctuation curve group based on the historical carbon fixation data and the historical dates, and import the soil region number into the data change fluctuation curve group. At the same time, call the agricultural weather detection database, obtain a weather factor group based on the weather data in the agricultural weather detection database, and obtain a carbon fixation-time change fluctuation curve group based on the data change fluctuation curve group, the weather factor group and the soil region number; S104: Establishing an agricultural soil carbon sequestration model, making a copy of the carbon sequestration-time variation fluctuation curve group, importing the copy as a training sample into the agricultural soil carbon sequestration model, and deriving an agricultural soil carbon sequestration prediction model based on iterative training. The agricultural soil carbon sequestration prediction model predicts the carbon sequestration of different sub-regions in the next year based on the monthly change rate of the carbon sequestration-time variation fluctuation curve group, thereby obtaining carbon sequestration prediction data matching the soil region number; S105: designing and establishing a carbon sequestration verification model, obtaining carbon sequestration verification data matching the soil region number based on the carbon sequestration verification model, performing data verification and calibration operations based on the carbon sequestration verification data and the carbon sequestration prediction data, and verifying the carbon sequestration prediction data based on the data verification and calibration operations; S106: Based on the verified carbon sequestration prediction data, carbon enhancement optimization operations are performed on sub-regions with different soil area numbers.

2. The artificial intelligence-based soil carbon sequestration and carbon enhancement optimization method according to claim 1 is characterized in that: Design and establish a carbon sequestration verification model, and obtain carbon sequestration verification data matching the soil area number based on the carbon sequestration verification model, including: A1: Construct a small-scale regional digital map, which is constructed based on the actual geographical data of the actual soil area; construct a small-scale agricultural planting soil area digital sand table based on the small-scale regional digital map, which contains planted plants. The small-scale agricultural planting soil area digital sand table is used to simulate the environment of the agricultural planting soil area, divide the small-scale agricultural planting soil area digital sand table into regions, and number the soil areas at the same time; A2: A virtual weather observation station and a virtual weather generation station are established in the digital sandbox of the small agricultural planting soil area. The virtual weather generation station is used to dynamically simulate the actual weather, and the virtual weather observation station is used to detect the weather in the sandbox of the small agricultural planting soil area and obtain virtual weather data; A3: Construct a carbon sequestration verification model. The carbon sequestration verification model is constructed using the DAYCENT model; A4: Import virtual weather data and agricultural soil area related parameters into the carbon sequestration verification model, perform verification simulation through the carbon sequestration verification model, and generate carbon sequestration verification data matching the soil area number.

3. The artificial intelligence-based soil carbon sequestration and carbon enhancement optimization method according to claim 2 is characterized in that: The carbon fixation verification model is simulated based on the carbon cycle function, which is expressed as: SOC T =SOC T-1 +GPP T -R T -M T ; Among them, SOC T Represents the soil carbon content at time T, SOC T-1 Indicates the soil carbon content at the last moment T, GPP T represents the total primary productivity at time T, R T represents the amount of carbon released by the respiration of plants and soil at time T, M T It is expressed as the amount of carbon released from soil organic matter; During the simulation, virtual weather data is imported into the carbon fixation verification model as a virtual weather factor. After the virtual weather factor is imported, the carbon fixation verification model uses dynamic calculation simulation to perform simulation.

4. The artificial intelligence-based soil carbon fixation and carbon enhancement optimization method according to claim 1 or 2, characterized in that: The regional division operation specifically includes: B1: Obtaining rectangular NDVI images of agricultural planting soil areas based on remote sensing technology; B2: De-noising the rectangular NDVI image based on wavelet transform technology, and then sharpening and contrast enhancing the rectangular NDVI image after denoising; B3: The Canny edge detection algorithm is used to detect the boundary of the soil area based on the brightness gradient of the rectangular NDVI image. At the same time, the Canny edge detection algorithm is used to remove useless images from the rectangular NDVI image to obtain a curved NDVI image. The useless images are the image parts of the rectangular image that do not contain the agricultural soil area. B4: Image segmentation of curved NDVI images is performed based on the region growing algorithm, and the areas with similar spectral characteristics in the NDVI images are divided into multiple sub-areas; B5: Create a color gradient table that matches the character numbers of different plant species. The color of the bare soil roads in the area is defined as white. The NDVI values ​​in the sub-areas of the curved NDVI image are mapped to different colors based on the color mapping technology. The color mapping is based on the color gradient table.

5. The artificial intelligence-based soil carbon sequestration and carbon enhancement optimization method according to claim 4 is characterized in that: Based on the Canny edge detection algorithm, the useless images in the rectangular NDVI image are removed to obtain the curved NDVI image, including: After obtaining the curved NDVI image, the curved NDVI image is corrected based on the area of ​​the agricultural planting soil region and the boundary of the agricultural planting soil region.

6. The artificial intelligence-based soil carbon sequestration and carbon enhancement optimization method according to claim 5 is characterized in that: Obtain rectangular images of agricultural planting soil areas based on remote sensing technology, including: The rectangular image of the agricultural planting soil area obtained is a high-resolution image.

7. The artificial intelligence-based soil carbon sequestration and carbon enhancement optimization method according to claim 1 is characterized in that: Based on the matching combination of digital numbers and character numbers, the combination is a soil area number, including: When numbering, use the numbers 1, 2, 3, 4 to n, where n represents the number of agricultural planting soil sub-regions; When performing character numbering, the names of different plant species are obtained, and the first letters are extracted based on the pinyin of the different plant species, and character encoding is performed based on the pinyin first letter combination of the different plant species. If the character codes are the same, a number is added to the end of the pinyin first letter combination, and at the same time, the character codes are assigned to different plant species for character numbering; The digital number and the character number are matched and combined to form a soil area number. The soil area number is expressed as (count, name), where count is a digital number and name is a character number.

8. The artificial intelligence-based soil carbon sequestration and carbon enhancement optimization method according to claim 1 is characterized in that: Carbon addition optimization operations specifically include: Carbon is added to the soil, and when adding carbon, a combined carbon addition method is used; A large amount of first carbon-increasing materials are used to increase carbon in the soil, wherein the first carbon-increasing materials include decomposed straw and harmlessly treated feces; Using a second carbon-enhancing material in an appropriate amount to increase carbon in the soil, wherein the second carbon-enhancing material is a strain of bacteria that meets the standards of food, feed and fertilizer; A small amount of accurate third carbon-enhancing material is added to the soil to increase carbon. The third carbon-enhancing material is represented by mineral elements that meet the standards of organic cultivation; The user can adjust the ratio and total usage of the first carbon-increasing material, the second carbon-increasing material and the third carbon-increasing material based on the carbon fixation prediction data.

9. The artificial intelligence-based soil carbon sequestration and carbon enhancement optimization method according to claim 1, characterized in that: The data verification and validation operations specifically include: Obtaining a verification threshold, the verification threshold is used to verify the carbon fixation prediction data and the carbon fixation verification data, subtracting the carbon fixation prediction data from the carbon fixation verification data and taking the absolute value to obtain a carbon fixation comparison value, and comparing the carbon fixation comparison value with the verification threshold; If the carbon fixation comparison value is less than or equal to the verification threshold, the carbon fixation prediction data is derived as the verified carbon fixation prediction data; If the carbon sequestration comparison value is greater than the verification threshold, a re-prediction is performed based on the agricultural soil carbon sequestration prediction model.

10. The soil carbon sequestration and carbon enhancement optimization system based on artificial intelligence is characterized by: include: An acquisition module is used to acquire the area and boundary of the agricultural planting soil area, acquire different plant species planted in the agricultural planting soil area, and acquire historical carbon sequestration data and historical dates of different sub-regions; Weather module, the weather module includes agricultural weather detection stations and agricultural weather detection database; Agricultural weather monitoring stations are used to monitor the weather in agricultural planting soil areas and obtain weather data; Agricultural weather detection database, used to store weather data detected by agricultural weather detection stations; A region module, used to perform region division operations; A numbering module, used for character numbering of different plant species, digital numbering of different sub-regions, and generating soil region numbers based on the character numbers and digital numbers; A processing module is used to obtain a data change fluctuation curve group based on historical carbon fixation data and historical dates, import the soil area number into the data change fluctuation curve group, and call the agricultural weather detection database to obtain a weather factor group, and obtain a carbon fixation-time change fluctuation curve group based on the weather factor group, the data change fluctuation curve group and the soil area number; A model module is used to establish an agricultural soil carbon sequestration model and generate an agricultural soil carbon sequestration prediction model based on a carbon sequestration-time variation fluctuation curve group, thereby obtaining carbon sequestration prediction data matching the soil area number based on the agricultural soil carbon sequestration prediction model; A verification simulation module is used to design and establish a carbon sequestration verification model, and obtain carbon sequestration verification data matching the soil area number based on the carbon sequestration verification model; The verification module is used to perform data verification and calibration operations on the carbon fixation verification data and the carbon fixation prediction data.

Images