Soil data automatic processing method and system

Abstract

The invention relates to a soil data automatic processing method and system, and the method comprises the following steps: obtaining the basic information and monitoring demands of target soil, and carrying out the matching to generate a soil monitoring scheme; arranging a data acquisition end according to the data processing flow, and acquiring soil data of the target soil; preprocessing the collected soil data according to a data cleaning rule to obtain soil data with a uniform data format; and sending the preprocessed soil data to a preset soil data analysis model, carrying out data monitoring analysis according to a data processing flow, and sending alarm information when abnormal data is monitored. The method has the effect of improving the soil data processing accuracy and efficiency.

Description

Technical Field

[0001] This application relates to the field of soil data processing, and in particular to a method and system for automatic soil data processing. Background Art

[0002] With the growth of the population and the acceleration of the urbanization process, land resources have become more precious. To achieve the sustainable use of land resources, it is necessary to conduct long-term monitoring of soil quality and fertility. Automatic soil data processing can complete this task more efficiently and help managers formulate reasonable land use plans.

[0003] Soil is a complex multiphase system, and its physical, chemical, and biological properties are intertwined. Under different soil types (such as sandy soil, loam soil, clay soil) and environmental conditions (such as different climate zones, vegetation coverage), the applicability of data processing methods will vary.

[0004] Existing soil data collection and processing mostly adopt manual timing collection, which not only consumes a large amount of manpower and material resources, especially when dealing with large areas of land, but also cannot achieve high-frequency data collection. This leads to the inability to capture key information in some studies sensitive to soil dynamic changes in a timely manner. Moreover, during manual collection, there are likely to be differences in collection standards and operations, resulting in data errors. Although there are some cases of soil data monitoring using automated solutions, not only do the design schemes require a large amount of manpower and material resources, but the designed schemes are highly targeted and lack universality, which leads to data compatibility problems in the soil data of each plot and makes it impossible to effectively process and analyze soil data. Summary of the Invention

[0005] To solve the above problems, this application provides a method and system for automatic soil data processing.

[0006] In the first aspect, this application provides a method for automatic soil data processing, adopting the following technical solution: A method for automatic soil data processing includes the following steps: Obtain the basic information and monitoring requirements of the target soil, and match and generate a soil monitoring plan; the basic information of the target soil includes soil geographical information and soil type information; the monitoring requirements of the target soil include one or more of the monitoring requirements for soil physical property data, soil chemical property data, and soil biological property data; the soil monitoring plan includes at least one data collection end, data cleaning rules, and data processing flow information; Arrange the data collection end according to the data processing flow, and collect the soil data of the target soil; Preprocess the collected soil data according to the data cleaning rules to obtain soil data with a unified data format; Send the preprocessed soil data to a preset soil data analysis model, perform data monitoring and analysis according to the data processing flow, and send an alarm message when abnormal data is detected. The soil data analysis model is obtained by deep learning of historical soil monitoring data samples through a neural network model.

[0007] Preferably, it further includes: extracting soil data and analysis results within a specified time period according to user requirements or regularly, performing visualization processing to draw soil data charts, and generating a visualization report.

[0008] Preferably, the data acquisition terminal can be any one of a sensor, an automatic sampling device, an automatic remote sensing device, and an automatic observation device.

[0009] Preferably, the steps of obtaining the basic information and monitoring requirements of the target soil and matching and generating a soil monitoring plan specifically include the following steps: Obtain the basic information and monitoring requirements of the target soil, and match and generate multiple alternative monitoring plans that meet the monitoring requirements through a preset plan matching model; the plan matching model is obtained by iterative training of historical soil monitoring plan data through a machine learning model; Calculate the plan scores of each alternative monitoring plan through a preset plan score calculation formula; Select the alternative monitoring plan with the highest plan score as the soil monitoring plan for the target soil.

[0010] Preferably, the plan score calculation formula is specifically: ; where n is the total number of data acquisition terminals in the alternative plan, is the cost of the i-th data acquisition terminal in the alternative monitoring plan, is the average service life of the i-th data acquisition terminal in the alternative monitoring plan, Z1 is a preset cost benchmark, and X1 is a preset cost scoring coefficient; is the measurement error percentage of the acquisition accuracy of the i-th data acquisition terminal in the alternative monitoring plan, Z2 is a preset accuracy error benchmark, and X2 is a preset accuracy scoring coefficient; The data error rate of the i-th data acquisition terminal in the alternative monitoring plan, Z3 is a preset data transmission error benchmark, and X3 is a preset data scoring coefficient; and X1, X2, and X3 are all set by the management personnel.

[0011] Preferably, the steps of preprocessing the collected soil data according to the data cleaning rules to obtain soil data with a unified data format specifically include the following steps: Clean the collected soil data according to the data cleaning rules, remove abnormal data and fill in missing data; Normalize the cleaned data to obtain soil data with a unified data format.

[0012] Preferably, step 7 of sending an alarm message when abnormal data is detected specifically includes the following steps: After detecting abnormal data, control the video acquisition device to acquire the image information of the soil monitored by the data acquisition end to which the abnormal data belongs, and identify whether there is a temporary abnormal source at the soil monitored by the data acquisition end through a pre-set image recognition model; If there is, generate an abnormal cleaning instruction, package it into the alarm message and send it to the management personnel; If not, obtain the geographical information and building information around the target soil, and analyze and determine whether there is a fixed abnormal source through the soil data analysis model; If there is, generate an abnormal source prompt message, package it into the alarm message and send it to the management personnel; If not, generate a calibration request for the acquisition end, package it into the alarm message and send it to the management personnel.

[0013] In a second aspect, the present application provides a soil data automatic processing system, adopting the following technical solutions: A soil data automatic processing system, comprising: A scheme generation module, configured to obtain the basic information and monitoring requirements of the target soil, and match and generate a soil monitoring scheme; the basic information of the target soil includes soil geographical information and soil type information; the monitoring requirements of the target soil include one or more of the monitoring requirements for soil physical property data, soil chemical property data, and soil biological property data; the soil monitoring scheme includes at least one data acquisition end, data cleaning rules, and data processing flow information; A data acquisition module, configured to deploy a data acquisition end according to the data processing flow and acquire the soil data of the target soil; A data processing module, configured to preprocess the acquired soil data according to the data cleaning rules to obtain soil data with a unified data format; A data analysis module, configured to send the preprocessed soil data to a pre-set soil data analysis model, perform data monitoring and analysis according to the data processing flow, and send an alarm message when abnormal data is detected, and the soil data analysis model is obtained by deep learning through historical soil monitoring data samples.

[0014] Preferably, the scheme generation module includes: A solution matching unit is configured to obtain the basic information and monitoring requirements of the target soil, and generate multiple alternative monitoring solutions that meet the monitoring requirements through a pre-set solution matching model; the solution matching model is obtained by iterative training of a machine learning model using historical soil monitoring solution data; A solution scoring unit is configured to calculate the solution scores of each alternative monitoring solution through a pre-set solution scoring calculation formula; A solution selection unit is configured to select the alternative monitoring solution with the highest solution score as the soil monitoring solution for the target soil.

[0015] Preferably, the data analysis module sending an alarm message when abnormal data is detected specifically includes the following steps: After detecting abnormal data, control the video acquisition device to acquire the image information of the soil monitored by the data acquisition end to which the abnormal data belongs, and identify whether there is a temporary abnormal source at the soil monitored by the data acquisition end through a pre-set image recognition model; If there is, generate an abnormal cleaning instruction, package it into the alarm message, and send it to the management personnel; If not, obtain the geographical information and building information around the target soil, and analyze and determine whether there is a fixed abnormal source through the soil data analysis model; If there is, generate an abnormal source prompt message, package it into the alarm message, and send it to the management personnel; If not, generate a calibration request for the acquisition end, package it into the alarm message, and send it to the management personnel.

[0016] In summary, the present application includes at least one of the following beneficial technical effects: 1. According to the actual situation of the target soil and combined with the user's soil monitoring requirements, intelligently and personalized customize the soil monitoring solution through the solution matching model, select the best data acquisition end, ensure that the collected data is closely related to the actual requirements, provide a reliable basis for subsequent accurate analysis and decision-making, and there is no need for manual consumption of a large amount of manpower and material resources to customize the soil monitoring solution; then automatically arrange the data acquisition end according to the data processing process, which can realize continuous and automatic acquisition of soil data, greatly reducing the workload and time cost of manual sampling; then preprocess the collected soil data, which helps to realize the efficient and automatic processing of soil data, achieving the effect of improving the accuracy and efficiency of data processing; 2. Score each of the matched alternative monitoring solutions from three aspects: data acquisition cost, data acquisition accuracy, and data acquisition stability, which can accurately and scientifically select the soil monitoring solution suitable for the target soil, realize the efficient and automatic processing of soil data, and achieve the effect of improving the accuracy and efficiency of data processing; 3. According to the soil anomaly data, timely cleaning up the temporary anomaly sources, investigating the fixed anomaly sources, and calibrating the data acquisition terminals can keep the monitoring system in a good running state continuously, realize the benign and efficient automatic processing of soil data, and reduce the data loss or errors caused by external interference or equipment failures. Description of the Drawings

[0017] Figure 1 is the flowchart of a method for automatic processing of soil data in an embodiment of the present application; Figure 2 is the flowchart of a method for matching and generating a soil monitoring plan in an embodiment of the present application; Figure 3 is the flowchart of a method for preprocessing soil data in an embodiment of the present application; Figure 4 is the flowchart of a method for determining anomaly sources when abnormal data is monitored in an embodiment of the present application; Figure 5 is the system block diagram of a soil data automatic processing system in an embodiment of the present application.

[0018] Description of the reference numerals: 1. Scheme generation module; 11. Scheme matching unit; 12. Scheme scoring unit; 13. Scheme selection unit; 2. Data acquisition module; 3. Data processing module; 4. Data analysis module. Detailed Description of the Embodiment

[0019] The following further describes the present application in detail Figures 1-5 with reference to the accompanying drawings.

[0020] An embodiment of the present application discloses a method for automatic processing of soil data. Referring to Figure 1 , a method for automatic processing of soil data includes the following steps: S1. Obtain a soil monitoring plan: Obtain the basic information and monitoring requirements of the target soil, and match and generate a soil monitoring plan; the basic information of the target soil includes soil geographical information and soil type information; the monitoring requirements of the target soil include one or more of the monitoring requirements for soil physical property data, soil chemical property data, and soil biological property data; the soil monitoring plan includes at least one data acquisition terminal, data cleaning rules, and data processing flow information; S2. Collect soil data: Arrange data acquisition terminals according to the data processing flow, and collect the soil data of the target soil; S3. Preprocess the soil data: Preprocess the collected soil data according to the data cleaning rules to obtain soil data with a unified data format; S4. Soil data monitoring and analysis: Send the preprocessed soil data to a preset soil data analysis model, conduct data monitoring and analysis according to the data processing flow, and send an alarm message when abnormal data is detected. The soil data analysis model is obtained by deep learning of the neural network model through historical soil monitoring data samples. It should be noted that the specific training steps of the neural network model are prior art and will not be elaborated here. S5. Visualization processing: Extract soil data and analysis results within a specified time period according to user requirements or at regular intervals, perform visualization processing to draw soil data charts, and generate a visualization report. According to the actual situation of the target soil and the user's soil monitoring requirements, intelligently and personalized customize the soil monitoring plan through the plan matching model, select the best data acquisition terminal to ensure that the collected data is closely related to the actual needs, providing a reliable basis for subsequent accurate analysis and decision-making, without the need for manual consumption of a large amount of manpower and material resources to customize the soil monitoring plan. Then, automatically deploy the data acquisition terminal according to the data processing flow, which can realize continuous and automatic collection of soil data, greatly reducing the workload and time cost of manual sampling. Then, preprocess the collected soil data, which helps to achieve efficient and automatic processing of soil data, achieving the effect of improving the accuracy and efficiency of data processing.

[0021] In addition, according to actual needs, display the complex soil data and analysis results in the form of intuitive charts through step S5, and generate a detailed visualization report. Different user groups, such as scientific researchers, agricultural technology extension personnel, or land managers, can quickly and clearly understand the change trends of various soil indicators, the relationships between data, and the conclusions and suggestions obtained through the analysis from the visualization report.

[0022] Among them, the data acquisition terminal can be any one of sensors, automatic sampling devices, automatic remote sensing devices, and automatic observation devices. For example, a soil moisture sensor in a sensor, an intelligent soil testing robot or an unmanned aerial vehicle sampling device in an automatic sampling device; satellite remote sensing devices and aerial remote sensing devices for obtaining large-area soil information in an automatic remote sensing device; a soil moisture and temperature observation system in an automatic observation device, such as the JC-GTS3 soil moisture and temperature observation system, which can collect soil moisture and temperature data in real time and continuously, and transmit the data to the central processing system through wireless transmission technology. Farmers can view the data at any time through terminal devices such as mobile phones and computers. By determining the soil monitoring plan and selecting the data acquisition terminal according to local conditions through the plan matching model, it can meet the monitoring needs of various soils and greatly improve the universality.

[0023] Refer to Figure 2 , the steps of obtaining the basic information and monitoring requirements of the target soil and matching and generating a soil monitoring plan specifically include the following steps: A1. Generate alternative monitoring plans: Obtain the basic information of the target soil and monitoring requirements, and generate multiple alternative monitoring plans that meet the monitoring requirements through a pre-set plan matching model; the plan matching model is obtained by iterative training of a machine learning model with historical soil monitoring plan data; it should be noted that the specific training steps of the machine learning model are prior art and will not be elaborated here; A2. Calculate plan scores: Calculate the plan scores of each alternative monitoring plan through a pre-set plan score calculation formula; A3. Select a soil monitoring plan: Select the alternative monitoring plan with the highest plan score as the soil monitoring plan for the target soil.

[0024] The above-mentioned plan score calculation formula is specifically: ; where n is the total number of data acquisition ends in the alternative plans, is the cost of the i-th data acquisition end in the alternative monitoring plan, is the average service life of the i-th data acquisition end in the alternative monitoring plan, Z1 is a pre-set cost benchmark, and X1 is a pre-set cost score coefficient; is the measurement error percentage of the acquisition accuracy of the i-th data acquisition end in the alternative monitoring plan, Z2 is a pre-set accuracy error benchmark, and X2 is a pre-set accuracy score coefficient; is the data error rate of the i-th data acquisition end in the alternative monitoring plan, Z3 is a pre-set data transmission error benchmark, and X3 is a pre-set data score coefficient; and X1, X2, and X3 are all set by the management personnel. By scoring each of the matched alternative monitoring plans from three aspects: data acquisition cost, data acquisition accuracy, and data acquisition stability, a soil monitoring plan suitable for the target soil can be accurately and scientifically selected, realizing the efficient and automated processing of soil data, and achieving the effect of improving the accuracy and efficiency of data processing.

[0025] Refer to Figure 3 and the specific steps for preprocessing the collected soil data according to the data cleaning rules to obtain soil data with a unified data format are as follows: B1. Data cleaning: Clean the collected soil data according to the data cleaning rules, remove abnormal data, and fill in missing data; B2. Normalization processing: Perform normalization processing on the cleaned data to obtain soil data with a unified data format. The scheme matching model determines the soil monitoring scheme according to the actual situation of the target soil and the monitoring requirements, and selects data acquisition terminals with consistent data specifications and accurate data acquisition as much as possible. Set data cleaning rules to preprocess the collected soil data, which can effectively improve data accuracy, enhance data integrity, facilitate data comparison and integration, optimize data analysis efficiency and accuracy, and help the soil data analysis model analyze and process soil data efficiently, achieving the effect of improving the accuracy and efficiency of soil data processing..

[0026] Refer to Figure 4 , The specific steps for sending an alarm message when abnormal data is detected are as follows: C1. Determine whether there is a temporary abnormal source at the soil monitored by the data acquisition terminal: After detecting abnormal data, control the video acquisition device to collect the image information of the soil monitored by the data acquisition terminal where the abnormal data belongs, and identify whether there is a temporary abnormal source at the soil monitored by the data acquisition terminal through a pre-set image recognition model; The image recognition model is based on the ChatGPT4.0 model and is trained using soil pollution source samples; The video acquisition device can be a fixed video acquisition device such as a monitor, or a movable video acquisition device such as a drone or an automatic patrol vehicle; C2. If there is, generate an abnormal cleaning instruction, package it into the alarm message, and send it to the management personnel; C3. Analyze and determine whether there is a fixed abnormal source: If not, obtain the geographical information and building information around the target soil, and analyze and determine whether there is a fixed abnormal source through the soil data analysis model; C4. If there is, generate an abnormal source prompt message, package it into the alarm message, and send it to the management personnel; C5. If not, generate a calibration request for the data acquisition end, package it into an alarm message, and send it to the management personnel. Through the above steps, when the soil data is abnormal, capture the soil images in the abnormal area. With the powerful image recognition ability of the ChatGPT4.0 model, determine whether there is a temporary abnormal source, such as animal activities (feces, damage), recent human construction, or sudden natural disasters (such as local small debris flows covering soil monitoring points). After excluding the temporary abnormal source, further analyze whether there is a fixed abnormal source by combining the surrounding geographical information and building information using the soil data analysis model, such as long-term pollutant emissions from nearby factories causing abnormal soil chemical properties, underground pipeline leaks affecting soil humidity, or the soil being in a special terrain (such as low-lying waterlogged areas causing the soil to be overly wet for a long time). This helps to accurately locate the root cause of the abnormal data, providing a key basis for subsequent targeted soil improvement, pollution control, or adjustment of monitoring points. And generate alarm information accordingly, which helps to achieve an efficient problem handling response; adopting this processing flow not only focuses on the abnormal data itself but also on the maintenance of the entire soil monitoring system. By promptly clearing the temporary abnormal source, investigating the fixed abnormal source, and calibrating the data acquisition end, the monitoring system can be continuously in a good operating state, realizing the benign and efficient automated processing of soil data, and reducing data loss or errors caused by external interference or equipment failures.

[0027] This embodiment of the present application also discloses an automated soil data processing system. Referring to Figure 5 , an automated soil data processing system includes: A scheme generation module 1, configured to obtain the basic information and monitoring requirements of the target soil, and match and generate a soil monitoring scheme; the basic information of the target soil includes soil geographical information and soil type information; the monitoring requirements of the target soil include one or more of the monitoring requirements for soil physical property data, soil chemical property data, and soil biological property data; the soil monitoring scheme includes at least one data acquisition end, data cleaning rules, and data processing flow information; A data acquisition module 2, configured to deploy data acquisition ends according to the data processing flow and acquire the soil data of the target soil; A data processing module 3, configured to preprocess the acquired soil data according to the data cleaning rules to obtain soil data with a unified data format; A data analysis module 4, configured to send the preprocessed soil data to a preset soil data analysis model, perform data monitoring and analysis according to the data processing flow, and send an alarm message when abnormal data is detected. The soil data analysis model is obtained by deep learning through historical soil monitoring data samples; A visualization module 5 is used to extract soil data and analysis results within a specified time period according to user requirements or at regular intervals, perform visualization processing to draw soil data charts, and generate visualization reports. According to the actual situation of the target soil and the user's soil monitoring requirements, a soil monitoring plan is intelligently and personalized customized through a plan matching model, and the best data acquisition terminal is selected to ensure that the collected data is closely related to the actual requirements, providing a reliable basis for subsequent accurate analysis and decision-making, without the need for manual consumption of a large amount of manpower and material resources to customize the soil monitoring plan; then, according to the data processing process, the data acquisition terminal is automatically arranged, which can realize continuous and automatic acquisition of soil data, greatly reducing the workload and time cost of manual sampling; then, the collected soil data is preprocessed, which helps to realize the efficient and automatic processing of soil data, achieving the effect of improving the accuracy and efficiency of data processing.

[0028] In addition, according to actual requirements, the visualization module displays complex soil data and analysis results in the form of intuitive charts and generates detailed visualization reports. Different user groups, such as scientific researchers, agricultural technology extension personnel, and land managers, can quickly and clearly understand the change trends of various soil indicators, the relationships between data, and the conclusions and suggestions obtained from the analysis through the visualization reports.

[0029] Refer to Figure 5 , the scheme generation module 1 includes: A scheme matching unit 11 is used to obtain the basic information and monitoring requirements of the target soil, and match and generate multiple alternative monitoring schemes that meet the monitoring requirements through a pre-set scheme matching model; the scheme matching model is an iterative training of a machine learning model through historical soil monitoring scheme data; A scheme scoring unit 12 is used to calculate the scheme scores of each alternative monitoring scheme through a pre-set scheme scoring calculation formula; A scheme selection unit 13 is used to select the alternative monitoring scheme with the highest scheme score as the soil monitoring scheme for the target soil.

[0030] The above scheme scoring calculation formula is specifically: ; where n is the total number of data acquisition terminals in the alternative scheme, is the cost of the i-th data acquisition terminal in the alternative monitoring scheme, is the average service life of the i-th data acquisition terminal in the alternative monitoring scheme, Z1 is a pre-set cost benchmark, and X1 is a pre-set cost scoring coefficient; is the measurement error percentage of the acquisition accuracy of the i-th data acquisition terminal in the alternative monitoring scheme, Z2 is a pre-set accuracy error benchmark, and X2 is a pre-set accuracy scoring coefficient; The bit error rate of the i-th data acquisition terminal in the alternative monitoring scheme, Z3 is the preset data transmission error benchmark, and X3 is the preset data scoring coefficient; and X1, X2, and X3 are all set by the management personnel. By scoring each of the matched alternative monitoring schemes from three aspects: data acquisition cost, data acquisition accuracy, and data acquisition stability, a soil monitoring scheme suitable for the target soil can be accurately and scientifically selected, realizing the efficient and automated processing of soil data, and achieving the effect of improving the accuracy and efficiency of data processing.

[0031] The above data analysis module 4 sends an alarm message when detecting abnormal data, which specifically includes the following steps: After detecting abnormal data, control the video acquisition device to acquire the image information of the soil monitored by the data acquisition terminal to which the abnormal data belongs, and identify whether there is a temporary abnormal source at the soil monitored by the data acquisition terminal through the preset image recognition model; If there is, generate an abnormal cleaning instruction, package it into the alarm message, and send it to the management personnel; If not, obtain the geographical information and building information around the target soil, and analyze and determine whether there is a fixed abnormal source through the soil data analysis model; If there is, generate an abnormal source prompt message, package it into the alarm message, and send it to the management personnel; If not, generate a calibration request for the data acquisition terminal, package it into the alarm message, and send it to the management personnel. Through the above steps, when abnormal soil data appears, collect the soil images in the abnormal area, and rely on the powerful image recognition ability of the ChatGPT4.0 model to judge whether there is a temporary abnormal source, such as animal activities (feces, damage), recent human construction, or sudden natural disasters (such as local small mudslides covering soil monitoring points), etc. After excluding the temporary abnormal source, further combine the surrounding geographical information and building information, and use the soil data analysis model to analyze whether there is a fixed abnormal source, such as long-term pollutant emissions from nearby factories causing abnormal soil chemical properties, underground pipeline leaks affecting soil moisture, or the soil being in a special terrain (such as low-lying waterlogged areas causing the soil to be overly wet for a long time). This helps to accurately locate the root cause of abnormal data, providing a key basis for subsequent targeted soil improvement, pollution control, or monitoring point adjustment. And generate alarm messages accordingly, which helps to achieve an efficient problem handling response; adopting this processing flow not only focuses on the abnormal data itself but also on the maintenance of the entire soil monitoring system. By promptly cleaning up temporary abnormal sources, investigating fixed abnormal sources, and calibrating the data acquisition terminal, the monitoring system can be continuously in a good operating state, realizing the benign and efficient automated processing of soil data, and reducing data loss or errors caused by external interference or equipment failures.

[0032] The embodiments of the present application also disclose a computer-readable storage medium, which stores a computer program that can be loaded and executed by a processor and is the same as the method described above. The computer-readable storage medium includes, for example, various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0033] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the protection scope of the invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on these embodiments, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art can still, without conflict and without creative efforts, combine, add, delete, or make other adjustments to the features in the embodiments of the present invention according to the situation, so as to obtain different technical solutions that are essentially not divorced from the concept of the present invention, and these technical solutions also fall within the protection scope of the present invention.

Claims

1. A soil data automatic processing method, characterized in that: The following steps are involved: Obtain basic information and monitoring requirements of the target soil, and match and generate a soil monitoring plan; the basic information of the target soil includes soil geographic information and soil type information; the monitoring requirements of the target soil include one or more of soil physical property data monitoring requirements, soil chemical property data monitoring requirements, and soil biological property data monitoring requirements; the soil monitoring plan includes at least one data collection terminal, data cleaning rules, and data processing flow information; According to the data processing flow, a data collection terminal is arranged to collect soil data of the target soil; Preprocessing the collected soil data according to data cleaning rules to obtain soil data with a unified data format; The preprocessed soil data is sent to a preset soil data analysis model, and data monitoring and analysis are performed according to the data processing process, and an alarm message is sent when abnormal data is monitored. The soil data analysis model is a neural network model obtained by deep learning of historical soil monitoring data samples.

2. The method for automatic soil data processing according to claim 1, characterized in that: Also includes: According to user needs or timing, soil data and analysis results within a specified time period are extracted, and soil data charts are drawn for visualization to generate visualization reports.

3. The method for automatic soil data processing according to claim 1, characterized in that: The data acquisition end may be any one of a sensor, an automatic sampling device, an automatic remote sensing device, and an automatic observation device.

4. The method for automatic soil data processing according to claim 1, characterized in that: The acquisition of basic information of the target soil and monitoring requirements and matching and generating a soil monitoring plan specifically include the following steps: Obtain basic information and monitoring requirements of the target soil, and generate multiple alternative monitoring plans that meet the monitoring requirements through a preset plan matching model; the plan matching model is a machine learning model obtained by iterative training of historical soil monitoring plan data; Calculate the scheme score of each alternative monitoring scheme through the preset scheme score calculation formula; The alternative monitoring scheme with the highest scheme score is selected as the soil monitoring scheme for the target soil.

5. The method for automatic soil data processing according to claim 4, characterized in that: The specific calculation formula for the scheme score is: ； Where n is the total number of data collection terminals in the alternative solution, is the cost of the i-th data collection terminal in the alternative monitoring solution, is the average service life of the ith data collection terminal in the alternative monitoring scheme, Z1 is the preset cost benchmark, and X1 is the preset cost scoring coefficient; is the measurement error percentage of the collection accuracy of the i-th data collection terminal in the alternative monitoring scheme, Z2 is the preset accuracy error benchmark, and X2 is the preset accuracy scoring coefficient; The data bit error rate of the i-th data acquisition terminal in the alternative monitoring scheme, Z3 is the preset data transmission error benchmark, X3 is the preset data scoring coefficient; and X1, X2, and X3 are all set by the management personnel.

6. The method for automatic soil data processing according to claim 1, characterized in that: The method of preprocessing the collected soil data according to the data cleaning rules to obtain soil data with a unified data format specifically includes the following steps: Clean the collected soil data according to the data cleaning rules, remove abnormal data and fill in missing data; The cleaned data are normalized to obtain soil data with a unified data format.

7. The method for automatic soil data processing according to claim 1, characterized in that: The sending of alarm information when abnormal data is detected specifically includes the following steps: After abnormal data is detected, the video acquisition device is controlled to collect image information of the soil monitored by the data acquisition end to which the abnormal data belongs, and a preset image recognition model is used to identify whether there is a temporary abnormal source at the soil monitored by the data acquisition end; If it exists, an exception cleanup instruction is generated and packaged into an alarm message and sent to the management staff; If it does not exist, obtain the geographic information and building information around the target soil, and use the soil data analysis model to analyze and determine whether there is a fixed abnormal source; If it exists, the abnormal source prompt information is generated and packaged into the alarm information and sent to the management personnel; If it does not exist, a calibration request for the acquisition end is generated and packaged into an alarm message and sent to the management personnel.

8. A soil data automatic processing system, characterized in that include: A scheme generation module (1) is used to obtain basic information and monitoring requirements of target soil, and match and generate a soil monitoring scheme; the basic information of the target soil includes soil geographic information and soil type information; the monitoring requirements of the target soil include one or more of soil physical property data monitoring requirements, soil chemical property data monitoring requirements, and soil biological property data monitoring requirements; the soil monitoring scheme includes at least one data collection terminal, data cleaning rules, and data processing flow information; A data acquisition module (2) is used to arrange a data acquisition terminal according to a data processing flow to acquire soil data of the target soil; A data processing module (3) is used to pre-process the collected soil data according to data cleaning rules to obtain soil data in a unified data format; The data analysis module (4) is used to send the pre-processed soil data to a preset soil data analysis model, perform data monitoring and analysis according to the data processing process, and send an alarm message when abnormal data is monitored. The soil data analysis model is a neural network model obtained by deep learning through historical soil monitoring data samples.

9. The soil data automatic processing system according to claim 8, characterized in that: The solution generation module (1) comprises: A scheme matching unit (11) is used to obtain basic information of the target soil and monitoring requirements, and to generate a plurality of alternative monitoring schemes that meet the monitoring requirements through matching with a preset scheme matching model; the scheme matching model is a machine learning model obtained through iterative training of historical soil monitoring scheme data; A scheme scoring unit (12), used to calculate the scheme score of each alternative monitoring scheme using a preset scheme scoring calculation formula; The scheme selection unit (13) is used to select the alternative monitoring scheme with the highest scheme score as the soil monitoring scheme for the target soil.

10. The soil data automatic processing system according to claim 8, characterized in that: The data analysis module (4) sends an alarm message when abnormal data is detected, specifically comprising the following steps: After abnormal data is detected, the video acquisition device is controlled to collect image information of the soil monitored by the data acquisition end to which the abnormal data belongs, and a preset image recognition model is used to identify whether there is a temporary abnormal source at the soil monitored by the data acquisition end; If it exists, an exception cleanup instruction is generated and packaged into an alarm message and sent to the management staff; If it does not exist, obtain the geographic information and building information around the target soil, and use the soil data analysis model to analyze and determine whether there is a fixed abnormal source; If it exists, the abnormal source prompt information is generated and packaged into the alarm information and sent to the management personnel; If it does not exist, a calibration request for the acquisition end is generated and packaged into an alarm message and sent to the management personnel.

Images