A method and system for ecological management of a wasteland

Abstract

The application discloses a kind of abandoned land ecological management method and system, belong to ecological management technical field, the specific steps of the management method are as follows: (1) collecting abandoned land information and analysis;(2) construct abandoned land model to carry out recovery prediction;(3) construct scheme database to carry out scheme replacement;(4) real-time record analysis is carried out to recovery information;The application can improve the accuracy of staff on abandoned land recovery prediction by constructing prediction neural network, facilitate staff to carry out abandoned land analysis, improve staff analysis quality, can be updated and optimized to management scheme, improve abandoned land management efficiency, while simplifying system operation steps, facilitate staff to use.

Description

Technical Field

[0001] This invention relates to the field of ecological governance technology, and in particular to a method and system for ecological governance of abandoned land. Background Technology

[0002] Abandoned land refers to land that cannot be used due to excavation, subsidence, occupation, pollution, or damage caused by mining, industrial, and construction activities, as well as natural disasters. Given that land issues in my country are currently a major constraint on resources, strengthening the ecological management of abandoned land is an important way to alleviate the contradiction between land resource supply and demand. There are many methods for ecological restoration of abandoned land, which can be divided into engineering restoration methods, chemical restoration methods, and biological restoration methods according to the technology used. With the continuous enhancement of my country's comprehensive national strength, abandoned land management has also become one of the important means to improve comprehensive production capacity, promote modern residential construction, and strive to increase residents' income. Existing methods and systems for ecological restoration of abandoned land have low accuracy in predicting restoration, making it inconvenient for staff to analyze abandoned land. In addition, existing methods and systems cannot automatically update and optimize restoration plans, reducing the efficiency of abandoned land restoration. Therefore, we propose a method and system for ecological restoration of abandoned land. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method and system for the ecological management of abandoned land.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: An ecological restoration method for abandoned land, the specific steps of which are as follows: (1) Collect and analyze abandoned land information: The server controls the collection and processing module to collect relevant abandoned land information based on the selection of staff, and analyzes and classifies the collected information, and constructs an information record table to record the abandoned land information; (2) Constructing a model of abandoned land for restoration prediction: Constructing a three-dimensional model of abandoned land based on the collected information on abandoned land, while communicating with the external Internet and capturing relevant restoration information for restoration prediction; (3) Construct a solution database for solution replacement: The solution database receives the governance solutions uploaded by staff and stores them in the solution database. At the same time, an analysis and adjustment model is constructed to adjust and modify the recovery solutions in the solution database. (4) Real-time recording and analysis of recovery information: Real-time collection of the recovery status of abandoned sites, and at the same time, generating visualization charts based on the collected information. Then, the server makes a secondary prediction of the recovery time of abandoned sites based on the visualization charts and feeds back the visualization charts and prediction results to the staff.

[0005] As a further aspect of the present invention, the specific steps of the analysis and classification in step (1) are as follows: Step 1: The acquisition and processing module connects with the remote sensing satellite and receives regional images sent by the remote sensing satellite in real time, and selects the regional images of the corresponding area according to the selection information sent by the staff; Step 2: The acquisition and processing module performs a missing image analysis on the region. If there are missing images in the region, the server communicates with the drone and external monitoring equipment to collect image data of the missing areas and supplement the regional images. Step 3: Classify the collected regional images into mining wastelands, polluted wastelands, disaster wastelands, and vacant wastelands. At the same time, construct soil profiles for each region to analyze soil fertility levels, suitability, and productivity. Also, generate the same number of information record tables as the number of regions collected, and enter the wasteland type, soil fertility level, suitability, and productivity into the information record tables.

[0006] As a further aspect of the present invention, the recovery prediction in step (2) specifically involves the following steps: Step 1: Construct and train a predictive neural network, and import the constructed 3D models of each abandoned site into the predictive neural network. At the same time, the 3D models of abandoned sites receive information recording tables and extract feature information from the abandoned sites. The second step: The predictive neural network converts the non-binary data in the extracted feature information into binary, and performs normalization and feature dimensionality reduction on the converted information to make the features range from 0 to 1. The third step is to convert the processed feature information into image format features to form sample feature maps, then manually label the sample feature maps, perform recovery evaluation on each group of labeled sample feature maps, and draw corresponding recovery curves based on the evaluation results.

[0007] As a further aspect of the present invention, the specific steps for training the predictive neural network in the first step are as follows: S1.1: Predict the communication connection between the neural network and the external Internet, and simultaneously capture information on the restoration of various types of abandoned sites in the past, and integrate and summarize the captured information as a set of simulated data into a simulated dataset; S1.2: Select a simulated data as validation data, and use the validation data repeatedly to verify the accuracy of the predictive neural network; S1.3: Select any subset as the test set, then take the remaining subset as the training set, and make a prediction for each set of data, and output the data with the best prediction result as the optimal parameters; S1.4: Standardize the training set according to the optimal parameters, and finally feed the training samples into the prediction neural network. At the same time, the prediction neural network is trained based on the training samples. S1.5: Draw a curve based on the training results of the predictive neural network, record any abnormal curves, and provide feedback to staff for manual adjustment.

[0008] As a further aspect of the present invention, the specific steps for adjustment and modification in step (3) are as follows: S2.1: The solution database receives the governance solutions uploaded by the staff, matches each group of governance solutions with the corresponding type of abandoned land, and feeds back to the corresponding staff. At the same time, the server receives governance information in real time. S2.2: Simultaneously, the solution analysis module communicates with the external Internet and retrieves existing waste land remediation solutions. It also filters each group of remediation solutions based on waste land suitability assessment, waste residue treatment, soil construction, fertilization and improvement, and landscape restoration. S2.3: Analyze and adjust the model to receive governance information and the selected governance solutions, and learn and train the two sets of data through input, convolution, pooling, fully connected and output to update the governance solutions. At the same time, synchronize it to the mobile devices of governance personnel and prompt the staff to check.

[0009] An ecological governance system for abandoned land includes a server, a predictive analysis module, a data acquisition and processing module, a scheme database, a scheme analysis module, a governance and control module, and a management platform; The server is used to receive instruction information issued by the management platform, control each sub-module, and receive data sent by each sub-module for storage. The data collection and processing module is used to collect and classify the information on abandoned sites selected by the staff. The predictive analysis module is used to predict the restoration cycle of abandoned land. The solution analysis module is used to capture external governance solutions and update and optimize the staff's governance solutions; The governance and control module is used to collect staff work information in real time and calculate the restoration efficiency of abandoned sites for each staff member; The management platform is used for staff to log in and verify their identity, and to provide feedback on relevant data regarding the remediation of abandoned land based on staff's operation instructions.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Compared with previous methods, this method for ecological restoration of abandoned land uses a predictive analysis module to construct and train a predictive neural network. The constructed three-dimensional models of each abandoned land are then imported into this predictive neural network. Simultaneously, the three-dimensional models receive information recording tables, extract feature information from the abandoned land, and perform sign-value conversion, normalization, and feature dimensionality reduction on each set of feature information. The feature information is then converted into image-format features to form sample feature maps. These sample feature maps are manually labeled, and a restoration assessment is performed on each labeled set of sample feature maps. Based on the assessment results, corresponding restoration curves are plotted. By constructing a predictive neural network, the accuracy of abandoned land restoration predictions can be improved, facilitating abandoned land analysis and enhancing the quality of analysis. 2. This wasteland ecological management system includes a scheme analysis module. The scheme database receives management schemes uploaded by staff and matches each group of management schemes with the corresponding types of wasteland. Simultaneously, the server receives management information in real time. The scheme analysis module communicates with the external internet, retrieves existing wasteland management schemes, and filters each group of management schemes. The system then analyzes and adjusts the received management information and the filtered management schemes. It learns and trains the two sets of data through input, convolution, pooling, fully connected layers, and output to update the management schemes. This information is simultaneously synchronized to the management personnel's mobile devices, prompting staff to view the updates. This allows for self-updating and optimization of management schemes, improving wasteland management efficiency and simplifying system operation for easier use by staff. Attached Figure Description

[0011] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0012] Figure 1 This is a flowchart of an ecological remediation method for abandoned land proposed in this invention; Figure 2 This is a system block diagram of an ecological management system for abandoned land proposed in this invention. Detailed Implementation

[0013] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0014] Example 1 Reference Figure 1 This embodiment discloses a method for ecological restoration of abandoned land, the specific steps of which are as follows: Collect and analyze abandoned site information: The server controls the data collection and processing module to collect relevant abandoned site information based on the information selected by the staff. At the same time, the collected information is analyzed and classified, and an information record table is built to record the abandoned site information.

[0015] Specifically, the server communicates with remote sensing satellites and receives regional images transmitted by the satellites in real time. Based on the selection information sent by the staff, it selects regional images of the corresponding areas. Then, it performs a missing image analysis on the collected regional images. If there are missing regional images, the server communicates with drones and external monitoring equipment to collect image data of the missing areas and supplement the regional images. The collected regional images are then classified into mining wastelands, polluted wastelands, disaster wastelands, and vacant wastelands. At the same time, soil profiles are constructed for each region to analyze the soil fertility level, suitability, and productivity. In addition, an equal number of information record tables are generated based on the number of regions collected, and the wasteland type, soil fertility level, suitability, and productivity are entered into the information record tables.

[0016] Constructing abandoned site models for restoration prediction: Based on the collected abandoned site information, construct a three-dimensional model of the abandoned site, and connect it to the external Internet to capture relevant restoration information for restoration prediction.

[0017] Specifically, the predictive analysis module constructs and trains a predictive neural network, and imports the constructed 3D models of each abandoned site into the predictive neural network. Simultaneously, the 3D models of the abandoned sites receive information recording tables and extract feature information from the abandoned sites. The predictive neural network then converts the non-binary data in the extracted feature information into binary data, and performs normalization and feature dimensionality reduction processing on the converted information to reduce the features to the range of 0 to 1. The processed feature information is then converted into image-format features to form sample feature maps. These sample feature maps are then manually labeled, and each set of labeled sample feature maps undergoes a recovery evaluation. Based on the evaluation results, corresponding recovery curves are plotted. By constructing a predictive neural network, the accuracy of abandoned site recovery predictions can be improved, facilitating abandoned site analysis and enhancing the quality of analysis.

[0018] It should be further explained that the predictive neural network communicates with the external Internet and simultaneously captures information on the restoration of various types of abandoned sites. This captured information is then integrated and summarized into a set of simulated datasets. One set of simulated data is selected as validation data and used repeatedly to verify the accuracy of the predictive neural network. An arbitrary subset is then selected as the test set, and the remaining subset is used as the training set. A prediction is made for each set of data, and the best prediction is output as the optimal parameters. The training set is then standardized based on the optimal parameters. Finally, the training samples are fed into the predictive neural network, which learns and trains based on these samples. A graph is then plotted based on the training results, and any abnormal curves are recorded and fed back to staff for manual adjustments.

[0019] A solution database is built for solution replacement: The solution database receives governance solutions uploaded by staff and stores them in the database. At the same time, an analysis and adjustment model is built to adjust and modify the recovery solutions in the solution database.

[0020] Specifically, the solution database receives remediation plans uploaded by staff, matches each group of remediation plans with corresponding types of wasteland, and provides feedback to the relevant staff. Simultaneously, the server receives remediation information in real time. The solution analysis module communicates with the external internet, retrieves existing wasteland remediation plans, and filters each group of remediation plans based on wasteland suitability assessment, waste disposal, soil construction, fertilization and improvement, and landscape restoration. The analysis and adjustment model receives the remediation information and the filtered plans, and learns and trains on the two sets of data through input, convolution, pooling, fully connected layers, and output to update the remediation plans. This information is also synchronized to the remediation personnel's mobile devices, prompting staff to view the updates. This allows for self-updating and optimization of remediation plans, improving wasteland remediation efficiency while simplifying system operation and making it easier for staff to use.

[0021] Real-time recording and analysis of recovery information: Real-time collection of the recovery status of abandoned sites, and generation of visualization charts based on the collected information. The server then makes a secondary prediction of the recovery time of abandoned sites based on the visualization charts, and then feeds back the visualization charts and prediction results to the staff.

[0022] Example 2 Reference Figure 2 This embodiment discloses an ecological governance system for abandoned land, including a server, a predictive analysis module, a data acquisition and processing module, a scheme database, a scheme analysis module, a governance and control module, and a management platform; The server is used to receive instructions from the management platform, control each sub-module, and receive data from each sub-module for storage. The data acquisition and processing module is used to collect and classify the information on abandoned sites selected by staff. The predictive analysis module is used to predict the restoration cycle of abandoned sites; The solution analysis module is used to capture external governance solutions and update and optimize the staff's governance solutions; The governance and control module is used to collect staff work information in real time and calculate the restoration efficiency of abandoned sites for each staff member; The management platform is used for staff to log in and verify their identity, and to provide feedback on relevant data regarding the remediation of abandoned land based on staff's operation instructions.

[0023] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for ecological remediation of abandoned land, characterized in that, The specific steps of this treatment method are as follows: (1) Collect and analyze abandoned land information: The server controls the collection and processing module to collect relevant abandoned land information based on the selection of staff, and analyzes and classifies the collected information, and constructs an information record table to record the abandoned land information; (2) Constructing a model of abandoned land for restoration prediction: Constructing a three-dimensional model of abandoned land based on the collected abandoned land information, communicating with the external Internet, and capturing relevant restoration information for restoration prediction; (3) Construct a solution database for solution replacement: The solution database receives the governance solutions uploaded by staff and stores them in the solution database. At the same time, an analysis and adjustment model is constructed to adjust and modify the recovery solutions in the solution database. (4) Real-time recording and analysis of recovery information: Real-time collection of the recovery status of abandoned sites, and at the same time, generating visualization charts based on the collected information. Then, the server makes a secondary prediction of the recovery time of abandoned sites based on each set of visualization charts, and then feeds back each set of visualization charts and prediction results to the staff. The specific steps for recovery prediction in step (2) are as follows: Step 1: Construct and train a predictive neural network, and import the constructed 3D models of each abandoned site into the predictive neural network. At the same time, the 3D models of abandoned sites receive information recording tables and extract feature information from the abandoned sites. The second step: The predictive neural network converts the non-binary data in the extracted feature information into binary, and performs normalization and feature dimensionality reduction on the converted information to make the features range from 0 to 1. The third step is to convert the processed feature information into image format features to form sample feature maps, then manually label the sample feature maps, perform recovery evaluation on each group of labeled sample feature maps, and draw corresponding recovery curves based on the evaluation results. The specific steps for training the predictive neural network described in the first step are as follows: S1.1: Predict the communication connection between the neural network and the external Internet, and simultaneously capture information on the restoration of various types of abandoned sites in the past, and integrate and summarize the captured information as a set of simulated data into a simulated dataset; S1.2: Select a simulated data as validation data, and use the validation data repeatedly to verify the accuracy of the predictive neural network; S1.3: Select any subset as the test set, then take the remaining subset as the training set, and make a prediction for each set of data, and output the data with the best prediction result as the optimal parameters; S1.4: Standardize the training set according to the optimal parameters, and finally feed the training samples into the prediction neural network. At the same time, the prediction neural network is trained based on the training samples. S1.5: Draw a curve based on the training results of the predictive neural network, record any abnormal curves, and provide feedback to staff for manual adjustment; The specific steps for adjustment and modification described in step (3) are as follows: S2.1: The solution database receives the governance solutions uploaded by the staff, matches each group of governance solutions with the corresponding type of abandoned land, and feeds back to the corresponding staff. At the same time, the server receives governance information in real time. S2.2: Simultaneously, the solution analysis module communicates with the external Internet and retrieves existing waste land remediation solutions. It also filters each group of remediation solutions based on waste land suitability assessment, waste residue treatment, soil construction, fertilization and improvement, and landscape restoration. S2.3: Analyze and adjust the model to receive governance information and the selected governance solutions, and learn and train the two sets of data through input, convolution, pooling, fully connected and output to update the governance solutions. At the same time, synchronize them to the mobile devices of governance personnel and prompt staff to check.

2. The method for ecological restoration of abandoned land according to claim 1, characterized in that, The specific steps of the analysis and classification described in step (1) are as follows: Step 1: The acquisition and processing module connects with the remote sensing satellite and receives regional images sent by the remote sensing satellite in real time, and selects the regional images of the corresponding area according to the selection information sent by the staff; Step 2: The acquisition and processing module performs a missing image analysis on the region. If there are missing images in the region, the server communicates with the drone and external monitoring equipment to collect image data of the missing areas and supplement the regional images. Step 3: Classify the collected regional images into mining wastelands, polluted wastelands, disaster wastelands, and vacant wastelands. At the same time, construct soil profiles for each region to analyze soil fertility levels, suitability, and productivity. Also, generate the same number of information record tables as the number of regions collected, and enter the wasteland type, soil fertility level, suitability, and productivity into the information record tables.

3. An ecological management system for implementing the ecological management method for abandoned land as described in claim 1, characterized in that, It includes servers, predictive analysis modules, data acquisition and processing modules, solution databases, solution analysis modules, governance and control modules, and a management platform; The server is used to receive instruction information issued by the management platform, control each sub-module, and receive data sent by each sub-module for storage. The data collection and processing module is used to collect and classify the information on abandoned sites selected by the staff. The predictive analysis module is used to predict the restoration cycle of abandoned land. The solution analysis module is used to capture external governance solutions and update and optimize the staff's governance solutions; The governance and control module is used to collect staff work information in real time and calculate the restoration efficiency of abandoned sites for each staff member; The management platform is used for staff to log in and verify their identity, and to provide feedback on relevant data regarding the remediation of abandoned land based on staff's operation instructions.

Images