Waste site comprehensive evaluation and risk early warning system based on big data

Abstract

The invention relates to the technical field of data analysis, in particular to an abandoned site comprehensive evaluation and risk early warning system based on big data, and the system comprises a data analysis module which carries out the analysis of an abandoned site through a spatial analysis method according to basic geographic information, historical land information, pollution record information and environmental impact factors, generates a spatial analysis result, and carries out the early warning of the risk of the abandoned site. Evaluating the risk level of the abandoned site, and generating repair suggestion information according to the spatial analysis result and the risk level; and the early warning module carries out corresponding risk early warning according to the risk level of the abandoned site. According to the scheme, waste site comprehensive evaluation and risk early warning can be carried out according to the data information of the waste site, the current situation of the waste site is fed back, and a risk early warning mechanism is formed, so that personnel safety is guaranteed, environment protection is carried out, and sustainable development is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and particularly to a comprehensive evaluation and risk warning system for abandoned sites based on big data. Background Art

[0002] With the acceleration of the industrialization and urbanization processes, a large number of lands that were once used for industrial production, agricultural activities, or other purposes have gradually been abandoned or left idle. These abandoned sites often carry complex historical land use information and may hide various environmental problems, such as soil pollution, groundwater pollution, etc. These problems not only affect the ecological functions of the land itself but also pose potential threats to the environmental quality and the health of residents in the surrounding areas. Moreover, the idleness of abandoned sites not only results in resource waste, but also the polluted abandoned sites pose great safety hazards, and there may be further occurrences or spread of environmental pollution incidents, which may cause harm to human health and result in social and economic losses.

[0003] Therefore, how to utilize the data information of abandoned sites, effectively evaluate the current situation of abandoned sites, and establish a risk warning mechanism to protect personnel and the environment and promote sustainable development. Summary of the Invention

[0004] The present invention aims to provide a comprehensive evaluation and risk warning system for abandoned sites based on big data, which can conduct a comprehensive evaluation and risk warning of abandoned sites according to the data information of abandoned sites, feedback the current situation of abandoned sites, form a risk warning mechanism to ensure personnel safety, protect the environment, and promote sustainable development.

[0005] The present invention provides the following basic solution: A comprehensive evaluation and risk warning system for abandoned sites based on big data, including: a data acquisition module, a data processing module, a data storage module, a data analysis module, and a warning module;

[0006] The data acquisition module is used to collect the basic geographical information, historical land use information, pollution record information, and environmental impact factors of the abandoned site to be analyzed and send them to the data storage module;

[0007] The data processing module is used to process the obtained basic geographical information, historical land use information, pollution record information, and environmental impact factors and send them to the data storage module;

[0008] The data storage module is used to store the basic geographical information, historical land use information, pollution record information, and environmental impact factors;

[0009] The data analysis module is used to analyze the abandoned site by using a spatial analysis method according to the basic geographical information, historical land use information, pollution record information, and environmental impact factors and generate a spatial analysis result;

[0010] It is also used to evaluate the risk level of the abandoned site according to the spatial analysis results, and generate remediation advice information according to the spatial analysis results and the risk level;

[0011] The early warning module is used to perform corresponding risk warnings according to the risk level of the abandoned site.

[0012] Furthermore, the basic geographic information is the topographic information of the abandoned site;

[0013] The historical land use information is the land use situation data of the abandoned site in a preset time period before the current time point;

[0014] The pollution record information includes: soil pollution monitoring data, water pollution monitoring data, and air pollution monitoring data of the abandoned site; and each type of pollution monitoring data includes: pollutant type, pollution concentration, and pollution distribution range;

[0015] The environmental impact factors are natural and socio-economic factors that identify and quantify the factors affecting the site, including but not limited to: physical factors, chemical factors, biological factors, human factors, and socio-economic factors;

[0016] Among them, the environmental impact factors also include the environmental impact factors of the abandoned site and the environmental impact factors of the area within a preset distance range from the abandoned site.

[0017] Furthermore, the data processing includes:

[0018] Data format conversion, integrating data from different sources into the same geographic coordinate system;

[0019] Layer construction, according to the data after data format conversion, using the GIS platform to build the spatial database required for the project, and importing the data to create a layer corresponding to each type of data.

[0020] Furthermore, the data storage module includes a first database and a second database;

[0021] The first database is used to store the basic geographic information, historical land use information, pollution record information, and environmental impact factors of the abandoned site to be analyzed collected by the data collection module, as well as the basic geographic information, historical land use information, pollution record information, and environmental impact factors after data format conversion by the data processing module;

[0022] The second database is a spatial database created using the GIS platform and containing all layers.

[0023] Furthermore, the analysis of the abandoned site includes: overlay analysis, buffer generation, network simulation, hot spot detection, and trend surface modeling;

[0024] Perform overlay analysis. Use the overlay analysis tool of the GIS platform to overlay different types of historical land use maps with the pollution record layer to identify the affected areas;

[0025] Generate buffers. Based on the pollution record information, obtain the locations of pollution sources, and use the Buffer Analysis of the GIS platform to create buffers within a preset range to estimate the affected areas;

[0026] Conduct network simulations. If the pollutant type is other flowing media, use the pollutant transport model to predict the diffusion path of the pollutants;

[0027] Perform hotspot detection. Use statistical methods to query the areas where the pollution concentration is higher than the preset pollution concentration value;

[0028] Conduct trend surface modeling. Use regression analysis to establish a trend model of the pollution concentration varying with space.

[0029] Furthermore, the method for evaluating the risk level of the abandoned site according to the spatial analysis results and generating repair recommendation information based on the spatial analysis results and risk level includes:

[0030] According to the spatial analysis results, use the multi-criteria evaluation method to set the weights of each spatial analysis result and calculate the risk scores of the abandoned site and its regions;

[0031] According to the risk scores, set the threshold ranges of risk scores corresponding to different risk levels, and divide the abandoned site and its regions into different risk levels;

[0032] Call the corresponding preset repair recommendation information according to the different risk levels of each region, and integrate and generate the repair recommendation information for the abandoned site.

[0033] Furthermore, the method for evaluating the risk level of the abandoned site according to the spatial analysis results and generating repair recommendation information based on the spatial analysis results and risk level includes:

[0034] According to the spatial analysis results, use a deep learning model to evaluate the risk level of the abandoned site and generate repair recommendation information based on the spatial analysis results and risk level; where the deep learning model uses a CNN convolutional neural network model, takes the spatial analysis results as input, and outputs the risk levels of the abandoned site and its regions.

[0035] Furthermore, the architecture process of the CNN convolutional neural network model is as follows:

[0036] Construct a training set and a test set using historical spatial analysis results and corresponding risk levels;

[0037] Build a CNN convolutional neural network model, including: an input layer, an output layer, and several convolutional layers, pooling layers, and fully connected layers; the input layer inputs the spatial analysis results, and sequentially performs image feature extraction through the convolutional layers, image feature compression through the pooling layers, risk level analysis through the fully connected layers, and the output layer outputs the risk levels of the abandoned site and its various regions;

[0038] Use the training set to train the CNN convolutional neural network model;

[0039] Use the test set to test the trained CNN convolutional neural network model, and judge whether the CNN convolutional neural network model meets the preset requirements according to the test results. If so, output the CNN convolutional neural network model for risk level assessment of the abandoned site; if not, continue to train the CNN convolutional neural network model.

[0040] Furthermore, during the training process of the CNN convolutional neural network model, an optimization algorithm is used to update the weights of the nodes in the fully connected layer, including:

[0041] S1. Randomly generate several individual solutions of weights to form an initial population; specifically, the weight solution is [w 1 , w 2 , …, w m , where w m is the m-th weight value;

[0042] S2. Construct a weight evaluation function according to the training evaluation value of the CNN convolutional neural network model;

[0043] S3. According to the weight evaluation function, select the weight solutions in the initial population, perform iterative optimization, and obtain the optimal weight solution.

[0044] Furthermore, the weight evaluation function is:

[0045]

[0046] where P is the weight evaluation value and a is an adjustment constant; is the risk level of the abandoned site and its various regions output by the CNN convolutional neural network model with the input data in the i-th group of training data in the training set as the input, and y i is the actual result in the i-th group of training data in the training set.

[0047] Beneficial effects of the basic solution: First, this solution constructs a data acquisition module to collect a large amount of data on abandoned sites, including: basic geographical information, historical land use information, pollution record information, and environmental impact factors, and constructs a database (data storage module) for storage to ensure the traceability of subsequent original data;

[0048] Secondly, a data processing module is constructed to process the collected data, ensuring unified format for subsequent utilization. Moreover, a GIS platform is used to build the spatial database required for the project, import data, create layers corresponding to each type of data, and store them for subsequent spatial analysis;

[0049] Next, a data analysis module is constructed. Based on basic geographical information, historical land use information, pollution record information, and environmental impact factors, spatial analysis methods are adopted to analyze the abandoned site, generate spatial analysis results, evaluate the risk level of the abandoned site according to the spatial analysis results, and generate repair suggestion information according to the spatial analysis results and risk level;

[0050] Finally, a warning module conducts corresponding risk warnings according to the risk level of the abandoned site, thus completing the comprehensive evaluation and risk warning of the abandoned site;

[0051] Among them, the comprehensive evaluation of the abandoned site not only focuses on the visible pollution phenomena on the surface, but more emphasizes starting from multiple dimensions and comprehensively considering the natural and socio-economic factors of the site and its surrounding areas. By integrating historical land use information, existing pollution records, and environmental impact factors, a more three-dimensional and realistic site portrait can be constructed, which helps to identify areas that seem harmless but may actually have high risks, providing a solid data basis for subsequent risk management. At the same time, as part of preventive measures, risk warning plays a crucial role in early detection and intervention. Through the results of comprehensive evaluation, it can timely reflect the change trend of pollutants in the environment, issue early warnings, and thus avoid the occurrence or spread of environmental pollution incidents. This not only helps to protect the ecological environment, but also reduces the harm to human health and social and economic losses.

[0052] In summary, this solution can conduct comprehensive evaluation and risk warning of the abandoned site according to the data information of the abandoned site, feedback the current situation of the abandoned site, form a risk warning mechanism to ensure personnel safety, protect the environment, and promote sustainable development. Brief Description of the Drawings

[0053] Figure 1 It is a logic block diagram of an embodiment of the comprehensive evaluation and risk warning system for abandoned sites based on big data of the present invention. Detailed Description of the Preferred Embodiments

[0054] The following is a more detailed description through specific embodiments:

[0055] Embodiment 1

[0056] The embodiment is basically as shown in the appendix Figure 1Shown: A comprehensive evaluation and risk warning system for abandoned sites based on big data, including: a data collection module, a data processing module, a data storage module, a data analysis module, and a warning module;

[0057] The data collection module is used to collect the basic geographical information, historical land use information, pollution record information, and environmental impact factors of the abandoned site to be analyzed, and send them to the data storage module; among them, the environmental impact factors also include the environmental impact factors of the area within a preset distance range from the abandoned site.

[0058] Specifically, the basic geographical information, which is the topographic information of the abandoned site, can be obtained by consulting maps.

[0059] The historical land use information is the land use situation data of the abandoned site in a preset time period before the current time point, such as industrial, agricultural, or residential use, and is obtained by consulting urban planning archives, historical maps, and other relevant documents.

[0060] The pollution record information includes, but is not limited to: soil pollution monitoring data, water pollution monitoring data, and air pollution monitoring data of the abandoned site; and each type of pollution monitoring data includes: pollutant type, pollution concentration, and pollution distribution range.

[0061] The environmental impact factors are the natural and socio-economic factors that identify and quantify the impact on the site, including, but not limited to: physical factors, chemical factors, biological factors, human factors, and socio-economic factors;

[0062] Among them, the physical factors include, but are not limited to: topography and landform, such as slope and altitude; climate conditions, such as temperature, precipitation, and wind speed; soil characteristics, such as soil type (sandy, clay, etc.), permeability, and pH value; hydrological conditions, such as groundwater level, river flow velocity and direction, and tidal changes of lakes and oceans;

[0063] The chemical factors include, but are not limited to: pollutants, such as specific pollutant types (heavy metals, organic substances, etc.), concentration and their distribution; chemical reactions, such as the interaction between pollutants and their chemical reactions with other components (such as minerals) in the environment; acidity and alkalinity, such as the pH value of soil or water body;

[0064] The biological factors include, but are not limited to: vegetation cover, animal activities, and microbial communities;

[0065] The human factors include, but are not limited to: historical land use information; infrastructure, such as roads, drainage systems, buildings, etc.; management measures, such as the application of existing environmental protection regulations, waste treatment facilities and technologies;

[0066] The socio-economic factors include, but are not limited to: population density and economic development level.

[0067] The selection of environmental impact factors is based on the actual situation of the abandoned site.

[0068] A data processing module is used to process the acquired basic geographic information, historical land use information, pollution record information, and environmental impact factors, and send them to the data storage module.

[0069] Among them, data processing includes: data format conversion, integrating data from different sources (basic geographic information, historical land use information, pollution record information, and environmental impact factors) into the same geographic coordinate system for subsequent spatial analysis operations, such as converting all data to a unified spatial reference system (such as the WGS84 coordinate system), and formatting point, line, or polygon data as required (such as Shapefile, GeoJSON, etc.).

[0070] Layer construction: Based on the data after data format conversion, use a GIS platform to build the spatial database required for the project, import the data, and create layers corresponding to each type of data.

[0071] A data storage module is used to store basic geographic information, historical land use information, pollution record information, and environmental impact factors.

[0072] Specifically, the data storage module includes a first database and a second database.

[0073] The first database is used to store the basic geographic information, historical land use information, pollution record information, and environmental impact factors of the abandoned site to be analyzed collected by the data acquisition module, as well as the basic geographic information, historical land use information, pollution record information, and environmental impact factors after data format conversion by the data processing module.

[0074] The second database is a spatial database created using a GIS platform that contains all layers.

[0075] A data analysis module is used to analyze the abandoned site using spatial analysis methods based on basic geographic information, historical land use information, pollution record information, and environmental impact factors, and generate spatial analysis results.

[0076] Among them, the analysis of the abandoned site includes: overlay analysis, buffer generation, network simulation, hotspot detection, and trend surface modeling.

[0077] Specifically, for overlay analysis, use the overlay analysis tools (such as Union, Intersect, etc.) of the GIS platform to overlay different types of historical land use layers and pollution record layers to identify the affected areas.

[0078] Generate a buffer. Based on the pollution record information, obtain the location of the pollution source, and use the Buffer Analysis on the GIS platform to create a buffer within a preset range to estimate the affected area.

[0079] Conduct network simulation. If the pollutant type is other flowing media, use a pollutant transport model (such as SWAT, MODFLOW, etc.) to predict the diffusion path of the pollutant.

[0080] Conduct hot spot detection. Use statistical methods (such as Getis-Ord Gi* Hot Spot Analysis) to query areas where the pollution concentration is higher than the preset pollution concentration value.

[0081] Conduct trend surface modeling. Use regression analysis (such as ordinary least squares OLS, geographically weighted regression GWR) to establish a trend model of the pollution concentration varying with space.

[0082] The data analysis module is also used to evaluate the risk level of the abandoned site based on the spatial analysis results, and generate repair suggestion information according to the spatial analysis results and the risk level.

[0083] Specifically, according to the spatial analysis results, use a multi-criteria evaluation method (such as the analytic hierarchy process AHP, fuzzy comprehensive evaluation FCE, etc.) to set the weight of each spatial analysis result, and calculate the risk scores of the abandoned site and its various regions.

[0084] According to the risk scores, set the risk score range thresholds corresponding to different risk levels, and divide the abandoned site and its various regions into different risk levels, such as low, medium, and high.

[0085] Call the corresponding preset repair suggestion information according to the different risk levels of each region, and integrate and generate the repair suggestion information for the abandoned site.

[0086] The early warning module is used to conduct corresponding risk warnings according to the risk level of the abandoned site. In this embodiment, the risk levels are divided into low, medium, and high, and the corresponding risk warnings are set as mild risk warnings, moderate risk warnings, and severe risk warnings. Different risk warnings are set with different warning messages and pushed to the corresponding management departments.

[0087] This solution can conduct a comprehensive evaluation and risk warning of the abandoned site based on the data information of the abandoned site, feedback the current situation of the abandoned site, form a risk warning mechanism to ensure the safety of personnel, protect the environment, and promote sustainable development.

[0088] Embodiment 2

[0089] The difference between this embodiment and the above embodiment is:

[0090] The data analysis module is also used to evaluate the risk level of the abandoned site according to the spatial analysis result, and generate repair suggestion information according to the spatial analysis result and the risk level;

[0091] Specifically, according to the spatial analysis result, a deep learning model is used to evaluate the risk level of the abandoned site, and repair suggestion information is generated according to the spatial analysis result and the risk level; in this embodiment, a CNN convolutional neural network model is used, and the spatial analysis result is used as the input to output the risk levels of the abandoned site and its respective regions;

[0092] The specific process is as follows:

[0093] Perform rasterization processing on the spatial analysis result. If the spatial analysis result is in vector format, convert it to raster format (such as GeoTIFF);

[0094] Use the constructed CNN convolutional neural network model to analyze and output the risk levels of the abandoned site and its respective regions according to the rasterized spatial analysis result;

[0095] Call the corresponding preset repair suggestion information according to the different risk levels of each region, and integrate and generate the repair suggestion information of the abandoned site.

[0096] The construction process of the CNN convolutional neural network model is as follows:

[0097] Use the historical spatial analysis results and the corresponding risk levels to construct a training set and a test set;

[0098] Construct a CNN convolutional neural network model, including: an input layer, an output layer, and several convolutional layers, pooling layers, and fully connected layers; the input layer inputs the spatial analysis result, and sequentially performs image feature extraction through the convolutional layer, image feature compression through the pooling layer, risk level analysis through the fully connected layer, and the output layer outputs the risk levels of the abandoned site and its respective regions;

[0099] Use the training set to train the CNN convolutional neural network model;

[0100] Use the test set to test the trained CNN convolutional neural network model, and judge whether the CNN convolutional neural network model meets the preset requirements according to the test results. If so, output the CNN convolutional neural network model for risk level evaluation of the abandoned site; if not, continue to train the CNN convolutional neural network model;

[0101] In this embodiment, the CNN neural network model includes: an input layer, an output layer, two convolutional layers, a pooling layer, and a fully connected layer. The connection relationship is that the output of the input layer serves as the input of the first convolutional layer, the output of the first convolutional layer serves as the input of the first pooling layer, the output of the first pooling layer serves as the input of the second convolutional layer, the output of the second convolutional layer serves as the input of the second pooling layer, the output of the second pooling layer serves as the input of the fully connected layer, and the output of the fully connected layer is output through the output layer. Among them, the number of nodes in the input layer is m, which is equal to the number of types of spatial analysis results, and the number of nodes in the output layer is n, which is equal to the number of abandoned sites and their respective regions. The pooling layer adopts the maximum pooling strategy, and the fully connected layer uses the Leaky ReLU function for activation (non-linear mapping).

[0102] In this solution, the spatial analysis results can be effectively combined with the deep learning model to realize the intelligent evaluation of the risk level of abandoned sites, which helps to improve the prediction accuracy and provides strong technical support for environmental protection.

[0103] Embodiment III

[0104] This embodiment is basically the same as the above embodiment, except that:

[0105] During the training process of the CNN convolutional neural network model, an optimization algorithm is used to update the weights of the nodes in the fully connected layer, specifically including:

[0106] S1. Randomly generate several individual solutions of weights to form an initial population. Specifically, the weight solution is [w 1 , w 2 , …, w m , where w m is the m-th weight value;

[0107] S2. Construct a weight evaluation function according to the training evaluation value of the CNN convolutional neural network model;

[0108] Specifically, the weight evaluation function is:

[0109]

[0110] where P is the weight evaluation value and a is an adjustment constant; is the risk level of the abandoned site and its respective regions output by the CNN convolutional neural network model with the input data in the i-th group of training data in the training set as the input, y i is the actual result in the i-th group of training data in the training set, and v is the number of training times and also the number of groups of training data in the training set; Take the risk level of the abandoned site, y i Take the actual risk level of the abandoned site in the i-th group of training data in the training set;

[0111] S3. Select the weight scheme in the initial population according to the weight evaluation function, perform iterative optimization, and obtain the optimal weight scheme;

[0112] Specifically, S301. Initialize the iteration number k = 1;

[0113] S302. According to the individual schemes in the population, sequentially set the weights of the fully connected layers in the CNN convolutional neural network model, and use the same training set to train the CNN neural network models with different weights to obtain the corresponding outputs

[0114] S303. According to the weight evaluation function, calculate the weight evaluation value P of each individual scheme j , where is the weight evaluation value of the j-th individual scheme;

[0115] S304. According to the weight evaluation value P of each individual scheme j , use the roulette wheel method to select individual schemes in the initial population; the selection probability is proportional to the size of the weight evaluation value;

[0116] S305. Mutate and cross the selected individual schemes to form a new population;

[0117] S306. Determine whether the current iteration number is equal to the preset number. If so, obtain the individual scheme with the largest weight evaluation value as the optimal weight scheme. If not, update the iteration number k = k + 1 and go to S302.

[0118] In this solution, during the training process of the CNN convolutional neural network model, the weights of the nodes in the fully connected layer are updated using an optimization algorithm to find the optimal weight scheme, which can improve the accuracy of the CNN convolutional neural network model analysis and further ensure the accuracy of the comprehensive evaluation.

[0119] Example 4

[0120] This example is basically the same as the above example, except that is the risk level of the abandoned site and its various regions, and y i takes the actual risk level of the abandoned site and its various regions in the i-th group of training data in the training set;

[0121] The weight evaluation function is:

[0122]

[0123] where b r is the weight of the r-th risk level in the risk levels of the abandoned site and its various regions.

[0124] The mean square error is calculated for the risk levels of the abandoned site and its various regions, making the evaluation result of the weight evaluation function more accurate. And corresponding weights are set for the risk levels of the abandoned site and its various regions, so that the tendency of the weight evaluation function for the abandoned site and its various regions can be adjusted, enabling the weight evaluation to adapt to different requirements.

[0125] The above are only embodiments of the present invention. Specific structures and common knowledge such as characteristics well known in the art are not described in detail herein. Those of ordinary skill in the art know all the common general technical knowledge in the technical field to which the invention belongs before the filing date or the priority date, can know all the prior art in this field, and have the ability to apply conventional experimental means before this date. Those of ordinary skill in the art can, under the inspiration given in this application and in combination with their own abilities, improve and implement this solution. Some typical well-known structures or well-known methods should not be an obstacle to those of ordinary skill in the art in implementing this application. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can still be made, and these should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be based on the content of its claims, and the specific implementation manners and the like recorded in the specification can be used to interpret the content of the claims.

Claims

1. A comprehensive evaluation and risk warning system for abandoned sites based on big data, characterized by: include: Data acquisition module, data processing module, data storage module, data analysis module, and early warning module; A data collection module is used to collect basic geographic information, historical land use information, pollution record information and environmental impact factors of the abandoned site to be analyzed, and send it to the data storage module; A data processing module is used to process the acquired basic geographic information, historical land use information, pollution record information and environmental impact factors, and send them to the data storage module; Data storage module, used to store basic geographic information, historical land use information, pollution record information and environmental impact factors; The data analysis module is used to analyze the abandoned sites and generate spatial analysis results based on basic geographic information, historical land use information, pollution record information and environmental impact factors using spatial analysis methods; It is also used to assess the risk level of abandoned sites based on the spatial analysis results, and to generate restoration recommendation information based on the spatial analysis results and the risk level; The early warning module is used to issue corresponding risk warnings according to the risk level of the abandoned site.

2. The comprehensive evaluation and risk warning system for abandoned sites based on big data according to claim 1 is characterized by: The basic geographic information is the topographic information of the abandoned site; Historical land use information is the land use data of abandoned sites in a preset time period before the current time point; Pollution record information, including: soil pollution monitoring data, water pollution monitoring data, and air pollution monitoring data of abandoned sites; and all types of pollution monitoring data include: pollutant type, pollution concentration, and pollution distribution range; Environmental impact factors are the identification and quantification of natural and socio-economic factors that affect the site, including but not limited to: physical factors, chemical factors, biological factors, human factors, and socio-economic factors; The environmental impact factors also include the environmental impact factors of the abandoned site and the environmental impact factors of the area within a preset distance range from the abandoned site.

3. The comprehensive evaluation and risk warning system for abandoned sites based on big data according to claim 2 is characterized by: The data processing includes: Data format conversion, integrating data from different sources into the same geographic coordinate system; Layer construction: Based on the data after data format conversion, the GIS platform is used to build the spatial database required for the project, and the data is imported to create layers corresponding to each type of data.

4. The comprehensive evaluation and risk warning system for abandoned sites based on big data according to claim 3 is characterized by: The data storage module includes a first database and a second database; The first database is used to store the basic geographic information, historical land use information, pollution record information and environmental impact factors of the abandoned site to be analyzed collected by the data collection module, and the basic geographic information, historical land use information, pollution record information and environmental impact factors after data format conversion by the data processing module; The second database is a spatial database containing all layers created using the GIS platform.

5. The comprehensive evaluation and risk warning system for abandoned sites based on big data according to claim 4 is characterized by: The analysis of the abandoned site includes: overlay analysis, buffer generation, network simulation, hot spot detection, and trend surface modeling; Conduct overlay analysis, using the overlay analysis tool of the GIS platform to overlay different types of historical map layers with pollution record layers to identify affected areas; Generate buffer zones. Obtain the location of pollution sources based on pollution record information. Apply BufferAnalysis on the GIS platform to create buffer zones within a preset range and estimate the impact range. Conduct network simulation. If the pollutant type is other flow media, use the pollutant transport model to predict the diffusion path of the pollutant; Conduct hotspot detection and use statistical methods to query areas where pollution concentration is higher than the preset pollution concentration value; Trend surface modeling was performed and regression analysis was used to establish a trend model of pollution concentration changing with space.

6. The comprehensive evaluation and risk warning system for abandoned sites based on big data according to claim 1 is characterized by: The system is used to evaluate the risk level of the abandoned site according to the spatial analysis results, and generate restoration suggestion information according to the spatial analysis results and the risk level, including: Based on the spatial analysis results, a multi-criteria evaluation method is used to set the weight of each spatial analysis result and calculate the risk score of the abandoned site and its various areas; According to the risk score, the risk score range thresholds corresponding to different risk levels are set to divide the abandoned site and its areas into different risk levels; According to the different risk levels of each area, the corresponding preset restoration suggestion information is called up to integrate and generate the restoration suggestion information of the abandoned site.

7. The comprehensive evaluation and risk warning system for abandoned sites based on big data according to claim 1 is characterized by: The system is used to evaluate the risk level of the abandoned site according to the spatial analysis results, and generate restoration suggestion information according to the spatial analysis results and the risk level, including: According to the results of spatial analysis, a deep learning model is used to evaluate the risk level of the abandoned site, and based on the spatial analysis results and risk level, restoration recommendation information is generated; the deep learning model uses a CNN convolutional neural network model, which takes the spatial analysis results as input and outputs the risk level of the abandoned site and its various areas.

8. The comprehensive evaluation and risk warning system for abandoned sites based on big data according to claim 7 is characterized by: The construction process of the CNN convolutional neural network model is as follows: The training and test sets are constructed using historical spatial analysis results and corresponding risk levels; Construct a CNN convolutional neural network model, including: input layer, output layer, and several convolution layers, pooling layers and fully connected layers; the input layer inputs the spatial analysis results, and sequentially passes through the convolution layer to extract image features, the pooling layer to compress image features, the fully connected layer to analyze the risk level, and the output layer to output the risk level of the abandoned site and its various areas; Use the training set to train the CNN convolutional neural network model; The trained CNN convolutional neural network model is tested using the test set, and the test results are used to determine whether the CNN convolutional neural network model meets the preset requirements. If so, the CNN convolutional neural network model is output to conduct risk level assessment of the abandoned site; if not, the CNN convolutional neural network model continues to be trained.

9. The comprehensive evaluation and risk warning system for abandoned sites based on big data according to claim 8 is characterized by: During the training process of the CNN convolutional neural network model, an optimization algorithm is used to update the weights of the nodes in the fully connected layer, including: S1. Randomly generate several weighted individual schemes to form an initial population; specifically, the weight scheme is [w1,w2,…,w m ], where w m is the mth weight value; S2. Construct a weight evaluation function based on the training evaluation value of the CNN convolutional neural network model; S3. According to the weight evaluation function, select the weight scheme in the initial population, perform iterative optimization, and obtain the optimal weight scheme.

10. The waste site comprehensive evaluation and risk early warning system based on big data according to claim 9 is characterized by: The weight evaluation function is: Where P is the weight evaluation value, and a is the adjustment constant; is the risk level of the abandoned site and its various areas output by the CNN convolutional neural network model with the input data in the i-th group of training data in the training set, y i is the actual result in the i-th group of training data in the training set.

Images