Multi-source pollution analysis method and system based on spatio-temporal features

Through the spatial and temporal characteristics analysis method of multi-source pollution based on deep learning, the problems of soil pollution diffusion and multi-source pollution data analysis are solved, accurate prediction and efficient regulation of soil pollution are achieved, and pollution prevention and control efficiency is improved.

WO2024244279A1PCT designated stage expired Publication Date: 2024-12-05BCEG ENVIRONMENTAL REMEDIATION CO LTD +1

Patent Information

Application Number
PCT/CN2023/127873
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-06-02
Filing Date
2023-10-30
Publication Date
2024-12-05

AI Technical Summary

Technical Problem

It is difficult for the existing technology to effectively analyze and predict the spread of soil pollution and multi-source pollution data, which has led to the hindering of the precise prevention and control of soil pollution.

Method used

The multi-source pollution spatiotemporal feature analysis method based on deep learning is adopted. By obtaining monitoring data and environmental feature data of soil-polluted areas, a pollution diffusion model is constructed, spatial and temporal feature analysis and prediction are carried out, the preferred monitoring sub-regions are screened, and the regulation scheme is generated.

Benefits of technology

Accurate analysis and prediction of soil pollution areas has been achieved, reducing the consumption of manpower and material resources has been monitored, and the efficiency of pollution prediction and regulation has been improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2023127873_05122024_PF_FP_ABST
    Figure CN2023127873_05122024_PF_FP_ABST
Patent Text Reader

Abstract

Disclosed in the present application are a multi-source pollution analysis method and system based on deep learning and spatio-temporal features. The method comprises: within one monitoring cycle, importing soil pollution monitoring data and environment feature data into a pollution diffusion model for pollution analysis and pollution prediction based on spatio-temporal features, so as to obtain pollution prediction data and an optimal monitoring sub-region; within the next monitoring cycle, performing pollution analysis and pollution prediction on the optimal monitoring sub-region to obtain secondary pollution prediction data and a secondary optimal monitoring sub-region; and according to the secondary pollution prediction data, performing pollution regulation and control analysis on a soil pollution region and generating regulation and control scheme information. By means of the present application, accurate analysis and prediction of diffusion simulation can be performed on pollution regions in terms of spatio-temporal features, and an optimal monitoring point is selected in each monitoring cycle, such that manpower and material resources consumed during monitoring are reduced, and efficient analysis, pollution prediction and scheme regulation and control analysis are implemented on the pollution regions on the basis of optimal monitoring points.
Need to check novelty before this filing date? Find Prior Art

Description

A method and system for analyzing spatiotemporal characteristics of multi-source pollution

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to the Chinese patent application filed with the China Patent Office on June 2, 2023, with application number 202310644864.6 and invention name “A method and system for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning”, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present application relates to the field of deep learning, and more specifically, to a method and system for analyzing the spatiotemporal characteristics of multi-source pollution based on deep learning. Background Art

[0004] In recent years, with the deepening advancement of agricultural modernization, agriculture has achieved remarkable success, with both quality and yield increasing. However, this tremendous growth has also come at the expense of soil and environmental pollution. In particular, soil pollution from heavy metals and acidification is a prominent problem, leading to deterioration of soil quality and degradation. Continued soil pollution not only reduces soil fertility but also poses a significant threat to human health. Furthermore, the spread of soil pollution can further impact adjacent, uncontaminated areas.

[0005] However, due to the limitations of traditional pollution monitoring methods, soil pollution analysis is relatively simple. There is a lack of accurate analysis of soil pollution prediction and diffusion, pollution data characteristics, and multi-source pollution data analysis. This seriously hinders the effective analysis and precise prevention of soil pollution. Therefore, there is an urgent need for a multi-source pollution analysis method.

[0006] Summary of the Invention

[0007] This application overcomes the shortcomings of the existing technology and proposes a multi-source pollution spatiotemporal feature analysis method and system based on deep learning.

[0008] The first aspect of the present application provides a method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning, comprising:

[0009] Acquire a soil contaminated area, and divide the area into multiple contaminated sub-areas based on the soil contaminated area;

[0010] Acquire soil pollution monitoring data and environmental characteristic data in multiple polluted sub-areas within one monitoring cycle;

[0011] Importing the soil pollution monitoring data and environmental characteristic data into a pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, thereby obtaining pollution prediction data and preferred monitoring sub-areas;

[0012] In the next monitoring cycle, pollution analysis and pollution prediction are carried out on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas;

[0013] A pollution control analysis is performed on the soil contaminated area based on the secondary pollution prediction data, and control plan information is generated, which is then sent to a preset terminal device.

[0014] In this solution, the soil contaminated area is obtained, and the soil contaminated area is divided into multiple contaminated sub-areas based on the soil contaminated area, specifically:

[0015] Obtain regional area and regional contour information in soil contaminated areas;

[0016] Constructing a pollution area map model based on the area and area contour information;

[0017] Based on the pollution area map model, the pollution area is divided into grids to obtain multiple pollution sub-areas, and the area difference between each pollution sub-area is guaranteed to be within a preset range.

[0018] In this solution, the soil pollution monitoring data and environmental characteristic data in multiple polluted sub-areas are obtained within one monitoring cycle, specifically:

[0019] Conduct soil sampling and testing in each polluted sub-area, and obtain the sub-area pollution monitoring data corresponding to each polluted sub-area;

[0020] Integrate the pollution monitoring data of all sub-regions to obtain soil pollution monitoring data;

[0021] The soil pollution monitoring data includes pH, heavy metal content, soil moisture, and soil color information;

[0022] Obtain weather, humidity, and wind direction information in soil contaminated areas;

[0023] The weather, humidity and wind direction information are subjected to environmental information feature extraction to obtain environmental feature data.

[0024] In this solution, the soil pollution monitoring data and environmental characteristic data are imported into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, thereby obtaining pollution prediction data and preferred monitoring sub-areas, which includes:

[0025] Build a pollution diffusion model based on the preset deep learning algorithm;

[0026] Obtain historical soil monitoring data and historical environmental characteristic data within a preset historical time period;

[0027] Normalize, standardize and integrate historical soil monitoring data and historical environmental characteristic data to obtain a pollution dataset;

[0028] Divide the contaminated dataset into training dataset, validation dataset, and test dataset according to the preset ratio;

[0029] The training dataset, validation dataset, and test dataset are imported into the pollution diffusion model for repeated pollution prediction training until the obtained pollution prediction data passes all test datasets.

[0030] In this solution, the soil pollution monitoring data and environmental characteristic data are imported into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, and pollution prediction data and preferred monitoring sub-areas are obtained, specifically:

[0031] Importing the soil pollution monitoring data and environmental characteristic data into a pollution diffusion model;

[0032] Based on soil pollution monitoring data, the basic pollution degree of each polluted sub-area is analyzed to obtain the initial pollution index;

[0033] Calculate the predicted diffusion rate of each polluted sub-area based on the initial pollution index;

[0034] The pollution diffusion model predicts the pollution diffusion of all polluted sub-areas based on the predicted diffusion rate and environmental characteristic data to obtain pollution prediction data;

[0035] Analyze the predicted pollution index of each polluted sub-area based on the pollution prediction data;

[0036] The predicted pollution index and the predicted diffusion rate in the polluted sub-region are determined. If the predicted pollution index is greater than a first preset threshold and the predicted diffusion rate is greater than a second preset threshold, the corresponding polluted sub-region is marked as a preferred monitoring sub-region.

[0037] In this solution, during the next monitoring cycle, pollution analysis and pollution prediction are performed on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas, specifically:

[0038] In the next monitoring cycle, pollution monitoring is performed on the preferred monitoring sub-areas and preferred pollution monitoring data of each preferred monitoring sub-area is obtained;

[0039] During the next monitoring cycle, analyze and obtain secondary environmental characteristic data of soil contaminated areas;

[0040] The preferred pollution monitoring data and secondary environmental characteristic data are introduced into a pollution diffusion model to perform secondary pollution analysis and secondary pollution prediction, thereby obtaining secondary pollution prediction data and secondary preferred monitoring sub-areas.

[0041] In this solution, the pollution control analysis of the soil contaminated area is performed based on the secondary pollution prediction data, and control solution information is generated, and the control solution information is sent to the preset terminal device, specifically:

[0042] Divide all polluted sub-areas into control areas by taking the secondary optimal monitoring sub-area as the first control sub-area and marking the remaining polluted sub-areas as the second control sub-areas;

[0043] Based on the first control sub-region and the second control sub-region, the secondary pollution prediction data is divided into first prediction data and second prediction data;

[0044] Performing pollution control analysis on the first control sub-region and generating a first control plan based on the first prediction data;

[0045] Performing pollution control analysis on the second control sub-region and generating a second control plan based on the second prediction data;

[0046] The first control scheme and the second control scheme are integrated into control scheme information and sent to a preset terminal device.

[0047] A second aspect of the present application further provides a multi-source pollution spatiotemporal feature analysis system based on deep learning, the system comprising: a memory and a processor, wherein the memory comprises a multi-source pollution spatiotemporal feature analysis program based on deep learning, and when the multi-source pollution spatiotemporal feature analysis program based on deep learning is executed by the processor, the following steps are implemented:

[0048] Acquire a soil contaminated area, and divide the area into multiple contaminated sub-areas based on the soil contaminated area;

[0049] Acquire soil pollution monitoring data and environmental characteristic data in multiple polluted sub-areas within one monitoring cycle;

[0050] Importing the soil pollution monitoring data and environmental characteristic data into a pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, thereby obtaining pollution prediction data and preferred monitoring sub-areas;

[0051] In the next monitoring cycle, pollution analysis and pollution prediction are carried out on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas;

[0052] A pollution control analysis is performed on the soil contaminated area based on the secondary pollution prediction data, and control plan information is generated, which is then sent to a preset terminal device.

[0053] In this solution, the soil contaminated area is obtained, and the soil contaminated area is divided into multiple contaminated sub-areas based on the soil contaminated area, specifically:

[0054] Obtain regional area and regional contour information in soil contaminated areas;

[0055] Constructing a pollution area map model based on the area and area contour information;

[0056] Based on the pollution area map model, the pollution area is divided into grids to obtain multiple pollution sub-areas, and the area difference between each pollution sub-area is guaranteed to be within a preset range.

[0057] In this solution, the soil pollution monitoring data and environmental characteristic data in multiple polluted sub-areas are obtained within one monitoring cycle, specifically:

[0058] Conduct soil sampling and testing in each polluted sub-area, and obtain the sub-area pollution monitoring data corresponding to each polluted sub-area;

[0059] Integrate the pollution monitoring data of all sub-regions to obtain soil pollution monitoring data;

[0060] The soil pollution monitoring data includes pH, heavy metal content, soil moisture, and soil color information;

[0061] Obtain weather, humidity, and wind direction information in soil contaminated areas;

[0062] The weather, humidity and wind direction information are subjected to environmental information feature extraction to obtain environmental feature data.

[0063] The present application discloses a method and system for analyzing the spatiotemporal characteristics of multi-source pollution based on deep learning. In one monitoring cycle, soil pollution monitoring data and environmental characteristic data are imported into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, and pollution prediction data and preferred monitoring sub-areas are obtained. In the next monitoring cycle, pollution analysis and pollution prediction are performed on the preferred monitoring sub-areas to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas. Based on the secondary pollution prediction data, pollution control analysis is performed on the soil pollution area and control scheme information is generated. Through this application, it is possible to accurately analyze and predict the diffusion simulation of the pollution area from the perspective of spatiotemporal characteristics, and select preferred monitoring points in each monitoring cycle, thereby reducing the manpower and material resources consumed in monitoring, and realizing efficient analysis, pollution prediction and scheme control analysis of the pollution area based on the preferred monitoring points. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] FIG1 shows a flowchart of a method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning in the present application;

[0065] FIG2 shows a flowchart of obtaining multiple contaminated sub-areas of the present application;

[0066] FIG3 shows a flowchart of the environmental characteristic data acquisition process of the present application;

[0067] FIG4 shows a block diagram of a multi-source pollution spatiotemporal feature analysis system based on deep learning in the present application. DETAILED DESCRIPTION

[0068] In order to more clearly understand the above-mentioned objects, features and advantages of the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features therein can be combined with each other in the absence of conflict.

[0069] In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present application is not limited to the specific embodiments disclosed below.

[0070] FIG1 shows a flowchart of a multi-source pollution spatiotemporal feature analysis method based on deep learning in the present application.

[0071] As shown in FIG1 , the first aspect of the present application provides a method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning, including:

[0072] S102, obtaining a soil contaminated area, and dividing the soil contaminated area into multiple contaminated sub-areas;

[0073] S104, obtaining soil pollution monitoring data and environmental characteristic data in multiple polluted sub-areas within a monitoring cycle;

[0074] S106, importing the soil pollution monitoring data and environmental characteristic data into a pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, to obtain pollution prediction data and preferred monitoring sub-areas;

[0075] S108, in the next monitoring cycle, performing pollution analysis and pollution prediction on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas;

[0076] S110 , performing pollution control analysis on the soil contaminated area according to the secondary pollution prediction data and generating control plan information, and sending the control plan information to a preset terminal device.

[0077] It should be noted that the spatiotemporal feature analysis in this application is specifically to conduct a pollution analysis of the soil contaminated area in the spatiotemporal feature dimension through analysis of different monitoring periods and soil pollution diffusion analysis.

[0078] FIG2 shows a flowchart of obtaining multiple polluted sub-areas in the present application.

[0079] According to an embodiment of the present application, the soil contaminated area is obtained by dividing the soil contaminated area into multiple contaminated sub-areas, specifically:

[0080] S202, obtaining regional area and regional contour information in the soil contaminated area;

[0081] S204, constructing a pollution area map model based on the area and area outline information;

[0082] S206 , based on the pollution area map model, the pollution area is divided into grids to obtain a plurality of pollution sub-areas, and the area difference between each pollution sub-area is ensured to be within a preset range.

[0083] It should be noted that the contaminated area map model is specifically a three-dimensional visual map model, which can display the pollution status of the area and dynamically display the pollution spread to the user. The number of sub-areas in the multiple contaminated areas is determined by the area size of the soil contamination area. The larger the area, the greater the number of sub-areas. The area difference between each contaminated sub-area is ensured to be within a preset range, which is generally determined by the user based on the area size of the soil contamination area.

[0084] FIG3 shows a flow chart of obtaining environmental characteristic data of the present application.

[0085] According to an embodiment of the present application, the soil pollution monitoring data and environmental characteristic data in multiple polluted sub-areas are obtained within one monitoring cycle, specifically:

[0086] S302, soil sampling and testing is performed on each polluted sub-area, and sub-area pollution monitoring data corresponding to each polluted sub-area is obtained;

[0087] S304: Integrate the pollution monitoring data of all sub-regions to obtain soil pollution monitoring data;

[0088] S306, the soil pollution monitoring data includes pH, heavy metal content, soil moisture, and soil color information;

[0089] S308, obtaining weather, humidity, and wind direction information of the soil contaminated area;

[0090] S310: Extract environmental information features from the weather, humidity, and wind direction information to obtain environmental feature data.

[0091] It should be noted that the environmental characteristic data refers to the primary data features of the weather, humidity, and wind direction information, and can be incorporated into pollution diffusion models for feature analysis. The soil pollution monitoring data includes information such as pH, heavy metal content, soil moisture, and soil color. Specifically, soil color information is obtained by performing color feature analysis on soil image data obtained by drone aerial photography of each contaminated sub-area.

[0092] According to an embodiment of the present application, the soil pollution monitoring data and environmental characteristic data are imported into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics to obtain pollution prediction data and preferred monitoring sub-areas, which includes:

[0093] Build a pollution diffusion model based on the preset deep learning algorithm;

[0094] Obtain historical soil monitoring data and historical environmental characteristic data within a preset historical time period;

[0095] Normalize, standardize and integrate historical soil monitoring data and historical environmental characteristic data to obtain a pollution dataset;

[0096] Divide the contaminated dataset into training dataset, validation dataset, and test dataset according to the preset ratio;

[0097] The training dataset, validation dataset, and test dataset are imported into the pollution diffusion model for repeated pollution prediction training until the obtained pollution prediction data passes all test datasets.

[0098] It should be noted that the preset deep learning algorithm is specifically a combination of multiple prediction model algorithms, including linear regression, logistic regression, neural networks, Bayesian networks, and other prediction model algorithms, and the pollution diffusion model includes at least one prediction model algorithm. The preset historical time period is generally the last 1-3 years. The preset ratio is generally 2:2:1, that is, the training dataset and validation dataset each account for 40%, and the test dataset accounts for 20%.

[0099] According to an embodiment of the present application, the soil pollution monitoring data and environmental characteristic data are imported into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, and pollution prediction data and preferred monitoring sub-areas are obtained, specifically:

[0100] Importing the soil pollution monitoring data and environmental characteristic data into a pollution diffusion model;

[0101] Based on soil pollution monitoring data, the basic pollution degree of each polluted sub-area is analyzed to obtain the initial pollution index;

[0102] Calculate the predicted diffusion rate of each polluted sub-area based on the initial pollution index;

[0103] The pollution diffusion model predicts the pollution diffusion of all polluted sub-areas based on the predicted diffusion rate and environmental characteristic data to obtain pollution prediction data;

[0104] Analyze the predicted pollution index of each polluted sub-area based on the pollution prediction data;

[0105] The predicted pollution index and the predicted diffusion rate in the polluted sub-region are determined. If the predicted pollution index is greater than a first preset threshold and the predicted diffusion rate is greater than a second preset threshold, the corresponding polluted sub-region is marked as a preferred monitoring sub-region.

[0106] It should be noted that the initial pollution index is obtained by analyzing soil pollution monitoring data. The analysis dimensions include pH, heavy metal content, soil moisture, soil color information, etc. The larger the initial pollution index, the more serious the pollution level in the corresponding polluted sub-area, which is specifically manifested in lower soil pH values, higher heavy metal content, higher heavy metal concentration, and soil color deviating from normal color.

[0107] In addition, the predicted diffusion rate is specifically obtained by multiplying the initial pollution index by the pollution coefficient, that is, the initial pollution index is proportional to the predicted diffusion rate. The larger the initial pollution index, the greater the predicted diffusion rate, and the greater the pollution diffusion impact of the corresponding polluted sub-area. The first preset threshold and the second preset threshold are specifically set by the user, and the preset threshold is used to screen out the preferred monitoring sub-area. In the pollution monitoring and analysis tasks within a larger pollution area, the larger the area, the greater the monitoring work required, and the corresponding human and material resources consumed are also greater. The present application selects the preferred monitoring sub-area and monitors the preferred monitoring sub-area in the next monitoring cycle. While ensuring the data requirements of data monitoring and pollution diffusion model, fewer monitoring points can be used. By monitoring the preferred monitoring sub-area, the diffusion situation of the soil pollution area can be further accurately analyzed, thereby achieving accurate monitoring and diffusion analysis of the soil pollution area. The pollution prediction data includes prediction information such as soil pH, heavy metal content, soil moisture, and soil color.

[0108] According to an embodiment of the present application, during the next monitoring cycle, pollution analysis and pollution prediction are performed on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas, specifically:

[0109] In the next monitoring cycle, pollution monitoring is performed on the preferred monitoring sub-areas and preferred pollution monitoring data of each preferred monitoring sub-area is obtained;

[0110] During the next monitoring cycle, analyze and obtain secondary environmental characteristic data of soil contaminated areas;

[0111] The preferred pollution monitoring data and secondary environmental characteristic data are introduced into a pollution diffusion model to perform secondary pollution analysis and secondary pollution prediction, thereby obtaining secondary pollution prediction data and secondary preferred monitoring sub-areas.

[0112] It should be noted that this application predicts pollution spread by analyzing the pollution monitoring data within each monitoring cycle, and selects the preferred monitoring sub-area for the next cycle based on the predicted data. Due to the influence of pollution spread, the preferred monitoring sub-area for each cycle is generally different.

[0113] According to an embodiment of the present application, the pollution control analysis of the soil contaminated area based on the secondary pollution prediction data is performed and control scheme information is generated, and the control scheme information is sent to a preset terminal device, specifically:

[0114] Divide all polluted sub-areas into control areas by taking the secondary optimal monitoring sub-area as the first control sub-area and marking the remaining polluted sub-areas as the second control sub-areas;

[0115] Based on the first control sub-region and the second control sub-region, the secondary pollution prediction data is divided into first prediction data and second prediction data;

[0116] Performing pollution control analysis on the first control sub-region and generating a first control plan based on the first prediction data;

[0117] Performing pollution control analysis on the second control sub-region and generating a second control plan based on the second prediction data;

[0118] The first control scheme and the second control scheme are integrated into control scheme information and sent to a preset terminal device.

[0119] It should be noted that the preset terminal devices include mobile terminal devices and computer terminal devices. The secondary pollution prediction data include prediction data for all polluted sub-areas. The pollution control analysis specifically includes analysis of control schemes for soil acidification, heavy metal pollution and other pollution, and the schemes include the use of chemical reagents for control, microbial fertilizer application control, vegetation control and other methods. Since the first control sub-area is specifically a secondary preferred monitoring sub-area, its pollution level is generally high, and it is specifically a core monitoring point. Therefore, the first control sub-area is a key control area, and the corresponding control scheme generally includes a combination of multiple control methods. The second control sub-area is a non-key control area, generally corresponding to one control method.

[0120] According to an embodiment of the present application, it further includes:

[0121] Obtain pollution prediction data, soil pollution monitoring data, and environmental characteristic data within the first monitoring cycle;

[0122] Obtaining the preferred pollution monitoring data for the second cycle;

[0123] Compare and analyze the pollution prediction data with the optimal pollution monitoring data and calculate the data difference between the two;

[0124] Analyze the number of forecast cycles based on the data difference and obtain the optimal number of forecast cycles N;

[0125] Taking the first monitoring period as a reference point, obtaining environmental characteristic data of the previous N periods from the historical environmental characteristic data, and marking the environmental characteristic data as historical period environmental characteristic data;

[0126] Import soil pollution monitoring data, environmental characteristic data and historical period environmental characteristic data into the pollution diffusion model to perform reverse data prediction for N periods to obtain reverse pollution prediction data;

[0127] Based on N periods, the reverse pollution prediction data is divided into N monitoring data;

[0128] Use N monitoring data as control monitoring indicators for N periods;

[0129] Conduct pollution control on the soil contaminated area according to the control plan information, and obtain soil monitoring data for each cycle within N cycle time periods after the second cycle;

[0130] The soil monitoring data of each cycle is compared with the regulation and monitoring indicators of the corresponding cycle to obtain the regulation and evaluation information of each cycle.

[0131] It should be noted that the data difference can reflect the accuracy of the pollution diffusion model prediction data. Through the data difference, the optimal prediction cycle number N can be reasonably calculated, that is, the pollution diffusion model has a high accuracy in predicting data within N cycles. The reverse pollution prediction data is specifically a simulation analysis of the monitoring data of the soil pollution area in the first N cycles. According to the reverse pollution prediction data, the soil monitoring situation in the soil pollution area within the first cycle can be grasped, and the reverse pollution prediction data has a high reference and guidance value for soil pollution control. The control evaluation information of each cycle includes a total of N cycles of control evaluation information, that is, N control evaluation information. According to the N control evaluation information, the pollution control situation of the soil pollution area within N cycles can be accurately grasped.

[0132] According to an embodiment of the present application, it further includes:

[0133] Obtaining preferred pollution monitoring data for preferred monitoring sub-areas;

[0134] Based on the preferred pollution monitoring data, the basic pollution level analysis is conducted on each preferred monitoring sub-area to obtain the initial pollution index;

[0135] Calculating a predicted diffusion rate for each preferred monitoring sub-area based on the initial pollution index;

[0136] Obtain secondary pollution prediction data and secondary environmental characteristic data;

[0137] Analyze the predicted pollution index of each preferred monitoring sub-area based on the secondary pollution prediction data;

[0138] Determine the predicted pollution index and predicted diffusion rate of the preferred monitoring sub-area, and if the predicted pollution index is greater than the warning index and the predicted diffusion rate is greater than the warning rate, mark the corresponding preferred monitoring sub-area as a warning monitoring sub-area;

[0139] Taking the early warning monitoring sub-region as the center, the adjacent non-preferred monitoring sub-regions are marked as pollution early warning sub-regions;

[0140] Acquire the pollution monitoring data in the pollution warning sub-area within the next detection cycle and mark it as warning monitoring data;

[0141] Based on the early warning monitoring data, the initial pollution index of each pollution warning sub-area is calculated;

[0142] The pollution warning sub-area where the initial pollution index is greater than the warning index is marked as an emergency pollution source;

[0143] Obtain the location information of all sudden pollution sources and summarize them into sudden pollution source information;

[0144] The sudden pollution source information is sent to a preset terminal device.

[0145] It should be noted that the warning index and warning rate are user-set warning values, and the warning index is greater than a first preset threshold, and the warning rate is greater than a second preset threshold. When the predicted pollution index is greater than the warning index and the predicted diffusion rate is greater than the warning rate, it means that there is a sudden and severe pollution diffusion situation in the corresponding preferred monitoring sub-area, and the sudden pollution source is generally located in the non-preferred monitoring sub-area. At this time, the present application can accurately analyze the location of the sudden pollution source by analyzing the pollution situation in the non-preferred monitoring sub-area, and send the corresponding sudden pollution source information to the preset terminal device, so that the user can grasp the sudden pollution situation in the soil contaminated area in real time.

[0146] FIG4 shows a block diagram of a multi-source pollution spatiotemporal feature analysis system based on deep learning in the present application.

[0147] A second aspect of the present application further provides a multi-source pollution spatiotemporal feature analysis system 4 based on deep learning, the system comprising: a memory 41 and a processor 42, wherein the memory comprises a multi-source pollution spatiotemporal feature analysis program based on deep learning, and when the multi-source pollution spatiotemporal feature analysis program based on deep learning is executed by the processor, the following steps are implemented:

[0148] Acquire a soil contaminated area, and divide the area into multiple contaminated sub-areas based on the soil contaminated area;

[0149] Acquire soil pollution monitoring data and environmental characteristic data in multiple polluted sub-areas within one monitoring cycle;

[0150] Importing the soil pollution monitoring data and environmental characteristic data into a pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, thereby obtaining pollution prediction data and preferred monitoring sub-areas;

[0151] In the next monitoring cycle, pollution analysis and pollution prediction are carried out on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas;

[0152] A pollution control analysis is performed on the soil contaminated area based on the secondary pollution prediction data, and control plan information is generated, which is then sent to a preset terminal device.

[0153] According to an embodiment of the present application, the soil contaminated area is obtained by dividing the soil contaminated area into multiple contaminated sub-areas, specifically:

[0154] Obtain regional area and regional contour information in soil contaminated areas;

[0155] Constructing a pollution area map model based on the area and area contour information;

[0156] Based on the pollution area map model, the pollution area is divided into grids to obtain multiple pollution sub-areas, and the area difference between each pollution sub-area is guaranteed to be within a preset range.

[0157] It should be noted that the contaminated area map model is specifically a three-dimensional visual map model, which can display the pollution status of the area and dynamically display the pollution spread to the user. The number of sub-areas in the multiple contaminated areas is determined by the area size of the soil contamination area. The larger the area, the greater the number of sub-areas. The area difference between each contaminated sub-area is ensured to be within a preset range, which is generally determined by the user based on the area size of the soil contamination area.

[0158] According to an embodiment of the present application, the soil pollution monitoring data and environmental characteristic data in multiple polluted sub-areas are obtained within one monitoring cycle, specifically:

[0159] Conduct soil sampling and testing in each polluted sub-area, and obtain the sub-area pollution monitoring data corresponding to each polluted sub-area;

[0160] Integrate the pollution monitoring data of all sub-regions to obtain soil pollution monitoring data;

[0161] The soil pollution monitoring data includes pH, heavy metal content, soil moisture, and soil color information;

[0162] Obtain weather, humidity, and wind direction information in soil contaminated areas;

[0163] The weather, humidity and wind direction information are subjected to environmental information feature extraction to obtain environmental feature data.

[0164] It should be noted that the environmental characteristic data refers to the primary data features of the weather, humidity, and wind direction information, and can be incorporated into pollution diffusion models for feature analysis. The soil pollution monitoring data includes information such as pH, heavy metal content, soil moisture, and soil color. Specifically, soil color information is obtained by performing color feature analysis on soil image data obtained by drone aerial photography of each contaminated sub-area.

[0165] According to an embodiment of the present application, the soil pollution monitoring data and environmental characteristic data are imported into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics to obtain pollution prediction data and preferred monitoring sub-areas, which includes:

[0166] Build a pollution diffusion model based on the preset deep learning algorithm;

[0167] Obtain historical soil monitoring data and historical environmental characteristic data within a preset historical time period;

[0168] Normalize, standardize and integrate historical soil monitoring data and historical environmental characteristic data to obtain a pollution dataset;

[0169] Divide the contaminated dataset into training dataset, validation dataset, and test dataset according to the preset ratio;

[0170] The training dataset, validation dataset, and test dataset are imported into the pollution diffusion model for repeated pollution prediction training until the obtained pollution prediction data passes all test datasets.

[0171] It should be noted that the preset deep learning algorithm is specifically a combination of multiple prediction model algorithms, including linear regression, logistic regression, neural networks, Bayesian networks, and other prediction model algorithms, and the pollution diffusion model includes at least one prediction model algorithm. The preset historical time period is generally the last 1-3 years. The preset ratio is generally 2:2:1, that is, the training dataset and validation dataset each account for 40%, and the test dataset accounts for 20%.

[0172] According to an embodiment of the present application, the soil pollution monitoring data and environmental characteristic data are imported into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, and pollution prediction data and preferred monitoring sub-areas are obtained, specifically:

[0173] Importing the soil pollution monitoring data and environmental characteristic data into a pollution diffusion model;

[0174] Based on soil pollution monitoring data, the basic pollution degree of each polluted sub-area is analyzed to obtain the initial pollution index;

[0175] Calculate the predicted diffusion rate of each polluted sub-area based on the initial pollution index;

[0176] The pollution diffusion model predicts the pollution diffusion of all polluted sub-areas based on the predicted diffusion rate and environmental characteristic data to obtain pollution prediction data;

[0177] Analyze the predicted pollution index of each polluted sub-area based on the pollution prediction data;

[0178] The predicted pollution index and the predicted diffusion rate in the polluted sub-region are determined. If the predicted pollution index is greater than a first preset threshold and the predicted diffusion rate is greater than a second preset threshold, the corresponding polluted sub-region is marked as a preferred monitoring sub-region.

[0179] It should be noted that the initial pollution index is obtained by analyzing soil pollution monitoring data. The analysis dimensions include pH, heavy metal content, soil moisture, soil color information, etc. The larger the initial pollution index, the more serious the pollution level in the corresponding polluted sub-area, which is specifically manifested in lower soil pH values, higher heavy metal content, higher heavy metal concentration, and soil color deviating from normal color.

[0180] In addition, the predicted diffusion rate is specifically obtained by multiplying the initial pollution index by the pollution coefficient, that is, the initial pollution index is proportional to the predicted diffusion rate. The larger the initial pollution index, the greater the predicted diffusion rate, and the greater the pollution diffusion impact of the corresponding polluted sub-area. The first preset threshold and the second preset threshold are specifically set by the user, and the preset threshold is used to screen out the preferred monitoring sub-area. In the pollution monitoring and analysis tasks within a larger pollution area, the larger the area, the greater the monitoring work required, and the corresponding human and material resources consumed are also greater. The present application selects the preferred monitoring sub-area and monitors the preferred monitoring sub-area in the next monitoring cycle. While ensuring the data requirements of data monitoring and pollution diffusion model, fewer monitoring points can be used. By monitoring the preferred monitoring sub-area, the diffusion situation of the soil pollution area can be further accurately analyzed, thereby achieving accurate monitoring and diffusion analysis of the soil pollution area. The pollution prediction data includes prediction information such as soil pH, heavy metal content, soil moisture, and soil color.

[0181] According to an embodiment of the present application, during the next monitoring cycle, pollution analysis and pollution prediction are performed on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas, specifically:

[0182] In the next monitoring cycle, pollution monitoring is performed on the preferred monitoring sub-areas and preferred pollution monitoring data of each preferred monitoring sub-area is obtained;

[0183] During the next monitoring cycle, analyze and obtain secondary environmental characteristic data of soil contaminated areas;

[0184] The preferred pollution monitoring data and secondary environmental characteristic data are introduced into a pollution diffusion model to perform secondary pollution analysis and secondary pollution prediction, thereby obtaining secondary pollution prediction data and secondary preferred monitoring sub-areas.

[0185] It should be noted that this application predicts pollution spread by analyzing the pollution monitoring data within each monitoring cycle, and selects the preferred monitoring sub-area for the next cycle based on the predicted data. Due to the influence of pollution spread, the preferred monitoring sub-area for each cycle is generally different.

[0186] According to an embodiment of the present application, the pollution control analysis of the soil contaminated area based on the secondary pollution prediction data is performed and control scheme information is generated, and the control scheme information is sent to a preset terminal device, specifically:

[0187] Divide all polluted sub-areas into control areas by taking the secondary optimal monitoring sub-area as the first control sub-area and marking the remaining polluted sub-areas as the second control sub-areas;

[0188] Based on the first control sub-region and the second control sub-region, the secondary pollution prediction data is divided into first prediction data and second prediction data;

[0189] Performing pollution control analysis on the first control sub-region and generating a first control plan based on the first prediction data;

[0190] Performing pollution control analysis on the second control sub-region and generating a second control plan based on the second prediction data;

[0191] The first control scheme and the second control scheme are integrated into control scheme information and sent to a preset terminal device.

[0192] It should be noted that the preset terminal devices include mobile terminal devices and computer terminal devices. The secondary pollution prediction data include prediction data for all polluted sub-areas. The pollution control analysis specifically includes analysis of control schemes for soil acidification, heavy metal pollution and other pollution, and the schemes include the use of chemical reagents for control, microbial fertilizer application control, vegetation control and other methods. Since the first control sub-area is specifically a secondary preferred monitoring sub-area, its pollution level is generally high, and it is specifically a core monitoring point. Therefore, the first control sub-area is a key control area, and the corresponding control scheme generally includes a combination of multiple control methods. The second control sub-area is a non-key control area, generally corresponding to one control method.

[0193] According to an embodiment of the present application, it further includes:

[0194] Obtain pollution prediction data, soil pollution monitoring data, and environmental characteristic data within the first monitoring cycle;

[0195] Obtaining the preferred pollution monitoring data for the second cycle;

[0196] Compare and analyze the pollution prediction data with the optimal pollution monitoring data and calculate the data difference between the two;

[0197] Analyze the number of forecast cycles based on the data difference and obtain the optimal number of forecast cycles N;

[0198] Taking the first monitoring period as a reference point, obtaining environmental characteristic data of the previous N periods from the historical environmental characteristic data, and marking the environmental characteristic data as historical period environmental characteristic data;

[0199] Import soil pollution monitoring data, environmental characteristic data and historical period environmental characteristic data into the pollution diffusion model to perform reverse data prediction for N periods to obtain reverse pollution prediction data;

[0200] Based on N periods, the reverse pollution prediction data is divided into N monitoring data;

[0201] Use N monitoring data as control monitoring indicators for N periods;

[0202] Conduct pollution control on the soil contaminated area according to the control plan information, and obtain soil monitoring data for each cycle within N cycle time periods after the second cycle;

[0203] The soil monitoring data of each cycle is compared with the regulation and monitoring indicators of the corresponding cycle to obtain the regulation and evaluation information of each cycle.

[0204] It should be noted that the data difference can reflect the accuracy of the pollution diffusion model prediction data. Through the data difference, the optimal prediction cycle number N can be reasonably calculated, that is, the pollution diffusion model has a high accuracy in predicting data within N cycles. The reverse pollution prediction data is specifically a simulation analysis of the monitoring data of the soil pollution area in the first N cycles. According to the reverse pollution prediction data, the soil monitoring situation in the soil pollution area within the first cycle can be grasped, and the reverse pollution prediction data has a high reference and guidance value for soil pollution control. The control evaluation information of each cycle includes a total of N cycles of control evaluation information, that is, N control evaluation information. According to the N control evaluation information, the pollution control situation of the soil pollution area within N cycles can be accurately grasped.

[0205] The present application discloses a method and system for analyzing the spatiotemporal characteristics of multi-source pollution based on deep learning. In one monitoring cycle, soil pollution monitoring data and environmental characteristic data are imported into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, and pollution prediction data and preferred monitoring sub-areas are obtained. In the next monitoring cycle, pollution analysis and pollution prediction are performed on the preferred monitoring sub-areas to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas. Based on the secondary pollution prediction data, pollution control analysis is performed on the soil pollution area and control scheme information is generated. Through this application, it is possible to accurately analyze and predict the diffusion simulation of the pollution area from the perspective of spatiotemporal characteristics, and select preferred monitoring points in each monitoring cycle, thereby reducing the manpower and material resources consumed in monitoring, and realizing efficient analysis, pollution prediction and scheme control analysis of the pollution area based on the preferred monitoring points.

[0206] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0207] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0208] In addition, all functional units in the embodiments of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the above-mentioned integrated units can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0209] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0210] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0211] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning, characterized in that: include: Acquire a soil contaminated area, and divide the area into multiple contaminated sub-areas based on the soil contaminated area; Acquire soil pollution monitoring data and environmental characteristic data in multiple pollution sub-areas within one monitoring cycle; Importing the soil pollution monitoring data and environmental characteristic data into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, and obtaining pollution prediction data and preferred monitoring sub-areas; In the next monitoring cycle, pollution analysis and pollution prediction are carried out on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas; A pollution control analysis is performed on the soil contaminated area according to the secondary pollution prediction data, and control scheme information is generated, and the control scheme information is sent to a preset terminal device.

2. According to the method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning in claim 1, it is characterized in that: The soil contaminated area is obtained, and a plurality of contaminated sub-areas are obtained based on the soil contaminated area. Specifically, Obtaining regional area and regional contour information in soil contaminated areas; Constructing a pollution area map model based on the area and contour information of the area; Based on the pollution area map model, the pollution area is divided into grids to obtain multiple pollution sub-areas, and the area difference between each pollution sub-area is guaranteed to be within a preset range.

3. According to the method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning in claim 1, it is characterized in that: The soil pollution monitoring data and environmental characteristic data in multiple pollution sub-areas are obtained within one monitoring cycle, specifically: Conduct soil sampling and testing in each polluted sub-area, and obtain the sub-area pollution monitoring data corresponding to each polluted sub-area; Integrate the pollution monitoring data of all sub-regions to obtain soil pollution monitoring data; The soil pollution monitoring data includes pH, heavy metal content, soil moisture, and soil color information; Obtain weather, humidity, and wind direction information for soil contaminated areas; The weather, humidity and wind direction information are subjected to environmental information feature extraction to obtain environmental feature data.

4. According to the method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning in claim 1, it is characterized in that: The step of importing the soil pollution monitoring data and environmental characteristic data into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics to obtain pollution prediction data and preferred monitoring sub-areas includes: Build a pollution diffusion model based on the preset deep learning algorithm; Obtain historical soil monitoring data and historical environmental characteristic data within a preset historical time period; The historical soil monitoring data and historical environmental characteristic data are normalized, standardized and integrated to obtain the pollution data set; Divide the contaminated data set into a training data set, a validation data set, and a test data set according to a preset ratio; The training data set, validation data set, and test data set are imported into the pollution diffusion model for repeated pollution prediction training until the obtained pollution prediction data passes all test data sets.

5. According to the method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning in claim 4, it is characterized in that: The soil pollution monitoring data and environmental characteristic data are imported into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, and pollution prediction data and preferred monitoring sub-areas are obtained, specifically: Importing the soil pollution monitoring data and environmental characteristic data into a pollution diffusion model; Based on soil pollution monitoring data, the basic pollution degree of each polluted sub-area is analyzed to obtain the initial pollution index; The predicted diffusion rate of each polluted sub-area is calculated based on the initial pollution index; The pollution diffusion model predicts the pollution diffusion of all polluted sub-areas based on the predicted diffusion rate and environmental characteristic data to obtain pollution prediction data; Analyze the predicted pollution index of each polluted sub-area based on the pollution prediction data; The predicted pollution index and the predicted diffusion rate in the polluted sub-region are determined. If the predicted pollution index is greater than a first preset threshold and the predicted diffusion rate is greater than a second preset threshold, the corresponding polluted sub-region is marked as a preferred monitoring sub-region.

6. According to a method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning according to claim 1, it is characterized in that: In the next monitoring cycle, pollution analysis and pollution prediction are performed on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas, specifically: In the next monitoring cycle, pollution monitoring is performed on the preferred monitoring sub-areas and preferred pollution monitoring data of each preferred monitoring sub-area is obtained; In the next monitoring cycle, analyze and obtain secondary environmental characteristic data of soil contaminated areas; The preferred pollution monitoring data and secondary environmental characteristic data are introduced into the pollution diffusion model to perform secondary pollution analysis and secondary pollution prediction, so as to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas.

7. The method for analyzing spatiotemporal characteristics of multi-source pollution based on deep learning according to claim 6 is characterized in that: The pollution control analysis of the soil contaminated area is performed according to the secondary pollution prediction data, and control scheme information is generated, and the control scheme information is sent to a preset terminal device, specifically: Divide all polluted sub-areas into control areas, and the division method is to use the secondary optimal monitoring sub-area as the first control sub-area and mark the remaining polluted sub-areas as the second control sub-areas; Based on the first regulation sub-region and the second regulation sub-region, the secondary pollution prediction data is divided into first prediction data and second prediction data; According to the first prediction data, a pollution control analysis is performed on the first control sub-region and a first control scheme is generated; According to the second prediction data, performing pollution control analysis on the second control sub-region and generating a second control plan; The first control scheme and the second control scheme are integrated into control scheme information and sent to a preset terminal device.

8. A multi-source pollution spatiotemporal feature analysis system based on deep learning, characterized in that: The system includes: a memory and a processor, wherein the memory includes a multi-source pollution spatiotemporal feature analysis program based on deep learning, and when the multi-source pollution spatiotemporal feature analysis program based on deep learning is executed by the processor, the following steps are implemented: Acquire a soil contaminated area, and divide the area into multiple contaminated sub-areas based on the soil contaminated area; Acquire soil pollution monitoring data and environmental characteristic data in multiple pollution sub-areas within one monitoring cycle; Importing the soil pollution monitoring data and environmental characteristic data into the pollution diffusion model to perform pollution analysis and pollution prediction based on spatiotemporal characteristics, and obtaining pollution prediction data and preferred monitoring sub-areas; In the next monitoring cycle, pollution analysis and pollution prediction are carried out on the preferred monitoring sub-area to obtain secondary pollution prediction data and secondary preferred monitoring sub-areas; A pollution control analysis is performed on the soil contaminated area according to the secondary pollution prediction data, and control scheme information is generated, and the control scheme information is sent to a preset terminal device.

9. The multi-source pollution spatiotemporal feature analysis system based on deep learning according to claim 8 is characterized in that: The obtaining of the soil contaminated area, and performing regional division based on the soil contaminated area to obtain a plurality of contaminated sub-areas, are specifically: Obtaining regional area and regional contour information in soil contaminated areas; Constructing a pollution area map model based on the area and contour information of the area; Based on the pollution area map model, the pollution area is divided into grids to obtain multiple pollution sub-areas, and the area difference between each pollution sub-area is guaranteed to be within a preset range.

10. The multi-source pollution spatiotemporal feature analysis system based on deep learning according to claim 8, characterized in that: The soil pollution monitoring data and environmental characteristic data in multiple pollution sub-areas are obtained within one monitoring cycle, specifically: Conduct soil sampling and testing in each polluted sub-area, and obtain the sub-area pollution monitoring data corresponding to each polluted sub-area; Integrate the pollution monitoring data of all sub-regions to obtain soil pollution monitoring data; The soil pollution monitoring data includes pH, heavy metal content, soil moisture, and soil color information; Obtain weather, humidity, and wind direction information for soil contaminated areas; The weather, humidity and wind direction information are subjected to environmental information feature extraction to obtain environmental feature data.

Citation Information

Patent Citations

  • Risk assessment method based on soil heavy metal pollution pathway analysis

    CN110969345A

  • Post-natural-disaster drinking water source pollution monitoring and early warning method and system

    CN116109462A

  • Method and system for monitoring perfluorinated compound pollution in river based on Internet of Things

    CN116168771A

  • Diffusion prediction method and system based on deep learning

    CN116186566A

  • Multi-source pollution spatial-temporal characteristic analysis method and system based on deep learning

    CN116362571A

Cited By

  • Soil remediation monitoring method and system

    CN119380206A

  • Ecological environment dynamic monitoring system based on machine learning

    CN119493983A

  • Groundwater environment monitoring well whole process data management method and system

    CN119671074A

  • Smart city data processing method and system based on big data

    CN119919268A

  • Resistivity parameter-based soil organic pollution concentration prediction method and system

    CN120217908A