A method for optimizing the degradation of new soil pollutants

Abstract

The present application provides a method for optimizing the degradation and control of new pollutants in soil, which relates to the technical field of pollutant degradation. The method includes: according to the target area, setting the soil detection time point cloud and the soil detection location point cloud, obtaining a real-time soil detection data set, determining the historical neighborhood soil detection time node, and retrieving the historical neighborhood soil detection data set, obtaining the analysis results of new soil pollutants, obtaining the prediction results of the spread of new soil pollutants, obtaining a soil degradation control plan, and controlling the degradation of new pollutants in the target area. This application can solve the technical problem in the prior art that the lack of effective prediction models and control strategies for soil pollution control leads to unsatisfactory control effects. By acquiring real-time soil detection data sets, retrieving historical neighborhood data, analyzing new soil pollutants, and combining the spread prediction results to obtain a soil degradation control plan, the technical effect of improving degradation efficiency is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of pollutant degradation, and in particular to a method for optimizing the degradation and control of new soil pollutants. Background Art

[0002] With the acceleration of industrialization and urbanization, soil pollution is becoming increasingly serious, and the types and quantities of new pollutants are increasing, posing a threat to the ecological environment and human health. At present, traditional remediation methods have limited effects when facing new pollutants, and more precise and efficient governance technologies are needed.

[0003] In summary, there are technical problems in the prior art that the soil pollution control lacks effective prediction models and control strategies, resulting in unsatisfactory control effects. Summary of the invention

[0004] The purpose of this application is to provide a method for optimizing the degradation of new pollutants in soil, so as to solve the technical problem in the prior art that the soil pollution control lacks effective prediction models and control strategies, resulting in unsatisfactory control effects.

[0005] In view of the above problems, the present application provides a method for optimizing the degradation control of new soil pollutants, wherein the method includes: setting a soil detection time point cloud and a soil detection position point cloud according to a target area; based on the soil detection time point cloud and the soil detection position point cloud, performing soil detection on the target area according to a soil detection system to obtain a real-time soil detection data set corresponding to the real-time soil detection time node; determining a historical neighborhood soil detection time node of the real-time soil detection time node according to the soil detection time point cloud, and retrieving a historical neighborhood soil detection data set corresponding to the historical neighborhood soil detection time node according to the soil detection system; performing new soil pollutant analysis based on the historical neighborhood soil detection data set and the real-time soil detection data set to obtain new soil pollutant analysis results; performing spread prediction based on the new soil pollutant analysis results to obtain new soil pollutant spread prediction results; performing degradation control decisions based on the new soil pollutant analysis results and the new soil pollutant spread prediction results to obtain a soil degradation control plan; and performing new pollutant degradation control and management on the target area according to the soil degradation control plan.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] According to the target area, a soil detection time point cloud and a soil detection location point cloud are set; based on the soil detection time point cloud and the soil detection location point cloud, soil detection is performed on the target area according to the soil detection system to obtain a real-time soil detection data set corresponding to the real-time soil detection time node; according to the soil detection time point cloud, the historical neighborhood soil detection time node of the real-time soil detection time node is determined, and the historical neighborhood soil detection data set corresponding to the historical neighborhood soil detection time node is retrieved according to the soil detection system; new soil pollutants are analyzed based on the historical neighborhood soil detection data set and the real-time soil detection data set to obtain new soil pollutant analysis results; spread prediction is performed based on the new soil pollutant analysis results to obtain new soil pollutant spread prediction results; degradation control decisions are made based on the new soil pollutant analysis results and the new soil pollutant spread prediction results to obtain a soil degradation control plan; new pollutant degradation control is performed on the target area according to the soil degradation control plan. That is, by acquiring the real-time soil detection data set, retrieving the historical neighborhood data, analyzing new soil pollutants, and combining the spread prediction results to obtain a soil degradation control plan, the technical effect of improving degradation efficiency is achieved.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 A schematic diagram of a process for optimizing the degradation and control of new soil pollutants in this application;

[0011] Figure 2 This is a schematic diagram of the process of obtaining the analysis results of new soil pollutants in a new soil pollutant degradation optimization control method of this application. DETAILED DESCRIPTION

[0012] This application provides a method for optimizing the degradation and control of new soil pollutants, which solves the technical problem in the prior art that the lack of effective prediction models and control strategies for soil pollution control leads to unsatisfactory control effects. By acquiring real-time soil detection data sets, retrieving historical neighborhood data, analyzing new soil pollutants, and combining the spread prediction results to obtain a soil degradation control plan, the technical effect of improving degradation efficiency is achieved.

[0013] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0014] Please refer to the attached Figure 1 The present application provides a method for optimizing the degradation of new pollutants in soil, wherein the method specifically comprises the following steps:

[0015] Step 1: Set the soil detection time point cloud and soil detection location point cloud according to the target area.

[0016] Specifically, the target area for monitoring is clearly defined, including a specific industrial area, agricultural area or urban area. According to the soil pollution characteristics and monitoring needs of the target area, a series of soil testing time points are set, which can be fixed (such as daily, weekly, monthly) or dynamically adjusted (such as increasing the monitoring frequency after pollution). The time point cloud ensures the continuity and regularity of monitoring. According to the topography, land use type, distribution of pollution sources and other factors in the target area, a series of soil testing time points are set, which can be fixed or adjusted according to needs. The location point cloud ensures the comprehensiveness and representativeness of the monitoring. By setting the soil testing time point cloud and the soil testing location point cloud, comprehensive monitoring of soil pollution in the target area can be achieved.

[0017] Step 2: Based on the soil detection time point cloud and the soil detection position point cloud, soil detection is performed on the target area according to a soil detection system to obtain a real-time soil detection data set corresponding to a real-time soil detection time node.

[0018] Specifically, the time nodes of real-time soil detection are determined according to the soil detection time point cloud, based on seasonal changes, new pollutant transmission characteristics, crop growth cycle and other factors. Next, the soil detection location corresponding to each time node is determined according to the soil detection location point cloud to ensure that the sampling point can represent the soil conditions in the target area. Ensure that the soil detection system can accurately collect and analyze soil samples, including sampling equipment, analytical instruments, etc. Soil sampling is carried out at designated locations, and sampling is carried out strictly in accordance with the operating procedures of the soil detection system to ensure the accuracy and representativeness of the samples. The collected soil samples are sent to the laboratory for analysis to obtain real-time soil detection data sets, including the content of new pollutants in the soil, soil physical and chemical properties, microbial indicators, etc. Through soil testing, real-time data on soil pollution in the target area is obtained. By obtaining real-time soil detection data sets, real-time monitoring of the target area is achieved to ensure the accuracy and timeliness of soil detection.

[0019] Step three: Determine the historical neighborhood soil detection time node of the real-time soil detection time node according to the soil detection time point cloud, and retrieve the historical neighborhood soil detection data set corresponding to the historical neighborhood soil detection time node according to the soil detection system.

[0020] Specifically, according to the soil detection time point cloud, the previous soil detection time node is selected as the historical neighborhood soil detection time node, which is close to the real-time detection time point in time, so as to analyze the long-term change trend and seasonal pattern of soil pollution. Using the soil detection system, according to the determined historical neighborhood soil detection time node, the corresponding historical soil detection data is retrieved, including information such as the types and concentrations of new pollutants in the soil samples collected at these time points. The retrieved historical soil detection data is processed, and the processed data is compared and analyzed with the real-time soil detection data. Statistical methods, data mining techniques, etc. are used to compare the soil pollution status of different periods, identify the change trend of new pollutants, and evaluate the migration and transformation process of new pollutants. For example, suppose that an annual soil detection time point cloud is set, that is, a test is conducted on the first day of each year. The real-time soil detection time node is June 2024, and the historical neighborhood soil detection time node determined based on this is the detection time point of the previous year, that is, January 1, 2023. The soil sample data collected on this day is retrieved and compared with the soil detection data at the current time to analyze the change trend of new pollutants. Based on real-time soil detection time nodes, historical data is obtained and analyzed to more accurately understand the evolution of soil pollution.

[0021] Step 4: Perform new soil pollutant analysis based on the historical neighborhood soil detection dataset and the real-time soil detection dataset to obtain new soil pollutant analysis results.

[0022] Specifically, the historical neighborhood soil detection dataset and the real-time soil detection dataset are integrated. The historical neighborhood soil detection dataset contains soil sample data collected at the same or similar location before, and the real-time soil detection dataset contains soil sample data collected currently. The integrated data is cleaned and normalized, and the processed data is trained using a soil new pollutant detection learning model, such as a BP neural network. Through model training and optimization, key features in soil pollution data are extracted to identify the types and concentrations of new pollutants. Then, through the application of the soil new pollutant detection learning model, new pollutants in the soil are analyzed, including identifying the types of new pollutants, analyzing the sources of new pollutants, and evaluating the scope and extent of the impact of new pollutants. Then, the results of soil new pollutant analysis are generated, including information such as the types, concentrations, and distribution ranges of new pollutants. By obtaining the results of soil new pollutant analysis, changes and trends in soil pollution conditions can be more comprehensively understood.

[0023] Step 5: Perform a spread prediction based on the analysis results of the new soil pollutants to obtain the spread prediction results of the new soil pollutants.

[0024] Specifically, analyze the results of new soil pollutant analysis, evaluate the characteristics of new pollutants, the impact of environmental factors, etc. Select appropriate prediction models and methods to predict the spread of new soil pollutants, including neural networks, decision trees, support vector machines, etc., to simulate and predict the spread process of new pollutants. Use the selected prediction models and methods to predict the spread of new soil pollutants based on the results of new soil pollutants analysis to obtain the prediction results of new soil pollutant spread. Input parameters such as the characteristics of new pollutants, environmental factors, and soil characteristics to predict the spread speed, range, and trend of new pollutants. By predicting the spread based on the results of new soil pollutants analysis, accurate predictions of future spread trends and impact ranges of new soil pollutants can be achieved, thereby improving the accuracy of soil pollution monitoring.

[0025] Step 6: Make a degradation control decision based on the analysis results of the new soil pollutants and the prediction results of the spread of the new soil pollutants to obtain a soil degradation control plan.

[0026] Specifically, the soil samples are analyzed through the soil new pollutant detection learning model to identify the type, concentration, distribution range and other information of new pollutants, and the analytical results of new soil pollutants are obtained. Using the soil new pollutant spread prediction channel, according to the characteristics of new soil pollutants, environmental factors and monitoring needs, the spread trend and impact range of pollutants are predicted, and the prediction results of new soil pollutant spread are obtained. Select appropriate explanation methods, determine monitoring frequency and scope, formulate pollution control strategies, etc., compare the effects and costs of different governance plans, and select the best plan. According to the decision results, the soil degradation control plan is obtained, which will guide the actual soil pollution control work, including the selection of governance technology, formulation of governance plans, and arrangement of governance resources. For example, suppose that in the soil pollution monitoring of the industrial park, the analytical results and spread prediction results of new soil pollutants are collected. By analyzing these data, biodegradation technology is used to treat soil pollution. A detailed governance plan is formulated, including monitoring the soil pollution status, introducing appropriate microorganisms, and regularly evaluating the governance effect. Then, according to this plan, soil degradation control measures are implemented to control and reduce soil pollution. By making degradation control decisions based on the analysis results of new soil pollutants and the prediction results of the spread of new soil pollutants, a soil degradation control plan can be obtained, which can achieve intelligent management of soil pollution.

[0027] Step 7: Control the degradation of new pollutants in the target area according to the soil degradation control plan.

[0028] Specifically, based on the results of the analysis of new soil pollutants and the prediction results of the spread of new soil pollutants, as well as the recommendations of the output of the soil degradation control decision model, select appropriate degradation technologies, determine monitoring plans, formulate governance schedules, arrange governance resources, etc. Implement soil degradation control programs, including physical remediation, chemical remediation, biological remediation, or a combination of multiple methods. For example, physical remediation involves the excavation and replacement of soil; chemical remediation involves the addition of chemicals to neutralize or degrade pollutants; biological remediation involves the introduction of specific microorganisms or plants to degrade pollutants. The governance measures were implemented according to the plan, and the concentration changes of pollutants in the soil were regularly monitored to evaluate the effectiveness of the governance measures. According to the monitoring results, the governance plan was adjusted to improve the governance effect. The data, measures taken and governance effects in the governance process were recorded in detail, and a report was formed to facilitate subsequent evaluation and decision-making. By controlling the degradation of new pollutants in the target area according to the soil degradation control plan, we can effectively control and remediate soil pollution.

[0029] Furthermore, step one of this application includes:

[0030] Obtain a regional layout map and a soil pollution monitoring level of the target area; set a soil detection unit space grid based on the regional layout map and the soil pollution monitoring level; grid the target area according to the soil detection unit space grid to obtain a regional grid division result; arrange soil detection points according to the regional grid division result to obtain the soil detection position point cloud; collect pollution source information based on the soil detection position point cloud to obtain pollution source distribution information at each position; construct the soil detection time point cloud according to the pollution source distribution information at each position.

[0031] Specifically, obtain a regional layout map of the target area, that is, the spatial structure and characteristics, including land use type, topography, hydrological conditions, etc. For example, the soil pollution characteristics and risks in industrial and agricultural areas are completely different, so it is necessary to adjust the grid settings according to regional characteristics. Near the pollution source or in environmentally sensitive areas, the grid can be set smaller to obtain more accurate monitoring data. The soil pollution monitoring level reflects the severity and changing trend of soil pollution in the target area. The monitoring level is determined based on factors such as the concentration, type, and distribution range of new pollutants in the soil. High-level monitoring areas require more frequent inspections and stricter control measures.

[0032] According to the size of the target area, the degree of soil pollution and the purpose of detection, determine the appropriate grid size, set the soil detection unit space grid, use each grid unit as an independent monitoring point, collect real-time soil detection data, and realize the refined monitoring of soil pollution in the target area. According to the size and distribution of the soil detection unit space grid, subdivide the target area into a series of smaller, interconnected grid units, and record the coordinates and number of each grid unit. Each grid unit represents an independent monitoring area, and its size and shape can be adjusted according to the specific conditions and monitoring needs of the area. Set sampling points within each grid unit or at the junction of the units to ensure that the sampling points can represent the soil conditions of the entire grid. The selection of sampling points should take into account a variety of factors, such as historical data on soil pollution, land use type, topography, etc. In areas with higher pollution risks, the density of detection points can be increased.

[0033] The specific location of each sampling point is recorded to form a soil detection location point cloud. The soil detection location point cloud is the spatial distribution of all soil detection points in the target area. It is obtained by converting the regional grid division results into specific soil detection points. Each point corresponds to a specific geographical location. These points together constitute a point set in a three-dimensional space, namely a point cloud. Pollution source information is collected based on the soil detection location point cloud, including collecting soil samples, recording the types and concentrations of new pollutants, analyzing the sources and migration paths of new pollutants, etc., and establishing the association between each point and potential pollution sources. The collected soil samples are sent for inspection to analyze the distribution of pollution sources, including heavy metals, new organic pollutants, etc. According to the pollution source distribution information and soil characteristics, the monitoring time points are determined, such as different stages of crop growth, before and after the rainy season, and other key periods. The time points are combined with the soil detection location point cloud to form a soil detection time point cloud, which is used to guide future monitoring activities. The soil detection time point cloud represents the time distribution of all soil detection points in the target area. It is constructed based on the pollution source distribution information of each location. Each point corresponds to a specific time point. These time points together constitute a time series, namely a time point cloud.

[0034] For example, assume that the target area is an industrial park, which contains multiple factories and warehouses, and needs to monitor and control a new pollutant discharged from industrial activities in the soil in the area. First, obtain a regional layout map, including the geographical location of the industrial park, the distribution of factories and warehouses, the road network, and the surrounding environmental characteristics, such as rivers and residential areas. Then, based on historical data and real-time monitoring results, evaluate the concentration and distribution of new pollutants, divide the industrial park into different pollution level areas, and the area close to the factory is divided into high pollution level areas. Set smaller grid cells around each factory to more accurately monitor the emission of new pollutants. In areas far away from the factory, the grid cells are set larger. According to the set grid cells, the target area is gridded to obtain a detailed grid map showing the location and size of all grid cells. Soil detection points are arranged in each grid cell, soil samples are collected at each point, and the types and concentrations of new pollutants are analyzed to obtain the distribution information of pollution sources at each location, such as which factories emit new pollutants and the diffusion range of new pollutants. According to the pollution source distribution information, track the change trend of new pollutants over time. By setting up a spatial grid of soil detection units and deploying detection points in each grid unit, refined monitoring of soil pollution can be achieved.

[0035] Furthermore, the present application also includes the following steps:

[0036] According to the soil pollution monitoring level, the basic computing power for soil pollution monitoring is matched; according to the distribution information of the pollution sources at each location, the soil pollution depth is predicted to obtain the soil pollution depth prediction results at each location; according to the soil pollution depth prediction results at each location, the basic computing power for soil pollution monitoring is optimized respectively to obtain the soil pollution monitoring computing power at each location; according to the soil pollution monitoring computing power at each location, the soil detection time is set to generate the soil detection time point cloud.

[0037] Specifically, the required monitoring frequency and data volume are evaluated based on the soil pollution monitoring level. Areas with high pollution levels require more frequent monitoring and more data points, and therefore require stronger computing power to process and analyze this data. According to the determined computing power requirements, the corresponding soil pollution monitoring basic computing power is matched, and the computing power to process and analyze soil pollution monitoring data includes resources in data storage, processing, and analysis.

[0038] Collect the distribution information of pollution sources at each location, including the type, concentration, and distribution range of new pollutants, and integrate relevant data such as the physical and chemical properties of the soil, hydrogeological conditions, and land use history. Based on these data, and considering the migration and transformation mechanism of new pollutants in the soil, as well as various factors affecting the depth distribution of new pollutants, establish a soil pollution depth prediction model. Use historical data to train the model, and use an independent data set to verify the model to ensure the prediction accuracy of the model. Input the distribution information of pollution sources at each location into the prediction model to predict the depth of soil pollution, obtain the prediction results, and show the depth distribution of new pollutants in the soil at different locations. Analyze the prediction results of soil pollution depth at each location, understand the vertical distribution of new pollutants in the soil at different locations, and adjust and optimize the allocation of monitoring computing power. Optimize the soil pollution monitoring computing power by improving the data storage structure, optimizing the data processing algorithm, or introducing parallel computing and distributed computing technologies to obtain the soil pollution monitoring computing power at each location. For example, in some areas, new pollutants have penetrated into deeper soil layers, increase the monitoring frequency and data points, and improve computing resources to ensure that the data in these areas can be effectively processed and analyzed.

[0039] Determine the optimal soil detection time based on soil pollution monitoring computing power and pollution source distribution information. Considering the dynamic changes of new soil pollutants, it may be necessary to monitor at different time points to capture seasonal changes in new pollutants or the impact of emergencies. Combine the optimized monitoring computing power with the soil detection location point cloud to generate a soil detection time point cloud that contains the time distribution information of all monitoring points. By matching and optimizing the basic computing power for soil pollution monitoring, monitoring data can be processed and analyzed more efficiently, and dynamic monitoring of soil pollution can be achieved.

[0040] Further, as attached Figure 2As shown, step 4 of this application includes:

[0041] According to the real-time soil detection data set, extract the first real-time soil detection data sequence corresponding to the first soil detection point; according to the historical neighborhood soil detection data set, extract the first historical neighborhood soil detection data sequence corresponding to the first soil detection point; perform deviation analysis based on the first real-time soil detection data sequence and the first historical neighborhood soil detection data sequence to obtain a first point soil detection deviation data sequence; construct a soil new pollutant detection channel; input the first point soil detection deviation data sequence into the soil new pollutant detection channel to obtain a first point soil new pollutant identification result, and add the first point soil new pollutant identification result to the soil new pollutant analysis result; according to the historical neighborhood soil detection data set and the real-time soil detection data set, in combination with the soil new pollutant detection channel, continue to perform soil new pollutant analysis, and generate the soil new pollutant analysis result.

[0042] Specifically, the first point is selected from a series of soil detection points in the real-time soil detection data set, and the real-time soil detection data of the point at the current time point is obtained to form the first real-time soil detection data sequence, including parameters such as the type and concentration of new pollutants. The data closest to the current first soil detection point is selected from the historical neighborhood soil detection data set to form the first historical neighborhood soil detection data sequence, including parameters such as the type and concentration of new pollutants collected previously. The deviation between the two data sequences is calculated, that is, the difference between the real-time data and the historical data, and the calculated deviation data is used to form the first point soil detection deviation data sequence, reflecting the changes in the pollution status of the first soil detection point between the current time point and the historical time point, including the increase or decrease in the concentration of new pollutants, the change in the type of new pollutants, etc.

[0043] In the target area, according to the potential distribution of soil pollution and environmental factors, a new soil pollutant detection channel is constructed. It can be fixed or adjusted as needed. The first point soil detection deviation data sequence is input into the soil new pollutant detection channel, including the comparison results of real-time detection data and historical detection data, as well as information such as the change trend of new pollutant concentration. Use specific detection technology, algorithms or models to identify and analyze new pollutants, determine whether there are new types of pollutants or concentration changes, obtain the first point soil new pollutant identification results, and add them to the previously obtained soil new pollutant analysis results. Use the historical neighborhood soil detection dataset and the real-time soil detection dataset, combined with the soil new pollutant detection channel, continue to analyze new pollutants at other points, and generate the final soil new pollutant analysis results containing information such as new pollutant types, concentrations, and distribution ranges. By constructing a soil new pollutant detection channel and generating soil new pollutant analysis results, intelligent monitoring and identification of new soil pollutants can be achieved.

[0044] Furthermore, the present application also includes the following steps:

[0045] Load the sample point soil detection deviation data sequence records and the sample point soil new pollutant identification result records; activate the soil new pollutant detection learning model; use the sample point soil detection deviation data sequence records as input data and the sample point soil new pollutant identification result records as output supervision data to train the soil new pollutant detection learning model, and obtain the soil new pollutant detection loss coefficient after each predetermined number of trainings; if the soil new pollutant detection loss coefficient is less than the predetermined soil new pollutant detection loss coefficient, generate the soil new pollutant detection channel.

[0046] Specifically, collect the data sequence records of soil detection deviations at the sample points, including the comparison results of the real-time data and historical data of the soil samples collected at the sample points, to reflect the changing trend of the concentration of new pollutants. Then, collect the results records of soil new pollutant identification at the sample points, including the information of new pollutants identified through the soil new pollutant detection channel, such as type, concentration, distribution range, etc. According to the research purpose and available data, select a suitable soil new pollutant detection learning model, such as BP neural network, and activate the model. Use the data sequence records of soil detection deviations at the sample points as input data and the results records of soil new pollutant identification at the sample points as output supervision data to train the soil new pollutant detection learning model. The model will learn how to predict the output new pollutant identification result records based on the input soil detection data sequence records. During the training process, the model will continuously adjust the parameters to reduce the error between the predicted results and the actual output supervision data.

[0047] After each training for a predetermined number of times, the loss coefficient of soil new pollutant detection is obtained. The loss coefficient is calculated by cross-validation or other evaluation methods during the training process. It is an indicator to measure the performance of the soil new pollutant detection model, reflecting the error between the model and the true value when predicting new pollutants in the soil. During the training process, the model predicts the identification results of soil new pollutants at the sample point and compares them with the actual output supervision data. The loss function calculates the difference between the predicted results and the actual results and uses these differences as losses. The loss coefficient is the average of these losses, usually a numerical value, indicating the accuracy of the model prediction. The predetermined soil new pollutant detection loss coefficient refers to an expected error threshold set before model training. The purpose of model training is to make the loss coefficient as close to this predetermined value as possible. This threshold is not fixed, but can be adjusted according to actual conditions, including the complexity of the model, the quality of training data, and the needs of actual applications. If the soil new pollutant detection loss coefficient is less than the predetermined soil new pollutant detection loss coefficient, it means that the error between the model's prediction results and the actual output supervision data is small, and the model performance is good. This shows that the model has learned the key features in the soil pollution data accurately enough and can be used for the detection and identification of new soil pollutants. For example, the loss coefficient of new soil pollutant detection is set to 0.1, which means that the prediction error of the model is expected to be no more than 10%. If the loss coefficient is always lower than 0.1 during the training process, it can be considered that the model has reached the predetermined performance standard and a new soil pollutant detection channel can be generated. Through these steps, an accurate and reliable new soil pollutant detection channel can be generated to monitor new pollutants in the soil in real time.

[0048] Further, step five of this application includes:

[0049] Obtain a spread prediction time zone window; perform soil environment prediction on the target area according to the spread prediction time zone window to obtain a regional soil environment prediction result; build a new soil pollutant spread prediction channel; input the new soil pollutant analysis results and the regional soil environment prediction results into the new soil pollutant spread prediction channel to generate the new soil pollutant spread prediction result.

[0050] Specifically, according to the characteristics of new soil pollutants, environmental conditions and time series analysis results of monitoring data, the length of the prediction time zone window is determined, which can be set as a time window of one month or one year in the future. The rationality of this time zone window is verified using historical data and simulation experiment results. According to the set spread prediction time zone window, the soil environment of the target area is predicted to obtain the prediction results of the regional soil environment in the future time period, including the spread trend, the scope of impact, and the changes in soil environmental quality. Build a new soil pollutant spread prediction channel, including specific algorithms, models or software tools, to analyze and predict the spread trend of new soil pollutants.

[0051] The results of soil new pollutant analysis and regional soil environment prediction are input into the soil new pollutant spread prediction channel. Through the processing of the channel, the results of soil new pollutant spread prediction are generated, including the spread trend of pollutants, the scope of influence, the change of soil environmental quality, etc., to provide a scientific basis for pollution control and governance. For example, a serious soil pollution incident once occurred in an industrial area. In order to predict the diffusion and spread trend of pollutants in the future, a spread prediction time zone window is first obtained, and the prediction time range for the next year is set. The soil environment change trend of the target area in the next year is predicted using historical data and models, and a soil new pollutant spread prediction channel is built. The soil new pollutant analysis result record set, the regional soil environment record set and the soil new pollutant spread result record set are collected, and K soil new pollutant spread prediction learning models are supervised and learned to generate K models that meet the soil new pollutant spread prediction accuracy. Then, the output information of these models is collected to obtain the spread prediction integrated learning information. Finally, with the integrated learning information as input and the soil new pollutant spread result record set as output, a soil new pollutant spread prediction integrated model that meets the soil new pollutant spread prediction accuracy is trained. By building a channel to predict the spread of new soil pollutants, the spread trend of new soil pollutants can be monitored in real time.

[0052] Furthermore, the present application also includes the following steps:

[0053] Obtain a new soil pollutant analysis result record set, a regional soil environment record set and a new soil pollutant spread result record set; perform supervised learning on K new soil pollutant spread prediction learning models based on the new soil pollutant analysis result record set, the regional soil environment record set and the new soil pollutant spread result record set to generate K new soil pollutant spread prediction models that meet the new soil pollutant spread prediction accuracy, wherein K is a positive integer greater than 1; collect the output information of the K new soil pollutant spread prediction models to obtain spread prediction integrated learning information; use the spread prediction integrated learning information as input information and the new soil pollutant spread result record set as output information to train a new soil pollutant spread prediction integrated model that meets the new soil pollutant spread prediction accuracy; merge the K new soil pollutant spread prediction models as parallel independent nodes to generate a new soil pollutant spread prediction processing layer; merge the new soil pollutant spread prediction processing layer with the input layer of the new soil pollutant spread prediction integrated model to generate the new soil pollutant spread prediction channel.

[0054] Specifically, collect records of new soil pollutant analysis results, including information such as the type, concentration, and distribution range of new pollutants. Collect regional soil environment records, including environmental factors such as soil physical and chemical properties, topography, and hydrological conditions, as well as other information related to soil pollution. Collect records of new soil pollutant spread results, including the spread trend and impact range of new pollutants. Use known records of new soil pollutant analysis results, regional soil environment records, and new soil pollutant spread results to train the prediction model. Choose a suitable learning model for predicting the spread of new soil pollutants, which can be a variety of machine learning algorithms, such as neural networks, decision trees, support vector machines, etc.

[0055] Input the training data into the model to let the model learn the spread law of new soil pollutants and the influencing factors of regional soil environment. During the training process, the model will continuously adjust the parameters to minimize the error between the predicted results and the actual results. Through supervised learning, K new soil pollutant spread prediction models that meet the prediction accuracy of new soil pollutants are generated, which means that the prediction error of the model is within an acceptable range, such as the prediction error is less than a predetermined threshold. Collect the output information of K new soil pollutant spread prediction models to obtain spread prediction integrated learning information, including the prediction results of all models, which can be used to evaluate the prediction performance of the model and the diversity of prediction results. Take the spread prediction integrated learning information as input information and output the new soil pollutant spread result record set, which contains the actual new soil pollutant spread data. Select appropriate integrated learning algorithms, such as random forests, gradient boosting machines or stacking models, to train the new soil pollutant spread prediction integrated model, and combine the prediction results of multiple models to improve the accuracy and robustness of the prediction. During the training process, the model will continuously adjust the parameters to reduce the error between the prediction results and the actual output supervision data.

[0056] Select K soil new pollutant spread prediction models and merge them as parallel independent nodes. Each model is regarded as an independent prediction unit, which can run simultaneously and produce prediction results. These models are independent of each other and do not affect each other. These parallel independent nodes are combined into a processing layer, namely the soil new pollutant spread prediction processing layer, which can process multiple input data at the same time and produce multiple prediction results. The soil new pollutant spread prediction processing layer is merged with the input layer of the soil new pollutant spread prediction integrated model to generate a soil new pollutant spread prediction channel. Exemplarily, it is assumed that K = 3 soil new pollutant spread prediction models are constructed, based on BP neural network, random forest and support vector machine algorithms respectively. These models are used to predict the same set of soil new pollutant spread prediction data and collect output information. Select the random forest integrated learning algorithm, take the prediction results of these K models as input information, and the actual soil new pollutant spread data as output information to train the soil new pollutant spread prediction integrated model. Through training, an integrated model that meets the soil new pollutant spread prediction accuracy is generated. By building a soil new pollutant spread prediction channel, the spread trend of soil new pollutants can be monitored in real time.

[0057] Furthermore, step six of this application includes:

[0058] Based on the knowledge graph, a new soil pollutant degradation control decision model is constructed; the new soil pollutant analysis results and the new soil pollutant spread prediction results are input into the new soil pollutant degradation control decision model, and the soil degradation control plan is output.

[0059] Specifically, the knowledge graph is a structured semantic knowledge base that represents and links various knowledge through entities, attributes and relationships. Entities include soil, pollutants, degradation processes, environmental factors, etc. Attributes include types, concentrations, and degradation efficiency of pollutants, etc. Relationships include interactions between pollutants and soil, steps in the degradation process, etc. The knowledge graph, machine learning or other data analysis techniques are used to construct a new soil pollutant degradation control decision model, including using graph query language or graph reasoning algorithm to analyze the data in the knowledge graph to generate suggestions or decisions for new soil pollutant degradation control. The parsing results and spread prediction results of new soil pollutants are input into the new soil pollutant degradation control decision model, including the types, concentrations, distribution ranges, spread trends and impact ranges of pollutants. The model will analyze the soil pollution status and predicted spread trends based on the input data and the information in the knowledge graph, so as to generate an optimal soil degradation control plan, including selecting appropriate degradation methods, determining monitoring frequency and range, formulating pollution control strategies, etc., to provide guidance for the governance and control of soil pollution. For example, a new soil pollutant knowledge graph containing information such as types, degradation processes, and environmental factors of new soil pollutants. Using this knowledge graph, combined with the analysis results and spread prediction results of new soil pollutants, a new soil pollutant degradation control decision model is constructed. Through this model, according to the specific situation of soil pollution, appropriate degradation methods are selected, monitoring frequency and scope are determined, and pollution control strategies are formulated. Based on these decisions, soil degradation control measures are implemented to control and reduce soil pollution. By constructing a new soil pollutant degradation control decision model based on the knowledge graph, a soil degradation control plan is output to improve the effect of soil pollution control.

[0060] In summary, the method for optimizing the degradation of new soil pollutants provided in this application has the following technical effects:

[0061] According to the target area, a soil detection time point cloud and a soil detection location point cloud are set; based on the soil detection time point cloud and the soil detection location point cloud, soil detection is performed on the target area according to the soil detection system to obtain a real-time soil detection data set corresponding to the real-time soil detection time node; according to the soil detection time point cloud, the historical neighborhood soil detection time node of the real-time soil detection time node is determined, and the historical neighborhood soil detection data set corresponding to the historical neighborhood soil detection time node is retrieved according to the soil detection system; new soil pollutants are analyzed based on the historical neighborhood soil detection data set and the real-time soil detection data set to obtain new soil pollutant analysis results; spread prediction is performed based on the new soil pollutant analysis results to obtain new soil pollutant spread prediction results; degradation control decisions are made based on the new soil pollutant analysis results and the new soil pollutant spread prediction results to obtain a soil degradation control plan; new pollutant degradation control is performed on the target area according to the soil degradation control plan. That is, by acquiring the real-time soil detection data set, retrieving the historical neighborhood data, analyzing new soil pollutants, and combining the spread prediction results to obtain a soil degradation control plan, the technical effect of improving degradation efficiency is achieved.

[0062] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0063] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A method for optimizing the degradation of new soil pollutants, characterized in that: The method comprises: According to the target area, set the soil detection time point cloud and soil detection location point cloud; Based on the soil detection time point cloud and the soil detection position point cloud, soil detection is performed on the target area according to a soil detection system to obtain a real-time soil detection data set corresponding to a real-time soil detection time node; Determine, according to the soil detection time point cloud, a historical neighborhood soil detection time node of the real-time soil detection time node, and retrieve, according to the soil detection system, a historical neighborhood soil detection data set corresponding to the historical neighborhood soil detection time node; Performing new soil pollutant analysis based on the historical neighborhood soil detection dataset and the real-time soil detection dataset to obtain new soil pollutant analysis results; Performing a spread prediction based on the new soil pollutant analysis results to obtain a new soil pollutant spread prediction result; Make a degradation control decision based on the analysis results of the new soil pollutants and the prediction results of the spread of the new soil pollutants to obtain a soil degradation control plan; Conducting new pollutant degradation control in the target area according to the soil degradation control plan; Wherein, performing a spread prediction based on the analysis results of the new soil pollutants to obtain a spread prediction result of the new soil pollutants includes: Obtain the spread prediction time zone window; Perform soil environment prediction on the target area according to the spread prediction time zone window to obtain a regional soil environment prediction result; Build a channel to predict the spread of new soil pollutants; Inputting the new soil pollutant analysis results and the regional soil environment prediction results into the new soil pollutant spread prediction channel to generate the new soil pollutant spread prediction results; Build a channel to predict the spread of new soil pollutants, including: Obtaining a record set of new soil pollutant analysis results, a record set of regional soil environment, and a record set of new soil pollutant spread results; According to the new soil pollutant analysis result record set, the regional soil environment record set and the new soil pollutant spread result record set, supervised learning is performed on K new soil pollutant spread prediction learning models to generate K new soil pollutant spread prediction models that meet the new soil pollutant spread prediction accuracy, wherein K is a positive integer greater than 1; Collecting output information of the K new soil pollutant spread prediction models to obtain spread prediction integrated learning information; Using the spread prediction integrated learning information as input information and the new soil pollutant spread result record set as output information, training a new soil pollutant spread prediction integrated model that meets the new soil pollutant spread prediction accuracy; The K soil new pollutant spread prediction models are merged as parallel independent nodes to generate a soil new pollutant spread prediction processing layer; The new soil pollutant spread prediction processing layer is merged with the input layer of the new soil pollutant spread prediction integrated model to generate the new soil pollutant spread prediction channel.

2. A method for optimizing the degradation of new soil pollutants as claimed in claim 1, characterized in that: According to the target area, set the soil detection time point cloud and soil detection location point cloud, including: Obtaining a regional layout map and soil pollution monitoring level of the target area; Based on the regional layout map and the soil pollution monitoring level, setting a soil detection unit spatial grid; Gridding the target area according to the soil detection unit spatial grid to obtain a regional gridding result; Arrange soil detection points according to the regional grid division result to obtain the soil detection position point cloud; Collect pollution source information based on the soil detection location point cloud to obtain pollution source distribution information at each location; The soil detection time point cloud is constructed according to the pollution source distribution information at each location.

3. A method for optimizing the degradation of new soil pollutants as claimed in claim 2, characterized in that: According to the pollution source distribution information at each location, the soil detection time point cloud is constructed, including: According to the soil pollution monitoring level, matching the soil pollution monitoring basic computing power; Predicting soil pollution depth at each location based on the pollution source distribution information at each location to obtain soil pollution depth prediction results at each location; According to the soil pollution depth prediction results of each location, the soil pollution monitoring basic computing power is optimized respectively to obtain the soil pollution monitoring computing power of each location; The soil detection time is set according to the soil pollution monitoring computing power of each location to generate the soil detection time point cloud.

4. A method for optimizing the degradation of new soil pollutants as claimed in claim 1, characterized in that: New soil pollutant analysis is performed based on the historical neighborhood soil detection dataset and the real-time soil detection dataset to obtain new soil pollutant analysis results, including: Extracting a first real-time soil detection data sequence corresponding to a first soil detection point according to the real-time soil detection data set; Extracting a first historical neighborhood soil detection data sequence corresponding to the first soil detection point according to the historical neighborhood soil detection data set; Perform deviation analysis on the first real-time soil detection data sequence and the first historical neighborhood soil detection data sequence to obtain a first point soil detection deviation data sequence; Build a new soil pollutant detection channel; Inputting the first point soil detection deviation data sequence into the soil new pollutant detection channel to obtain the first point soil new pollutant identification result, and adding the first point soil new pollutant identification result to the soil new pollutant analysis result; According to the historical neighborhood soil detection dataset and the real-time soil detection dataset, the new soil pollutant analysis is continued in combination with the new soil pollutant detection channel to generate the new soil pollutant analysis result.

5. A method for optimizing the degradation of new soil pollutants as claimed in claim 4, characterized in that: Construct a new soil pollutant detection channel, including: Load the sample point soil detection deviation data series record and the sample point soil new pollutant identification result record; Activate the learning model for detecting new soil pollutants; Using the soil detection deviation data sequence record of the sample point as input data and the soil new pollutant identification result record of the sample point as output supervision data, the soil new pollutant detection learning model is trained, and after each predetermined number of trainings, the soil new pollutant detection loss coefficient is obtained; If the soil new pollutant detection loss coefficient is less than the predetermined soil new pollutant detection loss coefficient, the soil new pollutant detection channel is generated.

6. A method for optimizing the degradation of new soil pollutants as claimed in claim 1, characterized in that: According to the analysis results of the new soil pollutants and the prediction results of the spread of the new soil pollutants, a degradation control decision is made to obtain a soil degradation control plan, including: Based on the knowledge graph, a new soil pollutant degradation control decision model is constructed; The new soil pollutant analysis results and the new soil pollutant spread prediction results are input into the new soil pollutant degradation control decision model, and the soil degradation control plan is output.

Images