Contaminated soil remediation technology-material-equipment collaborative configuration decision method and system

Abstract

The application discloses a contaminated soil remediation technology-material-equipment collaborative configuration decision method and system, and the method comprises the following steps: constructing a remediation case database; sorting out the core and general materials and equipment of each remediation technology; screening the material and equipment combination of the target technology, determining the final technology card, calculating the unit remediation cost and environmental impact index of each technology card, and constructing a technology card library; performing cluster analysis on the contaminated site information, associating the remediation scheme after clustering to form a scene library; performing spatial block on the contaminated site information of the remediation case, and constructing a prediction model; and matching the target contaminated site with the scene library and calling the prediction model to determine the candidate remediation scheme of the target contaminated site, so as to select the optimal comprehensive remediation scheme. The application can solve the problems of lacking multi-technology collaborative analysis and not taking the materials and equipment into the engineering application decision in the existing contaminated site remediation decision method.

Description

Technical Field

[0001] This invention relates to a collaborative configuration decision-making method and system for contaminated soil remediation technology, materials, and equipment, belonging to the field of environmental protection. Background Technology

[0002] Soil is the material foundation upon which humankind depends for survival, an indispensable natural resource for economic and social development, and a fundamental guarantee for agricultural production and food security. However, among the ten major threats facing soil, soil pollution is listed as one of the major threats in the "State of World Soil Resources." Therefore, taking effective measures to prevent and control soil pollution is an essential requirement for ensuring soil health and achieving sustainable development.

[0003] The development of scientific remediation plans is a key breakthrough for achieving green and sustainable soil remediation. On the one hand, the scientific development of soil remediation plans relies on a large amount of historical remediation case data. However, existing data suffers from problems such as high dispersion and low sharing, necessitating systematic collection, integration, and key feature extraction to build a database system to support decision-making. On the other hand, due to the complexity of contaminated sites, the remediation process usually involves the combined application of multiple technologies. However, a remediation plan is not simply a combination of these technologies; it also requires quantitative analysis and optimization of the synergistic relationships between technologies. Furthermore, the same remediation technology, in practical applications, needs to be combined with appropriate materials and equipment configurations to achieve optimal remediation results.

[0004] In existing technologies, various information-based and intelligent methods have been proposed to address the decision-making problem of contaminated site remediation. For example, Chinese patent application CN107767032A discloses a decision-making system and method for heavy metal pollution in farmland soil. This system utilizes intelligent data acquisition terminals or sensors to acquire heavy metal pollution monitoring data of farmland soil from various monitoring points. It employs spatial analysis, factor analysis, and other methods to achieve coordinated graded early warning for soil and crops. Based on risk and early warning levels, it uses case reasoning algorithms to match similar cases from a case library as emergency response plans for soil. It also adjusts crop planting based on public opinion surveys, heavy metal adsorption experiments on crops, or empirical knowledge. Another Chinese patent application CN115330153A discloses a decision-making method for the remediation of heavy metal contaminated soil. This method uses feedforward neural network technology to estimate the area and volume of soil remediation and to formulate remediation technology solutions for soil pollutants. In addition, comprehensive decision-making systems still focus on recommending single technologies and relying on expert scoring; the literature "Acta Pedologica Sinica, 2012, 49(6), 1088-1094" adopts expert scoring of technical, economic and environmental indicators, and uses the Analytic Hierarchy Process (AHP) and the Top-Order Solution Approximation Method (TOPSIS) to screen contaminated site remediation technologies. In 2018, the Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, developed a WebGIS-based auxiliary decision-making platform for contaminated site remediation. This platform can comprehensively manage various information related to contaminated sites, and through online expert evaluation and scoring, select the optimal remediation technology through preliminary selection and comparison to assist users in decision-making. In recent years, with the development of artificial intelligence technology, patents have proposed contaminated site remediation decision-making methods based on intelligent algorithms or big data. For example, in Chinese patent application CN121094212A, a multi-media intelligent decision-making system for contaminated site remediation based on a large language model is proposed. This system realizes contaminated site feature analysis, remediation technology matching and scheme optimization by constructing a multi-source data input and standardization module, a site feature analysis module, a large language model knowledge reasoning engine and a dynamic matrix scoring system. It also generates remediation schemes by combining multi-objective optimization algorithms and provides a certain decision interpretation capability.Overall, existing technologies have the following potential in contaminated site remediation decision-making: 1) They can process and standardize multi-source heterogeneous data to build a data-driven decision-making foundation; 2) They can achieve contaminated site characteristic analysis and technology matching through machine learning or large language models; 3) They have a certain degree of automated decision-making capability. However, there are still some shortcomings: 1) Existing case matching methods mostly adopt the direct matching of target contaminated sites with historical cases, without systematically mining the common characteristics and patterns in the case data, and also lacking statistical analysis of the frequency of use of remediation schemes in sites with similar characteristics. Therefore, there are certain limitations in terms of pattern extraction and decision support; 2) Although current decision-making systems can generate technology combination schemes, they do not systematically model the synergistic relationship and synergy level between multiple technologies; 3) They mainly focus on the selection and matching of remediation technologies, without incorporating materials and equipment as core decision-making elements into a unified analysis framework; they lack quantitative assessment of the multi-dimensional differences of the same technology under different material and equipment combinations, and lack the ability to construct complete schemes for engineering implementation.

[0005] Therefore, it is necessary to propose a soil remediation decision-making method oriented towards engineering applications, which can comprehensively consider the synergistic relationship between remediation technologies, as well as the matching and optimization mechanism between technologies, materials, and equipment, and conduct multi-dimensional quantitative evaluation and optimization of multi-factor combination schemes, thereby realizing the scientific generation and selection of remediation schemes. Summary of the Invention

[0006] In view of this, the present invention provides a collaborative configuration decision-making method, system, computer equipment and storage medium for contaminated soil remediation technology-materials-equipment, which can solve the problems of lack of multi-technology collaborative analysis and failure to incorporate materials and equipment into engineering application decision-making in existing contaminated site remediation decision-making methods.

[0007] The first objective of this invention is to provide a collaborative configuration decision-making method for contaminated soil remediation technology, materials, and equipment.

[0008] The second objective of this invention is to provide a collaborative configuration decision system for contaminated soil remediation technology, materials, and equipment.

[0009] A third objective of this invention is to provide a computer device.

[0010] A fourth objective of this invention is to provide a storage medium.

[0011] The first objective of this invention can be achieved by adopting the following technical solution:

[0012] A collaborative configuration decision-making method for contaminated soil remediation technology, materials, and equipment, characterized in that the method includes:

[0013] Acquire information on contaminated sites, remediation technologies and related information, engineering application data of materials and equipment, and build a remediation case database;

[0014] Based on restoration cases and relevant literature, we have sorted out the materials and equipment involved in various restoration techniques and their corresponding detailed information, and identified the core and common materials and equipment for each restoration technique.

[0015] The materials and equipment combinations are screened to determine the final technical cards, the unit repair cost and environmental impact indicators of each technical card are calculated, and a technical card library is constructed.

[0016] Cluster analysis is performed on contaminated site information, and typical pollution scenarios and remediation solutions are summarized based on the clustered categories to form a scenario library;

[0017] Based on the repair cases, a feature matrix of spatial block field information and corresponding repair technologies is constructed. Through machine learning model training, a repair technology prediction model is obtained to predict spatial block repair technologies.

[0018] The target contaminated site is matched with the scenario library and the similarity is calculated. If there is a scenario with a similarity greater than or equal to a preset percentage in the matching results, the scenario with the highest similarity is selected and the set of remediation schemes corresponding to the scenario is taken as one of the candidate remediation schemes for the target contaminated site. At the same time, the remediation technology prediction model is called to determine other candidate remediation schemes for the target contaminated site.

[0019] Based on the technical card library and candidate repair solutions, the optimal comprehensive repair solution is selected.

[0020] Furthermore, the acquisition of contaminated site information, remediation technologies and related information, engineering application data of materials and equipment, and the construction of a remediation case database include:

[0021] Based on various remediation plans and risk assessment reports related to contaminated site remediation, a case database is constructed by using a large language model to extract site characteristic parameters, remediation technologies, overall remediation cost, remediation days from start to completion, and engineering application data of materials and equipment.

[0022] Furthermore, the process involves clustering the contaminated site information and summarizing typical pollution scenarios and remediation solutions based on the clustered categories to form a scenario library, including:

[0023] Based on contaminated site information, a feature matrix is ​​constructed, cluster analysis is performed on the feature matrix, and statistical analysis is conducted on the clustered categories. The mean or mode of each feature variable is used as the cluster center to summarize typical pollution scenarios and remediation solutions, thereby forming a scenario library.

[0024] Furthermore, the feature matrix of spatial block field information and corresponding repair technologies constructed based on repair cases is trained through a machine learning model to obtain a repair technology prediction model for predicting spatial block repair technologies, including:

[0025] Based on the permeability of the formation, the type of pollutant, the depth range, the medium, and the degree of pollution, the site information of the remediation case is spatially divided into blocks. The blocks are used as the smallest remediation decision units. A feature matrix is ​​constructed for each block. The feature matrix is ​​used as the training feature, and all the remediation technologies included in the remediation case are used as the training labels to train the machine learning model, thus obtaining a remediation technology prediction model based on spatial block division.

[0026] Furthermore, the invocation of the remediation technology prediction model to determine other candidate remediation schemes for the target contaminated site includes:

[0027] Based on the permeability of the formation, the type of pollutant, the depth range, the medium and the degree of pollution, the target contaminated site is divided into spatial blocks, and a remediation technology prediction framework with the block as the smallest remediation decision unit is constructed.

[0028] Based on the pollution characteristics and site conditions of each block, a variety of potential feasible remediation technologies and their corresponding recommended probabilities are obtained using remediation technology prediction models.

[0029] By utilizing the principle of technological synergy, local spatial solutions are obtained, and by utilizing the principle of spatial synergy, these local spatial solutions are integrated to further form other candidate remediation solutions for the target contaminated site.

[0030] Furthermore, the step of selecting the optimal comprehensive repair solution based on the technology card library and candidate repair solutions includes:

[0031] For each remediation technique in each candidate remediation scheme, retrieve all corresponding candidate technique cards from the technique card library;

[0032] Based on the set weights of each evaluation dimension, calculate the comprehensive score of each technical card, and select the technical card with the highest score as the optimal matching technical card.

[0033] Based on the optimal matching technology card, the index values ​​of each candidate repair scheme are calculated under each evaluation dimension;

[0034] Based on the set weights of each evaluation dimension, the multi-evaluation dimension indicators of each candidate repair scheme are weighted and summarized to obtain the overall score of each candidate repair scheme.

[0035] The overall scores of all candidate repair schemes are ranked, and the candidate repair scheme with the highest score is selected as the optimal comprehensive repair scheme.

[0036] Furthermore, the step of selecting the optimal comprehensive repair solution based on the technology card library and candidate repair solutions includes:

[0037] For each repair technology in each candidate repair scheme, obtain all corresponding candidate technology cards from the technology card library; select the one with the lowest economic cost as the optimal matching technology card; calculate the total economic cost of each candidate repair scheme based on the optimal matching technology card; select the candidate repair scheme with the lowest economic cost as the optimal comprehensive repair scheme;

[0038] Alternatively, for each remediation technology in each candidate remediation scheme, retrieve all corresponding candidate technology cards from the technology card library; select the one with the lowest environmental impact as the optimal matching technology card; calculate the total environmental impact of each candidate remediation scheme based on the optimal matching technology card; and select the candidate remediation scheme with the lowest environmental impact as the optimal comprehensive remediation scheme.

[0039] The second objective of this invention can be achieved by adopting the following technical solution:

[0040] A collaborative configuration decision-making system for contaminated soil remediation technology, materials, and equipment, the system comprising:

[0041] The remediation case database construction module is used to acquire information on contaminated sites, remediation technologies and related information, and engineering application data of materials and equipment to build a remediation case database.

[0042] The Materials and Equipment Classification Module is used to sort out the materials and equipment involved in various restoration techniques and their corresponding detailed information based on restoration cases and relevant literature, and to classify the core and common materials and equipment for each restoration technique.

[0043] The technology card library construction module is used to screen material and equipment combinations, determine the final technology cards, calculate the unit repair cost and environmental impact indicators of each technology card, and build the technology card library.

[0044] The scenario library construction module is used to perform cluster analysis on contaminated site information and summarize typical pollution scenarios and remediation solutions based on the clustered categories to form a scenario library.

[0045] The prediction model building module is used to construct a feature matrix of spatial block field information and corresponding repair technologies based on repair cases. Through machine learning model training, a repair technology prediction model is obtained to predict spatial block repair technologies.

[0046] The candidate remediation scheme determination module is used to match the target contaminated site with the scenario library and calculate the similarity. If there is a scenario with a similarity greater than or equal to a preset percentage in the matching results, the scenario with the highest similarity is selected, and the remediation scheme set corresponding to the scenario is used as one of the candidate remediation schemes for the target contaminated site. At the same time, the remediation technology prediction model is called to determine other candidate remediation schemes for the target contaminated site.

[0047] The comprehensive repair solution selection module is used to select the optimal comprehensive repair solution based on the technical card library and candidate repair solutions.

[0048] The third objective of this invention can be achieved by adopting the following technical solution:

[0049] A computer device includes a processor and a memory for storing a processor-executable program, wherein when the processor executes the program stored in the memory, it implements the above-described collaborative configuration decision-making method for contaminated soil remediation technology-materials-equipment.

[0050] The fourth objective of this invention can be achieved by adopting the following technical solution:

[0051] A storage medium storing a program that, when executed by a processor, implements the aforementioned collaborative configuration decision-making method for contaminated soil remediation technology, materials, and equipment.

[0052] The present invention has the following advantages over the prior art:

[0053] This invention aims at green and sustainable soil remediation. Focusing on the issues of multi-technology synergy and optimized material and equipment configuration, it constructs an integrated decision-making framework of "technology-materials-equipment." Through structured extraction and cluster analysis of multi-source remediation case data, a typical pollution scenario library is formed. The nearest neighbor algorithm is used to match scenarios, and based on the frequency statistics of technology combinations associated with scenarios, high-maturity technology combinations are selected. Simultaneously, machine learning models are combined to spatially segment complex sites and predict remediation technologies, and a technology synergy mechanism is introduced to qualitatively analyze the synergistic relationships of multiple technologies. Furthermore, based on the technology-materials-equipment screening mechanism, a technology card library of material and equipment combinations corresponding to each technology is obtained, and the economic costs and environmental impacts of different combinations are quantitatively evaluated. This enables the scientific generation and optimization of remediation solutions that include materials and equipment, improving the scientific rigor, systematic nature, and engineering feasibility of remediation solutions, and providing strong support for green and sustainable soil remediation. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0055] Figure 1 This is a technical architecture diagram of the collaborative configuration decision-making method for contaminated soil remediation technology, materials, and equipment in Embodiment 1 of the present invention.

[0056] Figure 2 This is a flowchart of the collaborative configuration decision-making method for contaminated soil remediation technology, materials, and equipment in Embodiment 1 of the present invention.

[0057] Figure 3 This is a flowchart of the material screening process in Embodiment 1 of the present invention.

[0058] Figure 4 This is a flowchart of the equipment selection process in Embodiment 1 of the present invention.

[0059] Figure 5 This is a scene library matching diagram after cluster analysis in Embodiment 1 of the present invention.

[0060] Figure 6 This is a flowchart illustrating the construction process of the repair technology prediction model in Embodiment 1 of the present invention.

[0061] Figure 7 This is a flowchart illustrating the implementation of the intelligent inference and repair scheme in Embodiment 1 of the present invention.

[0062] Figure 8 This is a diagram of the input interface for basic information of the target contaminated site in Embodiment 2 of the present invention.

[0063] Figure 9 This is a structural block diagram of the collaborative configuration decision system for contaminated soil remediation technology, materials, and equipment in Embodiment 4 of the present invention.

[0064] Figure 10 This is a structural block diagram of the computer device according to Embodiment 5 of the present invention. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] Example 1:

[0067] like Figure 1 and Figure 2 As shown in the figure, this embodiment provides a collaborative configuration decision-making method for contaminated soil remediation technology, materials, and equipment. The method includes the following steps:

[0068] S201. Obtain information on contaminated sites, remediation technologies and related information, and engineering application data of materials and equipment to construct a remediation case database.

[0069] To ensure the scientific validity and practical applicability of the remediation plan, this embodiment first downloads various remediation plans, risk assessment reports, and other related reports. Domestic remediation cases are retrieved from the latest list of construction land soil pollution risk control and remediation projects and the latest removal list. International remediation cases are primarily sourced from the US EPA website. Engineering application data mainly comes from: 1) Past key site remediation R&D programs: including major remediation projects over the years, covering examples and effectiveness of technology applications; 2) Literature and patents: obtaining the latest technological advancements and application cases from research papers and patents in related fields; 3) Remediation company cases: extracting experience from actual application cases in the industry and obtaining feedback on technology applications; 4) Pollution control engineering databases and publicly available government data: utilizing publicly available data released by the government and enterprises to ensure the reliability and comprehensiveness of the data; and 5) International remediation cases.

[0070] Based on various remediation plans and risk assessment reports, this embodiment utilizes a large language model (such as ChatGPT) to extract contaminated site information, remediation technologies and related information, and engineering application data of materials and equipment. Contaminated site information may include hydrogeological parameters of the contaminated site, and remediation-related information may include overall remediation costs and the number of days from the start of remediation to achieving the required standards. These parameters specifically include: 1) Hydrogeological conditions: collecting key hydrogeological parameters such as aquifer characteristics, stratigraphic structure, and corresponding depths; 2) Pollution characteristics: primarily collecting information on pollutant types, concentration distribution, and distribution depth; 3) Remediation targets: extracting the remediation target values ​​set in each remediation report; 4) Land use patterns: extracting the planned land use after site remediation to ensure that the long-term use of the remediated site meets planning requirements; 5) Remediation technologies, materials, and equipment: collecting the remediation technologies mentioned in the remediation plan, the names of the materials and equipment used, and the applicable temperature, pH, and other conditions for the materials, as well as the specifications and performance parameters of the equipment; 6) Remediation technology data includes key parameters such as the scope of application of the technology, treatment effects, operating conditions, economic costs, and remediation efficiency.

[0071] S202. Based on restoration cases and relevant literature, sort out the materials and equipment involved in various restoration techniques and their corresponding detailed information, and identify the core and common materials and equipment for each restoration technique.

[0072] This embodiment, based on the remediation cases collected in S201, summarizes and organizes the remediation technologies involved in these cases. For each remediation technology, the materials and equipment involved in its actual engineering application are reviewed one by one, and categorized into core materials and equipment and general materials and equipment according to their role in the implementation of the technology. Core materials and equipment refer to materials or equipment that directly participate in the removal, transformation, or stabilization of pollutants in a specific remediation technology and play a decisive role in the remediation effect. General materials and equipment refer to materials or equipment that are widely used in various remediation technologies and mainly play an auxiliary or supporting role. The technology-materials-equipment database contains six types of core data tables: technology-functional unit-equipment type / equipment system table, equipment system-specific equipment combination table, specific equipment information table, technology-functional unit-material type combination table, material type combination-specific material combination table, and specific material information table, which are explained in detail below:

[0073] (1) Technology-Functional Unit-Equipment Type / Equipment System Table: This table is used to establish the correspondence between the repair technology and the functional units in its implementation process, and further associate the equipment types or equipment systems required to realize the functional unit. A functional unit refers to the basic operational steps required to realize a certain repair technology process, such as excavation, screening, reagent addition reaction, and waste gas treatment. Since some functional units can be completed by a single type of equipment, while some functional units require multiple equipment to work together, both equipment types and equipment systems are recorded in this table. When a functional unit can be realized by a single equipment type, only the equipment type is recorded; when a functional unit requires multiple equipment to work together, it is recorded as an equipment system.

[0074] (2) Equipment System - Equipment Type Combination Table: This table records the specific equipment type combination relationships contained in the equipment system. For an equipment system, multiple devices need to work together to realize functional units. For example, a sewage treatment system may include wastewater collection equipment, wastewater reaction equipment, etc. This table describes the correspondence between the equipment system and its internal equipment type composition, thereby realizing the standardized expression of the system equipment structure.

[0075] (3) Specific Equipment Information Table: This table stores the attribute information of specific equipment, including equipment type, equipment name, processing capacity, applicable environmental conditions, equipment cost, operating efficiency, and environmental impact. When the system calls up equipment combinations, it can further obtain specific equipment information through this table and filter according to environmental conditions or user needs.

[0076] (4) Technology-Functional Unit-Material Type Combination Table: This table is used to establish the correspondence between repair technologies, functional units, and material types. Material type refers to the category of materials required in a certain functional unit, such as oxidants, reducing agents, carbon sources, stabilizers, or adsorbents. Since some functional units require the synergistic effect of multiple materials, this table records the material type combination information corresponding to the functional unit.

[0077] (5) Material Type Combination - Specific Material Combination Table: This table records the correspondence between specific material combinations within a material type combination. Although there are combination relationships between different material types, not all materials of the same type can be used in any combination. Therefore, a dedicated material combination table is needed to record the specific material combinations that can be practically applied. For example, a material type combination might be "carbon source + reducing agent," and its specific combinations might include "lactate + zero-valent iron" or "starch + zero-valent iron," etc. This table ensures that the database only calls material combinations that have been verified through engineering practice.

[0078] (6) Specific Material Information Table: This table stores the attribute information of specific materials, including material name, material type, applicable pollutant type, applicable environmental conditions (such as pH, soil type or groundwater conditions), material cost, environmental impact and remediation efficiency. After the system determines the material combination, it can further filter specific materials that meet specific environmental conditions through this table.

[0079] S203. Screen the combination of materials and equipment, determine the final technical cards, calculate the unit repair cost and environmental impact indicators of each technical card, and build a technical card library.

[0080] In this embodiment, the combination of remediation technology materials and equipment is optimized based on the differentiated conditions of the contaminated site and other variable requirements, and a standardized remediation technology application card is constructed, such as... Figure 3 and Figure 4 As shown, during the database call process, the system first queries the technology-functional unit table based on the input remediation technology name to obtain the corresponding functional unit information. Then, based on the functional unit, it queries the equipment system table and material type combination table to obtain the equipment system and material type combination required to implement that functional unit. For the equipment part, the system further obtains the equipment combination through the equipment system-equipment type combination table and filters specific equipment that meets the environmental conditions through the specific equipment information table. For the materials part, the system obtains the specific material combination through the material type combination-specific material combination table and filters materials suitable for specific environmental conditions through the specific material information table. Finally, it outputs a material and equipment configuration scheme suitable for the target contaminated site, i.e., a technology card. Each technology card contains the following: remediation technology name, combination scheme of core materials and general materials, configuration scheme of core equipment and general equipment, etc.

[0081] This embodiment is based on technical cards and calculates the unit repair cost corresponding to each technical card. The unit repair cost consists of material cost, equipment cost, and costs excluding material and equipment costs. By analyzing multiple cases that adopt the same repair technology, the unit volume repair cost after deducting material and equipment costs is calculated, and the average value is taken to obtain the average unit volume water and soil cost of repair excluding material and equipment costs.

[0082] Material costs are calculated by multiplying the unit material usage by its unit cost. When the unit material usage is unknown, the average material usage for each application case using the same technology is calculated. With ample material usage data, the relationship between the volume of soil and water restored and the material usage can be explored. A function fitting method can be used to obtain a calculation function for material usage with respect to the volume of soil and water restored, thus revealing the trend of material usage as the volume of soil and water restored increases. This function can then be used to calculate the material usage for the corresponding volume of soil and water restored, thereby obtaining the material cost for restoring that volume. Equipment costs are calculated based on the equipment's rental or purchase price, service life, and actual usage time per unit restoration scale. After summing all costs, the unit restoration cost for the corresponding technology application card is obtained.

[0083] The environmental impact of the repair technology application process only considers the environmental impact of material consumption and energy consumption during equipment operation, excluding the environmental impact of the equipment production process itself. For the environmental impact at the material level, the system employs a Life Cycle Assessment (LCA) method, specifically using SimaPro software and the ReCiPe2016 model to calculate the midpoint and endpoint environmental impact indicators for materials under unit usage conditions. For special materials not covered by the Ecoinvent database in SimaPro software, the system identifies and extracts the core chemical components, obtains the environmental impact data for each core chemical component, and calculates the unit environmental impact indicator for the special material based on the proportion of each component in the material, thereby estimating the environmental impact of this type of material. For the environmental impact at the equipment level, the system first calculates the total energy consumption of the equipment under a unit repair scale based on the equipment's rated power and the cumulative operating time corresponding to a unit workload. Subsequently, based on the environmental impact factor of unit energy consumption, the system calculates the environmental impact value corresponding to the equipment's energy consumption, thereby achieving a quantitative evaluation of the equipment's environmental contribution.

[0084] By calculating the unit remediation cost and environmental impact indicators of each technical card, a technical card library is ultimately constructed, extending from the remediation technology level to the specific materials and equipment level. Each technical card in the library contains a corresponding material and equipment combination scheme, unit remediation cost, and environmental impact indicators. This technical card library can provide quantitative and comparable technical support for the comparison, optimization, and decision-making of subsequent contaminated site remediation schemes.

[0085] S204. Perform cluster analysis on contaminated site information, and summarize typical pollution scenarios and remediation solutions based on the clustered categories to form a scenario library.

[0086] This embodiment constructs a feature matrix for contaminated site clustering analysis based on the contaminated site information extracted in S201. The feature matrix includes at least pollutant characteristic parameters, site characteristic parameters, and the remediation technologies used. The remediation technologies used do not participate in clustering but are only used for subsequent association with the clustering results. Among them, the pollutant characteristic parameters include the types of different types of pollutants and their corresponding pollution concentration information. The pollutant types include at least heavy metals, halogenated hydrocarbons, petroleum hydrocarbons, etc., and the distribution depth of pollutants. The site characteristic parameters include the spatial location of the contaminated medium (vadose zone or saturation zone) and the soil structure type. The soil structure type includes at least sand / silt, silty clay, and clay.

[0087] For the above-mentioned different types of variable parameters, continuous variables and categorical variables are subjected to unified structural processing to construct a site feature vector suitable for distance calculation of mixed variables.

[0088] Next, based on the feature matrix, five clustering algorithms were compared, including K-Medoids based on Gower distance, DBSCAN, Hierarchical clustering, as well as K-Prototypes and K-Modes algorithms, to perform cluster analysis on the actual site to identify typical contaminated site types. The clustering effects of different algorithms were systematically evaluated using indicators such as the profile coefficient, and the K-Medoids clustering algorithm based on Gower distance was finally confirmed to perform the best.

[0089] Then, statistical analysis is performed on the characteristic parameters of contaminated sites within each cluster category. By averaging continuous variables and calculating the mode of categorical variables, the cluster centers corresponding to each cluster category are determined. These cluster centers, as representative features of the contaminated sites in that category, characterize a typical contamination scenario. Through this method, typical scenarios of contaminated sites are summarized, forming a set of typical contamination scenarios composed of multiple cluster centers, i.e., a scenario library. Figure 5 As shown.

[0090] The scenario library in this embodiment includes not only site feature information corresponding to each typical pollution scenario, but also multiple remediation technology solutions associated with that typical scenario. Specifically, the scenario library is defined as the association mapping relationship from the feature space of typical polluted sites to the set of remediation technology solutions. For any typical pollution scenario s represented by a cluster center in the scenario library, the mapping relationship is f: s → {t1, t2, ……, t n The typical pollution scenario is associated with multiple feasible technical remediation solutions. Based on historical remediation cases belonging to the same cluster category as the typical pollution scenario, corresponding remediation technical solutions are extracted and statistically summarized. When multiple remediation technical solutions exist within the same cluster category, the remediation technical solutions with higher frequency and stronger engineering feasibility are selected as the candidate remediation solution set for that typical pollution scenario. This achieves the association of multiple technically available remediation solutions for a single typical scenario. Through this method, the applicability problem caused by recommending a single solution is avoided, and the flexibility and engineering adaptability of the remediation solution recommendation results are improved.

[0091] S205. Based on the repair cases, construct a feature matrix of spatial block field information and corresponding repair technologies, and train it through a machine learning model to obtain a repair technology prediction model for predicting spatial block repair technologies.

[0092] This embodiment spatially segments the site information of remediation cases. Based on the spatial structure of the contaminated site, the site is divided into several blocks with relatively consistent characteristics. The segmentation is based on information such as formation permeability, pollutant type, depth range, medium, and degree of contamination. Each block serves as the smallest remediation decision unit. A feature matrix is ​​constructed for each block, and this feature matrix is ​​used as training features. All remediation technologies included in the remediation cases are used as training labels to construct a dataset. A machine learning model is then trained to obtain a remediation technology prediction model based on spatial segmentation. The machine learning model used includes at least one of the random forest algorithm and the gradient boosting-based XGBoost algorithm. Through learning from historical case data, the model can establish a mapping relationship between spatial block features and remediation technology selection. The model construction process is as follows: Figure 6 As shown.

[0093] Because the information density varies across different cases, and the presence of missing values ​​can severely impact the model's training performance, the following adjustments and optimizations were made to the training features in the dataset:

[0094] 1) Formation permeability: The main basis is the permeability coefficient k (m / s) of each formation mentioned in the report text. Low permeability is 10-8~10-6 m / s, medium permeability is 10-6~10-4 m / s, high permeability is 104~10-2 m / s, and extremely high permeability is >10-2 m / s.

[0095] 2) Pollutant types: Since there are thousands of pollutants, it is necessary to classify them roughly. Referring to the classification in USEPA and remediation cases, the pollutants are finally identified as follows: petroleum hydrocarbons, halogenated organic compounds, benzene series compounds, polycyclic aromatic hydrocarbons, organic pesticides, other organic compounds, heavy metals and non-heavy metal inorganic compounds. When performing machine learning, the input variables are the major pollutant categories and the names of the dominant pollutants.

[0096] 3) Depth range: Since the range value is not suitable as an input variable for machine learning, the upper and lower limits of the range are used as training features. For example, if the depth range is 0~5m, the input values ​​are 0 and 5 respectively.

[0097] 4) Medium: namely, the vadose zone / saturation zone, which is mainly determined by the groundwater level. If the polluted area is above the groundwater level, it is the vadose zone; otherwise, it is the saturation zone.

[0098] 5) Pollution level: Used to determine the remediation priority of different blocks. Since the remediation target values, risk screening values, and control values ​​of different pollutants are different, the exceedance multiple is used as the standard for judging the pollution level. According to the collected case data, an exceedance multiple of 1-2 indicates low pollution, 2-5 indicates medium pollution, 5-25 indicates high pollution, and >25 indicates extremely high pollution.

[0099] It's worth noting that the technical decisions in real-world engineering cases involve comparing and selecting from technologies after the initial screening. This means that there is often more than one technology suitable for the site, and therefore the prediction model will ultimately output more than one technology label; currently, it is set to output three technology labels. For model performance validation, five-layer cross-validation is used, a resampling technique commonly used in model evaluation and selection. The process is as follows:

[0100] 1) Data partitioning: The entire dataset is randomly and uniformly divided into five mutually exclusive subsets, each of which is called a "fold". During the partitioning process, a stratified sampling strategy is usually used to keep the distribution of samples in each fold consistent with the original dataset.

[0101] 2) Iterative training and validation: Each fold is used as the validation set, and the remaining four folds are used as the training set to build the model and make predictions. This process is repeated five times, each time using a different validation fold.

[0102] 3) Performance summary: The evaluation metrics (such as accuracy, mean squared error, etc.) obtained from the five validations are statistically analyzed. Usually, the mean is taken as the estimated value of the final performance of the model. At the same time, the standard deviation can be calculated to evaluate the stability of the performance.

[0103] Based on the above steps, different models are trained, focusing primarily on performance metrics such as accuracy, F1 score, and recall. The performance metrics of different models are compared, and the model with the best performance is selected as the final prediction model, which can be used to recommend feasible remediation technologies for land parcels with given pollution characteristics.

[0104] S206. Match the target contaminated site with the scenario library and calculate the similarity. If there is a scenario in the matching result with a similarity greater than or equal to a preset percentage, select the scenario with the highest similarity and take the set of remediation schemes corresponding to the scenario as one of the candidate remediation schemes for the target contaminated site. At the same time, call the remediation technology prediction model to determine other candidate remediation schemes for the target contaminated site.

[0105] 1) Analogy matching method based on typical pollution scenario library: The preset percentage in this embodiment is 85%. When there are scenarios with similarity greater than or equal to 85% in the matching results, the scenario with the highest similarity is selected, and its corresponding remediation scheme set is used as one of the candidate remediation schemes for the target site; at the same time, the remediation technology prediction model is called to determine other candidate remediation schemes for the target contaminated site.

[0106] In this embodiment, the matching of the target contaminated site with typical scenarios in the scenario library also adopts the k-Nearest Neighbor (k-NN) algorithm and the analytic hierarchy process (AHP). KNN usually assigns different weights to different features and measures similarity by calculating the distance between features (such as Euclidean distance, Manhattan distance, or cosine similarity). Among them, the analytic hierarchy process is used to determine the weight of each site feature index.

[0107] 2) Intelligent Inference of Remediation Schemes: In this embodiment, the remediation technology prediction model is invoked to determine candidate remediation schemes for the target contaminated site. Specifically, based on the formation permeability, pollutant type, depth range, medium, and degree of contamination, the target contaminated site is divided into spatial blocks. The spatial block division of the target contaminated site is the same as in step S205. A remediation technology prediction framework with blocks as the smallest remediation decision unit is constructed. For each block's contamination characteristics and site conditions, the remediation technology prediction model is used to obtain multiple potential feasible remediation technologies and their corresponding recommendation probabilities. Specifically, a corresponding feature vector is constructed for each block and input into the trained remediation technology prediction model to obtain the prediction results of one or more remediation technologies corresponding to that block. This achieves remediation technology prediction with spatial blocks as the smallest decision unit, thereby supporting the decision to adopt differentiated remediation technologies for different blocks in the same contaminated site. Based on this, the principle of technology synergy is used to obtain local spatial schemes, and the principle of spatial synergy is used to integrate the local spatial schemes to further form other candidate remediation schemes for the target contaminated site, such as... Figure 7As shown, this method can effectively overcome the decision-making limitations caused by the lack of scenarios highly similar to the target contaminated site in the typical scenario library. By expanding the candidate remediation technology space, it significantly improves the diversity and selectivity of remediation scheme combinations, thereby avoiding local optimal selection in a limited and suboptimal set of technical solutions, and providing more globally optimal decision support for the realization of the whole site remediation goal.

[0108] The combination of the two methods described above is used to make technically feasible repair solutions. It can not only refer to actual repair solutions, but also expand the range of feasible repair solutions for subsequent selection of the optimal repair solution.

[0109] The specific details regarding technological and spatial synergy are as follows:

[0110] 1) Technological Synergy: After obtaining a set of candidate remediation technologies at the block level, the concept of technological synergy is introduced to optimize the combination of multiple remediation technologies within different spatial blocks. Technological synergy refers to the complementary or synergistic relationship between different remediation technologies in terms of their mechanism of action, target objects, or engineering implementation stages for different blocks. Through reasonable combination, effective removal of pollutants in a unified block can be achieved. The determination of technological synergy is based on collected engineering case studies. These case studies primarily focus on the synergy between remediation efficiency and engineering measures under the condition of meeting remediation objectives. Common synergistic scenarios include:

[0111] a) Series: This refers to the sequential use of technology A and technology B. There are usually three scenarios: First, in cases of combined pollution of heavy metals and organic matter, technology A needs to treat the organic matter first, and then technology B can treat the heavy metals. Second, in cases of single high-concentration pollution, technology A needs to quickly reduce the pollution load, and then technology B is responsible for the subsequent residual pollution. Third, when pollutants exist in hard-to-reach areas, technology A is used to change the physical structure of the medium or the state of the pollutants to create a mass transfer channel for technology B, such as thermal desorption + gas phase extraction.

[0112] b) Parallel operation: This means that technology A and technology B are implemented simultaneously. There are usually three scenarios: First, different regions have different types of pollutants, so technology A treats one type of pollutant while technology B treats another. Second, different regions have different levels of pollution. In high-concentration scenarios, technology A is mostly used for remediation, while in low-concentration scenarios, technology B is mostly used for risk management. Third, the synergy between active and passive remediation. A common technology combination is extraction treatment + monitoring of natural decay.

[0113] In remediation technology decision-making, technological synergy is not simply the superposition of multiple technologies, but a comprehensive match between engineering adaptability and operational mechanisms. By summarizing and generalizing the technological synergy mechanisms in engineering cases, a set of technological synergy rules is constructed. Based on these rules, a set of block-level candidate remediation technologies is screened and combined to form several block-level remediation technology solutions. Each solution can further calculate its comprehensive applicability score for subsequent ranking and selection.

[0114] 2) Spatial Synergy: After obtaining the block-level remediation technology solutions for each spatial area, the principle of spatial synergy is introduced to integrate multiple block solutions into a comprehensive remediation technology solution for the entire contaminated site. Spatial synergy mainly refers to the judgment made through several hard constraint modules, including whether the site is in operation, the status of underground pipelines, the existence of sensitive land use in the surrounding area, the spatial adjacency of blocks, and the direction of groundwater flow. These constraints are used to determine the spatial coordination and compatibility of different block remediation technology solutions, avoiding the failure of the overall remediation solution due to construction conflicts, migration control imbalances, or superimposed environmental disturbances, thus providing a basis for forming a comprehensive technology solution for the entire contaminated site.

[0115] For example, if the contaminated site is in production, then ex-situ remediation technology should be ruled out first, and related technical solutions should not be included in the final decision-making process; if the underground pipelines are very complex, then the priority of some in-situ remediation technologies that require large-area well placement should be reduced, and technologies with fewer well points and less interference should be given priority.

[0116] Based on the above constraints, the remediation technology solutions for each block are coordinated, merged, or adjusted at the spatial level to ultimately form a comprehensive remediation technology solution covering the entire site.

[0117] In this way, even when there is a lack of scenarios highly similar to the target contaminated site in the typical scenario library, the system can still expand the space of candidate remediation technologies through model prediction.

[0118] S207. Based on the technical card library and candidate repair solutions, select the optimal comprehensive repair solution.

[0119] In this embodiment, the repair schemes are initially screened based on the obtained multiple feasible repair schemes; when the repair scheme sets recommended by the two methods are consistent, the repair technology is given priority; when there are differences, multiple repair technologies can be output as candidate schemes for further technology screening and optimization decisions.

[0120] In this embodiment, based on the constructed technology card library, the corresponding material and equipment combinations are determined for each remediation technology scheme that has passed the initial screening. Since the same remediation technology may correspond to multiple different material and equipment combinations in the technology card library, and different material and equipment combinations differ in terms of economic cost and environmental impact, the system needs to match an applicable technology card for each remediation technology scheme before comparing the remediation schemes. According to the basic parameters of the target contaminated site, material and equipment combination cards that match the conditions of the target site are selected from the technology card library as the candidate technology card set for the remediation technology scheme.

[0121] For multiple candidate technology cards corresponding to the same remediation technology solution, screening is conducted from the dimensions of economic cost and environmental impact. For example, the unit remediation cost corresponding to each candidate technology card is compared, and the material and equipment combination card with the best economic cost is selected as the representative technology card of the remediation technology in terms of economics; based on the environmental impact indicators recorded in the technology cards, the environmental impact of different material and equipment combinations is compared, and the material and equipment combination card with the least environmental impact is selected as the representative technology card of the remediation technology in terms of environmental friendliness.

[0122] In this embodiment, the remediation schemes that have passed the initial screening are ranked based on multi-dimensional evaluation indicators such as economic cost and environmental impact.

[0123] After completing the initial technical screening of each candidate repair scheme, assuming that several alternative repair schemes (e.g., three repair schemes) are obtained, and each repair scheme includes one or more repair technologies, the following evaluation and selection mechanisms are constructed for different decision preferences:

[0124] 1) Economic Priority Decision-Making Mechanism: The economic priority decision-making mechanism sets the weight of economic cost to 1 and the weight of other dimension indicators to 0; for each candidate solution, for each repair technology it contains, the card with the best economic efficiency is matched from the technical material equipment card library; based on the cost parameters of the matched card, the overall economic cost of the solution is calculated; finally, the total cost of all candidate solutions is compared, and the solution with the lowest economic cost is selected as the optimal solution.

[0125] 2) Environmental Priority Decision-Making Mechanism: Similar to the economic priority mechanism, the environmental priority decision-making mechanism sets the weight of environmental impact to 1 and the weight of other dimensions to 0; for each candidate scheme, the material and equipment card with the best environmental performance is matched for each remediation technology; the environmental impact index of each scheme is calculated based on the matching results; by comparing the environmental indicators of each scheme, the scheme with the least environmental impact is selected as the optimal scheme.

[0126] 3) Multi-dimensional Comprehensive Decision-Making Mechanism: In comprehensive decision-making scenarios, a multi-dimensional evaluation weighting system is introduced. The system defaults to equal weighting for each evaluation dimension, but users can also customize the weighting according to specific site restoration needs. The specific steps are as follows:

[0127] 3.1) For each remediation technique in each candidate remediation scheme, retrieve all corresponding candidate technique cards from the technique card library;

[0128] 3.2) Calculate the comprehensive score of each technical card according to the set weights of each evaluation dimension, and select the technical card with the highest score as the optimal matching technical card;

[0129] 3.3) Based on the optimal matching technology card, calculate the index value of each candidate repair scheme under each evaluation dimension;

[0130] 3.4) Based on the set weights of each evaluation dimension, the multi-evaluation dimension indicators of each candidate repair scheme are weighted and summarized to obtain the overall score of each candidate repair scheme;

[0131] 3.5) Sort the overall scores of all candidate repair schemes and select the candidate repair scheme with the highest score as the optimal comprehensive repair scheme.

[0132] It should be noted that although the method operations of the above embodiments are described in a specific order, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. On the contrary, the described steps may be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0133] Example 2:

[0134] The target contaminated site involved in this embodiment is located in a former chemical plant area of ​​a coastal industrial city, and is intended for redevelopment into residential land. Based on preliminary site investigation and risk assessment, the following basic information about the site was obtained:

[0135] 1. Polluting media: soil and groundwater.

[0136] 2. Main pollutant types: Petroleum hydrocarbons (C10–C40) and benzene series compounds are the main pollutants in soil and groundwater; some areas are accompanied by certain concentrations of heavy metals (mainly nickel and lead).

[0137] 3. Pollution distribution characteristics: Pollution is mainly concentrated in the range of 0-6 m below the surface; 0-3 m is the vadose zone, and 3-6 m is the saturation zone.

[0138] 4. Soil structure and corresponding pollutants: 0-2 m: sand / silt; petroleum hydrocarbons, heavy metals; 2-5 m: silty clay; petroleum hydrocarbons, benzene compounds; 5-6 m: clay; petroleum hydrocarbons, benzene compounds.

[0139] Users input information on the input page based on the above information, such as... Figure 8 As shown in Table 1, the top five scenarios with a similarity greater than 85% based on cluster similarity matching are listed below.

[0140] Table 1. Top five scenes with a similarity greater than 85%

[0141]

[0142] Therefore, the scheme with the highest similarity was selected as the recommended scheme for the target site, which utilizes ex-situ chemical oxidation + ex-situ solidification / stabilization + extraction treatment technologies. From a comprehensive perspective, the materials and equipment information required for each technology was obtained after screening based on the technical materials and equipment database, and the results are shown in Table 2 below.

[0143] Table 2. Information on materials and equipment required for each technology

[0144]

[0145] Example 3:

[0146] The contaminated site involved in this embodiment is an operating chemical industrial park. Due to historical production activities, it has developed soil-groundwater complex pollution, mainly composed of benzene series compounds and petroleum hydrocarbons. The pollution exhibits clear spatial zoning characteristics: the source area has a high degree of contamination in the vadose zone, the migration corridor area is mainly contaminated in the saturated zone, and the site boundary area has low-concentration residual pollution. Simultaneously, the site faces constraints such as being in operation, complex underground pipelines, and surrounding sensitive land use, making remediation conditions quite complex, as detailed below:

[0147] 1. The input condition is the site pollution status, as shown in Table 3 below.

[0148] Table 3 Site Contamination Status

[0149]

[0150] 2. The prediction model output results are shown in Table 4 below. There are a total of 3×3×3=27 technology combinations.

[0151] Table 4. Prediction Model Output Results

[0152]

[0153] The reasonable combinations after screening for technological synergy are not described in detail here, but only the theoretically valid combinations of technologies and their synergistic logic are given.

[0154] Combination 1: Thermal desorption + PRB + risk management, logic: active remediation in high concentration areas + migration channel blockage + risk control in low concentration areas;

[0155] Combination 2: Thermal desorption + extraction treatment + MNA, logic: rapid reduction of source area + active control of groundwater + long-term support of boundary area;

[0156] Combination 3: SVE + PRB + Risk Management, Logic: Volatile pollutant source area remediation + migration channel interception + boundary area low-disturbance control;

[0157] Combination 4: Thermal desorption + PRB + MNA, logic: enhanced repair in high concentration areas + control of migration channels + natural attenuation in boundary areas.

[0158] 4. Spatial synergy is further screened, and the advantages and disadvantages of the above feasible combinations are further compared. The key hard constraints in this case are as follows:

[0159] Z1 in production: Large-scale, high-disturbance technology is not prioritized, therefore in-situ thermal desorption < SVE;

[0160] Z2 has complex underground pipelines: technologies that rely on dense well placement or injection are not preferred, therefore extraction treatment < PRB;

[0161] Z3 is near a school: Prioritize low-noise, low-odor, and low-disturbance technologies; risk management is approximately MNA.

[0162] In summary, the final selection result is: combination 3 > combination 1 ≈ combination 4 > combination 2.

[0163] 5. Final solution generation

[0164] After the final technology combination is determined, the database further outputs the material and equipment information corresponding to each technology to form a complete repair plan, as shown in Table 5.

[0165] Table 5 Complete Repair Plan

[0166]

[0167] The implementation logic for each block is Z1→(Z2∥Z3), that is, Z1 prioritizes the implementation of SVE, while Z2 and Z3 simultaneously implement PRB and risk management based on site conditions.

[0168] Previous studies have shown that SVE has a high removal efficiency for benzene series / light hydrocarbon pollution, with an overall extraction efficiency exceeding 92% in sandy soil systems, and the remediation time for benzene can be controlled within hours (Albergaria et al., Journal of Environmental Management, 2012, DOI: 10.1016 / j.jenvman.2012.03.033). PRB also has a clear interception and reduction effect on benzene in groundwater. In the field case, the benzene concentration in the reactive barrier was consistently below 0.04 µg / L, and the downstream concentration was also significantly lower than the upstream concentration. At the same time, it can delay the response time of pollutant migration (Gao et al., Water, 2021, DOI: 10.3390 / w13233336).

[0169] Example 4:

[0170] like Figure 9 As shown, this embodiment provides a collaborative configuration decision-making system for contaminated soil remediation technology, materials, and equipment. The system includes a remediation case database construction module 901, a materials and equipment classification module 902, a technology card library construction module 903, a scenario library construction module 904, a prediction model construction module 905, a candidate remediation scheme determination module 906, and a comprehensive remediation scheme selection module 907. Specific descriptions of each module are as follows:

[0171] The remediation case database construction module 901 is used to acquire information on contaminated sites, remediation technologies and related information, and engineering application data of materials and equipment to construct a remediation case database.

[0172] The Materials and Equipment Classification Module 902 is used to sort out the materials and equipment involved in various repair techniques and their corresponding detailed information based on repair cases and relevant literature, and to classify the core and common materials and equipment for each repair technique.

[0173] The technology card library construction module 903 is used to screen material and equipment combinations, determine the final technology cards, calculate the unit repair cost and environmental impact indicators of each technology card, and construct the technology card library.

[0174] The scenario library construction module 904 is used to perform cluster analysis on contaminated site information and summarize typical pollution scenarios and remediation solutions based on the clustered categories to form a scenario library.

[0175] The prediction model building module 905 is used to build a feature matrix of spatial block field information and corresponding repair technologies based on repair cases, and to obtain a repair technology prediction model for predicting spatial block repair technologies through machine learning model training.

[0176] The candidate remediation scheme determination module 906 is used to match the target contaminated site with the scenario library and calculate the similarity. If there is a scenario with a similarity greater than or equal to a preset percentage in the matching results, the scenario with the highest similarity is selected, and the remediation scheme set corresponding to the scenario is used as one of the candidate remediation schemes for the target contaminated site. At the same time, the remediation technology prediction model is called to determine other candidate remediation schemes for the target contaminated site.

[0177] The comprehensive repair solution selection module 907 is used to select the optimal comprehensive repair solution based on the technical card library and candidate repair solutions.

[0178] The specific implementation of each module in this embodiment can be found in Embodiment 1 above, and will not be repeated here. It should be noted that the system provided in this embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure can be divided into different functional modules to complete all or part of the functions described above.

[0179] Example 5:

[0180] This embodiment provides a computer device, such as... Figure 10 As shown, it includes a processor 1002, a memory, and a network interface 1003 connected via a system bus 1001. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium 1004 and internal memory 1005. The non-volatile storage medium 1004 stores an operating system, computer programs, and a database. The internal memory 1005 provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. When the processor 1002 executes the computer programs stored in the memory, it implements the contaminated soil remediation technology-material-equipment collaborative configuration decision of Embodiment 1 above, as follows:

[0181] This process involves acquiring information on contaminated sites, remediation technologies and related information, and engineering application data of materials and equipment to construct a remediation case database. Based on these cases and relevant literature, the process identifies the materials and equipment involved in various remediation technologies, along with their detailed information, and categorizes the core and common materials and equipment for each technology. Material and equipment combinations are then screened to determine the final technology cards. The unit remediation cost and environmental impact indicators for each technology card are calculated, and a technology card library is constructed. Contaminated site information is clustered, and typical pollution scenarios and remediation schemes are summarized based on the clustered categories to form a scenario library. A feature matrix of spatial block site information and corresponding remediation technologies is constructed based on the remediation cases. A remediation technology prediction model is trained using a machine learning model to predict spatial block remediation technologies. The target contaminated site is matched with the scenario library, and similarity is calculated. If a scenario with a similarity greater than or equal to a preset percentage exists in the matching results, the scenario with the highest similarity is selected, and the remediation scheme set corresponding to that scenario is used as one of the candidate remediation schemes for the target contaminated site. Simultaneously, the remediation technology prediction model is invoked to determine other candidate remediation schemes for the target contaminated site. Finally, based on the technology card library and candidate remediation schemes, the optimal comprehensive remediation scheme is selected.

[0182] Example 6:

[0183] This embodiment provides a storage medium, which is a computer-readable storage medium, storing a computer program. When the program is executed by a processor, the processor executes the computer program stored in the memory to realize the collaborative configuration decision of contaminated soil remediation technology-materials-equipment in Embodiment 1 above, as follows:

[0184] This process involves acquiring information on contaminated sites, remediation technologies and related information, and engineering application data of materials and equipment to construct a remediation case database. Based on these cases and relevant literature, the process identifies the materials and equipment involved in various remediation technologies, along with their detailed information, and categorizes the core and common materials and equipment for each technology. Material and equipment combinations are then screened to determine the final technology cards. The unit remediation cost and environmental impact indicators for each technology card are calculated, and a technology card library is constructed. Contaminated site information is clustered, and typical pollution scenarios and remediation schemes are summarized based on the clustered categories to form a scenario library. A feature matrix of spatial block site information and corresponding remediation technologies is constructed based on the remediation cases. A remediation technology prediction model is trained using a machine learning model to predict spatial block remediation technologies. The target contaminated site is matched with the scenario library, and similarity is calculated. If a scenario with a similarity greater than or equal to a preset percentage exists in the matching results, the scenario with the highest similarity is selected, and the remediation scheme set corresponding to that scenario is used as one of the candidate remediation schemes for the target contaminated site. Simultaneously, the remediation technology prediction model is invoked to determine other candidate remediation schemes for the target contaminated site. Finally, based on the technology card library and candidate remediation schemes, the optimal comprehensive remediation scheme is selected.

[0185] It should be noted that the computer-readable storage medium in this embodiment can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor device, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0186] In this embodiment, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used or combined with an instruction execution device, apparatus, or device. In this embodiment, the computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable storage medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use or combined with an instruction execution device, apparatus, or device. The computer program contained on the computer-readable storage medium can be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0187] The computer-readable storage medium described above can be used to write computer programs for executing this embodiment in one or more programming languages ​​or combinations thereof. These programming languages ​​include object-oriented programming languages—such as Java, Python, and C++—and conventional procedural programming languages—such as C or similar programming languages. The program can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0188] In summary, this invention aims at green and sustainable soil remediation. Focusing on the issues of multi-technology synergy and optimized material and equipment configuration, it constructs an integrated decision-making framework encompassing "technology-materials-equipment." Through structured extraction and cluster analysis of multi-source remediation case data, a typical pollution scenario library is formed. The nearest neighbor algorithm is used to match scenarios, and based on the frequency statistics of technology combinations associated with scenarios, high-maturity technology combinations are selected. Simultaneously, machine learning models are combined to spatially segment complex sites and predict remediation technologies, and a technology synergy mechanism is introduced to qualitatively analyze the synergistic relationships of multiple technologies. Furthermore, based on the technology-materials-equipment screening mechanism, a technology card library of material and equipment combinations corresponding to each technology is obtained, and the economic costs and environmental impacts of different combinations are quantitatively evaluated. This enables the scientific generation and optimization of remediation solutions that include materials and equipment, improving the scientific rigor, systematicity, and engineering feasibility of remediation solutions, and providing strong support for green and sustainable soil remediation.

[0189] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope disclosed in the present invention, based on the technical solution and inventive concept of the present invention, shall fall within the scope of protection of the present invention.

Claims

1. A collaborative configuration decision-making method for contaminated soil remediation technology, materials, and equipment, characterized in that, The method includes: Acquire information on contaminated sites, remediation technologies and related information, engineering application data of materials and equipment, and build a remediation case database; Based on restoration cases and relevant literature, we have sorted out the materials and equipment involved in various restoration techniques and their corresponding detailed information, and identified the core and common materials and equipment for each restoration technique. The materials and equipment combinations are screened to determine the final technical cards, the unit repair cost and environmental impact indicators of each technical card are calculated, and a technical card library is constructed. Cluster analysis is performed on contaminated site information, and typical pollution scenarios and remediation solutions are summarized based on the clustered categories to form a scenario library; Based on the repair cases, a feature matrix of spatial block field information and corresponding repair technologies is constructed. Through machine learning model training, a repair technology prediction model is obtained to predict spatial block repair technologies. The target contaminated site is matched with the scenario library and the similarity is calculated. If there is a scenario with a similarity greater than or equal to a preset percentage in the matching results, the scenario with the highest similarity is selected and the set of remediation schemes corresponding to the scenario is taken as one of the candidate remediation schemes for the target contaminated site. At the same time, the remediation technology prediction model is called to determine other candidate remediation schemes for the target contaminated site. Based on the technical card library and candidate repair solutions, the optimal comprehensive repair solution is selected.

2. The synergistic configuration decision-making method for contaminated soil remediation technology-materials-equipment according to claim 1, characterized in that, The acquisition of contaminated site information, remediation technologies and related information, and engineering application data of materials and equipment, to construct a remediation case database, includes: Based on various remediation plans and risk assessment reports related to contaminated site remediation, a case database is constructed by using a large language model to extract site characteristic parameters, remediation technologies, overall remediation cost, remediation days from start to completion, and engineering application data of materials and equipment.

3. The synergistic configuration decision-making method for contaminated soil remediation technology-materials-equipment according to claim 1, characterized in that, The process involves clustering contaminated site information and summarizing typical contamination scenarios and remediation solutions based on the clustered categories to form a scenario library, including: Based on contaminated site information, a feature matrix is ​​constructed, cluster analysis is performed on the feature matrix, and statistical analysis is conducted on the clustered categories. The mean or mode is used as the cluster center to summarize typical pollution scenarios and remediation solutions, thereby forming a scenario library.

4. The collaborative configuration decision-making method for contaminated soil remediation technology-materials-equipment according to claim 1, characterized in that, The feature matrix of spatial block field information and corresponding repair technologies constructed based on repair cases is then trained using a machine learning model to obtain a model for predicting spatial block repair technologies, including: Based on the permeability of the formation, the type of pollutant, the depth range, the medium, and the degree of pollution, the site information of the remediation case is spatially divided into blocks. The blocks are used as the smallest remediation decision units. A feature matrix is ​​constructed for each block. The feature matrix is ​​used as the training feature, and all the remediation technologies included in the remediation case are used as the training labels to train the machine learning model, thus obtaining a remediation technology prediction model based on spatial block division.

5. The synergistic configuration decision-making method for contaminated soil remediation technology-materials-equipment according to claim 1, characterized in that, The aforementioned remediation technology prediction model is invoked to determine other candidate remediation schemes for the target contaminated site, including: Based on the permeability of the formation, the type of pollutant, the depth range, the medium and the degree of pollution, the target contaminated site is divided into spatial blocks, and a remediation technology prediction framework with the block as the smallest remediation decision unit is constructed. Based on the pollution characteristics and site conditions of each block, a variety of potential feasible remediation technologies and their corresponding recommended probabilities are obtained using remediation technology prediction models. By utilizing the principle of technological synergy, local spatial solutions are obtained, and by utilizing the principle of spatial synergy, these local spatial solutions are integrated to further form other candidate remediation solutions for the target contaminated site.

6. The synergistic configuration decision-making method for contaminated soil remediation technology-materials-equipment according to claim 1, characterized in that, The step of selecting the optimal comprehensive repair solution based on the technology card library and candidate repair solutions includes: For each remediation technique in each candidate remediation scheme, retrieve all corresponding candidate technique cards from the technique card library; Based on the set weights of each evaluation dimension, calculate the comprehensive score of each technical card, and select the technical card with the highest score as the optimal matching technical card. Based on the optimal matching technology card, the index values ​​of each candidate repair scheme are calculated under each evaluation dimension; Based on the set weights of each evaluation dimension, the multi-evaluation dimension indicators of each candidate repair scheme are weighted and summarized to obtain the overall score of each candidate repair scheme. The overall scores of all candidate repair schemes are ranked, and the candidate repair scheme with the highest score is selected as the optimal comprehensive repair scheme.

7. The collaborative configuration decision-making method for contaminated soil remediation technology-materials-equipment according to claim 1, characterized in that, The step of selecting the optimal comprehensive repair solution based on the technology card library and candidate repair solutions includes: For each candidate repair scheme and its various repair technologies, the most economical technology card is matched from the technology card library; based on the cost parameters of the matched technology card, the overall economic cost of the candidate repair scheme is calculated; the total cost of all candidate repair schemes is compared, and the scheme with the lowest economic cost is selected as the optimal comprehensive repair scheme. Alternatively, for each candidate remediation scheme and its constituent remediation technologies, the technology card with the best environmental performance can be matched from the technology card library; the environmental impact indicators of each scheme can be calculated based on the matching results; and by comparing the environmental indicators of each candidate remediation scheme, the candidate remediation scheme with the least environmental impact can be selected as the optimal comprehensive remediation scheme.

8. A collaborative configuration decision-making system for contaminated soil remediation technology, materials, and equipment, characterized in that, The system includes: The remediation case database construction module is used to acquire information on contaminated sites, remediation technologies and related information, and engineering application data of materials and equipment to build a remediation case database. The Materials and Equipment Classification Module is used to sort out the materials and equipment involved in various restoration techniques and their corresponding detailed information based on restoration cases and relevant literature, and to classify the core and common materials and equipment for each restoration technique. The technology card library construction module is used to screen material and equipment combinations, determine the final technology cards, calculate the unit repair cost and environmental impact indicators of each technology card, and build the technology card library. The scenario library construction module is used to perform cluster analysis on contaminated site information and summarize typical pollution scenarios and remediation solutions based on the clustered categories to form a scenario library. The prediction model building module is used to construct a feature matrix of spatial block field information and corresponding repair technologies based on repair cases. Through machine learning model training, a repair technology prediction model is obtained to predict spatial block repair technologies. The candidate remediation scheme determination module is used to match the target contaminated site with the scenario library and calculate the similarity. If there is a scenario with a similarity greater than or equal to a preset percentage in the matching results, the scenario with the highest similarity is selected, and the remediation scheme set corresponding to the scenario is used as one of the candidate remediation schemes for the target contaminated site. At the same time, the remediation technology prediction model is called to determine other candidate remediation schemes for the target contaminated site. The comprehensive repair solution selection module is used to select the optimal comprehensive repair solution based on the technical card library and candidate repair solutions.

9. A computer device comprising a processor and a memory for storing a processor-executable program, characterized in that, When the processor executes the program stored in the memory, it implements the collaborative configuration decision-making method for contaminated soil remediation technology-materials-equipment as described in any one of claims 1-7.

10. A storage medium storing a program, characterized in that, When the program is executed by the processor, it implements the collaborative configuration decision-making method for contaminated soil remediation technology-materials-equipment as described in any one of claims 1-7.

Images