A knowledge graph-based soil health key heavy metal identification and treatment decision method

Abstract

This invention discloses a knowledge graph-based method for identifying and making remediation decisions regarding key heavy metals in soil health, comprising the following steps: Step 1, data acquisition and preprocessing: collecting spectral data of heavy metal content in different soils to form a dataset; Step 2, knowledge graph construction: extracting spectral data features from soil testing and corresponding remediation schemes, classifying and storing them to form a knowledge graph; Step 3, heavy metal identification: matching and identifying soil samples with the knowledge graph after spectral detection; Step 4, heavy metal concentration determination: determining the heavy metal concentration in soil samples and assessing pollution risk based on the heavy metal concentration; Step 5, remediation decision generation: comparing and finding the most similar case to the current situation and generating a remediation decision based on historical cases. This invention solves the problems of low detection efficiency, reliance on expert experience, and fragmented and outdated knowledge systems in traditional soil heavy metal remediation by constructing a knowledge graph that integrates spectral features and remediation cases.

Description

Technical Field

[0001] This invention relates to the field of interdisciplinary technology of environmental governance and information technology, and in particular to a knowledge graph-based method for identifying and remediating key heavy metals in soil health. Background Technology

[0002] Heavy metal pollution in soil refers to the phenomenon where, due to human activities or natural processes, the levels of highly toxic heavy metals such as mercury, cadmium, lead, chromium, and arsenic in soil exceed background levels, causing ecological damage and environmental degradation. These heavy metals mainly enter the soil through industrial wastewater discharge, mining, pesticide and fertilizer application, and atmospheric deposition. They are difficult for microorganisms to degrade, easily accumulate in the soil, and alter their activity and toxicity through adsorption-desorption, precipitation-dissolution, and other chemical processes, thereby weakening soil fertility, affecting crop growth, and accumulating in organisms through the food chain, ultimately threatening human health and causing chronic poisoning, organ damage, and even cancer. The remediation of heavy metal pollution in soil has become a major global environmental challenge, stemming from the large-scale discharge and long-term accumulation of pollutants during industrialization and agricultural modernization.

[0003] Industrial and agricultural production activities lead to heavy metals such as arsenic, cadmium, lead, and chromium entering the soil system through wastewater discharge and atmospheric deposition. Due to the hidden, cumulative, and irreversible nature of soil pollution, pollutants remain for extended periods and accumulate in crops, ultimately threatening agricultural product safety and human health. Traditional methods for heavy metal remediation in soil suffer from low detection efficiency, reliance on expert experience, and fragmented and outdated knowledge bases. Therefore, this invention proposes a knowledge graph-based method for identifying and remediating key heavy metals in soil to address these problems in existing technologies. Summary of the Invention

[0004] To address the aforementioned problems, the present invention aims to propose a knowledge graph-based method for identifying and remediating key heavy metals in soil health. This knowledge graph-based method constructs a knowledge graph that integrates spectral features and remediation cases, solving the problems of low detection efficiency, reliance on expert experience, and fragmented and outdated knowledge systems in traditional soil heavy metal remediation. It achieves intelligent, precise, and adaptive optimization from pollution identification and risk assessment to remediation decision-making, significantly improving remediation efficiency and scientific rigor.

[0005] To achieve the objectives of this invention, the invention is implemented through the following technical solution: a knowledge graph-based method for identifying and remediating key heavy metals in soil health, comprising the following steps: Step 1: Data Acquisition and Preprocessing. Collect spectral data of heavy metal content in different soils, integrate soil physicochemical properties, and form a dataset with a unified spatiotemporal scale through spectral preprocessing. Step 2: Knowledge graph construction. Extract spectral data features of soil testing and corresponding remediation solutions from scientific literature, reports and structured datasets. Classify and store the spectral features according to the corresponding heavy metal pollution categories to form a knowledge graph. Step 3: Heavy metal identification. After sampling the soil sample, perform spectral detection to obtain spectral data, and then match and identify the spectral data with the knowledge graph. Step 4: Heavy metal concentration determination. After pre-processing the soil samples collected in Step 3, the heavy metal concentration is determined, and the pollution risk is assessed based on the heavy metal concentration, distinguishing between low risk, medium risk, and high risk. Step 5: Governance Decision Generation. The heavy metal pollution and heavy metal concentration measured in Step 4 are compared with the cases stored in the knowledge graph. The case most similar to the current situation is found and a governance decision is generated based on the historical cases. When historical cases are not available, an early warning is issued based on the measured risk.

[0006] A further improvement is that the data preprocessing in step one includes spectral denoising, spectral smoothing, and spectral transformation correction.

[0007] A further improvement is that, in step two, the data information within the knowledge graph is classified, integrated, and stored in the form of entity, relation, and attribute triples.

[0008] A further improvement is that the matching and identification in step three involves matching the sample spectral data with known contamination patterns in the knowledge graph. If the matching is successful, the identification result is output directly; if the matching fails, the sample spectral data is input into the knowledge graph for optimization and updating.

[0009] A further improvement is made in that the specific steps for determining the soil heavy metal concentration in step four include: S1. Sample pretreatment: After air-drying, grinding and sieving the soil sample to obtain uniform soil particles, aqua regia is poured into the sample for digestion to extract heavy metal elements into the solution. S2. Concentration determination: The fusion was analyzed by inductively coupled plasma mass spectrometry to determine the concentration of key heavy metals. S3. Pollution risk assessment: Based on the type and concentration of heavy metals, an assessment is conducted from three aspects: hazard, exposure, and toxicity.

[0010] A further improvement is that the sample pretreatment in S1 completely transfers heavy metals from the soil solid phase to the liquid phase.

[0011] A further improvement is that the hazard in S3 is determined by the type of heavy metal, the exposure is determined by the environment, and the toxicity is directly related to the concentration of heavy metals.

[0012] The beneficial effects of this invention are as follows: By constructing a knowledge graph that integrates spectral features and remediation cases, this invention solves the problems of low detection efficiency, reliance on expert experience, and fragmented and outdated knowledge systems in traditional soil heavy metal remediation. It achieves intelligent, precise, and adaptive optimization from pollution identification and risk assessment to remediation decision-making, significantly improving remediation efficiency and scientific rigor. Attached Figure Description

[0013] Figure 1 This is a flowchart of the steps of the present invention. Detailed Implementation

[0014] To enhance understanding of the present invention, the present invention will be further described in detail below with reference to embodiments. These embodiments are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0015] Traditional methods for assessing heavy metal pollution in soil are often limited to analysis of a single medium or scale, lacking a systematic consideration of the synergistic effects of multi-source pollution. For example, existing technologies typically employ methods such as chemical mass balance and principal component analysis for source apportionment, but these methods struggle to address the problem of insufficient nonnegativity constraints, and the apportionment results are highly dependent on empirical judgment, leading to significant subjectivity in source tracing outcomes. Furthermore, current remediation decisions are largely based on empirical rules, lacking dynamic simulation and optimization of the entire chain from pollution source to migration pathway to risk receptor, making it difficult to achieve precise prevention and control.

[0016] Based on this, according to Figure 1 As shown, this embodiment provides a knowledge graph-based method for identifying and managing key heavy metals in soil health, including the following steps: Step 1: Data Acquisition and Preprocessing. Spectral data of heavy metal content in different soils were collected, and soil physicochemical properties were integrated. Through spectral preprocessing, a dataset with a unified spatiotemporal scale was formed. Data preprocessing included spectral denoising, spectral smoothing, and spectral transformation correction. Spectral data of soil samples representing different regions and pollution levels were systematically collected through field sampling and laboratory analysis, and corresponding soil physicochemical properties were simultaneously recorded or measured. Subsequently, a series of preprocessing operations were performed on the acquired raw spectral data, including spectral denoising and smoothing to eliminate random errors, and spectral transformation corrections such as standard normal variable transformation or multivariate scattering correction to eliminate baseline drift and light scattering effects, ultimately forming a unified, high-quality dataset with a unified spatiotemporal scale. Transforming multi-source, heterogeneous raw data into clean, standardized, and computable high-quality data provides reliable input for subsequent knowledge graph construction and model training, fundamentally ensuring the accuracy and reliability of subsequent analytical results.

[0017] Step two involves knowledge graph construction. This step extracts spectral data features from soil testing and corresponding remediation plans from scientific literature, reports, and structured datasets. The spectral features are then categorized and stored according to their corresponding heavy metal pollution types to form a knowledge graph. Within the knowledge graph, data information is categorized, integrated, and stored as triples of entities, relationships, and attributes. This transforms scattered, implicit domain knowledge into an interconnected, reasonable whole, providing strong knowledge support and reasoning capabilities for subsequent rapid identification and intelligent decision-making.

[0018] Step 3: Heavy Metal Identification. After soil sample collection, spectral data is obtained through spectral analysis. This spectral data is then matched against a knowledge graph. Matching involves comparing the sample's spectral data with known pollution patterns in the knowledge graph. Successful matches result in the output; unsuccessful matches input the sample's spectral data into the knowledge graph for optimization and updates, enabling continuous learning. Utilizing a knowledge graph allows for rapid and low-cost preliminary screening and qualitative identification, significantly improving efficiency. Furthermore, the feedback mechanism enables the system to dynamically evolve, allowing it to handle novel or complex pollution scenarios.

[0019] Step 4: Heavy Metal Concentration Determination. After pre-processing the soil samples collected in Step 3, the heavy metal concentration is determined, and pollution risk is assessed based on the concentration, classifying them as low, medium, and high risk. The specific steps for soil heavy metal concentration determination include: S1: Sample Pre-processing. The soil sample is air-dried, ground, and sieved to obtain uniform soil particles. Aqua regia is added to the sample for digestion, extracting heavy metal elements into the solution. S2: Concentration Determination. The fusion is analyzed using inductively coupled plasma mass spectrometry to determine the concentration of key heavy metals. S3: Pollution Risk Assessment. An assessment is conducted based on the heavy metal category and concentration, considering hazard, exposure, and toxicity. The sample pre-processing in S1 completely transfers heavy metals from the soil solid phase to the liquid phase. In S3, hazard is determined by the heavy metal category, exposure by the environment, and toxicity is directly related to the heavy metal concentration. This provides accurate quantitative evidence for remediation decisions. Spectroscopic identification provides qualitative screening, while laboratory determination provides authoritative quantitative analysis; the combination of both constitutes a complete cognitive chain.

[0020] Step 5, Governance Decision Generation: The heavy metal pollution and concentration data measured in Step 4 are compared with cases stored in the knowledge graph. The most similar case to the current situation is identified, and governance decisions are generated based on historical cases. In the absence of historical cases, early warnings are issued based on the measured risks. This process transforms preliminary detection and diagnostic knowledge into concrete action plans.

[0021] This knowledge graph-based method for identifying and managing key heavy metals in soil health involves multi-source data acquisition and preprocessing through spectral detection and soil physicochemical property testing to create a high-quality dataset. A knowledge graph is then constructed to store spectral features, heavy metal categories, and successful remediation cases. Subsequently, spectral detection is performed on soil samples to be tested, and the samples are matched against the knowledge graph to quickly identify heavy metal categories. Unmatched samples are used to optimize the graph. After identification, samples undergo laboratory pretreatment and inductively coupled plasma mass spectrometry for precise concentration determination, and the pollution risk level is assessed based on hazard, exposure, and toxicity. Finally, the system intelligently matches the current pollution situation with historical cases in the knowledge graph to generate customized remediation plans. If no matching cases are found, an early warning is activated based on the risk level, thus completing a closed loop from data acquisition to intelligent decision-making.

[0022] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A knowledge graph-based soil health key heavy metal identification and management decision method, characterized in that, Includes the following steps: Step 1: Data Acquisition and Preprocessing. Collect spectral data of heavy metal content in different soils, integrate soil physicochemical properties, and form a dataset with a unified spatiotemporal scale through spectral preprocessing. Step 2: Knowledge graph construction. Extract spectral data features of soil testing and corresponding remediation solutions from scientific literature, reports and structured datasets. Classify and store the spectral features according to the corresponding heavy metal pollution categories to form a knowledge graph. Step 3: Heavy metal identification. After sampling the soil sample, perform spectral detection to obtain spectral data, and then match and identify the spectral data with the knowledge graph. Step 4: Heavy metal concentration determination. After pre-processing the soil samples collected in Step 3, the heavy metal concentration is determined, and the pollution risk is assessed based on the heavy metal concentration, distinguishing between low risk, medium risk, and high risk. Step 5: Governance Decision Generation. The heavy metal pollution and heavy metal concentration measured in Step 4 are compared with the cases stored in the knowledge graph. The case most similar to the current situation is found and a governance decision is generated based on the historical cases. When historical cases are not available, an early warning is issued based on the measured risk.

2. The knowledge graph-based soil health key heavy metal identification and management decision-making method according to claim 1, characterized in that: The data preprocessing in step one includes spectral denoising, spectral smoothing, and spectral transformation correction.

3. The knowledge graph-based soil health key heavy metal identification and management decision-making method according to claim 1, characterized in that: In step two, the data information within the knowledge graph is classified, integrated, and stored in the form of entity, relation, and attribute triples.

4. The method for identifying and remediating key heavy metals in soil health based on knowledge graphs according to claim 1, characterized in that: The matching and identification in step three involves matching the sample spectral data with known contamination patterns in the knowledge graph. If the match is successful, the identification result is output directly; if the match fails, the sample spectral data is input into the knowledge graph for optimization and updating.

5. The method for identifying and remediating key heavy metals in soil health based on knowledge graphs according to claim 1, characterized in that: The specific steps for determining the soil heavy metal concentration in step four include: S1. Sample pretreatment: After air-drying, grinding and sieving the soil sample to obtain uniform soil particles, aqua regia is poured into the sample for digestion to extract heavy metal elements into the solution. S2. Concentration determination: The fusion was analyzed by inductively coupled plasma mass spectrometry to determine the concentration of key heavy metals. S3. Pollution risk assessment: Based on the type and concentration of heavy metals, an assessment is conducted from three aspects: hazard, exposure, and toxicity.

6. The method for identifying and remediating key heavy metals in soil health based on knowledge graphs according to claim 5, characterized in that: The sample pretreatment in S1 completely transfers heavy metals from the soil solid phase to the liquid phase.

7. The method for identifying and remediating key heavy metals in soil health based on knowledge graphs according to claim 5, characterized in that: The hazard in S3 is determined by the type of heavy metal, the exposure is determined by the environment, and the toxicity is directly related to the concentration of heavy metal.

Images