Multistage early warning method and system for leakage of fly ash landfill

Abstract

The invention discloses a multi-stage early warning method and system for leakage of a fly ash landfill, and the method comprises the steps: obtaining environment background data, and carrying out the preprocessing of the environment background data, so as to obtain background well historical data and fly ash entrance detection data; based on dynamic baseline modeling and heavy metal feature decoupling, obtaining a conductivity baseline, a sulfate baseline and a PH value baseline according to the background well historical data and the fly ash admission detection data, and constructing a fly ash heavy metal fingerprint database; acquiring and processing underground water detection data to obtain comparison data; leakage risk multi-level perception and collaborative decision making are carried out by utilizing early warning, verification and confirmation three-level logic; and completing mean value updating and covariance matrix updating to optimize the conductivity, sulfate, PH value base line and fly ash heavy metal fingerprint database. The leakage risk of the fly ash landfill can be evaluated as soon as possible, and the operation economic cost is reduced while the environmental protection requirement is met.

Description

Technical Field

[0001] This application belongs to the field of hazardous waste treatment and environmental monitoring technology, specifically relating to a multi-level early warning method and system for fly ash landfill leakage. Background Technology

[0002] Fly ash, a byproduct of municipal solid waste incineration, is classified as HW18 hazardous waste in the National Hazardous Waste List (2021 edition) due to its content of heavy metals such as mercury (Hg), cadmium (Cd), and lead (Pb), as well as high concentrations of chloride ions. Currently, fly ash is mainly disposed of by chelating and solidifying it before being disposed of separately or in sections in sanitary landfills that commonly use high-density polyethylene (HDPE) membranes as impermeable barriers, in order to meet the requirements for long-term stable landfilling.

[0003] However, in actual landfill operation, HDPE membranes are susceptible to damage from multiple factors such as geological stress, aging corrosion, and construction defects, leading to a significant increase in the risk of leachate leakage. Existing fly ash landfill leakage detection technologies mainly rely on geophysical exploration methods such as high-voltage direct current (HVDC) and groundwater index analysis, but all have significant drawbacks: geophysical exploration technologies require the pre-installation of sensor networks during the construction phase, resulting in high initial investment and susceptibility to stray current interference, leading to high maintenance costs; groundwater detection suffers from long cycles (7-14 days / time), difficulty in tracing pollution sources (inability to distinguish between local leakage and external migration), and ambiguous thresholds (high overlap between heavy metal concentration and background values). These combined shortcomings result in high technical barriers, high operating costs, and insufficient reliability, making it difficult to meet the needs of refined landfill operation. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies, this application provides a multi-level early warning method and system for fly ash landfill leakage. Based on dynamic baseline and heavy metal fingerprint coupling, it is particularly suitable for early warning of leakage risks in municipal solid waste incineration fly ash landfills. It can achieve economical, efficient, reliable, and easy-to-operate early warning, so as to help landfill operators assess leakage risks early, meet environmental protection requirements, and reduce operating costs.

[0005] To achieve the above objectives, this application adopts the following technical solution:

[0006] A multi-level early warning method for fly ash landfill leakage based on dynamic baseline and heavy metal fingerprint coupling includes:

[0007] Environmental baseline data is acquired and preprocessed to obtain historical data of baseline wells and fly ash entry detection data;

[0008] Based on dynamic baseline modeling and heavy metal feature decoupling, baselines for conductivity, sulfate, and pH value are obtained from the historical data of the base well and the fly ash entry detection data, and a fly ash heavy metal fingerprint database is constructed.

[0009] Obtain and process groundwater monitoring data to obtain comparative data;

[0010] Utilizing a three-tiered logic of early warning, verification, and confirmation for multi-level perception and collaborative decision-making regarding leakage risks; and

[0011] The mean and covariance matrix were updated to optimize the baselines for conductivity, sulfate, pH, and fly ash heavy metal fingerprints.

[0012] As a preferred embodiment, the historical data of the baseline well includes pH, TDS, and sulfate data; the historical data of the baseline well covers at least one calendar year, must cover January to December and include the entire dry and wet seasons;

[0013] The fly ash entry detection data includes heavy metal data detected after the leachate preparation is completed;

[0014] The preprocessing includes preliminary calculation, grouping, and data rounding; the preliminary calculation is to convert total dissolved solids to conductivity using the empirical formula TDS=0.55K; the grouping is to statistically analyze the data of pH, conductivity, sulfate, and heavy metals in tabular form and distinguish them by natural month; the data rounding is to retain the data to the thousandths place to unify the number of decimal places caused by different index detection methods.

[0015] As a preferred embodiment, the method based on dynamic baseline modeling and heavy metal feature decoupling, obtaining baselines for conductivity, sulfate, and pH based on the baseline well historical data and the fly ash entry detection data, and constructing a fly ash heavy metal fingerprint database includes:

[0016] Calculate the monthly mean and monthly standard deviation of the pH, conductivity, and sulfate data;

[0017] Calculate the baseline ranges for pH, conductivity, and sulfate each month; these baseline ranges are interval values ​​constrained by a 95% confidence interval.

[0018] The heavy metal data will be integrated and standardized on a monthly basis.

[0019] The covariance matrix of the processed heavy metal data will be calculated monthly; and

[0020] PCA-based dimensionality reduction calculations were used to obtain heavy metal fingerprint features from fly ash to differentiate between fly ash landfill leaks and external pollution.

[0021] As a preferred embodiment, the calculation of the monthly mean and monthly standard deviation of the pH, conductivity, and sulfate data includes:

[0022] The mean calculation method is as follows: ,in This represents the monthly average of each indicator, where n is the number of days in the month. Monthly values ​​for pH, conductivity, and sulfate.

[0023] The formula for calculating the standard deviation is: and will Data outside the range is removed as outliers;

[0024] The interval value is .

[0025] As a preferred embodiment, the monthly integration and standardization of heavy metal data includes:

[0026] The integration process involves matrixing the heavy metal data to obtain the original data matrix. ,in To calculate the heavy metal data matrix for the current month, This represents the total number of batches for the month. This represents the total number of heavy metal items tested each month. batch The concentration of heavy metals is ,in The maximum value is , The maximum value is ;

[0027] The standardization process involves converting the heavy metal concentration of each batch of fly ash into a ratio, calculated using the following formula: , that is, the first Batch No. The ratios of various heavy metals were used to eliminate the influence of absolute concentration values, resulting in a standardized ratio matrix. The sum of all heavy metal concentrations after conversion to ratios is 100%, meaning the sum of each row in the standardized ratio matrix should be 1.

[0028] As a preferred embodiment, the step of calculating the covariance matrix of the processed heavy metal data on a monthly basis includes:

[0029] The covariance matrix calculation includes mean vector calculation, centering matrix, coefficient of variation calculation, and construction of the covariance matrix.

[0030] The mean vector is according to calculate, It is the first The historical average proportions of various heavy metals were determined, and a standardized average proportion matrix was constructed. ;

[0031] The centered matrix is ​​the result of subtracting the standardized mean from each column, as shown in the equation. calculate;

[0032] The formula for calculating the covariance matrix is: ,in It is the transpose of a centralized matrix with dimension . , Each element Indicates the first and the The proportional covariance of the heavy metals is calculated using the following formula: .

[0033] As a preferred embodiment, the step of obtaining fly ash heavy metal fingerprint features based on PCA dimensionality reduction calculation includes:

[0034] The PCA dimensionality reduction includes feature decomposition, variance contribution rate calculation, cumulative variance contribution rate calculation, selection of the number of principal components, and principal component projection.

[0035] The eigenvalue decomposition is the process of finding the eigenvalues ​​of the covariance matrix. and eigenvectors The calculation formula is , , It is the first 1 eigenvalue, It is the unit eigenvector corresponding to the eigenvalue;

[0036] The variance contribution rate is , by formula Calculated;

[0037] The cumulative variance contribution rate is , by formula Calculated;

[0038] The selection of principal components involves retaining heavy metals with a cumulative contribution rate >85% as principal components;

[0039] The principal component projection is the process of projecting standardized data and eigenvalue decomposition data onto the principal component space, and the calculation formula is as follows: ,in It was before A matrix composed of eigenvectors.

[0040] As a preferred embodiment, the acquisition and processing of groundwater detection data to obtain comparative data includes:

[0041] The groundwater testing data includes groundwater conductivity, pH, and sulfate data;

[0042] Calculate the correlation coefficient between sulfate and pH in groundwater using the Pearson method; and

[0043] Heavy metal data from pollution diffusion wells and pollution monitoring wells are obtained from regularly completed groundwater monitoring reports at landfills, and ratio vector calculations are performed to obtain heavy metal fingerprint data of groundwater.

[0044] As a preferred embodiment, the multi-level perception and collaborative decision-making of leakage risk using a three-level logic of early warning, verification, and confirmation includes:

[0045] If the average conductivity of groundwater exceeds the limit for three consecutive days or the daily average exceeds the limit twice within three days, a Level 1 alarm will be triggered.

[0046] The sulfate concentration exceeded the baseline range for the month, and the pH value and sulfate value for the month were negatively correlated, confirming an abnormal association and passing the secondary verification.

[0047] The cosine similarity of the obtained groundwater heavy metal fingerprint data with the fingerprint database for the current month is calculated, and a three-level confirmation is completed; the cosine similarity calculation is performed according to the formula. Perform calculations, where The first step in the groundwater heavy metal fingerprint data step Groundwater heavy metal fingerprint data; the three-level confirmation refers to confirming that the leakage source originates from a landfill when the calculated cosine similarity is higher than a preset limit; and

[0048] After completing the three-level confirmation, the environmental emergency response plan will be activated.

[0049] The process of completing mean and covariance matrix updates to optimize conductivity, sulfate, pH baselines, and fly ash heavy metal fingerprint databases includes:

[0050] While acquiring groundwater monitoring data, the pH, sulfate, and conductivity of the baseline wells are measured, and the baseline is updated monthly. The updated baseline uses the same processing method as the initial baseline calculated from historical data and is updated monthly.

[0051] The average value and covariance matrix of heavy metals in fly ash are updated batch by batch; the batch refers to the batch of fly ash entering the site; the average value update includes calculating and updating the average value of the new batch; the average value calculation is performed using the arithmetic mean method. The covariance matrix update uses a recursive method to avoid recalculating the entire matrix, and the calculation formula is as follows: get Update complete.

[0052] In addition, this application also provides a system for implementing the multi-level early warning method for fly ash landfill leakage based on dynamic baseline and heavy metal fingerprint coupling as described above, comprising:

[0053] The acquisition module is used to acquire and preprocess environmental baseline data to obtain historical data of baseline wells and fly ash entry detection data;

[0054] The baseline and fingerprint database construction module is used for dynamic baseline modeling and heavy metal feature decoupling. Based on the historical data of the base well and the fly ash entry detection data, it obtains baselines for conductivity, sulfate, and pH value and constructs a fly ash heavy metal fingerprint database.

[0055] The comparison data acquisition module is used to acquire and process groundwater detection data to obtain comparison data;

[0056] The perception and collaborative decision-making module is used for multi-level perception and collaborative decision-making regarding leakage risks using a three-level logic of early warning, verification, and confirmation; and

[0057] The update module is used to perform mean and covariance matrix updates to optimize the baselines for conductivity, sulfate, pH, and fly ash heavy metal fingerprints.

[0058] Compared with the prior art, this application has the following advantages:

[0059] Utilizing the leachate , Plasma (concentration typically at 10) 4 -10 5 This makes the conductivity (10-15) ) significantly higher than groundwater (<2 The characteristics of the impermeable layer allow for early warning of pollution by utilizing the step change in electrical conductivity caused by damage to the impermeable layer.

[0060] Combined with the abundant sulfates in fly ash ( The concentration in the leachate can reach 3×10 4 (Above) and the specificity of the differential diagnosis, which is usually negatively correlated with pH, ​​to further verify the source of leakage and distinguish geological background interference;

[0061] A fingerprint spectrum can be formed using the metal ion concentration in the fly ash chelate entry test report to confirm early warning information; groundwater monitoring wells are already constructed during the initial construction phase of the landfill, eliminating the need for additional well construction costs; the conductivity and other indicators of groundwater can be measured using portable devices such as conductivity meters, ion-selective electrodes, or test strips, which is economical, quick, and does not significantly increase operating costs; a test report containing metal ion concentration is required when fly ash chelates enter the landfill, allowing for the establishment of a characteristic fingerprint spectrum without relying on sophisticated laboratory instruments; during the operation period, the fly ash landfill needs to conduct regular groundwater monitoring in accordance with GB16889 requirements, and the data contains rich information for analysis. Attached Figure Description

[0062] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of the application and, together with their description, serve to explain the application, but do not constitute an undue limitation of the application. In the drawings:

[0063] Figure 1 This is a flowchart of the method in this application;

[0064] Figure 2 The process of multi-level perception and collaborative decision-making for leakage risk in the three-level logic of this application; Figure 3 This is a system framework diagram of this application. Detailed Implementation

[0065] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0066] In the description of this application, it should be understood that the relationship between the method steps can be sequential or non-sequential, as long as it does not affect the overall technical effect, and therefore should not be construed as a limitation of this application. The following description of this application is merely a description of individual embodiments of the technical solution of this application; other embodiments are not shown in the following description, but this does not mean that this application excludes these other embodiments, nor is the technical solution of this application limited to the specific implementations described below, and the scope of protection of this application is not limited to the specific implementations described below. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0067] It should be noted that if the terms "first," "second," etc., appear in the specification, claims, and accompanying drawings of this application, such descriptions are only used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0068] In some embodiments, such as Figure 1 As shown, this application provides a multi-level early warning method for fly ash landfill leakage based on dynamic baseline and heavy metal fingerprint coupling, characterized by including:

[0069] S1: Obtain environmental baseline data and preprocess it to obtain baseline well historical data and fly ash entry detection data;

[0070] S2: Based on dynamic baseline modeling and heavy metal feature decoupling, the conductivity, sulfate, and pH baselines are obtained according to the historical data of the base well and the fly ash entry detection data, and a fly ash heavy metal fingerprint database is constructed.

[0071] S3: Acquire and process groundwater monitoring data to obtain comparative data;

[0072] S4: Utilizing a three-tiered logic of early warning, verification, and confirmation for multi-level perception and collaborative decision-making regarding leakage risks; and

[0073] S5: Complete mean and covariance matrix updates to optimize conductivity, sulfate, pH baselines, and fly ash heavy metal fingerprint database.

[0074] In some embodiments, the method of step S1 is specifically as follows:

[0075] S1-1, Collect environmental baseline data and preprocess it to obtain baseline well historical data and fly ash entry detection data;

[0076] The baseline well historical data includes pH, total dissolved solids (TDS), and sulfate; the historical period is at least one calendar year, covering January to December, including the entire dry and wet seasons, and can provide a monthly dynamic baseline adapted to seasonal changes;

[0077] S1-2, Collect and preprocess the fly ash entry detection data to obtain fly ash entry detection data;

[0078] The test data for fly ash entering the site are heavy metal data obtained after the leachate is prepared according to standard methods such as HJ300.

[0079] S1-3, Preprocessing the data from S1-1 and S1-2; the preprocessing includes preliminary calculation, grouping, and data rounding.

[0080] The preliminary calculation formula is based on empirical formulas. Complete the conversion calculation between total dissolved solids (TDS) and electrical conductivity (K);

[0081] The grouping involves statistically analyzing parameters such as pH, conductivity, sulfate, and heavy metals in tabular form and separating them by calendar month.

[0082] The data rounding is performed in accordance with GB8170 and is retained to the thousands place to unify the number of decimal places caused by different index methods.

[0083] In some embodiments, the method of step S2 is specifically as follows:

[0084] S2-1, Calculate the monthly mean and monthly standard deviation of the above-mentioned pH, conductivity, and sulfate indicators;

[0085] The mean calculation method is as follows: ,in This represents the monthly average of each indicator, where n is the number of days in the month. The values ​​of pH, conductivity, and sulfate for each month after S1-3 treatment;

[0086] The formula for calculating the standard deviation is: , External data is removed as outliers and not included in the calculation;

[0087] S2-2, calculate the baseline range of pH, conductivity, and sulfate levels for each month;

[0088] The baseline range is an interval value constrained by a 95% confidence interval. ;

[0089] S2-3, The obtained heavy metal data will be integrated and standardized on a monthly basis;

[0090] The data integration involves matrixing the heavy metal data from the fly ash leachate test reports obtained in S1-3 to obtain the original data matrix. ,in To calculate the heavy metal data matrix for the current month, This represents the total number of batches for the month. This represents the total number of heavy metal items tested each month. batch The concentration of heavy metals is ,in The maximum value is , The maximum value is ;

[0091] The standardization process involves converting the heavy metal concentration of each batch of fly ash into a ratio, calculated using the following formula: , that is, the first Batch No. The ratios of various heavy metals were used to eliminate the influence of absolute concentration values, resulting in a standardized ratio matrix. The sum of all heavy metal concentrations after conversion to ratios is 100%, meaning the sum of each row in the matrix should be 1.

[0092] S2-4, calculate the covariance matrix of the obtained heavy metal concentration ratios on a monthly basis; the covariance matrix calculation includes mean vector calculation, centering matrix, coefficient of variation calculation and construction of covariance matrix;

[0093] The mean vector represents historical average characteristics, according to... calculate, It is the first Historical average ratios of various heavy metals were determined, and a standardized average ratio matrix was constructed. ;

[0094] The centered matrix is ​​formed by subtracting the standardized mean from each column to improve comparability, as shown in the equation. calculate;

[0095] The formula for calculating the covariance matrix is: ,in It is the transpose (dimension) of the centralized matrix. ), Each element Indicates the first and the The ratio covariance of the heavy metals is calculated using the following formula: ;

[0096] S2-5, Principal Component Analysis (PCA) dimensionality reduction calculation to obtain heavy metal fingerprint features of fly ash to distinguish between landfill leaks and external pollution; the PCA dimensionality reduction includes feature decomposition, variance contribution rate calculation, cumulative variance contribution rate calculation, selection of the number of principal components, and principal component projection; heavy metal principal component analysis (PCA) can be calculated using preset tools in environments such as Python, and more preferably, heavy metal fingerprint analysis can be performed using the chemometrics software SIMCA;

[0097] The eigenvalue decomposition is the process of finding the eigenvalues ​​of the covariance matrix. and eigenvectors The calculation formula is , , It is the first 1 eigenvalue (arranged in descending order) ), It is the unit eigenvector corresponding to the eigenvalue;

[0098] The variance contribution rate is given by the formula The cumulative variance contribution rate is calculated as follows: Calculated;

[0099] The selection of principal components involves retaining heavy metals with a cumulative contribution rate >85% as principal components (as previously stated). indivual);

[0100] The principal component projection is the process of projecting standardized data and eigenvalue decomposition data onto the principal component space, and the calculation formula is as follows: ,in It was before A matrix composed of eigenvectors;

[0101] The above calculations can be performed using a combination of functions such as AVERAGE, STEDEV, and PERCENTILE in Excel. More preferably, they can be performed using pre-set tools in environments such as Python.

[0102] In some embodiments, the method of step S3 is specifically as follows:

[0103] S3-1, to obtain data on the conductivity, pH, and sulfate content of groundwater;

[0104] The data acquisition refers to the manual measurement of conductivity, pH, and sulfate values ​​in monitoring wells and diffusion wells by personnel using portable equipment during daily inspections of the landfill. Data outside one hour before or after rainfall exceeding 10 mm / 24h (light rain) are considered valid data. The conductivity, pH, and sulfate data can be measured using online equipment to increase the amount of basic data and reduce the workload of personnel.

[0105] S3-2, Calculate the correlation coefficient between sulfate and pH value using the Pearson method;

[0106] S3-3: Obtain groundwater heavy metal data and complete ratio vector calculation to obtain groundwater fingerprint data;

[0107] The heavy metal data is obtained from the groundwater heavy metal data of the pollution diffusion wells and pollution monitoring wells in the groundwater test reports that are regularly completed at the landfill.

[0108] In some embodiments, the method of step S4 is specifically as follows:

[0109] S4-1, if the average groundwater conductivity exceeds the limit for 3 consecutive days or the daily average exceeds the limit twice within 3 days, a Level 1 alarm is triggered.

[0110] S4-2, the groundwater sulfate concentration exceeded the baseline range for the month, and the pH value and sulfate value for the month showed a significant negative correlation, confirming an abnormal association and passing the secondary verification;

[0111] The significant negative correlation refers to a correlation coefficient calculated using the Pearson method that is <0 and has an absolute value greater than 0.5, meaning the correlation is above moderate negative correlation, which increases suspicion of leakage.

[0112] S4-3, calculate the cosine similarity between the groundwater heavy metal fingerprint data obtained in step S3-3 and the fingerprint database data for the current month, and complete the three-level confirmation.

[0113] The cosine similarity calculation is based on Perform calculations, where The first one obtained in step S3-3 Groundwater heavy metal fingerprint data, This refers to the fingerprint database data obtained in steps 2-6;

[0114] The three-level confirmation refers to confirming that the leakage source originates from a landfill when the calculated cosine similarity is >0.9;

[0115] For details, please refer to Figure 2 As shown, Figure 2 This example demonstrates the process of multi-level perception and collaborative decision-making regarding leakage risk using the three-level logic of this application.

[0116] In some embodiments, the method of step S5 is specifically as follows:

[0117] S5-1 involves measuring the pH, sulfate, and conductivity of the baseline well while implementing field monitoring as in S3-1, and updating the baseline monthly.

[0118] The update is processed in the same way as the initial baseline calculated from historical data, i.e., it is updated monthly;

[0119] S5-2, update the mean value and covariance matrix of heavy metals in fly ash in batches;

[0120] The term "by batch" refers to the batch of fly ash entering the facility, typically 10 times per batch; the term "average update" includes calculating and updating the average value for the new batch; the average calculation is performed using the arithmetic mean method. ;

[0121] The covariance matrix is ​​updated using a recursive method to avoid recalculating the entire matrix. The calculation formula is as follows: get Update complete.

[0122] In some embodiments, this application also provides a system 1 for implementing the multi-level early warning method for fly ash landfill leakage based on dynamic baseline and heavy metal fingerprint coupling as described above, such as... Figure 3 As shown, it includes:

[0123] The acquisition module 11 is used to acquire environmental background data and perform preprocessing to obtain historical data of background wells and fly ash entry detection data;

[0124] The baseline and fingerprint database construction module 12 is used for dynamic baseline modeling and heavy metal feature decoupling. Based on the historical data of the base well and the fly ash entry detection data, it obtains the conductivity, sulfate, and pH baselines and constructs a fly ash heavy metal fingerprint database.

[0125] The comparison data acquisition module 13 is used to acquire and process groundwater detection data to obtain comparison data;

[0126] The perception and collaborative decision-making module 14 is used to perform multi-level perception and collaborative decision-making on leakage risks using a three-level logic of early warning, verification, and confirmation; and

[0127] The update module 15 is used to complete the mean update and covariance matrix update to optimize the baseline of conductivity, sulfate, pH value and fly ash heavy metal fingerprint database.

[0128] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims. The selected and described embodiments are intended to best elucidate the principles of this application and its practical application, thereby enabling other those skilled in the art to best utilize this application with various modifications suitable for the contemplated specific purpose, as well as the various described embodiments.

Claims

1. A multi-level early warning method for fly ash landfill leakage based on dynamic baseline and heavy metal fingerprint coupling, characterized in that, include: Acquire and preprocess environmental baseline data to obtain historical data of baseline wells and fly ash entry detection data; Based on dynamic baseline modeling and heavy metal feature decoupling, baselines for conductivity, sulfate, and pH value are obtained from the historical data of the base well and the fly ash entry detection data, and a fly ash heavy metal fingerprint database is constructed. Obtain and process groundwater monitoring data to obtain comparative data; Utilize a three-tiered logic of early warning, verification, and confirmation for multi-level perception and collaborative decision-making regarding leakage risks; as well as The mean and covariance matrix were updated to optimize the baselines for conductivity, sulfate, pH, and fly ash heavy metal fingerprints.

2. The multi-level early warning method for fly ash landfill leakage according to claim 1, characterized in that: The historical data of the baseline wells includes pH, TDS, and sulfate data; the historical data of the baseline wells covers at least one calendar year, must cover January to December and include the entire dry and wet seasons; The fly ash entry detection data includes heavy metal data detected after the leachate preparation is completed; The preprocessing includes preliminary calculation, grouping, and data rounding; the preliminary calculation is to convert total dissolved solids to conductivity using the empirical formula TDS=0.55K; the grouping is to statistically analyze the data of pH, conductivity, sulfate, and heavy metals in tabular form and distinguish them by natural month; the data rounding is to retain the data to the thousandths place to unify the decimal places caused by different index detection methods.

3. The multi-level early warning method for fly ash landfill leakage according to claim 2, characterized in that, The method based on dynamic baseline modeling and heavy metal feature decoupling, obtains baselines for conductivity, sulfate, and pH value according to the historical data of the baseline well and the fly ash entry detection data, and constructs a fly ash heavy metal fingerprint database, including: Calculate the monthly mean and monthly standard deviation of the pH, conductivity, and sulfate data; Calculate the baseline ranges for pH, conductivity, and sulfate each month; these baseline ranges are interval values ​​constrained by a 95% confidence interval. The heavy metal data will be integrated and standardized on a monthly basis. The covariance matrix of the processed heavy metal data will be calculated monthly; and PCA dimensionality reduction calculation is used to obtain the heavy metal fingerprint characteristics of fly ash to distinguish between fly ash landfill leakage and external pollution.

4. The multi-level early warning method for fly ash landfill leakage according to claim 3, characterized in that, The calculation of the monthly mean and monthly standard deviation of the pH, conductivity, and sulfate data includes: The mean calculation method is as follows: ,in This represents the monthly average of each indicator, where n is the number of days in the month. Monthly values ​​for pH, conductivity, and sulfate. The formula for calculating the standard deviation is: and will Data outside the range is removed as outliers; The interval value is .

5. The multi-level early warning method for fly ash landfill leakage according to claim 4, characterized in that, The process of integrating and standardizing heavy metal data on a monthly basis includes: The integration process involves matrixing the heavy metal data to obtain the original data matrix. ,in To calculate the heavy metal data matrix for the current month, This represents the total number of batches for the month. This represents the total number of heavy metal items tested each month. batch The concentration of heavy metals is ,in The maximum value is , The maximum value is ; The standardization process involves converting the heavy metal concentration of each batch of fly ash into a ratio, calculated using the following formula: , that is, the first Batch No. The ratios of various heavy metals were used to eliminate the influence of absolute concentration values, resulting in a standardized ratio matrix. The sum of all heavy metal concentrations after conversion to ratios is 100%, meaning the sum of each row in the standardized ratio matrix should be 1.

6. The multi-level early warning method for fly ash landfill leakage according to claim 5, characterized in that, The step of calculating the covariance matrix of the processed heavy metal data on a monthly basis includes: The covariance matrix calculation includes mean vector calculation, centering matrix, coefficient of variation calculation, and construction of the covariance matrix. The mean vector is according to calculate, It is the first The historical average proportions of various heavy metals were determined, and a standardized average proportion matrix was constructed. ; The centered matrix is ​​the result of subtracting the standardized mean from each column, as shown in the equation... calculate; The formula for calculating the covariance matrix is: ,in It is the transpose of a centralized matrix with dimension . , Each element Indicates the first and the The proportional covariance of the heavy metals is calculated using the following formula: .

7. The multi-level early warning method for fly ash landfill leakage according to claim 6, characterized in that, The method for obtaining fly ash heavy metal fingerprint features based on PCA dimensionality reduction calculation includes: The PCA dimensionality reduction includes feature decomposition, variance contribution rate calculation, cumulative variance contribution rate calculation, selection of the number of principal components, and principal component projection. The eigenvalue decomposition is the process of finding the eigenvalues ​​of the covariance matrix. and eigenvectors The calculation formula is , , It is the first 1 eigenvalue, It is the unit eigenvector corresponding to the eigenvalue; The variance contribution rate is , by formula Calculated; The cumulative variance contribution rate is , by formula Calculated; The selection of principal components involves retaining heavy metals with a cumulative contribution rate >85% as principal components; The principal component projection is the process of projecting standardized data and eigenvalue decomposition data onto the principal component space, and the calculation formula is as follows: ,in It was before A matrix composed of eigenvectors.

8. The multi-level early warning method for fly ash landfill leakage according to claim 7, characterized in that, The process of acquiring and processing groundwater detection data to obtain comparative data includes: The groundwater testing data includes groundwater conductivity, pH, and sulfate data; Calculate the correlation coefficient between sulfate and pH in groundwater using the Pearson method; and Heavy metal data from pollution diffusion wells and pollution monitoring wells are obtained from regularly completed groundwater monitoring reports at landfills, and ratio vector calculations are performed to obtain heavy metal fingerprint data of groundwater.

9. The multi-level early warning method for fly ash landfill leakage according to claim 8, characterized in that: The method of using a three-level logic of early warning, verification, and confirmation for multi-level perception and collaborative decision-making regarding leakage risks includes: If the average conductivity of groundwater exceeds the limit for three consecutive days or the daily average exceeds the limit twice within three days, a Level 1 alarm will be triggered. The sulfate concentration exceeded the baseline range for the month, and the pH value and sulfate value for the month were negatively correlated, confirming an abnormal association and passing the secondary verification. The cosine similarity of the obtained groundwater heavy metal fingerprint data with the fingerprint database for the current month is calculated, and a three-level confirmation is completed; the cosine similarity calculation is performed according to the formula. Perform calculations, where The first step in the groundwater heavy metal fingerprint data step Groundwater heavy metal fingerprint data; the three-level confirmation refers to confirming that the leakage source originates from a landfill when the calculated cosine similarity is higher than a preset limit; and After completing the three-level confirmation, the environmental emergency response plan will be activated. The process of completing mean and covariance matrix updates to optimize conductivity, sulfate, pH baselines, and fly ash heavy metal fingerprint databases includes: While acquiring groundwater monitoring data, the pH, sulfate, and conductivity of the baseline well are measured, and the baseline is updated monthly. The updated baseline uses the same processing method as the initial baseline calculated from historical data and is updated monthly. The average value and covariance matrix of heavy metals in fly ash are updated batch by batch; the batch refers to the batch of fly ash entering the site; the average value update includes calculating and updating the average value of the new batch; the average value calculation is performed using the arithmetic mean method. The covariance matrix update uses a recursive method to avoid recalculating the entire matrix, and the calculation formula is as follows: get Update complete.

10. A system for implementing the multi-level early warning method for fly ash landfill leakage based on dynamic baseline and heavy metal fingerprint coupling as described in any one of claims 1-9, characterized in that, include: The acquisition module is used to acquire and preprocess environmental baseline data to obtain historical data of baseline wells and fly ash entry detection data; The baseline and fingerprint database construction module is used for dynamic baseline modeling and heavy metal feature decoupling. Based on the historical data of the base well and the fly ash entry detection data, it obtains baselines for conductivity, sulfate, and pH value and constructs a fly ash heavy metal fingerprint database. The comparison data acquisition module is used to acquire and process groundwater detection data to obtain comparison data; The perception and collaborative decision-making module is used to perform multi-level perception and collaborative decision-making on leakage risks using a three-level logic of early warning, verification, and confirmation. as well as The update module is used to perform mean and covariance matrix updates to optimize the baselines for conductivity, sulfate, pH, and fly ash heavy metal fingerprints.

Images