A multi-dimensional holographic data analysis method and system for hazardous waste regulation

Abstract

The application belongs to the technical field of data processing, and particularly relates to a multi-dimensional holographic data analysis method and system for hazardous waste supervision, which comprises the following steps: obtaining fluorescence spectrum data and mass spectrum data; preprocessing the obtained data, performing peak detection on the fluorescence intensity matrix after preprocessing based on the first derivative of peak intensity; combining the fluorescence spectrum of a standard sample to determine the weight of each peak point, complete the matching of peak pairs, and calculate the peak matching distance and the distribution difference of the characteristic vector; comprehensively considering the peak matching distance and the distribution difference of the characteristic vector, combining the matching ratio and the weight ratio, calculating the similarity score of the spectrum; calculating the weighted cosine similarity and the Euclidean distance between the processed mass spectrum data and the mass spectrum data of the standard sample, and completing the calculation of the similarity score of the mass spectrum between samples; and tracing the source of the hazardous waste according to the similarity score of the spectrum and the similarity score of the mass spectrum. Precise tracing is realized through qualitative and quantitative analysis.

Description

Technical Field

[0001] The present invention belongs to the field of data processing technology, and in particular relates to a multi-dimensional holographic data analysis method and system for hazardous waste supervision. Background Art

[0002] With the acceleration of industrialization, the generation of hazardous waste is increasing, and the new pollutants it contains pose a serious threat to the ecological environment and human health. In the field of environmental regulation, due to the difficulty of supervision and the sensitivity of the environment, outdoor areas have become a hotbed for illegal dumping and leakage of hazardous waste, posing a serious threat to the ecological environment and human health. In this situation, achieving accurate traceability and detection of hazardous materials in the wild is crucial.

[0003] In the wild, multiple hazardous substances such as organic pollutants, heavy metal pollutants, and radioactive substances may exist simultaneously. Most existing traceability algorithms simply compare the similarity indicators of one type of data. For example, they only trace the source based on the composition data of chemical substances. They are unable to comprehensively utilize multiple data information. When processing hazardous substance data in complex wild environments, misjudgments are prone to occur, resulting in a large deviation between the traceability results and the actual situation.

[0004] How to develop a multidimensional holographic data analysis method for hazardous waste supervision, realize efficient traceability detection of hazardous materials in the wild, and provide strong guarantees for wild environmental protection and ecological security is an urgent problem to be solved at this stage. Summary of the Invention

[0005] In order to achieve efficient tracing of hazardous objects in the wild, the present invention provides a multi-dimensional holographic data analysis method and system for hazardous waste supervision.

[0006] In a first aspect, the technical solution of the present invention provides a multi-dimensional holographic data analysis method for hazardous waste supervision, comprising:

[0007] The hazardous waste samples are tested using a three-dimensional fluorescence spectrometer and a mass spectrometer to obtain fluorescence spectrum data and mass spectrum data;

[0008] The acquired fluorescence spectrum data is preprocessed to obtain a preprocessed fluorescence intensity matrix, and the mass spectrum data is aligned and normalized;

[0009] Peak detection is performed using the first-order derivative of the peak intensity based on the preprocessed fluorescence intensity matrix;

[0010] The weight of each peak point is determined by combining the fluorescence spectrum of the standard sample to complete the peak pair matching, and the peak matching distance and the distribution difference of the eigenvector are calculated;

[0011] The similarity score of the spectrum is calculated by combining the peak matching distance and the distribution difference of the eigenvectors, the matching ratio and the weight ratio;

[0012] Calculate the weighted cosine similarity and Euclidean distance between the processed mass spectrum data and the mass spectrum data of the standard sample to complete the calculation of the mass spectrum similarity score between the samples;

[0013] Trace the source of hazardous waste based on the similarity score of the spectrum and the similarity score of the mass spectrum.

[0014] As a further limitation of the technical solution of the present invention, the step of preprocessing the acquired fluorescence spectrum data to obtain a preprocessed fluorescence intensity matrix includes:

[0015] Performing dot matrix data extraction and filtering on the fluorescence spectrum data to generate a fluorescence intensity matrix with the excitation wavelength as the row index, the emission wavelength as the column index, and the corresponding fluorescence intensity value as the matrix element;

[0016] Gaussian smoothing is performed on the fluorescence intensity matrix after dot matrix data extraction and filtering to obtain a preprocessed fluorescence intensity matrix.

[0017] As a further limitation of the technical solution of the present invention, the steps of aligning and normalizing the mass spectrometry data include:

[0018] Convert mass spectrometry data at different time steps into a unified format;

[0019] Analyze the stability of mass spectrometry data at each time step, and select the time step where the number of peaks is within the set range and the peak intensity fluctuation range is less than the set value as the reference time step;

[0020] Determine the mass-to-charge ratio range window and perform preliminary matching on the peaks falling within the window in the mass spectra of the reference time step and other time steps;

[0021] By comparing the mass-to-charge ratio differences of multiple matching peaks, taking the average value as the global offset, and then performing overall translation on the mass spectrum data of other time steps to align the mass-to-charge ratios of the matching peaks;

[0022] For the mass spectrometry data of each time step, the intensities of all peaks are arranged in ascending order, the intensity value at the middle position is taken as the median intensity of the time step, and the original intensity of each peak in the time step is divided by the median intensity of the time step to obtain the normalized mass spectrometry data. If the number of peaks is even, the average of the two middle intensity values ​​is taken as the median intensity.

[0023] As a further limitation of the technical solution of the present invention, the step of performing peak detection based on the preprocessed fluorescence intensity matrix using the first-order derivative of the peak intensity includes:

[0024] The preprocessed fluorescence intensity matrix data is differentiated along the excitation wavelength axis and the emission wavelength axis. The characteristic that the first-order derivative is 0 is utilized, and the point sets in the two axis directions and the Euclidean distance judgment are combined to determine the data points with key features in the fluorescence intensity matrix, namely the peak points, and generate a peak point set.

[0025] As a further limitation of the technical solution of the present invention, the steps of determining the weight of each peak point in combination with the fluorescence spectrum of the standard sample to complete the peak pair matching, and calculating the peak matching distance and the distribution difference of the characteristic vector include:

[0026] Calculate the Euclidean distance between the detected peak points, and merge the peak points within the distance threshold range according to the set distance threshold;

[0027] Determine the weight of each peak point, calculate the distance between all point pairs of the two groups of peak points and sort them, select the point pairs that meet the distance constraints, and complete the peak pair matching;

[0028] For the selected matching peak pairs, calculate the Euclidean distance between the matching peak pairs, then use the linear allocation algorithm to find the optimal matching pair set, and accumulate the minimum cost to obtain the peak matching distance;

[0029] A set number of points are selected around each peak, the eigenvectors are calculated, and the distribution difference of the eigenvectors is calculated using the Euclidean distance.

[0030] As a further limitation of the technical solution of the present invention, the formula in the step of calculating the similarity score of the spectrum is as follows, combining the peak matching distance and the distribution difference of the eigenvector, and combining the matching ratio and the weight ratio:

[0031]

[0032] Where, is the weight of the peak matching distance, is the weight of the distribution difference of the eigenvector, is the distribution difference of the feature vector, the peak matching distance , The peak point of the corresponding matching The Euclidean distance between the two, and the match coefficient is the matching coefficient. Optimal Match represents the optimal matching pair set obtained by the linear assignment algorithm. The peak matching distance is normalized, the distribution difference of the feature vectors is normalized, and a weighted sum is performed according to the set weight coefficient to obtain a preliminary matching coefficient value. The preliminary matching coefficient value is multiplied by the matching ratio to obtain the match coefficient. The ratio of the peak point that successfully matches the peak point of the standard sample under the distance and width constraints is defined as the matching ratio.

[0033] As a further limitation of the technical solution of the present invention, the steps of calculating the weighted cosine similarity and Euclidean distance between the processed mass spectrum data and the mass spectrum data of the standard sample to complete the calculation of the mass spectrum similarity score between the samples include:

[0034] The processed mass spectrometry data and the mass spectrometry data of the standard sample are respectively organized into vector forms, and the weight of each mass-to-charge ratio position is set. The weight vector is ; Vector of processed mass spectrometry data , the vector of mass spectrum data of standard sample is ;

[0035] Weighted cosine similarity ;

[0036] Euclidean distance ;

[0037] Mass spectrum similarity score ;

[0038] Where, Indicates the The weight of each mass-to-charge ratio position, is the weight of weighted cosine similarity, is the weight of the Euclidean distance, is the maximum value of the Euclidean distance.

[0039] As a further limitation of the technical solution of the present invention, the step of tracing the source of hazardous waste based on the similarity score of the spectrum and the similarity score of the mass spectrum includes:

[0040] The weight distribution between the similarity score of the spectrum and the similarity score of the mass spectrum is calculated to obtain the final similarity score of the hazardous waste and the standard sample, and the source of the hazardous waste is traced based on the final similarity score.

[0041] Fluorescence spectral data is used to perform qualitative analysis on the traceability results, and the accuracy of mass spectrometry data is used to quantitatively calculate the sample similarity. This allows the intelligent traceability algorithm to not only quickly respond to traceability needs, but also further perform more reasonable quantitative analysis on the traceability results to achieve accurate traceability.

[0042] In a second aspect, the technical solution of the present invention also provides a multi-dimensional holographic data analysis system for hazardous waste supervision, including a detection data acquisition module, a preprocessing module, a spectral data processing module, a mass spectrometry data processing module and a traceability processing module;

[0043] A detection data acquisition module is used to detect hazardous waste samples using a three-dimensional fluorescence spectrometer and a mass spectrometer to obtain fluorescence spectrum data and mass spectrum data;

[0044] A preprocessing module is used to preprocess the acquired fluorescence spectrum data to obtain a preprocessed fluorescence intensity matrix, and to align and normalize the mass spectrum data;

[0045] The spectral data processing module is used to perform peak detection based on the first-order derivative of the peak intensity based on the preprocessed fluorescence intensity matrix. The module also determines the weight of each peak point in combination with the fluorescence spectrum of the standard sample to complete peak pair matching and calculates the peak matching distance and the distribution difference of the eigenvector. The module also calculates the spectral similarity score by combining the peak matching distance and the distribution difference of the eigenvector with the matching ratio and weight ratio.

[0046] The mass spectrometry data processing module is used to calculate the weighted cosine similarity and Euclidean distance between the processed mass spectrometry data and the mass spectrometry data of the standard sample, and complete the calculation of the mass spectrometry similarity score between the samples;

[0047] The source tracing processing module is used to trace the source of hazardous waste based on the similarity score of the spectrum and the similarity score of the mass spectrum.

[0048] As a further limitation of the technical solution of the present invention, the preprocessing module includes a spectral data preprocessing unit, which is used to perform dot matrix data extraction and filtering on the fluorescence spectral data to generate a fluorescence intensity matrix with the excitation wavelength as the row index, the emission wavelength as the column index, and the corresponding fluorescence intensity value as the matrix element; Gaussian smoothing is performed on the fluorescence intensity matrix after dot matrix data extraction and filtering to obtain a preprocessed fluorescence intensity matrix.

[0049] As a further limitation of the technical solution of the present invention, the preprocessing module also includes a mass spectrum data preprocessing unit, which converts the mass spectrum data of different time steps into a unified format; analyzes the stability of the mass spectrum data of each time step, and selects the time step in which the number of peaks is within the set range and the peak intensity fluctuation range is less than the set value as the reference time step; determines the mass-to-charge ratio range window, and preliminarily matches the peaks falling within the window in the mass spectra of the reference time step and other time steps; by comparing the mass-to-charge ratio differences of multiple matching peaks, takes the average value as the global offset, and then performs an overall translation on the mass spectrum data of other time steps to align the mass-to-charge ratios of the matching peaks; for the mass spectrum data of each time step, arranges the intensities of all peaks in ascending order, takes the intensity value in the middle position as the median intensity of the time step, divides the original intensity of each peak in the time step by the median intensity of the time step, and obtains the normalized mass spectrum data, wherein, if the number of peaks is an even number, takes the average of the two middle intensity values ​​as the median intensity.

[0050] As a further limitation of the technical solution of the present invention, the spectral data processing module includes a peak detection unit for taking the derivative of the preprocessed fluorescence intensity matrix data along the excitation wavelength axis and the emission wavelength axis, utilizing the characteristic that the first-order derivative is 0, combining the point sets in the two axis directions and the Euclidean distance judgment to determine the data points with key features in the fluorescence intensity matrix, i.e., the peak points, and generate a peak point set.

[0051] As a further limitation of the technical solution of the present invention, the spectral data processing module also includes a peak pair processing unit, which is used to calculate the Euclidean distance between the detected peak points, and merge the peak points within the distance threshold range according to the set distance threshold; determine the weight of each peak point, calculate the distance of all point pairs for the two groups of peak points and sort them, select the point pairs that meet the distance constraints, and complete the peak pair matching; for the screened matching peak pairs, calculate the Euclidean distance between the matching peak pairs, and then use the linear allocation algorithm to find the optimal matching pair set, and accumulate the minimum cost to obtain the peak matching distance; select a set number of points around each peak, calculate the eigenvector, and then calculate the distribution difference of the eigenvector through the Euclidean distance.

[0052] As a further limitation of the technical solution of the present invention, the spectral data processing module includes a first calculation unit for calculating the similarity score of the spectrum; the calculation formula is as follows:

[0053]

[0054] Where, is the weight of the peak matching distance, is the weight of the distribution difference of the eigenvector, is the distribution difference of the feature vector, the peak matching distance , The peak point of the corresponding matching The Euclidean distance between the two, and the match coefficient is the matching coefficient. Optimal Match represents the optimal matching pair set obtained by the linear assignment algorithm. The peak matching distance is normalized, the distribution difference of the feature vectors is normalized, and a weighted sum is performed according to the set weight coefficient to obtain a preliminary matching coefficient value. The preliminary matching coefficient value is multiplied by the matching ratio to obtain the match coefficient. The ratio of the peak point that successfully matches the peak point of the standard sample under the distance and width constraints is defined as the matching ratio.

[0055] As a further limitation of the technical solution of the present invention, the mass spectrum data processing module calculates the mass spectrum similarity score using the following formula:

[0056] The processed mass spectrometry data and the mass spectrometry data of the standard sample are respectively organized into vector forms, and the weight of each mass-to-charge ratio position is set. The weight vector is ; Vector of processed mass spectrometry data , the vector of mass spectrum data of standard sample is ;

[0057] Weighted cosine similarity ;

[0058] Euclidean distance ;

[0059] Mass spectrum similarity score ;

[0060] Where, Indicates the The weight of each mass-to-charge ratio position, is the weight of weighted cosine similarity, is the weight of the Euclidean distance, is the maximum value of the Euclidean distance.

[0061] As a further limitation of the technical solution of the present invention, the traceability processing module is specifically used to calculate the weight distribution between the spectral similarity score and the mass spectrum similarity score, obtain the final similarity score between the hazardous waste and the standard sample, and trace the source of the hazardous waste based on the final similarity score.

[0062] The beneficial effect of the technical solution of the present invention is that through the combination of the two methods, a multi-dimensional holographic analysis of hazardous waste is achieved, providing a rich and comprehensive data basis for subsequent accurate traceability and supervision. This multi-dimensional evaluation method can more comprehensively and objectively reflect the similarity between samples, avoiding the one-sidedness that may be caused by a single-dimensional evaluation.

[0063] The fluorescence spectral data is preprocessed to obtain the preprocessed fluorescence intensity matrix. At the same time, the acquired mass spectrometry data is preprocessed including alignment and normalization. The alignment operation targets the data distribution offset generated by multiple time steps of the mass spectrometry data, so that the algorithm can use mass spectrometry data of multiple time steps rather than a single time step, thereby improving the stability and accuracy of the algorithm. The normalization operation is to enable the algorithm to more stably examine the similarity between different mass spectrometry samples from the perspective of data distribution rather than data size, so that the algorithm can better handle the fluctuations in the analysis process caused by sample concentration.

[0064] Peak detection is performed using the first-order derivative of the peak intensity based on the preprocessed fluorescence intensity matrix. The weight of each peak point is determined in combination with the fluorescence spectrum of the standard sample to complete peak pair matching. This method accurately extracts characteristic peaks from the fluorescence spectral data and appropriately weights the peaks based on the information of the standard sample, thereby more accurately reflecting the characteristics of hazardous waste. Furthermore, the peak matching distance and the distribution difference of the eigenvectors are calculated, comprehensively considering the peak position and the overall distribution of the eigenvectors, further improving the accuracy and reliability of feature extraction.

[0065] By calculating the similarity scores of the spectrum and the mass spectrum respectively, and evaluating the similarity between hazardous waste samples and standard samples from the two dimensions of fluorescence spectrum and mass spectrum, the possible sources of hazardous waste can be accurately identified, providing strong support for the supervision and management of hazardous waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for the description. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0067] Figure 1 A schematic flow chart of a method provided in an embodiment of the present invention.

[0068] Figure 2 A schematic block diagram of a system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions of the present invention will be clearly and completely described below in conjunction with the drawings in the specific embodiments. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0070] like Figure 1 As shown, an embodiment of the present invention provides a multi-dimensional holographic data analysis method for hazardous waste supervision, comprising:

[0071] S1: The hazardous waste samples are tested using a three-dimensional fluorescence spectrometer and a mass spectrometer to obtain fluorescence spectrum data and mass spectrum data respectively;

[0072] For liquid hazardous waste samples, if there is suspended matter in the sample, it must be removed through a membrane filter to prevent scattering interference with the fluorescence signal. If the sample concentration is too high, it must be diluted with deionized water to bring the fluorescence intensity within the linear detection range of the instrument.

[0073] For solid hazardous waste samples, first grind them. Then, a certain amount of sample is weighed and extracted with an organic solvent or buffer solution to dissolve the target fluorescent substance in the extract. After filtering or centrifuging the extract, the supernatant is collected for testing. Parameters such as the excitation wavelength range, emission wavelength range, scan speed, and integration time are set based on the sample properties and the testing objectives. For example, for common organic pollutants, the excitation wavelength range can be set to 200-400 nm, and the emission wavelength range to 250-600 nm. The scan speed is generally set to around 1200 nm / min, and the integration time is adjusted appropriately based on the sample fluorescence intensity, typically 0.1-1 s. The processed sample is placed in a quartz cuvette and placed in the sample cell of the 3D fluorescence spectrometer. The instrument is started and scanned, automatically recording the fluorescence intensity at different excitation and emission wavelength combinations to generate 3D fluorescence spectral data. After the scan is complete, the data is saved and initially processed, such as by subtracting the fluorescence signal from a blank sample. It is possible that 3D fluorescence spectral data cannot be obtained at this step of the test. If this occurs, direct analysis and traceability using mass spectrometry data is performed.

[0074] The pre-treated sample is injected into the mass spectrometer, the sample is ionized in the ion source, the formed ions are separated according to the mass-to-charge ratio in the mass analyzer, and finally the detector detects and records the intensity and mass-to-charge ratio information of the ions to obtain mass spectrum data.

[0075] S2: Preprocessing the acquired fluorescence spectrum data to obtain a preprocessed fluorescence intensity matrix, and aligning and normalizing the mass spectrum data;

[0076] The original three-dimensional fluorescence spectral data may contain a large amount of redundant and irrelevant information, such as background noise from the instrument itself, scattered light unrelated to the target analyte, etc. Through dot matrix data extraction and filtering, it is possible to select data points that are valuable for analysis and remove irrelevant or interfering data. In addition, the amount of three-dimensional fluorescence spectral data is usually large, and direct processing may face problems such as high computational complexity and low analysis efficiency. Dot matrix data extraction can simplify and reduce the dimensionality of the data while retaining key information, converting complex three-dimensional data into a dot matrix form that is easier to process. The specific processing process in this step includes:

[0077] S21: performing dot matrix data extraction and filtering on the fluorescence spectrum data to generate a fluorescence intensity matrix with the excitation wavelength as the row index, the emission wavelength as the column index, and the corresponding fluorescence intensity value as the matrix element; performing Gaussian smoothing on the fluorescence intensity matrix after dot matrix data extraction and filtering to obtain a preprocessed fluorescence intensity matrix.

[0078] First, the matrix is ​​converted into dot matrix data. The dot matrix data is presented in the form of three-dimensional coordinates. Each point contains the excitation wavelength EX, emission wavelength EM and intensity (that is, weight). In order to highlight the effective information, the points with lower weight are filtered out by the formula To filter, set the excitation wavelength to , the emission wavelength is set to , the strength (weight) is set to , filter out valid data points based on the intensity (weight) threshold. is the maximum weight in the current point set, r is the weight threshold ratio. In the embodiment of the present invention, based on experience, r Set it between 0.2 and 0.5. This step can improve data processing efficiency and accuracy by removing data points that have little contribution to subsequent analysis. For example, it can reduce redundant information interference when processing large amounts of fluorescence spectral data.

[0079] Fluorescence intensity matrix F ( x, y ) Apply Gaussian smoothing to reduce noise, the formula is Where, is the Gaussian kernel function, is the smoothing parameter. By adjusting The value controls the degree of smoothing. This operation can effectively reduce the noise in the data, prevent the noise from interfering with peak detection, and improve the accuracy of peak detection.

[0080] During fluorescence spectroscopy measurements, various random noises, such as electronic noise and environmental interference, are inevitably introduced. These noises can cause spectral curves to fluctuate and become uneven, hindering the accurate identification and analysis of spectral features. Gaussian smoothing effectively suppresses this random noise by performing weighted averaging on the data, making the spectral curve smoother, highlighting the true spectral signal, and improving data quality and reliability.

[0081] After point-wise data extraction and filtering, the data points may be somewhat discrete or discontinuous. Gaussian smoothing can fill the gaps between data points to a certain extent, making the data more continuous and smooth in a local area. This facilitates subsequent spectral analysis, such as peak detection, and produces more accurate and stable results.

[0082] S22: Convert the mass spectrometry data of different time steps into a unified format to ensure that each data file contains the mass-to-charge ratio and the corresponding peak intensity information; analyze the stability of the mass spectrometry data of each time step, and select the time step in which the number of peaks is within the set range and the peak intensity fluctuation range is less than the set value as the reference time step; determine the mass-to-charge ratio range window, and preliminarily match the peaks that fall within the window in the mass spectra of the reference time step and other time steps; by comparing the mass-to-charge ratio differences of multiple matching peaks, take the average value as the global offset, and then perform an overall translation on the mass spectrometry data of other time steps to align the mass-to-charge ratios of the matching peaks; for the mass spectrometry data of each time step, arrange the intensities of all peaks in ascending order, take the intensity value in the middle position as the median intensity of the time step, divide the original intensity of each peak in the time step by the median intensity of the time step to obtain the normalized mass spectrometry data, where if the number of peaks is an even number, take the average of the two middle intensity values ​​as the median intensity.

[0083] S3: Peak detection is performed using the first-order derivative of the peak intensity based on the preprocessed fluorescence intensity matrix; specifically, the following steps are performed: differentiating the preprocessed fluorescence intensity matrix data along the excitation wavelength axis and the emission wavelength axis, utilizing the characteristic that the first-order derivative is 0, combining the point sets in the two axis directions and the Euclidean distance judgment to determine the data points with key features in the fluorescence intensity matrix, i.e., the peak points, and generating a peak point set.

[0084] Using the first-order derivative of the peak intensity, the data is differentiated along the EX and EM axes. Taking advantage of the fact that the first-order derivative is 0, the peak point is determined by combining the point sets in the two axis directions and the Euclidean distance judgment. The formula is as follows:

[0085] ,

[0086] Along the EX axis, according to the formula , traverse the data to determine whether Point set ;here Traverse different positions on the EX axis, Fixed at a certain position, find all the coordinate combinations that make the function value 0 and the corresponding F value to form a set ;

[0087] Along the EM axis, according to the formula ,akin, fixed, Traverse different positions on the EM axis and find the one that satisfies Point set ; is a collection and The spatial coordinates of the midpoint;

[0088] Traversal Set, Compute Set Each point in the set The Euclidean distance of all points in the ,select the point with Euclidean distance equal to 0 as the peak point.

[0089] S4: Determine the weight of each peak point based on the fluorescence spectrum of the standard sample to complete the peak pair matching, and calculate the peak matching distance and the distribution difference of the eigenvector;

[0090] This step specifically includes:

[0091] S41: Calculate the Euclidean distance between the detected peak points, and merge the peak points within the distance threshold range according to the set distance threshold;

[0092] For two points in the peak set and Calculate its two-dimensional Euclidean distance , providing data support for subsequent deduplication and merging operations. According to the set distance threshold , merge the peak points whose calculated distance is less than the distance threshold to obtain the new peak center, the coordinates are , the calculation formula is ,in n is the number of points that meet the distance constraint. This step avoids peak recounting caused by data errors or instrument noise, making the peak information more accurate.

[0093] S42: Determine the weight of each peak point, calculate the distances of all point pairs between the two groups of peak points, sort them, select the point pairs that meet the distance constraints, and complete the peak pair matching;

[0094] Find the corresponding weight (intensity value) through the detected points , is the fluorescence intensity matrix after Gaussian smoothing. Represents the final weight of the point after weight distribution, Indicates the current point Search within the neighborhood The maximum value of , where the neighborhood refers to the set of points within the set range centered on the current point;

[0095] For two sets of peak points and , calculate the distance of all point pairs and sort them, select the point pairs that meet the distance constraint, the formula is , providing matching point pairs for calculating peak matching distance. Specifically, in the two sets of peak point sets and , calculate all possible point pairs Euclidean distance . Set the distance to be less than or equal to the set distance threshold The point pairs are selected to form Matched Pairs, that is, the set of matched peak pairs.

[0096] S43: For the selected matching peak pairs, calculate the Euclidean distance between the matching peak pairs, then use the linear allocation algorithm to find the optimal matching pair set, and accumulate the minimum cost to obtain the peak matching distance;

[0097] For the two corresponding matching peak sets peaks1 and peaks2, first calculate the Euclidean distance between them ;

[0098] Then use the linear allocation algorithm to find the optimal matching pair set and accumulate the minimum cost to get the peak matching distance , The peak point of the corresponding matching Optimal Match represents the optimal matching pair set obtained by the linear assignment algorithm.

[0099] In the embodiment of the present invention, the process of obtaining the optimal matching pair set by the linear assignment algorithm is as follows:

[0100] For the two corresponding matching pairs of peaks peaks1 and peaks2, there are m peak points in peaks1 and n peak points in peaks2, calculate the Euclidean distance In the formula It is the first The coordinates of the peak points, Peaks2 The coordinates of the peak points are calculated to form a The cost matrix , the elements in the matrix .

[0101] For the cost matrix For each row of the matrix, find the minimum value in that row, then subtract this minimum value from each element in that row. The goal of this step is to ensure that each row has at least one zero element. For each column of the matrix after row reduction, find the minimum value in that column, then subtract this minimum value from each element in that column. Again, this will ensure that each column has at least one zero element.

[0102] Find as many independent zero elements as possible in a matrix (that is, zero elements that are located in different rows and columns). This can be done using notation:

[0103] First, check every zero element in the matrix. Mark any zero element whose row and column have no other marked zero elements as independent zero elements. Repeat the above steps until no more independent zero elements can be marked.

[0104] If the number of independent zero elements is equal to , it means that the optimal match has been found, and the peak point pairs corresponding to the rows and columns where these independent zero elements are located are the optimal matching pair sets.

[0105] If the number of independent zero elements is less than , then the matrix needs to be adjusted until the number of independent zero elements is equal to Here, the steps of the progressive matrix adjustment include: using the minimum number of lines (row lines and column lines) to cover all zero elements in the matrix, finding the minimum value K among the elements not covered by the lines, then subtracting the minimum value K from each element not covered by the lines, adding the minimum value K to each element covered by two lines, and then repeating the check and subsequent steps for each zero element in the matrix.

[0106] In one embodiment, the linear assignment algorithm is implemented using the linear_sum_assignment function in the scipy library. Furthermore, the linear_sum_assignment function in the scipy library processes the cost matrix to obtain the optimal matching row and column indices. Based on the obtained row and column indices, the corresponding peak coordinates are found from peaks1 and peaks2 to form an optimal matching pair set.

[0107] S44: Select a set number of points around each peak, calculate the eigenvector, and then calculate the distribution difference of the eigenvectors using Euclidean distance.

[0108] Select n points around each peak and calculate the eigenvector , use Euclidean distance to calculate the distribution difference of feature vectors .

[0109] Specifically, represents the eigenvector calculated from the points selected around each peak, Indicates the weights (intensity values) of n points selected around each peak, and calculates the average weight of these n points , and then normalize this average value to obtain the eigenvector .

[0110] Represents the distribution difference between two eigenvectors, which is used to measure the difference in the characteristics of the areas around the two peaks. and These are the eigenvectors corresponding to the two different peaks. Calculate the sum of the squares of the differences between the elements of the two eigenvectors and then take the square root. A larger sum indicates a greater difference in the characteristics of the areas surrounding the two peaks; conversely, a smaller difference indicates a smaller difference.

[0111] S5: Calculate the similarity score of the spectrum by combining the peak matching distance and the distribution difference of the eigenvectors, the matching ratio and the weight ratio;

[0112] The formula for calculating the similarity score of the spectrum is as follows:

[0113]

[0114] Where, is the weight of the peak matching distance, is the weight of the distribution difference of the eigenvector, is the distribution difference of the feature vectors; match coefficient is the matching coefficient. A preliminary matching coefficient value is obtained by normalizing the peak matching distance and the distribution difference of the feature vectors, and then weighted summing them according to the set weight coefficient. This preliminary matching coefficient value is multiplied by the matching ratio to obtain the matching coefficient. The ratio of successful matches between the detected peak points and the peak points of the standard sample is defined as the matching ratio. The detected peak points here refer to the peak points that meet the distance and width constraints during peak detection.

[0115] The distance constraint refers to the distance between peaks, that is, the spacing between different detected peaks in the coordinate space of the fluorescence intensity matrix. For example, if the distance between two adjacent peaks is less than a set threshold, they may be considered to be multiple pseudo-peaks caused by factors such as noise interference, and only one of them will be retained as the true peak.

[0116] The width constraint is the peak's half-maximum width, typically defined as the width of the horizontal coordinate (e.g., wavelength) corresponding to the peak at half the peak intensity. If the detected "peak" width is too narrow, it may be a spike caused by noise rather than a true fluorescence signal peak. If it is too wide, it may be due to overlapping peaks or signal broadening caused by instrument resolution issues, requiring further analysis or data processing to isolate the true peak.

[0117] S6: Calculate the weighted cosine similarity and Euclidean distance between the processed mass spectrum data and the mass spectrum data of the standard sample to complete the calculation of the mass spectrum similarity score between the samples;

[0118] In this step, the processed mass spectrometry data and the mass spectrometry data of the standard sample are organized into vectors, and the weight of each peak position is set. (Here, a weight is assigned to each mass-to-charge ratio position according to the actual situation. The weight can be determined based on factors such as the importance and stability of the peak. For example, some characteristic peaks have a high degree of discrimination for the sample and can be given a higher weight.) The weight vector is ; Vector of processed mass spectrometry data , the vector of mass spectrum data of standard sample is ;

[0119] Weighted cosine similarity , the value range of weighted cosine similarity is between [-1, 1]. The closer the value is to 1, the more similar the directions of the two vectors are;

[0120] Euclidean distance , the larger the value of Euclidean distance, the greater the difference between the two vectors;

[0121] Mass spectrum similarity score ;

[0122] Where, represents the weight of the i-th mass-to-charge ratio position, is the weight of weighted cosine similarity, is the weight of the Euclidean distance, It is the maximum value of the Euclidean distance and is used to normalize the Euclidean distance to the interval [0,1] so that it can be combined with the weighted cosine similarity on the same scale.

[0123] In an embodiment of the present invention, the preprocessing of the acquired mass spectrometry data includes alignment operation and normalization processing, wherein the alignment operation targets the data distribution offset generated by multiple time steps of the mass spectrometry data, so that the algorithm can use mass spectrometry data of multiple time steps rather than a single time step, thereby improving the stability and accuracy of the algorithm. The normalization processing is to enable the algorithm to more stably examine the similarity between different mass spectrometry samples from the perspective of data distribution rather than data size, so that the algorithm can better handle the fluctuations in the analysis process caused by the sample concentration. After the alignment operation and normalization processing, two feature vectors are obtained, representing two samples respectively, and the weighted cosine similarity and Euclidean distance between the two vectors are calculated; the weighted cosine similarity is used to calculate the similarity of the mass spectrometry shape, and the Euclidean distance is used to measure the absolute difference in the mass spectrometry data. Through this calculation method, the algorithm can complete the mass spectrometry similarity measurement calculation between samples from multiple aspects such as local shape (weighted) and global distribution.

[0124] S7: Trace the source of hazardous waste based on the similarity score of the spectrum and the similarity score of the mass spectrum.

[0125] Specifically, by calculating the weight distribution between the similarity score of the spectrum and the similarity score of the mass spectrum, the final similarity score of the hazardous waste and the standard sample is obtained, and the source of the hazardous waste is traced based on the final similarity score.

[0126] It should be noted that the relevant data of the standard samples used in the present invention are all collected through reasonable and compliant methods from hazardous wastes of various enterprises within the set detection range, and the detection data are pre-stored to facilitate subsequent monitoring of hazardous wastes for comparison, analysis and tracing.

[0127] like Figure 2 As shown, an embodiment of the present invention further provides a multi-dimensional holographic data analysis system for hazardous waste supervision, including a detection data acquisition module, a preprocessing module, a spectrum data processing module, a mass spectrum data processing module and a traceability processing module;

[0128] A detection data acquisition module is used to detect hazardous waste samples using a three-dimensional fluorescence spectrometer and a mass spectrometer to obtain fluorescence spectrum data and mass spectrum data;

[0129] A preprocessing module is used to preprocess the acquired fluorescence spectrum data to obtain a preprocessed fluorescence intensity matrix, and to align and normalize the mass spectrum data;

[0130] The spectral data processing module is used to perform peak detection based on the first-order derivative of the peak intensity based on the preprocessed fluorescence intensity matrix. The module also determines the weight of each peak point in combination with the fluorescence spectrum of the standard sample to complete peak pair matching and calculates the peak matching distance and the distribution difference of the eigenvector. The module also calculates the spectral similarity score by combining the peak matching distance and the distribution difference of the eigenvector with the matching ratio and weight ratio.

[0131] The mass spectrometry data processing module is used to calculate the weighted cosine similarity and Euclidean distance between the processed mass spectrometry data and the mass spectrometry data of the standard sample, and complete the calculation of the mass spectrometry similarity score between the samples;

[0132] The source tracing processing module is used to trace the source of hazardous waste based on the similarity score of the spectrum and the similarity score of the mass spectrum.

[0133] In some embodiments, the preprocessing module includes a spectral data preprocessing unit, which is used to perform dot matrix data extraction and filtering on the fluorescence spectral data to generate a fluorescence intensity matrix with the excitation wavelength as the row index, the emission wavelength as the column index, and the corresponding fluorescence intensity value as the matrix element; Gaussian smoothing is performed on the fluorescence intensity matrix after dot matrix data extraction and filtering to obtain a preprocessed fluorescence intensity matrix.

[0134] In some embodiments, the preprocessing module also includes a mass spectrum data preprocessing unit, which converts the mass spectrum data of different time steps into a unified format; analyzes the stability of the mass spectrum data of each time step, and selects the time step in which the number of peaks is within a set range and the peak intensity fluctuation range is less than a set value as the reference time step; determines the mass-to-charge ratio range window, and preliminarily matches the peaks that fall within the window in the mass spectrum of the reference time step and other time steps; by comparing the mass-to-charge ratio differences of multiple matching peaks, takes the average value as the global offset, and then performs an overall translation on the mass spectrum data of other time steps to align the mass-to-charge ratios of the matching peaks; for the mass spectrum data of each time step, arranges the intensities of all peaks in ascending order, takes the intensity value in the middle position as the median intensity of the time step, divides the original intensity of each peak in the time step by the median intensity of the time step, and obtains the normalized mass spectrum data, wherein, if the number of peaks is an even number, takes the average of the two middle intensity values ​​as the median intensity.

[0135] In some embodiments, the spectral data processing module includes a peak detection unit for taking the derivative of the preprocessed fluorescence intensity matrix data along the excitation wavelength axis and the emission wavelength axis, utilizing the characteristic that the first-order derivative is 0, combining the point sets in the two axis directions and the Euclidean distance judgment to determine the data points with key features in the fluorescence intensity matrix, i.e., the peak points, and generating a peak point set.

[0136] In some embodiments, the spectral data processing module also includes a peak pair processing unit for calculating the Euclidean distance between the detected peak points, and merging the peak points within the distance threshold range according to a set distance threshold; determining the weight of each peak point, calculating the distance of all point pairs for the two groups of peak points and sorting them, selecting the point pairs that meet the distance constraints, and completing the peak pair matching; for the screened matching peak pairs, calculating the Euclidean distance between the matching peak pairs, and then using the linear allocation algorithm to find the optimal matching pair set, and accumulating the minimum cost to obtain the peak matching distance; selecting a set number of points around each peak, calculating the eigenvector, and then calculating the distribution difference of the eigenvector through the Euclidean distance.

[0137] In some embodiments, the spectral data processing module includes a first calculation unit for calculating a spectral similarity score; the calculation formula is as follows:

[0138]

[0139] Where, is the weight of the peak matching distance, is the weight of the distribution difference of the eigenvector, is the distribution difference of the feature vector, the peak matching distance , The peak point of the corresponding matching The Euclidean distance between the two, and the match coefficient is the matching coefficient. Optimal Match represents the optimal matching pair set obtained by the linear assignment algorithm. The peak matching distance is normalized, the distribution difference of the feature vectors is normalized, and a weighted sum is performed according to the set weight coefficient to obtain a preliminary matching coefficient value. The preliminary matching coefficient value is multiplied by the matching ratio to obtain the match coefficient. The ratio of the peak point that successfully matches the peak point of the standard sample under the distance and width constraints is defined as the matching ratio.

[0140] In some embodiments, the mass spectrum data processing module calculates the mass spectrum similarity score using the following formula:

[0141] The processed mass spectrometry data and the mass spectrometry data of the standard sample are respectively organized into vector forms, and the weight of each mass-to-charge ratio position is set. The weight vector is ; Vector of processed mass spectrometry data , the vector of mass spectrum data of standard sample is ;

[0142] Weighted cosine similarity ;

[0143] Euclidean distance ;

[0144] Mass spectrum similarity score ;

[0145] Where, represents the weight of the i-th mass-to-charge ratio position, is the weight of weighted cosine similarity, is the weight of the Euclidean distance, is the maximum value of the Euclidean distance.

[0146] In some embodiments, the traceability processing module is specifically used to calculate the weight distribution between the spectral similarity score and the mass spectrum similarity score, obtain the final similarity score between the hazardous waste and the standard sample, and trace the source of the hazardous waste based on the final similarity score.

[0147] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multidimensional holographic data analysis method for hazardous waste supervision, characterized in that: include: The hazardous waste samples are tested using a three-dimensional fluorescence spectrometer and a mass spectrometer to obtain fluorescence spectrum data and mass spectrum data; The acquired fluorescence spectrum data is preprocessed to obtain a preprocessed fluorescence intensity matrix, and the mass spectrum data is aligned and normalized; Peak detection is performed using the first-order derivative of the peak intensity based on the preprocessed fluorescence intensity matrix; The weight of each peak point is determined by combining the fluorescence spectrum of the standard sample to complete the peak pair matching, and the peak matching distance and the distribution difference of the eigenvector are calculated; The similarity score of the spectrum is calculated by combining the peak matching distance and the distribution difference of the eigenvectors, the matching ratio and the weight ratio; Calculate the weighted cosine similarity and Euclidean distance between the processed mass spectrum data and the mass spectrum data of the standard sample to complete the calculation of the mass spectrum similarity score between the samples; Trace the source of hazardous waste based on the similarity score of the spectrum and the similarity score of the mass spectrum.

2. The multidimensional holographic data analysis method for hazardous waste supervision according to claim 1, characterized in that: The steps of preprocessing the acquired fluorescence spectrum data to obtain a preprocessed fluorescence intensity matrix include: Performing dot matrix data extraction and filtering on the fluorescence spectrum data to generate a fluorescence intensity matrix with the excitation wavelength as the row index, the emission wavelength as the column index, and the corresponding fluorescence intensity value as the matrix element; Gaussian smoothing is performed on the fluorescence intensity matrix after dot matrix data extraction and filtering to obtain a preprocessed fluorescence intensity matrix.

3. The multidimensional holographic data analysis method for hazardous waste supervision according to claim 2, characterized in that: The steps for aligning and normalizing mass spectrometry data include: Convert mass spectrometry data at different time steps into a unified format; Analyze the stability of mass spectrometry data at each time step, and select the time step where the number of peaks is within the set range and the peak intensity fluctuation range is less than the set value as the reference time step; Determine the mass-to-charge ratio range window and perform preliminary matching on the peaks falling within the window in the mass spectra of the reference time step and other time steps; By comparing the mass-to-charge ratio differences of multiple matching peaks, taking the average value as the global offset, and then performing overall translation on the mass spectrum data of other time steps to align the mass-to-charge ratios of the matching peaks; For the mass spectrometry data of each time step, the intensities of all peaks are arranged in ascending order, the intensity value at the middle position is taken as the median intensity of the time step, and the original intensity of each peak in the time step is divided by the median intensity of the time step to obtain the normalized mass spectrometry data. If the number of peaks is even, the average of the two middle intensity values ​​is taken as the median intensity.

4. The multidimensional holographic data analysis method for hazardous waste supervision according to claim 3, characterized in that: The steps of performing peak detection using the first derivative of the peak intensity based on the preprocessed fluorescence intensity matrix include: The preprocessed fluorescence intensity matrix data is differentiated along the excitation wavelength axis and the emission wavelength axis. The characteristic that the first-order derivative is 0 is utilized, and the point sets in the two axis directions and the Euclidean distance judgment are combined to determine the data points with key features in the fluorescence intensity matrix, namely the peak points, and generate a peak point set.

5. The multi-dimensional holographic data analysis method for hazardous waste supervision according to claim 4, characterized in that: The steps of determining the weight of each peak point by combining the fluorescence spectrum of the standard sample to complete the peak pair matching, and calculating the peak matching distance and the distribution difference of the characteristic vector include: Calculate the Euclidean distance between the detected peak points, and merge the peak points within the distance threshold range according to the set distance threshold; Determine the weight of each peak point, calculate the distance between all point pairs of the two groups of peak points and sort them, select the point pairs that meet the distance constraints, and complete the peak pair matching; For the selected matching peak pairs, calculate the Euclidean distance between the matching peak pairs, then use the linear allocation algorithm to find the optimal matching pair set, and accumulate the minimum cost to obtain the peak matching distance; A set number of points are selected around each peak, the eigenvectors are calculated, and the distribution difference of the eigenvectors is calculated using the Euclidean distance.

6. The multi-dimensional holographic data analysis method for hazardous waste supervision according to claim 5, characterized in that: The formula for calculating the similarity score of the spectrum is as follows, combining the peak matching distance and the distribution difference of the eigenvectors, the matching ratio and the weight ratio: Where, is the weight of the peak matching distance, is the weight of the distribution difference of the eigenvector, is the distribution difference of the feature vector, the peak matching distance , The peak point of the corresponding matching The Euclidean distance between them, match coefficient is the matching coefficient; Optimal Match represents the optimal matching pair set obtained by the linear assignment algorithm; Among them, by normalizing the peak matching distance, normalizing the distribution difference of the feature vector, and performing weighted summation according to the set weight coefficient, a preliminary matching coefficient value is obtained, and the preliminary matching coefficient value is multiplied by the matching ratio to obtain the match coefficient; the ratio of successful matching between the detected peak point and the peak point of the standard sample is defined as the matching ratio.

7. The multi-dimensional holographic data analysis method for hazardous waste supervision according to claim 6, characterized in that: The steps of calculating the weighted cosine similarity and Euclidean distance between the processed mass spectrum data and the mass spectrum data of the standard sample and completing the calculation of the mass spectrum similarity score between the samples include: The processed mass spectrometry data and the mass spectrometry data of the standard sample are respectively organized into vector forms, and the weight of each mass-to-charge ratio position is set. The weight vector is ; Vector of processed mass spectrometry data , the vector of mass spectrum data of standard sample is ; Weighted cosine similarity ; Euclidean distance ; Mass spectrum similarity score ; Where, Indicates the The weight of each mass-to-charge ratio position, is the weight of weighted cosine similarity, is the weight of the Euclidean distance, is the maximum value of the Euclidean distance.

8. The multi-dimensional holographic data analysis method for hazardous waste supervision according to claim 7, characterized in that: The steps for tracing the source of hazardous waste based on the similarity scores of spectra and mass spectra include: The weight distribution between the similarity score of the spectrum and the similarity score of the mass spectrum is calculated to obtain the final similarity score of the hazardous waste and the standard sample, and the source of the hazardous waste is traced based on the final similarity score.

9. A multi-dimensional holographic data analysis system for hazardous waste supervision, characterized in that: It includes detection data acquisition module, preprocessing module, spectrum data processing module, mass spectrum data processing module and traceability processing module; A detection data acquisition module is used to detect hazardous waste samples using a three-dimensional fluorescence spectrometer and a mass spectrometer to obtain fluorescence spectrum data and mass spectrum data; A preprocessing module is used to preprocess the acquired fluorescence spectrum data to obtain a preprocessed fluorescence intensity matrix, and to align and normalize the mass spectrum data; The spectral data processing module is used to perform peak detection based on the first-order derivative of the peak intensity based on the preprocessed fluorescence intensity matrix. The module also determines the weight of each peak point in combination with the fluorescence spectrum of the standard sample to complete peak pair matching and calculates the peak matching distance and the distribution difference of the eigenvector. The module also calculates the spectral similarity score by combining the peak matching distance and the distribution difference of the eigenvector with the matching ratio and weight ratio. The mass spectrometry data processing module is used to calculate the weighted cosine similarity and Euclidean distance between the processed mass spectrometry data and the mass spectrometry data of the standard sample, and complete the calculation of the mass spectrometry similarity score between the samples; The source tracing processing module is used to trace the source of hazardous waste based on the similarity score of the spectrum and the similarity score of the mass spectrum.

10. The multi-dimensional holographic data analysis system for hazardous waste management according to claim 9, characterized in that: The formula for calculating the similarity score of the spectrum in the spectral data processing module is as follows: Where, is the weight of the peak matching distance, is the weight of the distribution difference of the eigenvector, is the distribution difference of the feature vector, the peak matching distance , The peak point of the corresponding matching The Euclidean distance between the two, and the match coefficient is the matching coefficient; Optimal Match represents the optimal matching pair set obtained by the linear assignment algorithm; wherein, the peak matching distance is normalized, the distribution difference of the feature vector is normalized, and a weighted sum is performed according to the set weight coefficient to obtain a preliminary matching coefficient value, which is multiplied by the matching ratio to obtain the match coefficient; the ratio of successful matches between the detected peak points and the peak points of the standard sample is defined as the matching ratio.

Images