Method for calculating mountain packet signal volume in comprehensive two-dimensional chromatography

Abstract

The invention discloses a method for calculating the volume of a mountain packet signal in comprehensive two-dimensional chromatography, and the method comprises the following steps: S1, obtaining a sampling signal matrix U of which the size is M * N; s2, a convolution kernel A is set, the size of the convolution kernel A is m * n, m is the maximum peak width of all peaks of the comprehensive two-dimensional data sampling signals presented in the chromatogram on the Y axis, and n is the maximum peak width of all peaks on the X axis; s3, constructing a mountain signal matrix U1: carrying out class convolution calculation on the sampling signal matrix U by using the convolution kernel A, and obtaining the mountain signal matrix U1 according to a minimum element value z in a local matrix of the sampling signal matrix U corresponding to the size of the convolution kernel A; and S4, calculating the volume SU1 of the mountain, and further obtaining a quantitative analysis result of the target substance corresponding to the shape of the mountain. According to the method, for a substance chromatogram with bumps or humps, the bumps or humps can be effectively stripped from data, the volume of the bumps or humps can be effectively calculated, and a basis is provided for quantitative calculation of substances with peaks.

Description

[Technical field]

[0001] The invention belongs to the technical field of analytical chemistry, and particularly relates to a method for calculating the volume of a mountain signal in a full two-dimensional chromatogram. [Background technology]

[0002] Chromatograph is an instrument that uses chromatography analysis technology to conduct qualitative and quantitative analysis of substances, as well as study the physical and chemical properties of substances. Chromatography is a physical separation technology that uses the distribution difference between the components of a mixture in two immiscible phases (stationary phase and mobile phase) to perform multiple equilibriums in the two phases to achieve separation of the mixture. It can analyze almost all known substances and has a wide range of applications in many scientific fields.

[0003] Hydrocarbon mineral oil is an extremely complex hydrocarbon mixture, the number of carbon atoms of which is generally between 10 and 50, mainly including saturated hydrocarbon mineral oil (MOSH) composed of straight chain, branched chain and cyclic alkanes and aromatic hydrocarbon mineral oil (MOAH) composed of alkylated polyaromatic compounds. The composition of mineral oil is very complex, containing a large number of isomers, and even on the full two-dimensional spectrum, it is impossible to form a complete separation, showing a "bulge" shape. Some substances also contain some interfering substances (such as plasticizers) or standard samples, and these interfering substances will appear in the form of "spikes", some "spikes" will appear alone, and some "spikes" will appear on the "bulge", showing a "hump" shape. For this kind of signal data with "bulges" or "humps", how to calculate the content of the target substance has become a major problem. The prior art has not yet disclosed an integral calculation method for peaks, "bulges" or "humps" for the above signal data.

[0004] Therefore, it is necessary to provide a new method for calculating the volume of the mountain signal in the full two-dimensional chromatogram to solve the above technical problems. [Summary of the invention]

[0005] The main purpose of the present invention is to provide a method for calculating the volume of humps in a full two-dimensional chromatogram. For a chromatogram of a material with humps or humps, the humps or humps can be effectively stripped from the data and their volumes can be calculated, providing a basis for the quantitative calculation of materials showing sharp peaks.

[0006] The present invention achieves the above object through the following technical solution: a method for calculating the volume of a mountain signal in a full two-dimensional chromatogram, comprising the following steps:

[0007] S1. Obtain full two-dimensional data sampling signal u ij , we get the sampling signal matrix U = {u ij}, i = 1, 2, ..., M; j = 1, 2, ..., N; the size of the sampling signal matrix U is M × N;

[0008] S2. Set the convolution kernel A with a size of m×n, where m is the maximum peak width of all peaks of the two-dimensional data sampling signal on the Y-axis in the chromatogram, and n is the maximum peak width of all peaks of the two-dimensional data sampling signal on the X-axis in the chromatogram;

[0009] S3. Construct the hill signal matrix U1: The size of the hill signal matrix U1 is the same as that of the sampling signal matrix U. Perform a convolution-like calculation on the sampling signal matrix U with the convolution kernel A. The convolution-like calculation is the minimum element value z in the local matrix corresponding to the size of the convolution kernel A in the sampling signal matrix U, to obtain the hill signal matrix U1;

[0010] S4. Calculate the volume of the area where the hill signal matrix U1 appears in the chromatogram to obtain the hill volume S U1 , and further obtain the quantitative analysis result of the target substance corresponding to the hill shape.

[0011] Further, the S3 includes the following steps:

[0012] S31. Let the initial values p = 1 and q = 1;

[0013] S32. Extract the p-th to m + p - 1-th rows and the q-th to n + q - 1-th columns from the sampling signal matrix U to obtain the local matrix Z pq ;

[0014] S33. Take the minimum value z of all elements in the local matrix Z pq = min{u pq}, i = p, p + 1, ……, m + p - 1, j = q, q + 1, ……, n + q - 1, as the element value at the p-th row and p-th column position in the hill signal matrix U1; ij}

[0015] S34. Judge whether q < N - n + 1 holds. If it holds, then q++, and return to step S32. Otherwise, execute the next step;

[0016] S35. Judge whether p < M - m + 1 holds. If it holds, then p++, and return to step S32. Otherwise, execute the next step;

[0017] S36. Fill all element values in the (M - m + 2)-th to M-th rows and the (N - n + 2)-th to N-th columns in the hill signal matrix U1 with 0 to obtain the hill signal matrix U1;

[0018] Further, the calculation of the hill volume S U1 includes the following steps:

[0019] S41. Divide the XY-axis plane into several 1×1 cells;

[0020] S42, mapping each element value of the hill signal matrix U1 to the cell;

[0021] S43, taking the area of ​​the cell as the bottom area, and the element value corresponding to the cell is high, then the volume of each cell area corresponding to the hill signal is the bottom area multiplied by the height, which is the element value corresponding to the cell;

[0022] S44. Add the volumes of all cell areas to obtain the volume S of the area corresponding to the hill signal matrix U1. U1 ,Right now

[0023] Furthermore, it also includes:

[0024] S5. Calculate the volume S of the sampled signal matrix U U , using S U Subtract S U1 The volume of all peak signals is obtained, and then the quantitative analysis results of the target substances corresponding to the peaks are obtained.

[0025] Compared with the prior art, the method for calculating the volume of hill signal in the full two-dimensional chromatogram of the present invention has the beneficial effect of: by defining a convolution kernel A matrix, the number of rows of the convolution kernel A matrix is ​​set according to the maximum peak width of all peaks on the chromatogram on the Y axis, and the number of columns of the convolution kernel A matrix is ​​set according to the maximum peak width of all peaks on the chromatogram on the X axis, so that all peak signals can be framed by the convolution kernel A, so as to perform minimum value processing on the peak signals later; after the convolution kernel A is set, the original sampling signal matrix U is used to perform a convolution-like operation with the convolution kernel A, and the minimum element of the sampling signal matrix U corresponding to each convolution kernel A area is used as the new hill signal The element values ​​of the signal matrix U1 can be obtained by each convolution-like calculation to obtain a minimum element value, which is filled into the new hill signal matrix U1, and finally the hill signal matrix U1 is obtained. The element values ​​in this matrix have replaced all the peak signals with the minimum values ​​of the surrounding signals through convolution-like calculations. Therefore, what is retained is the bulge signal and the hump signal after removing the peak, which realizes the effective stripping of the hill signal. Then, the volume calculation of the hill information can be performed to obtain the quantitative analysis results of the substance corresponding to the hill signal, and then the quantitative analysis results of the substance corresponding to the peak signal can be obtained, which effectively solves the technical problem of difficulty in quantitative analysis of target substances in chromatograms with bulge or hump shapes.

Brief Description of the Drawings

[0026] Figure 1 A schematic diagram of a method flow of an embodiment of the present invention;

[0027] Figure 2 It is a schematic diagram of the method flow of hill signal stripping in an embodiment of the present invention;

[0028] Figure 3 A full two-dimensional data sampling signal chromatogram in an embodiment of the present invention;

[0029] Figure 4 This is a chromatogram of the mountain signal stripped out in an embodiment of the present invention. [Specific implementation method]

[0030] Embodiment 1:

[0031] Please refer to Figure 1-Figure 4 This embodiment is a method for calculating the volume of a mountain signal in a full two-dimensional chromatogram, which comprises the following steps:

[0032] S1. Obtain full two-dimensional data sampling signal u ij , we get the sampling signal matrix U = {u ij}, i = 1, 2, ..., M, j = 1, 2, ..., N, the size of the sampling signal matrix U is M × N, M is the number of rows of the sampling signal matrix U, and N is the number of columns of the sampling signal matrix U; the chromatogram presented by the sampling signal matrix U is as follows Figure 3 As shown;

[0033] S2. Set the convolution kernel A. The size of the convolution kernel A is m×n, where m is the maximum peak width of all the peaks of the full two-dimensional data sampling signal presented in the chromatogram on the Y axis, and n is the maximum peak width of all the peaks of the full two-dimensional data sampling signal presented in the chromatogram on the X axis. For example, if the maximum peak width of all the peaks of the full two-dimensional data sampling signal presented in the chromatogram on the Y axis is 3, and the maximum peak width on the X axis is also 3, then the convolution kernel A is a 3×3 matrix.

[0034] S3, construct a hill signal matrix U1, the size of the hill signal matrix U1 is the same as the size of the sampling signal matrix U, also with M rows and N columns; use the convolution kernel A to perform a quasi-convolution calculation on the sampling signal matrix U, the quasi-convolution calculation is the minimum element value z=min{u ij}, obtain the hill signal matrix U1; specifically comprising the following steps:

[0035] S31, initial value p=1, q=1;

[0036] S32, extract the pth to m+p-1th rows and the qth to n+q-1th columns from the sampled signal matrix U to obtain the local matrix Z pq ;

[0037] S33, take the local matrix Z pq The minimum value z of all elements in pq =min{u ij}, where \(i = p, p + 1,\cdots, m + p - 1\) and \(j = q, q + 1,\cdots, n + q - 1\), which is the element value at the position of the \(p\)-th row and the \(p\)-th column in the hill signal matrix \(U1\).

[0038] S34. Determine whether \(q \lt N - n + 1\) holds. If it holds, then \(q++\), and return to step S32; otherwise, execute the next step.

[0039] S35. Determine whether \(p \lt M - m + 1\) holds. If it holds, then \(p++\), and return to step S32; otherwise, execute the next step.

[0040] S36. Fill all the element values in the \( (M - m + 2)\)-th to \(M\)-th rows and the \((N - n + 2)\)-th to \(N\)-th columns of the hill signal matrix \(U1\) with 0 to obtain the hill signal matrix \(U1\).

[0041] The chromatogram presented by the hill signal matrix \(U1\) is as Figure 4 shown; the hill signal matrix \(U1\) obtained at this time is the bulge signal obtained by stripping all the spike data from the original signal data and the hump signal after removing the spikes. In this embodiment, the graph obtained after the spikes on the hump is called the hump bulge. Then, the hump signal after removing the spikes is correspondingly called the hump bulge signal. The bulge signal and the hump bulge signal are collectively called the hill signal.

[0042] When the convolution kernel \(A\) is used to perform box selection calculations row by row and column by column in the sampling signal matrix \(U\), the minimum element value of each local matrix region selected by the convolution kernel \(A\) is taken as the element value corresponding to the row and column of the hill signal matrix \(U1\), and the spike region values are replaced with the minimum values of the surrounding regions, thereby removing the spike data. However, when the convolution kernel \(A\) performs column-by-column dynamic box selection in the same row and moves to cover the last column of data, it cannot continue to move column by column in the same row. Therefore, there are no numerical values in the last \(n - 1\) columns of the hill signal matrix \(U1\); similarly, there is no data in the last \(m - 1\) rows of the hill signal matrix \(U1\). Therefore, in order to keep the size of the hill signal matrix \(U1\) the same as that of the sampling signal matrix \(U\) for convenient subsequent calculations, in this embodiment, all the element values in the last \(n - 1\) columns and the last \(m - 1\) rows of the hill signal matrix \(U1\) are filled with 0, which improves the calculation convenience without affecting the calculation results.

[0043] In other embodiments, if the calculation method for the volume of the subsequent hill signal matrix \(U1\) is used, even if the hill signal matrix \(U1\) does not have the last \(n - 1\) columns and the last \(m - 1\) rows, it does not affect the volume calculation, and the processing of step S36 can also be omitted.

[0044] S4. Calculate the volume of the region presented by the hill signal matrix \(U1\) on the chromatogram to obtain the hill volume \(S\) U1 ;

[0045] For the hill volume \(S\)U1 There are many ways to calculate , and the calculation methods in the prior art, such as integration method, geometric method, triangle wave method or oblique line method, etc., can be used, which is not limited in this embodiment.

[0046] This embodiment provides a geometric method to calculate the volume S of the hill. U1 The volume includes the following steps:

[0047] S41, dividing the XY axis plane into a plurality of 1×1 cells;

[0048] S42, mapping each element value of the hill signal matrix U1 to the cell;

[0049] S43, taking the area of ​​the cell as the bottom area, and the element value corresponding to the cell is high, then the volume of each cell area corresponding to the hill signal is the bottom area multiplied by the height, which is the element value corresponding to the cell;

[0050] S44. Add the volumes of all cell areas to obtain the volume S of the area corresponding to the hill signal matrix U1. U1 ,Right now

[0051] The hill shape that appears in the chromatogram reflects a substance that flows out within the corresponding time range. By calculating the volume of the hill, the content of the substance can be obtained, and the quantitative analysis result of the target substance can be obtained. If the target substance presents a peak shape feature, the volume S of the sampling signal matrix U is used U Subtract S U1 The volume of all peak signals can be obtained, and then the quantitative analysis results of the target substances corresponding to the peaks can be obtained.

[0052] The present embodiment provides a method for calculating the volume of a hill signal in a full two-dimensional chromatogram, by defining a convolution kernel A matrix, wherein the number of rows of the convolution kernel A matrix is ​​set according to the maximum peak width of all peaks on the chromatogram on the Y axis, and the number of columns of the convolution kernel A matrix is ​​set according to the maximum peak width of all peaks on the chromatogram on the X axis, so that all peak signals can be framed by the convolution kernel A, so as to perform minimum value processing on the peak signals later; after the convolution kernel A is set, a quasi-convolution operation is performed with the original sampling signal matrix U and the convolution kernel A, and the minimum element of the sampling signal matrix U corresponding to each convolution kernel A region is used as the element of the new hill signal matrix U1. Prime value, each convolution-like calculation can get a minimum element value, which is filled into the new hill signal matrix U1, and finally the hill signal matrix U1 is obtained. The element values ​​in this matrix have replaced all the peak signals with the minimum values ​​of the surrounding signals through the convolution-like calculation. Therefore, what is retained is the bulge signal and the hump signal after removing the peak, which realizes the effective stripping of the hill signal. Then, the volume calculation of the hill information can be performed to obtain the quantitative analysis results of the substance corresponding to the hill signal, and then the quantitative analysis results of the substance corresponding to the peak signal can be obtained, which effectively solves the technical problem of difficulty in quantitative analysis of target substances in chromatograms with bulge or hump shapes.

[0053] For those skilled in the art, several modifications and improvements can be made without departing from the inventive concept of the present invention, all of which fall within the protection scope of the present invention.

Claims

1. A method for obtaining the volume of the peak envelope signal in comprehensive two-dimensional chromatography, characterized in that: it includes the following steps: S1. Obtain full two-dimensional data sampling signal u ij , we get the sampling signal matrix U = {u ij }, i = 1, 2, ..., M; j = 1, 2, ..., N; the size of the sampling signal matrix U is M × N; S2. Set the convolution kernel A, the size of the convolution kernel A is m×n, where m is the maximum peak width of all the peaks presented by the comprehensive two-dimensional data sampling signal on the Y-axis in the chromatogram, and n is the maximum peak width of all the peaks presented by the comprehensive two-dimensional data sampling signal on the X-axis in the chromatogram; S3. Construct the peak envelope signal matrix U1: The size of the peak envelope signal matrix U1 is the same as the size of the sampling signal matrix U. Perform a convolution-like calculation on the sampling signal matrix U with the convolution kernel A. The convolution-like calculation is the minimum element value z in the local matrix corresponding to the size of the convolution kernel A in the sampling signal matrix U, to obtain the peak envelope signal matrix U1; S4. Calculate the area volume of the hill signal matrix U1 presented on the chromatogram to obtain the hill volume S U1 , and then obtain the quantitative analysis results of the target substance corresponding to the hill shape.

2. The method for obtaining the volume of the peak envelope signal in comprehensive two-dimensional chromatography according to claim 1, characterized in that: the S3 includes the following steps: S31. Let the initial values p = 1 and q = 1; S32, extract the pth to m+p-1th rows and the qth to n+q-1th columns from the sampled signal matrix U to obtain the local matrix Z pq ; S33, take the local matrix Z pq The minimum value z of all elements in pq =min{u ij }, i = p, p+1, ..., m+p-1, j = q, q+1, ..., n+q-1, as the element value at the p-th row and p-th column position in the hill signal matrix U1; S34. Judge whether q < N - n + 1 holds. If it holds, then q++, and return to step S32. Otherwise, execute the next step; S35. Judge whether p < M - m + 1 holds. If it holds, then p++, and return to step S32. Otherwise, execute the next step; S36. Fill all the element values in the peak envelope signal matrix U1 in the rows from M - m + 2 to M and the columns from N - n + 2 to N with 0, to obtain the peak envelope signal matrix U1.

3. The method for obtaining the volume of the peak envelope signal in comprehensive two-dimensional chromatography according to claim 1, characterized in that: The volume of the hill S U1 The calculation of includes the following steps: S41. Divide the XY-axis plane into several 1×1 cells; S42. Corresponding each element value of the peak envelope signal matrix U1 to the cell; S43. Taking the area of the cell as the base area and the element value corresponding to the cell as the height, then the volume of the peak envelope signal corresponding to each cell area is the base area multiplied by the height, that is, the element value corresponding to the cell; S44. Add the volumes of all cell areas to obtain the volume S of the area corresponding to the hill signal matrix U1. U1 ,Right now 4. The method for obtaining the volume of the peak envelope signal in comprehensive two-dimensional chromatography according to claim 1, characterized in that: it further includes: S5. Calculate the volume S of the sampled signal matrix U U , using S U Subtract S U1 The volume of all peak signals is obtained, and then the quantitative analysis results of the target substances corresponding to the peaks are obtained.

Images