A method for measuring nitrate in seawater based on spectral fingerprinting

Abstract

The application provides a seawater nitrate measurement method based on a spectral fingerprint, which comprises the following steps: step one, preparing an artificial seawater solution and constructing a spectral fingerprint S1 of the artificial seawater solution; step two, constructing spectral fingerprints S2, S3, S4 and S5 of organic matter-artificial seawater solutions and a nitrate-artificial seawater solution; step three, establishing a spectral reconstruction equation; step four, establishing a multivariate regression equation of a nitrate concentration C; step five, substituting an absorption spectrum of seawater into the spectral reconstruction equation in step three to obtain fitting parameters, and finally substituting the fitting parameters into the multivariate regression equation in step four to obtain the nitrate concentration of the seawater. The method effectively suppresses the interference of ions, organic matters and the like in seawater on the spectrum of nitrate, improves the measurement precision of nitrate in high-salinity water samples, improves the adaptability of a spectral analysis instrument to a marine environment, realizes rapid calibration, and improves the spectral model migration capability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rapid analysis of seawater quality by spectroscopy, and particularly relates to a method for measuring nitrate in seawater based on spectral fingerprint. BACKGROUND

[0002] The principle of the nitrate analyzer for seawater by spectroscopy is to measure the ultraviolet absorption spectrum data in the wave band of 210nm to 240nm by using a miniature ultraviolet light source and a spectral detector, and to establish a spectral nitrate concentration measurement model. Compared with surface fresh water, seawater contains high concentrations of chloride ions, bromide ions, organic matter and other substances, and their absorption spectra overlap with the nitrate spectrum in the ultraviolet wave band, increasing the difficulty of the seawater nitrate spectral model. How to effectively separate the nitrate spectrum from the interference spectrum is the key to solving the problem of measuring nitrate in seawater.

[0003] The commonly used method at present is to carry out large-scale orthogonal spectrum experiments, obtain a large amount of spectral data by preparing mixed solutions of chloride ions, bromide ions, nitrate and organic matter, and establish a complex algorithm model such as neural network to predict the content of nitrate in the sample. However, the above method still has the following problems: first, the time-consuming of orthogonal experiment is long; second, the spectral model migration problem, the complex model algorithm is not conducive to the rapid implantation of micro control chip, and the large amount of calculation also needs to consume more execution time, which is not easy to realize the in-situ real-time rapid analysis of nitrate.

[0004] In summary, there is a need to design a method for measuring nitrate in seawater based on spectral fingerprint to solve the above problems. SUMMARY

[0005] The present application provides a method for measuring nitrate in seawater based on spectral fingerprint, which solves the problem of interference of other substances in seawater on the nitrate spectrum and the problem of complicated calculation process.

[0006] To achieve the purpose of solving the above technical problems, the present application adopts the following technical scheme:

[0007] A method for measuring nitrate in seawater based on spectral fingerprint, comprising the following steps:

[0008] Step 1, preparing artificial seawater solution and constructing spectral fingerprint S1 of the artificial seawater solution;

[0009] Step 2, adding organic matter and nitrate to the artificial seawater solution respectively to prepare organic matter-artificial seawater solutions and nitrate-artificial seawater solutions with multiple concentrations; and constructing spectral fingerprint S2 of the organic matter-artificial seawater solution and spectral fingerprints S3, S4, S5 of the nitrate-artificial seawater solution;

[0010] Step three, using each spectrum fingerprint in step one and step two to establish a spectrum reconstruction equation ;

[0011] Wherein, sd is the second derivative spectrum of the artificial seawater solution; m is the number of wavelengths of the absorption spectrum of the artificial seawater solution; S0 is a number sequence with all element values being 1, and the length is consistent with the spectrum fingerprint S1; b 0- b 5 is a fitting parameter;

[0012] Step four, according to the spectrum reconstruction equation in step three, a multivariate regression equation of nitrate concentration C is established:

[0013] ;

[0014] Wherein k 0、 k 3、 k 4、 k 5 is a calibration coefficient;

[0015] Step five, according to the absorption spectrum of seawater sample, the second derivative spectrum sequence is calculated, and the second derivative spectrum sequence is substituted into the spectrum reconstruction equation in step three to obtain the fitting parameter b 3- b 5, finally, the fitting parameter b 3- b 5 is substituted into the multivariate regression equation in step four, and the nitrate concentration of the seawater sample is obtained.

[0016] In some embodiments of the present application, the step one specifically comprises the following steps:

[0017] 11. Preparing an artificial seawater solution;

[0018] 12. According to the absorption spectrum of the artificial seawater solution, the absorbance at each wavelength is calculated, and the absorbance is recorded as sequence one:

[0019]

[0020] 13. The second derivative spectrum sd of the absorbance is calculated, recorded as sequence two, and the sequence two is the spectrum fingerprint S1:

[0021] ;

[0022] Wherein, the superscript of sd is the sample number of the artificial seawater solution.

[0023] In some embodiments of the present application, the step 13 specifically comprises the following steps:

[0024] 13-1、According to the formula The first derivative of each of the absorbance in sequence one is calculated to obtain the following sequence: ;

[0025] 13-2、According to the formula The second derivative of each of the absorbance in sequence one is calculated to obtain the following sequence: .

[0026] In some embodiments of the present application, the step of constructing the spectral fingerprint S2 in step two comprises:

[0027] 21、Take n portions of the artificial seawater solution, and prepare n organic matter-artificial seawater solutions with different concentrations after adding organic matter;

[0028] 22、Repeat steps 12-13 to obtain n second derivative spectral sequences of the organic matter-artificial seawater solutions;

[0029] 23、Calculate the second derivative spectral difference between the organic matter-artificial seawater solution and the artificial seawater solution at each corresponding wavelength respectively to obtain a matrix .

[0030] 24、After singular value decomposition operation is performed on the matrix , a matrix is obtained, and the first column of the matrix is extracted as the spectral fingerprint S2 of the organic matter-artificial seawater solution:

[0031] .

[0032] In some embodiments of the present application, the step of constructing the spectral fingerprints S3-S5 in step two comprises:

[0033] 25、Take n portions of the artificial seawater solution, and prepare n nitrate-artificial seawater solutions with different concentrations after adding nitrate;

[0034] 26、Repeat steps 12-13 to obtain n second derivative spectral sequences of the nitrate-artificial seawater solutions;

[0035] 27、Calculate the second derivative spectral difference between the nitrate-artificial seawater solution and the artificial seawater solution at each corresponding wavelength respectively to obtain a matrix .

[0036] 24、After singular value decomposition operation is performed on the matrix , a matrix is obtained, and the first three columns of the matrix are extracted as the spectral fingerprints S3-S5 of the nitrate-artificial seawater solution: ​​​​

[0037] ;

[0038] ;

[0039] .

[0040] In some embodiments of the present application, the calculation process of each calibration coefficient in step four comprises the following steps:

[0041] 41. Take n portions of the artificial seawater solution, and after adding nitrate, prepare n portions of nitrate sample solution with gradient change in nitrate concentration, whose concentration C is recorded as sequence three:

[0042] ; wherein the superscript of concentration C represents the number of the nitrate sample solution;

[0043] 42. Repeat steps 12-13 to obtain a sequence of second derivative spectra of the n portions of nitrate sample solution, and arrange them into a matrix:

[0044] ;

[0045] wherein the superscript of sd represents the sample number of the nitrate sample solution, and the sequence of second derivative spectra corresponds to the sequence of nitrate concentration;

[0046] 43. Substitute each column of the matrix in step 42 into the spectral reconstruction equation, and sequentially solve to obtain fitting parameters b 3, b 4 and b 5, which are recorded as sequence four according to the sample number of the nitrate sample solution:

[0047] ;

[0048] Sequence five:

[0049] ;

[0050] Sequence six:

[0051] ;

[0052] 44. Substitute sequence three to sequence six into the multiple regression equation to solve the calibration coefficients k 0, k 3, k 4, k 5.

[0053] In some embodiments of the present application, the artificial seawater solution comprises chloride ions and bromide ions.

[0054] In some embodiments of the present application, the calibration coefficients and the fitting parameters are solved by using the least square method.

[0055] In some embodiments of the present application, the nitrate salt is selected from one or more of potassium nitrate, sodium nitrate or calcium nitrate.

[0056] In some embodiments of the present application, the organic matter is selected from one of yellow humic acid or sodium humate.

[0057] The technical solution of the present application has the following technical effects relative to the prior art:

[0058] The present application provides a seawater nitrate measurement method based on spectral fingerprints, which is used for a spectral method seawater nitrate analyzer. The spectral fingerprints of typical substances in seawater are used to implement fitting reconstruction on water sample spectra, and then the seawater nitrate concentration is measured. This method can obtain a plurality of effective spectral fingerprints, effectively suppress the interference of ions, organic matters and the like in seawater on the nitrate spectrum, improve the measurement accuracy of nitrate in high salinity water samples, improve the marine environment adaptability of spectral analysis instruments, realize rapid calibration of seawater nitrate analysis instruments, and improve the spectral model migration ability; thereby avoiding a large number of orthogonal experiments. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0060] Figure 1 Fig. 1 is a spectral graph of seawater spectral fingerprints S1-S5, in which the wavelength series is taken as the abscissa and the spectral fingerprint series is taken as the ordinate, and S1 is the spectral intensity after being enlarged by 100 times.

[0061] Figure 2 Fig. 4 is the correlation of the nitrate standard solution concentration and the predicted concentration.

[0062] Figure 3 Fig. 5 is the absorbance spectrum of seawater samples.

[0063] Figure 4 Fig. 6 is the second derivative spectrum of seawater samples and the reconstructed spectrum based on the spectral fingerprint library. DETAILED DESCRIPTION

[0064] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of the present application.

[0065] Embodiment 1

[0066] A method for measuring seawater nitrate based on spectral fingerprint, comprising the following steps:

[0067] Step 1, preparing artificial seawater solution and constructing spectral fingerprint S1 of the artificial seawater solution;

[0068] Specifically comprising the following steps:

[0069] 11, preparing artificial seawater solution; in this embodiment, 35 grams of sodium chloride and 86 milligrams of sodium bromide are contained in each liter of artificial seawater solution;

[0070] 12, calculating the absorbance at each wavelength according to the absorption spectrum of the artificial seawater solution:

[0071] The ultraviolet absorption spectrum of the pure water sample is measured by using a deuterium halogen lamp light source and an optical fiber spectrometer. In the wavelength range of 200 nm to 240 nm, the light intensity data of 125 pixel points can be recorded. In order of increasing wavelength, the transmitted light intensity of pure water is recorded as a sequence of pixels:

[0072] ;

[0073] The wavelength values corresponding to each pixel in the optical fiber spectrometer are recorded in order of increasing wavelength as a sequence:

[0074] ;

[0075] The ultraviolet absorption spectrum of the artificial seawater solution is recorded as a sequence of transmitted light intensity in order of increasing wavelength:

[0076] ;

[0077] Taking the pure water sample as a reference, the absorbance formula is:

[0078]

[0079] The absorbance at each wavelength is calculated one by one, and the absorbance data of the artificial seawater solution is recorded as a sequence one in order of increasing wavelength:

[0080]

[0081] 13. Calculate the second derivative spectrum sd of the absorbance:

[0082] The wavelength data corresponding to each pixel in the fiber spectrometer usually presents a uniform linear distribution, that is, the difference between adjacent wavelength data in the wavelength sequence is fixed and constant, and the wavelength data can be ignored when calculating the derivative spectrum, so the first derivative spectrum calculation formula can be written as:

[0083] ;

[0084] Using the numerical derivative method, the first derivative of the absorbance sequence is calculated one by one, and the first derivative is arranged into a sequence:

[0085] ;

[0086] According to the second derivative calculation formula:

[0087]

[0088] Using the numerical derivative method, the second derivative of the absorbance sequence is calculated one by one, and the second derivative is arranged into a sequence two, which is the spectral fingerprint S1:

[0089] ,

[0090] Among them, S1 contains 123 elements, and the superscript sd of the sample is the sample number of the artificial seawater solution.

[0091] Step two, add organic matter and nitrate to the artificial seawater solution respectively to prepare organic matter-artificial seawater solutions and nitrate-artificial seawater solutions with multiple concentrations; and construct the spectral fingerprints S2, S3, S4 and S5 of the organic matter-artificial seawater solution and the nitrate-artificial seawater solution;

[0092] The construction step of the spectral fingerprint S2 in the step two includes:

[0093] 21. Take 6 portions of artificial seawater solution, add humic acid to the artificial seawater solution, and prepare 6 kinds of humic acid-artificial seawater solutions with 6 concentrations, so that the concentration of humic acid in the solution is 1 mg / L, 2 mg / L, and each time increase by 1 mg / L until 6 mg / L, a total of 6 samples;

[0094] 22. Measure the absorbance curves of the 6 humic acid-artificial seawater samples respectively, and calculate the second derivative spectrum sequence respectively;

[0095] 23. Calculate the second derivative spectrum difference between the humic acid-artificial seawater sample and the artificial seawater solution at each corresponding wavelength:

[0096]

[0097] The second derivative spectrum difference series is written as:

[0098] ;

[0099] Arrange all the second derivative spectrum difference series into a matrix:

[0100] (13)

[0101] Wherein, the superscript of sd is the number of fulvic acid-artificial seawater sample, indicating that the second derivative spectrum value corresponds to the fulvic acid concentration.

[0102] 24, singular value decomposition operation is performed on the matrix

[0103]

[0104] Wherein The matrix is the basic spectrum, The matrix is the singular value diagonal matrix, The matrix is the intensity change of each basic spectrum. After singular value decomposition operation, the matrix can be obtained Each element value in the matrix contains 123 rows and 123 columns, which can be expressed as:

[0105]

[0106] Extract the first column of the matrix as the spectrum fingerprint S2 of the fulvic acid-artificial seawater solution:

[0107] . The construction step of the spectrum fingerprint S3-S5 in step two includes:

[0108] 25, take 20 portions of the artificial seawater solution, add potassium nitrate to prepare a plurality of different concentrations of nitrate-artificial seawater solution; the content of nitrate in the solution is 1 mg / L, 2 mg / L, 3 mg / L, and each time increases by 1 mg / L, gradually increases to 20 mg / L, a total of 20 samples;

[0109] 26, repeat steps 12-13 to obtain 20 second derivative spectrum series of the nitrate-artificial seawater solution;

[0110] 27, calculate the second derivative spectrum difference of the nitrate-artificial seawater solution and the artificial seawater solution at each corresponding wavelength respectively:

[0111]

[0112] ​​

[0113] The sequence is written as:

[0114]

[0115] The difference sequence of all second derivative spectra is arranged into a matrix:

[0116] (19)

[0117] wherein the superscript of sd is the number of nitrate-artificial seawater solution, indicating that the second derivative spectrum value corresponds to the concentration of the nitrate sample.

[0118] 28、The matrix is subjected to singular value decomposition operation:

[0119]

[0120] wherein The matrix is the basic spectrum, and after singular value decomposition operation, the Each element value in the matrix, containing 123 rows and 123 columns, can be expressed as:

[0121]

[0122] The first three columns of the matrix are extracted as the spectral fingerprints S3-S5 of the nitrate-artificial seawater solution:

[0123] ;

[0124] ;

[0125] .

[0126] Step three, using each spectral fingerprint in step one and step two to establish a spectral reconstruction equation:

[0127] 31、A sequence S0 with element values all being 1 is established:

[0128]

[0129] In this embodiment, S0 contains 123 elements in total, and the sequence length is consistent with the spectral fingerprint S1. S0 can be used as a unit constant spectral fingerprint, which functions to eliminate spectral baseline drift.

[0130] 32、The spectral fingerprints include: unit constant spectral fingerprint S0, artificial seawater spectral fingerprint S1, organic matter-artificial seawater solution spectral fingerprint S2, and nitrate-artificial seawater solution spectral fingerprints S3-S5, as shown in Figure 1 ​The linear combination method of the above spectral fingerprints is used to establish a spectral reconstruction equation, which can be expressed as:

[0131]

[0132] The spectral reconstruction equation is an overdetermined equation; b 0- b 5 is a fitting parameter, which can be solved by the least square method.

[0133] Step four, according to the spectral reconstruction equation in step three, a multivariate regression equation of nitrate concentration C is established:

[0134] 41, take 7 parts of the artificial seawater solution, add potassium nitrate in artificial seawater, respectively, prepare a plurality of nitrate-artificial seawater solution, in this embodiment, the content of nitrate in artificial seawater is 0 mg / L, 0.4 mg / L, 0.8 mg / L, 1.6 mg / L, 3.2 mg / L, 6.4 mg / L, 12.8 mg / L, a total of 7 samples as nitrate standard solution. The concentration of sample nitrate is recorded as a series of three:

[0135] ;

[0136] 42, repeat steps 12-13, get the second derivative spectrum series of 7 nitrate sample solutions, and arrange them into a matrix:

[0137] ;

[0138] Wherein sd superscript represents the sample number of the nitrate sample solution, and the second derivative spectrum series corresponds to the nitrate concentration series;

[0139] 43, each column of the matrix in step 42 is substituted into the spectral reconstruction equation, and then the fitting parameters b 3、 b 4 and b 5 are recorded as a series of four according to the sample number:

[0140] ;

[0141] Series five:

[0142] ;

[0143] Series six:

[0144] ;

[0145] 44、 b 3、 b 4 and bFive parameters are correlated with the intensity of the nitrate spectral fingerprint, and a system is established. b 3. b 4. b 5. Multiple regression equation for nitrate concentration:

[0146]

[0147] in k 0、 k 3. k 4. k 5 is the calibration coefficient;

[0148] Substitute the numbers from sequence 3 to sequence 6 into the multiple regression equation in sequence, and solve for the calibration coefficients using the least squares method. k 0、 k 3. k 4. k 5.

[0149] In this embodiment, the parameter value obtained is:

[0150] k 0 = -0.3621;

[0151] k 3 = -498.8070;

[0152] k 4 = 34.6966;

[0153] k 5 = -39.8213;

[0154] The results of the above analysis were used to measure the concentration of the nitrate standard solution as follows: Figure 2 As shown.

[0155] Step 5: Determine the nitrate concentration of the seawater sample:

[0156] 51. Take a seawater sample, repeat steps 12-13, and measure the absorption spectrum of the seawater sample, referring to... Figure 3 As shown, calculate its second derivative spectral sequence;

[0157] 52. Substitute the second-derivative spectral sequence into the spectral reconstruction equation in step three, and perform spectral reconstruction, referring to... Figure 4 As shown, the fitting parameters are obtained using the least squares method. b 3- b 5;

[0158] In this embodiment, the following is obtained: b 3- b The value of parameter 5 is:

[0159] b3 = -0.00054;

[0160] b 4 = -0.00298;

[0161] b 5 = -0.00917;

[0162] 53. Fitting parameters b 3- b Substituting 5 into the multiple regression equation in step four yields the nitrate concentration of the seawater sample. C It is 0.1707 mg / L.

[0163] The technical solution of the present invention has the following technical effects compared with the prior art:

[0164] This invention proposes a seawater nitrate measurement method based on spectral fingerprinting for use in spectroscopic seawater nitrate analyzers. The method reconstructs the water sample spectrum by fitting the spectral fingerprints of typical substances in seawater, thereby measuring the seawater nitrate concentration. This method can acquire multiple effective spectral fingerprints, effectively suppressing the interference of ion and organic matter spectra in seawater on the nitrate spectrum, improving the measurement accuracy of nitrate in high-salinity water samples, and enhancing the marine environmental adaptability of spectroscopic analyzers. Furthermore, it enables rapid calibration of seawater nitrate analyzers, improves the spectral model transfer capability, and thus avoids a large number of orthogonal experiments.

[0165] In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0166] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for measuring nitrate in seawater based on spectral fingerprinting, characterized in that, The method comprises the following steps: Step one, preparing artificial seawater solution and constructing the spectral fingerprint S1 of the artificial seawater solution; Step two, adding organic matter and nitrate into the artificial seawater solution respectively to prepare organic matter-artificial seawater solutions and nitrate-artificial seawater solutions with various concentrations; and constructing the spectral fingerprints S2, S3, S4 and S5 of the organic matter-artificial seawater solutions and the nitrate-artificial seawater solutions; Step three, establishing a spectral reconstruction equation by using the spectral fingerprints in step one and step two; ； Wherein, sd is the second derivative spectrum of the artificial seawater solution; m is the number of wavelengths of the absorption spectrum of the artificial seawater solution; S0 is a number sequence with all element values being 1, and the length of the number sequence is consistent with the spectrum fingerprint S1; b 0- b 5 is a fitting parameter; Step four, establishing a multivariate regression equation of the nitrate concentration C according to the spectral reconstruction equation in step three: ； wherein k 0, k 3, k 4, k 5 is a calibration coefficient; Step five, calculate the second derivative spectrum series of the seawater sample according to its absorption spectrum, and substitute the second derivative spectrum series into the spectrum reconstruction equation in step three to obtain the fitting parameters b 3- b 5, finally substitute the fitting parameters b 3- b 5 into the multiple regression equation in step four to obtain the nitrate concentration of the seawater sample.

2. The method according to claim 1, wherein, The step one specifically comprises the following steps:

11. Preparing artificial seawater solution; 12. Calculating the absorbance at each wavelength according to the absorption spectrum of the artificial seawater solution, and recording the absorbance as a sequence one: ； 13. Calculating the second derivative spectrum sd of the absorbance, and recording as a sequence two, which is the spectral fingerprint S1: ； Wherein, the superscript sd of the sample number of the artificial seawater solution.

3. The method according to claim 2, wherein, The step 13 specifically comprises the following steps: 13-1. According to the formula The first derivative of each of the absorbances in series one is calculated to give the following series: ； 13-2. According to the formula The second derivative of each of the absorbances in series one is calculated to obtain the following series: .

4. The method according to claim 1, wherein, The construction step of the spectral fingerprint S2 in the step two comprises:

21. Taking n portions of the artificial seawater solution, adding organic matter to prepare n organic matter-artificial seawater solutions with different concentrations; 22. Repeating steps 12-13 to obtain the second derivative spectrum sequence of the n organic matter-artificial seawater solutions; 23. Calculate the difference between the second derivative spectra of the organic- artificial seawater solution and the artificial seawater solution at each corresponding wavelength, respectively, to obtain a matrix 24、said matrix after performing a singular value decomposition operation matrix, extracting the first column of said matrix as the spectral fingerprint S2 of the organic-artificial seawater solution: 。 5. The method according to claim 1, wherein, The construction step of the spectral fingerprints S3-S5 in the step two comprises:

25. Taking n portions of the artificial seawater solution, adding nitrate to prepare n nitrate-artificial seawater solutions with different concentrations; 26. Repeating steps 12-13 to obtain the second derivative spectrum sequence of the n nitrate-artificial seawater solutions; 27. Calculate the difference between the second derivative spectra of the nitrate artificial seawater solution and the artificial seawater solution at each corresponding wavelength, respectively, to obtain a matrix ; 28、the matrix After performing the singular value decomposition operation, we obtain The matrix, the first three columns of which are extracted as the nitrate-artificial seawater solution spectral fingerprints S3-S5: The matrix, the first three columns of which are extracted as the nitrate-artificial seawater solution spectral fingerprints S3-S5: ； ； 。 6. The method according to claim 1, wherein, The calculation process of each calibration coefficient in the step four comprises the following steps:

41. Taking n portions of the artificial seawater solution, adding nitrate to prepare n nitrate sample solutions with gradient change in nitrate concentration, and recording the concentration C as a sequence three: ; wherein the superscript of the concentration C indicates the number of the nitrate sample solution; 42. Repeating steps 12-13 to obtain the second derivative spectrum sequence of the n nitrate sample solutions, and arranging them into a matrix: ； Wherein, the superscript sd of the sample number of the nitrate sample solution, the second derivative spectrum sequence corresponds to the nitrate concentration sequence; 43. Substitute each column of the matrix in step 42 into the spectral reconstruction equation, sequentially solve for the fitting parameters b 3、 b 4 and b 5 record as a number series four: sample number of nitrate salt sample solution sample ； Sequence five: ； Sequence six: ； 44. The calibration coefficients are solved by substituting the number series three-number series six into the multiple regression equation k 0, k 3, k 4, k 5.

7. The method according to claim 1, wherein, The artificial seawater solution comprises chloride ions and bromide ions.

8. The method according to claim 1, wherein, The calibration coefficients and the fitting parameters are solved by the least square method.

9. The method according to claim 1, wherein, The nitrate is selected from one or more of potassium nitrate, sodium nitrate or calcium nitrate.

10. The method according to claim 1, wherein, The organic matter is selected from one of yellow humic acid or sodium humate.

Images