Method for turbidity compensation in fluorescence detection and method for predicting concentration of phytoplankton in water body

Abstract

The present application relates to fluorescence detection methods, in particular to a turbidity compensation method in fluorescence detection and a concentration prediction method of water phytoplankton, to solve the problems of deviation of the characterization of turbidity interference, insufficient reliability and accuracy of the existing turbidity compensation method, or the problem of greater influence of the compensation accuracy of the neural network on the training set, and the problem of serious deviation of the measured value from the true value caused by the turbidity interference in the fluorescence detection of water phytoplankton. The turbidity compensation method in fluorescence detection comprises the following steps: obtaining standardized scattering spectrum S1 and standardized scattering spectrum S2, calculating weakening coefficient K i or weakening coefficient M i , eliminating the scattering enhancement of the fluorescence spectrum of the sample to be measured, calculating the fluorescence intensity of the sample to be measured after turbidity compensation, and completing the turbidity compensation. The concentration prediction method of water phytoplankton uses the above method to obtain the fluorescence intensity of the sample to be measured after turbidity compensation, and then substitutes it into the concentration prediction model to obtain more accurate concentration prediction value.

Description

Technical Field

[0001] This invention relates to fluorescence detection methods, specifically to turbidity compensation methods in fluorescence detection and methods for predicting the concentration of phytoplankton in water. Background Technology

[0002] Current turbidity compensation methods typically establish a linear relationship between turbidity values ​​and fluorescence intensity to assess the impact of different turbidities on sample fluorescence and then make corrections accordingly. However, this method does not consider the different degrees of enhancement and attenuation caused by turbidity, resulting in a bias in the characterization of turbidity interference. Furthermore, it uses the same coefficient to compensate for the effects across all turbidity ranges, leading to insufficient reliability and accuracy in turbidity compensation. In addition, some methods use neural networks for fitting, but neural networks, as a black box, can only simulate the changing patterns of the training data, and their compensation accuracy is greatly affected by the training set.

[0003] Phytoplankton are vital producers in the ecological environment and key hubs for energy flow. Fluorescence detection technology, with its advantages of rapid detection and flexibility, has become an important means of phytoplankton monitoring. However, in actual monitoring, turbidity in the water can significantly interfere with fluorescence detection. Due to differences in the absorption and scattering properties of suspended particulate matter, different suspended particulate matter also have significant differences in their impact on fluorescence. Therefore, when directly using the measured fluorescence spectrum to predict phytoplankton concentration, the measured values ​​will deviate severely from the true values. Summary of the Invention

[0004] The purpose of this invention is to address the problems of existing turbidity compensation methods having biases, insufficient reliability and accuracy in characterizing turbidity interference, or the fact that the compensation accuracy of neural networks is greatly affected by the training set, as well as the problem that turbidity interference in phytoplankton fluorescence detection causes a serious deviation between measured and true values. Therefore, this invention provides a turbidity compensation method for fluorescence detection and a method for predicting the concentration of phytoplankton in water.

[0005] To achieve the above objectives, the technical solution provided by this invention is as follows:

[0006] A turbidity compensation method for fluorescence detection, characterized by the following steps:

[0007] S1. Prepare multiple turbidity standard solutions with gradient concentrations and obtain their scattering spectra. The excitation wavelength of the scattering spectrum is Ex, and the emission interval is [Em1, Em2]. For each wavelength in the wavelength interval to the left of the excitation wavelength, fit a linear relationship between the turbidity value of the multiple turbidity standard solutions and the corresponding scattered light intensity, and select R. 2 Wavelength range L, R ≥ 0.99 2 The coefficient of determination for a linear relationship;

[0008] S2. Calculate the normalized scattering spectrum S1 in the wavelength range to the left of the excitation wavelength and the normalized scattering spectrum S2 in the wavelength range to the right of the excitation wavelength.

[0009] S3. Mix multiple standard samples of the target analyte with turbidity standard solutions of different concentrations to prepare p standard mixed solutions with turbidity gradients, and measure their fluorescence spectra.

[0010] S4. Using the standard mixed solution fluorescence spectrum in wavelength range L and the normalized scattering spectrum S1 and the normalized scattering spectrum S2 in the wavelength range to the right of the excitation wavelength, the scattering enhancement of the standard mixed solution fluorescence spectrum is eliminated, and the standard mixed solution fluorescence spectrum with scattering enhancement eliminated is obtained.

[0011] S5. For each of the p standard mixed solutions, calculate the attenuation coefficient K based on the turbidity value. i Or based on the attenuation coefficient M of the scattering spectrum i Where i takes values ​​of 1, ..., p, the weakening coefficient matrix K = [K1, ..., K...] is obtained. p Or the weakening coefficient matrix M = [M1, ..., M] p ];

[0012] S6. Eliminate the scattering enhancement of the measured fluorescence spectrum of the sample under test by using the standard mixed solution fluorescence spectrum and the standardized scattering spectrum S1 in the wavelength range L and the standardized scattering spectrum S2 in the wavelength range to the right of the excitation wavelength. Then, use the attenuation coefficient matrix K or the attenuation coefficient matrix M to calculate the fluorescence intensity of the sample under test after turbidity compensation to complete the turbidity compensation.

[0013] In steps S2 and S5, the excitation wavelength and emission range of the fluorescence spectrum are the same as in step S1.

[0014] Furthermore, step S2 specifically includes:

[0015] Normalize the scattering spectrum described in step S1, and then calculate the average intensity of scattered light from multiple turbidity standard solutions corresponding to each wavelength on the left and right sides of the excitation wavelength, respectively, to obtain the normalized scattering spectrum S1 on the left side of the excitation wavelength and the normalized scattering spectrum S2 on the right side of the excitation wavelength.

[0016] Furthermore, step S4 specifically includes:

[0017] S4.1 Calculate the enhancement coefficient H of the fluorescence spectrum of the standard mixed solution using the following formula;

[0018]

[0019] Among them, S cal(L) is the fluorescence spectrum of the standard mixed solution in the wavelength range L, and S1(L) is the normalized scattering spectrum of S1 in the wavelength range L.

[0020] S4.2. The scattering spectrum S to the right of the excitation wavelength of the standard mixed solution is estimated according to the following formula. e ;

[0021] S e =H*S2

[0022] S4.3. The fluorescence spectrum of the standard mixed solution after eliminating scattering enhancement is calculated using the following formula. el ;

[0023] S el =S cal -S e

[0024] Among them, S cal The fluorescence spectrum of the standard mixed solution obtained in step S3 is shown.

[0025] Furthermore, step S5 specifically includes:

[0026] Defined as a p×q fluorescence matrix F composed of the fluorescence spectra of multiple standard mixed solutions after scattering enhancement elimination, where q is the spectral number corresponding to each standard mixed solution, and the peak fluorescence intensities of the corresponding columns 1, ..., p-1, p are denoted as I1, ..., I... p-1 I p The turbidity of multiple standard mixed solutions constitutes a turbidity matrix N = [N1, ..., N2]. p-1 N p ], N1, ..., N p-1 N p The turbidity values ​​are the 1st, ..., p-1th, and pth standard mixed solutions arranged in ascending order of turbidity.

[0027] The weakening coefficient K is calculated using the following formula. i Or the weakening coefficient M i The value of i is 1, ..., p, thus obtaining the weakening coefficient matrix K = [K1, ..., K]. p Or the weakening coefficient matrix M = [M1, ..., M] p ];

[0028]

[0029]

[0030] Where F0 is the fluorescence intensity peak of the standard mixed solution of 0 NTU, and Sc i-1 and Sc iand are the average scattered light intensities of the (i-1)th and ith standard mixed solutions in the wavelength range L, respectively, where N is the average scattered light intensity when i = 1. i-1 and Sc i-1 The values ​​represent the turbidity of a standard mixed solution with a turbidity of 0 NTU and its average scattered light intensity in the wavelength range L, respectively.

[0031] Further, in step S6, the fluorescence intensity of the test sample after turbidity compensation is calculated using the attenuation coefficient matrix K or the attenuation coefficient matrix M as follows:

[0032] The compensated fluorescence intensity I of the test sample is calculated using the following formula. c1 or I c2 Complete turbidity compensation;

[0033]

[0034]

[0035] Among them, I e To eliminate the fluorescence peak intensity corresponding to the fluorescence spectrum of the sample after scattering enhancement; Δntu1, ..., Δntu m For 0NTU, N1, ..., N m The gradient of two adjacent turbidity values, N m Let p be the turbidity of the m-th standard mixed solution, where m ≤ p, and let N be the turbidity of the sample to be tested. sample N sample ≤N m ;ΔSc1、…、ΔSc m For 0NTU, N1, ..., N m The gradient of the scattered light intensity corresponding to two adjacent turbidity values.

[0036] Furthermore, in step S6, the scattering enhancement of the measured fluorescence spectrum of the sample to be tested is eliminated using the normalized scattering spectrum S1 and the normalized scattering spectrum S2, specifically as follows:

[0037] Calculate the enhancement coefficient of the fluorescence spectrum of the sample to be tested; then use the enhancement coefficient of the fluorescence spectrum of the sample to be tested to estimate the scattering spectrum in the wavelength range to the right of the excitation wavelength of the sample to be tested; then use the fluorescence spectrum of the sample to be tested to subtract the scattering spectrum of the sample to be tested to obtain the fluorescence spectrum of the sample to be tested after eliminating scattering enhancement.

[0038] Furthermore, before normalization in step S2, the saturation spectrum in the scattering spectrum is removed.

[0039] Meanwhile, this invention also provides a method for predicting the concentration of phytoplankton in water, which is characterized by including the following steps:

[0040] Step 1: Measure the fluorescence spectrum S of the sample to be tested. sample ;

[0041] Step 2: Use the turbidity compensation method in the fluorescence detection described above to perform turbidity compensation on the fluorescence spectrum of the sample to obtain the fluorescence intensity of the sample after turbidity compensation; the excitation wavelength and wavelength range of the fluorescence spectrum are the same as in Step 1.

[0042] Step 3: Substitute the fluorescence intensity of the turbidity-compensated sample into the concentration prediction model to predict the concentration of phytoplankton in the water.

[0043] Furthermore, step 2 specifically involves:

[0044] Determine whether the turbidity of the sample to be tested can be obtained;

[0045] If it can be obtained, the attenuation coefficient matrix K is calculated using the turbidity compensation method in the fluorescence detection described above, and the corresponding fluorescence intensity of the sample to be tested after turbidity compensation is calculated.

[0046] Otherwise, the attenuation coefficient matrix M is calculated using the turbidity compensation method in the fluorescence detection described above, and the corresponding fluorescence intensity of the sample after turbidity compensation is calculated.

[0047] Further, in step 3, the concentration prediction model is:

[0048] C=(Fb) / a

[0049] Where C is the concentration of phytoplankton in the water, a and b are the slope and intercept of the concentration prediction model, respectively, and F is the fluorescence intensity of the sample to be tested.

[0050] The beneficial effects of this invention are:

[0051] 1. This invention proposes a turbidity compensation method for fluorescence detection that first divides the influence of turbidity particles on fluorescence detection into scattering enhancement and absorption attenuation components, which are then eliminated and compensated for separately. Compared to commonly used linear models, this method has a wider range of applications and can more accurately compensate for turbidity in fluorescence detection, effectively improving the accuracy of fluorescence detection under turbidity particle interference after compensation.

[0052] 2. The turbidity compensation method in fluorescence detection of the present invention takes into account whether the actual turbidity value can be obtained directly, and sets attenuation coefficients K and M to ensure that turbidity compensation can be performed whether the turbidity is known or unknown, and the prediction results after compensation have good accuracy.

[0053] 3. The attenuation coefficient of the turbidity compensation method in fluorescence detection of the present invention is calculated based on different turbidity ranges, thereby improving the accuracy of the attenuation coefficient.

[0054] 4. The concentration prediction of phytoplankton in water bodies in this invention is based on turbidity compensation. It eliminates scattering enhancement from the fluorescence spectrum of the actual measured sample and calculates the fluorescence intensity of the sample after turbidity compensation using the turbidity attenuation coefficient. Substituting this into the concentration prediction model, the concentration of phytoplankton in water bodies is obtained, thus improving the accuracy of concentration prediction. Attached Figure Description

[0055] Figure 1 This is a flowchart of an embodiment of the method for predicting the concentration of phytoplankton in water according to the present invention;

[0056] Figure 2 This is a linear fitting graph of the concentration of a standard sample and the fluorescence intensity of the fluorescence spectrum in an embodiment of the method for predicting the concentration of phytoplankton in water according to the present invention.

[0057] Figure 3 This is a fluorescence spectrum of a standard mixed solution with different turbidities in an embodiment of the method for predicting the concentration of phytoplankton in water according to the present invention;

[0058] Figure 4 This is the fluorescence spectrum of a standard mixed solution after eliminating scattering enhancement in an embodiment of the method for predicting the concentration of phytoplankton in water according to the present invention;

[0059] Figure 5 This is a schematic diagram of the attenuation coefficient K (CofficientK) and attenuation coefficient M (CofficientM) in an embodiment of the method for predicting the concentration of phytoplankton in water in this invention.

[0060] Figure 6 These are the true values, the predicted values ​​without any processing, the predicted values ​​with only scattering eliminated, the predicted values ​​compensated by turbidity, and the predicted values ​​compensated by scattering, obtained in the embodiments of the water phytoplankton concentration prediction method of the present invention. Detailed Implementation

[0061] This invention provides a turbidity compensation method for fluorescence detection, which can correct fluorescence spectral deviations caused by turbidity. This method can be applied to the prediction of phytoplankton concentration in water bodies, thereby improving the accuracy of phytoplankton concentration prediction and providing technical support for the fluorescence detection of phytoplankton in actual water bodies.

[0062] In this embodiment, the method for predicting the concentration of phytoplankton in water is described in detail using *Plasmodium tsao-ko* as an example. Figure 1As shown, the specific steps include:

[0063] (1) Establish a concentration prediction model for *Lycoperdon perlatum* in Qingdao.

[0064] Nine groups of *Platycodon grandiflorus* samples of different concentrations were prepared as standard samples in the laboratory. The concentration of the standard samples was determined by microscopy, and the fluorescence spectrum was measured by spectrophotometer.

[0065] In this embodiment, the excitation wavelength Ex used in the fluorescence spectrum is 470 nm, and the emission range [Em1, Em2] is [200 nm, 750 nm].

[0066] The concentration of the standard sample was linearly fitted to the fluorescence intensity of the fluorescence spectrum, and the results are as follows: Figure 2 As shown, the concentration of the first six samples shows good linearity, while the latter three samples gradually become saturated. Therefore, the concentration prediction model of *Phytoplankton spp.* in Qingdao was established using the first six samples: C = (F - 2.1183) / 0.004, with a slope coefficient of 0.004 and an intercept of 2.1183, where F is the fluorescence intensity and C is the concentration of phytoplankton in the water.

[0067] (2) Measurement of fluorescence spectra under turbidity interference

[0068] Kaolin solutions with turbidities of 0 NTU, 10 NTU, 20 NTU, 40 NTU, 60 NTU, 80 NTU, 100 NTU, 150 NTU, 200 NTU, 300 NTU, 400 NTU, and 500 NTU were prepared, along with corresponding numbers of *Platycodon grandiflorus* samples of the same concentration. The kaolin solutions of different turbidities were mixed with *Platycodon grandiflorus* samples of the same concentration to obtain standard mixed solutions of different turbidities. The fluorescence spectra of the standard mixed solutions were measured using a spectrophotometer, with the excitation wavelength and emission range being the same as in step (1). Figure 3 As shown, the fluorescence intensity gradually decreases as the turbidity increases.

[0069] Defined as follows: the turbidity of multiple standard mixed solutions constitutes a turbidity matrix N = [N1, ..., N2]. p-1 N p ], where p is the quantity of the standard mixed solution. In this embodiment, p = 11, excluding the standard mixed solution corresponding to 0 NTU. N1, ..., N p-1 N p The turbidity values ​​are the 1st, ..., p-1th, and pth standard mixed solutions arranged in ascending order of turbidity.

[0070] (3) Calculation of normalized scattering spectra S1 and S2

[0071] Fluorescence spectra of turbidity standard solutions of 20 NTU, 50 NTU, 100 NTU, and 300 NTU (kaolin solutions) were measured. Saturated spectra in the wavelength range [450 nm, 490 nm] were removed. Fluorescence spectra in the wavelength range [200 nm, 450 nm] were normalized to [0, 1], and the average scattered light intensity for each wavelength was calculated to obtain the normalized scattering spectrum S1. Similarly, normalized fluorescence spectra in the wavelength range [490 nm, 750 nm] were obtained by calculating the average scattered light intensity for each wavelength to obtain the normalized scattering spectrum S2. Normalized scattering spectra S1 and S2 were used to eliminate fluorescence enhancement in phytoplankton turbidity compensation.

[0072] The fluorescence spectrum near Ex may be saturated. Therefore, in this invention, the upper limit of the wavelength range Ex1 of the normalized scattering spectrum S1 and the lower limit of the wavelength range Ex2 of the normalized scattering spectrum S2 are adjusted according to the saturation of the fluorescence spectrum. Their values ​​satisfy: Em1<Ex1≤Ex, Ex≤Ex2<Em2. In this embodiment, Ex1=450nm and Ex2=490nm.

[0073] By definition, the turbidity of multiple turbidity standard solutions constitutes a turbidity matrix T = [T1, ..., T2]. n-1 T n ], where T1, ..., T n-1 T n These are the turbidity values ​​of the 1st, ..., n-1th, and nth turbidity standard solutions, arranged in ascending order of turbidity value.

[0074] In other embodiments of the present invention, step (3) may be performed first, followed by step (2).

[0075] (4) Eliminate scattering enhancement

[0076] In fluorescence spectroscopy measurements, the fluorescence spectrum to the left of the excitation wavelength includes only Mie scattering, while the fluorescence spectrum to the right includes not only Mie scattering but also Raman scattering and fluorescence. Since Raman scattering and fluorescence do not overlap, only Mie scattering needs to be eliminated. To eliminate the influence of Mie scattering on fluorescence, the scattering spectrum to the right of the excitation wavelength needs to be estimated first. Specifically:

[0077] For each wavelength, a linear relationship between the turbidity values ​​of multiple turbidity standard solutions and their corresponding scattered light intensities is fitted, and the wavelength range L with good linearity, i.e., R, is selected. 2 Wavelength range L, R ≥ 0.99 2 L is the coefficient of determination for the linear relationship. In this embodiment, L is [380nm, 400nm].

[0078] The enhancement coefficient H is calculated using the fluorescence spectrum of the sample under test with a wavelength range of [380nm, 400nm] and the normalized scattering spectrum S1 of that wavelength range, according to the following formula.

[0079]

[0080] Among them, S cal (L) is the fluorescence spectrum of the standard mixed solution in the wavelength range L, and S1(L) is the normalized scattering spectrum S1 in the wavelength range L.

[0081] Then, the scattering spectrum S of the standard mixed solution with a wavelength range of [490 nm, 750 nm] to the right of the excitation wavelength is estimated according to the following formula. e :

[0082] S e =H*S2

[0083] Then, as shown in the following formula, the fluorescence spectrum of the standard mixed solution S is used. cal Subtract the scattering spectrum S to the right of the excitation wavelength e The fluorescence spectrum S of the standard mixed solution after eliminating scattering enhancement was obtained. el ,like Figure 4 As shown, the fluorescence spectrum of the standard mixed solution S was completed. cal The scattering enhancement is eliminated.

[0084] S el =S cal -S e

[0085] Defined as a p×q fluorescence matrix F composed of the fluorescence spectra of multiple standard mixed solutions after scattering enhancement elimination, where q is the spectral number corresponding to each standard mixed solution, and the peak fluorescence intensities of the corresponding columns 1, ..., p-1, p are denoted as I1, ..., I... p-1 I p ;

[0086] (5) Calculate the turbidity attenuation coefficient.

[0087] The attenuation coefficient K based on the turbidity value is calculated using the following formula. i Or based on the attenuation coefficient M of the scattering spectrum i The value of i is 1, ..., p, thus obtaining the weakening coefficient matrix K = [K1, ..., K]. p Or the weakening coefficient matrix M = [M1, ..., M] p ];

[0088]

[0089]

[0090] Where F0 is the fluorescence intensity peak of the standard mixed solution of 0 NTU, and Sc i-1 and Sc i and are the average scattered light intensities of the (i-1)th and ith standard mixed solutions in the wavelength range L, respectively; where, when i = 1, N i-1 and Sc i-1 The values ​​represent the turbidity of a standard mixed solution with a turbidity of 0 NTU and its average scattered light intensity in the wavelength range L, respectively.

[0091] Different turbidity ranges have different attenuation coefficients, representing different attenuation abilities for turbidity. Since Mie scattering in fluorescence spectra is usually caused by turbidity, there is a good correlation between Mie scattering and turbidity. When the turbidity value is unknown, for samples where the turbidity value cannot be directly obtained, the intensity of scattered light can represent the turbidity. Therefore, the average fluorescence intensity of the standard mixed solution to the left of the excitation wavelength, i.e., its average scattered light intensity, can be used to replace the true turbidity value to calculate the attenuation coefficient matrix M, which has the same function as the attenuation coefficient matrix K. In other embodiments of the present invention, depending on whether the turbidity value of the sample can be obtained, either the attenuation coefficient matrix K or the attenuation coefficient matrix M is selected for calculation. Then, the fluorescence intensity of the sample after turbidity compensation is calculated based on the selected attenuation coefficient matrix to predict the concentration. In this embodiment, both the attenuation coefficient matrix K and the attenuation coefficient matrix M are calculated simultaneously. The calculated attenuation coefficients are shown in [reference needed]. Figure 5 Specifically:

[0092] K=[0.0122027913895458 0.00494979142680698 0.00213167244651394 0.00264716061268164 0.00274628205950299 0.00153901471654787

[0094] 0.00123027345486870 8.28118900272714e-05 0.000681106839663058

[0095] 6.74071125504801e-05 0.000467392278315265]

[0096] M=[0.00511164842235465 0.00309326594232077 0.00101151542608054 0.00155060617550743 0.00120794093790155 0.00101696375318115

[0098] 0.000631057094617149 5.36177902209652e-05 0.000470445245678925

[0099] 4.68428762027859e-05 0.000507487394865828]

[0100] (6) Multiple mixed solutions containing different concentrations of *Leptochloa tsao-san* and different turbidities were prepared as test samples to verify the above method. The concentrations and turbidities of *Leptochloa tsao-san* in the test samples are shown in Table 1.

[0101] Table 1. Concentration and turbidity settings of test samples for *Platycodon grandiflorus* in Qingdao

[0102]

[0103]

[0104] Obtain the fluorescence spectrum S of the sample to be tested sample To perform scattering enhancement and elimination, specifically:

[0105] The enhancement coefficient H0 of the fluorescence spectrum of the sample to be tested is calculated using the following formula;

[0106]

[0107] Among them, S sample (L) represents the fluorescence spectrum of the sample to be tested in wavelength range L;

[0108] Then, the scattering spectrum S of the sample to the right of the excitation wavelength is estimated using the following formula. e0 ;

[0109] S e0 =H0*S2

[0110] Then, the fluorescence spectrum S of the sample after eliminating scattering enhancement is calculated using the following formula. el0 ;

[0111] S el0 =S sample -S e0 .

[0112] Then, according to the following formula, calculate the fluorescence intensity I of the turbidity-compensated sample based on the attenuation coefficient matrix K and the attenuation coefficient matrix M. c1 and I c2 :

[0113]

[0114]

[0115] Among them, I e To eliminate the fluorescence spectrum S of the sample after scattering enhancement sample The corresponding peak fluorescence intensities; Δntu1, ..., Δntu m For 0NTU, N1, N2, ..., N m The gradient of two adjacent turbidity values, N m Let p be the turbidity of the m-th standard mixed solution, where m ≤ p, and let N be the turbidity of the sample to be tested. sample N sample ≤N m ;ΔSc1、…、ΔSc m For 0NTU, N1, ..., N m The gradient of the scattered light intensity corresponding to two adjacent turbidity values.

[0116] Then, the fluorescence intensity I of the turbidity-compensated sample is... c1 and I c2 Substituting these values ​​into the concentration prediction model yields the predicted values ​​compensated by K and the predicted values ​​compensated by M.

[0117] The fluorescence intensity of the sample to be tested is directly substituted into the concentration prediction model to obtain the predicted value without any processing. The sample to be tested is processed through steps (1) to (4) above to obtain the predicted value with only scattering elimination. The prediction results after the above four processing are as follows: Figure 6 As shown in Table 2, the prediction results were evaluated using root mean square error (RMSE) and relative error (RE). It can be seen that the RMSE of the compensated prediction result is 7.27 × 10⁻⁶. 3 and 1.09×10 4 The relative error is less than 4%, which is significantly better than the uncompensated prediction results.

[0118] Table 2. RMSE and RE of the predicted values ​​using K and M compensation

[0119]

[0120] In other embodiments of the present invention, the turbidity compensation method in fluorescence detection of the present invention can also be used for fluorescence detection of polycyclic aromatic hydrocarbons, dissolved organic matter, etc., to prepare corresponding turbidity standard solutions and standard mixed solutions, and to obtain the compensated fluorescence intensity using the turbidity compensation method of the present invention, thereby improving the measurement accuracy.

Claims

1. A turbidity compensation method in fluorescence detection, characterized by, The method comprises the following steps: S1, prepare a plurality of turbidity standard solutions with gradient concentrations, obtain their scattering spectra, the excitation wavelength of the scattering spectrum is Ex, and the exit interval is [Em1, Em2]; for each wavelength in the wavelength interval on the left side of the excitation wavelength, fit the linear relationship between the turbidity values of the plurality of turbidity standard solutions and the corresponding scattering light intensities, and select the wavelength interval L with R 2 ≥0.99 as the wavelength interval of the linear relationship; S2, prepare a plurality of turbidity standard solutions with gradient concentrations, obtain their scattering spectra, the excitation wavelength of the scattering spectrum is Ex, and the exit interval is [Em1, Em2]; for each wavelength in the wavelength interval on the left side of the excitation wavelength, fit the linear relationship between the turbidity values of the plurality of turbidity standard solutions and the corresponding scattering light intensities, and select the wavelength interval L with R 2 ≥0.99 as the wavelength interval of the linear relationship; S2, calculating the normalized scattering spectrum S1 of the wavelength interval on the left of the excitation wavelength and the normalized scattering spectrum S2 of the wavelength interval on the right of the excitation wavelength; S3, mixing the standard samples of the target detection object with the same concentration with the turbidity standard solutions with different concentrations respectively to prepare p standard mixed solutions with turbidity gradient, and measuring the fluorescence spectrum of the standard mixed solutions; S4, eliminating the scattering enhancement of the fluorescence spectrum of the standard mixed solutions by using the fluorescence spectrum of the standard mixed solutions in the wavelength interval L and the normalized scattering spectrum S1 and the normalized scattering spectrum S2 of the wavelength interval on the right of the excitation wavelength, to obtain the fluorescence spectrum of the standard mixed solutions after eliminating the scattering enhancement; S5、corresponding to p standard mixed solution turbidity, the weakening coefficient K based on turbidity value is calculated respectively by the following formula i or the weakening coefficient M based on scattering spectrum i , wherein i takes the value of 1, …, p, to obtain the weakening coefficient matrix K = [K1, …, K p ] or the weakening coefficient matrix M = [M1, …, M p ]. Defined as a p×q fluorescence matrix F composed of the fluorescence spectra of multiple standard mixed solutions after scattering enhancement elimination, where q is the spectral number corresponding to each standard mixed solution, and the peak fluorescence intensities of the corresponding columns 1, ..., p-1, p are denoted as I1, ..., I... p-1 I p The turbidity of multiple standard mixed solutions constitutes a turbidity matrix N = [N1, ..., N2]. p-1 N p ], N1, ..., N p-1 N p The turbidity values ​​are the 1st, ..., p-1th, and pth standard mixed solutions arranged in ascending order of turbidity. Wherein, F0 is the fluorescence intensity peak value of the standard mixed solution with 0 NTU, Sc i-1 and Sc i are the average scattering light intensity of the i-1th and the ith standard mixed solution in the wavelength interval L, respectively, wherein, i=1, N i-1 and Sc i-1 are the turbidity of the standard mixed solution with 0 NTU and its average scattering light intensity in the wavelength interval L, respectively; S6, using the standard mixed solution fluorescence spectrum of the wavelength interval L and the normalized scattering spectrum S1 and the normalized scattering spectrum S2 of the wavelength interval right of the excitation wavelength, eliminating the scattering enhancement of the measured fluorescence spectrum of the sample to be measured, and then using the weakening coefficient matrix K or the weakening coefficient matrix M to calculate the fluorescence intensity I of the sample to be measured after turbidity compensation by the following formula c1 or I c2 , complete turbidity compensation; Wherein, I e is the fluorescence peak intensity corresponding to the fluorescence spectrum of the sample to be measured after eliminating the scattering enhancement; Δntu1, …, Δntu m is 0 NTU, N1, …, N m is the gradient of the adjacent two turbidity values, N m is the turbidity of the mth standard mixed solution, m≤p, and the turbidity of the sample to be measured is denoted as N sample ; N sample ≤N m ; ΔSc1, …, ΔSc m is 0 NTU, N1, …, N m is the gradient of the scattering light intensity corresponding to the adjacent two turbidity values; The excitation wavelength and the wavelength interval of the fluorescence spectrum in steps S2 and S5 are the same as those in step S1.

2. The method for turbidity compensation in fluorescence detection according to claim 1, wherein, Step S2 is specifically: The scattering spectrum in step S1 is normalized, and then the average scattering light intensity of the turbidity standard solutions corresponding to each wavelength on the left and right of the excitation wavelength is calculated respectively to obtain the normalized scattering spectrum S1 on the left of the excitation wavelength and the normalized scattering spectrum S2 on the right of the excitation wavelength.

3. The method for turbidity compensation in fluorescence detection according to claim 2, wherein, Step S4 is specifically: S4.1, the enhancement coefficient H of the fluorescence spectrum of the standard mixed solution is calculated by using the following formula: S cal (L) is the fluorescence spectrum of the standard mixture solution in the wavelength interval L, and S1(L) is the spectrum of the standardization scattering spectrum S1 in the wavelength interval L. S4.

2. The scattering spectrum S to the right of the excitation wavelength of the standard mixed solution is calculated according to the following formula e ; S e = H * S2 S4.

3. The standard mixed solution fluorescence spectrum S after eliminating scattering enhancement is calculated by the following formula el ; S el = S cal - S e S cal is the fluorescence spectrum of the standard mixed solution measured in step S3.

4. The turbidity compensation method in the fluorescence detection according to claim 1, characterized in that: In step S6, the scattering enhancement of the measured fluorescence spectrum of the sample to be detected is eliminated by using the normalized scattering spectrum S1 and the normalized scattering spectrum S2, which is specifically: The enhancement coefficient of the fluorescence spectrum of the sample to be detected is calculated, the scattering spectrum of the wavelength interval on the right of the excitation wavelength of the sample to be detected is estimated by using the enhancement coefficient, and then the fluorescence spectrum of the sample to be detected after eliminating the scattering enhancement is obtained by subtracting the scattering spectrum of the sample to be detected from the fluorescence spectrum of the sample to be detected.

5. The turbidity compensation method in the fluorescence detection according to claim 2, characterized in that: Before normalization in step S2, the saturated spectrum in the scattering spectrum is removed.

6. A method of predicting the concentration of phytoplankton in a body of water, characterized by, The method comprises the following steps: Step 1, measuring fluorescence spectrum S of the sample to be measured sample ; Step 2, the turbidity of the sample to be detected is compensated by using the turbidity compensation method in the fluorescence detection according to any one of claims 1-5 to obtain the fluorescence intensity of the sample to be detected after turbidity compensation, and the excitation wavelength and the wavelength interval of the fluorescence spectrum are the same as those in step 1; Step 3, the concentration of the phytoplankton in the water body is predicted by substituting the fluorescence intensity of the sample to be detected after turbidity compensation into the concentration prediction model.

7. The method for predicting the concentration of phytoplankton in water according to claim 6, characterized in that, Step 2 is specifically: It is judged whether the turbidity of the sample to be detected can be obtained; If yes, the weakening coefficient matrix K is calculated by using the turbidity compensation method in the fluorescence detection according to any one of claims 1-5, and the corresponding fluorescence intensity of the sample to be detected after turbidity compensation is calculated; Otherwise, the weakening coefficient matrix M is calculated by using the turbidity compensation method in the fluorescence detection according to any one of claims 1-5, and the corresponding fluorescence intensity of the sample to be detected after turbidity compensation is calculated.

8. The concentration prediction method of the phytoplankton in the water body according to claim 6 or 7, characterized in that: In step 3, the concentration prediction model is: C=(F-b) / a Wherein, C is the concentration of water phytoplankton, a, b are the slope and intercept of the concentration prediction model respectively, F is the fluorescence intensity of the sample to be measured.

Images