A water quality multi-parameter detection method and system based on ultraviolet-visible light spectroscopy

Abstract

The present invention provides a water quality multi-parameter detection method and system based on ultraviolet-visible light spectroscopy, which is used to detect the concentration of turbidity, chemical oxygen demand and nitrate nitrogen in water samples. The method first obtains the water sample spectral data and converts it into absorbance, then calculates the influence factors of turbidity, chemical oxygen demand and nitrate nitrogen on the mixed sample spectrum, and divides the spectral range into three characteristic spectral segments according to the influence factors, corresponding to turbidity, chemical oxygen demand and nitrate nitrogen respectively. Finally, a multi-parameter detection model is established using the partial least squares PLS method to predict the concentration of turbidity, chemical oxygen demand and nitrate nitrogen. The method avoids the interference of severe overlap of multi-parameter spectral lines by segmented analysis of the spectrum, and improves the detection accuracy by using the method of fusing first-order derivative spectra, especially in the detection of nitrate nitrogen. The method is simple to operate and low in cost, and provides a more effective technical means for water quality monitoring and water environment protection.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater environmental parameter monitoring, and in particular to a water quality multi-parameter detection method and system based on ultraviolet-visible light spectroscopy. Background Art

[0002] In recent years, UV-visible spectroscopy, as one of the water quality detection technologies of spectral analysis, has many advantages over chemical methods. For example, there is no need to add digestion agents, no secondary pollution, short cycle, easy operation and maintenance, low cost, and can achieve online and in-situ measurement. In the past, UV-visible spectroscopy was mainly used to detect water quality using single wavelength or dual wavelength methods. However, when the components of water samples are more complex, due to the selective absorption of different components at different wavelengths, single wavelength or dual wavelength modeling cannot achieve ideal results.

[0003] Turbidity, chemical oxygen demand, and nitrate are three important indicators for monitoring seawater quality. The higher the chemical oxygen demand, the higher the content of reducing substances (such as organic matter) in the water body, and reducing substances can reduce the content of dissolved oxygen in the water body, causing aquatic organisms to lack oxygen and even die, and the water quality to become corrupt and smelly. The increase in nitrate mass concentration will have a serious impact on the aquatic ecosystem, leading to eutrophication, toxic algal blooms, and hypoxia.

[0004] There is serious overlapping interference between the spectral lines of turbidity, chemical oxygen demand and nitrate nitrogen, which makes it difficult to determine the parameter concentration using UV-visible spectroscopy. Summary of the invention

[0005] The present invention overcomes the shortcomings of the prior art and provides a water quality multi-parameter detection method and system based on ultraviolet-visible light spectroscopy, which is used to detect the concentration of turbidity, chemical oxygen demand and nitrate nitrogen in water samples, avoiding the serious overlapping interference problem of multi-parameter spectral lines.

[0006] To achieve the above object, the technical solution adopted by the present invention is: a water quality multi-parameter detection method based on ultraviolet-visible light spectroscopy, comprising the following steps:

[0007] S1, obtaining the spectrum data of water samples and converting the spectrum data into absorbance;

[0008] S2. Calculate the influence factors of turbidity, chemical oxygen demand and nitrate nitrogen on the spectrum of mixed samples;

[0009] S3. According to the influencing factors, the spectral range is divided into three characteristic spectral segments, corresponding to turbidity, chemical oxygen demand and nitrate nitrogen respectively;

[0010] S4. Establish a multi-parameter detection model to predict the concentrations of turbidity, chemical oxygen demand, and nitrate nitrogen.

[0011] Further, step S1 includes:

[0012] S11. Scan the water sample using an ultraviolet-visible spectrometer to obtain absorbance data of the water sample in the wavelength range of 190nm-720nm.

[0013] S12. Use the Beer-Lambert law to convert the spectral data into absorbance using the formula: .in, is absorbance; is the outgoing light intensity; is the incident light intensity; is the transmittance, which is the ratio of the outgoing light intensity to the incident light intensity.

[0014] Further, step S2 includes:

[0015] S21. Calculate the influencing factor of turbidity using the formula: ,in, is the influence factor of turbidity on the spectrum of mixed solution; is the spectrum of mixed solution; It is the spectrum of a mixture containing equal concentrations of chemical oxygen demand and nitrate nitrogen and without turbidity.

[0016] S22. Calculate the influencing factor of chemical oxygen demand using the formula: ,in is the factor affecting the spectrum of mixed solution by chemical oxygen demand; is the spectrum of mixed solution; It is the spectrum of a mixture containing equal concentrations of turbidity and nitrate nitrogen and no chemical oxygen demand.

[0017] S23. Calculate the influence factor of nitrate nitrogen using the formula: ,in is the influence factor of nitrate nitrogen on the spectrum of mixed solution; is the spectrum of the mixed solution, It is the spectrum of a mixture containing equal concentrations of turbidity and chemical oxygen demand and without nitrate nitrogen.

[0018] Further, step S3 includes:

[0019] S31. Analyze the changing trend of impact factors;

[0020] S32. Divide the characteristic spectral bands, and use 310-720 nm as the characteristic spectral band of turbidity, 250-310 nm as the characteristic spectral band of chemical oxygen demand, and 190-250 nm as the characteristic spectral band of nitrate nitrogen.

[0021] Furthermore, step S4 includes: establishing a multi-parameter detection model using a partial least squares (PLS) method according to characteristic spectra of turbidity, chemical oxygen demand and nitrate nitrogen.

[0022] Furthermore, when the partial least squares PLS method is used to analyze nitrate nitrogen, the characteristic spectrum of nitrate nitrogen is processed by fusing the first-order derivative spectrum.

[0023] Furthermore, when analyzing turbidity using the partial least squares PLS method, the absorbance of the water sample in the 250–400 nm band was referenced;

[0024] When using the partial least squares PLS method to analyze chemical oxygen demand, the absorbance of the water sample in the 220-275 nm band is referenced;

[0025] When using the partial least squares PLS method to analyze nitrate nitrogen, the absorbance of the water sample in the 195-230nm band is used as a reference.

[0026] Preferably, the method for fusing the first-order derivative spectrum is direct fusion, which adds the first-order derivative spectrum data to the original spectrum data.

[0027] The present invention also provides a water quality multi-parameter detection system based on ultraviolet-visible light spectroscopy, which is used to implement the above detection method and includes the following modules:

[0028] Data acquisition and preprocessing module, used to scan water samples, obtain spectral data, and convert spectral data into absorbance data;

[0029] Impact factor calculation module, used to calculate the impact factors of turbidity, chemical oxygen demand and nitrate nitrogen on the spectrum of mixed solutions;

[0030] Characteristic spectrum segment identification module, used to analyze the changing trend of influencing factors and divide characteristic spectrum segments;

[0031] Multi-parameter detection model module, used to train the model for prediction using characteristic spectra of turbidity, chemical oxygen demand and nitrate nitrogen, as well as standard solutions of known concentrations;

[0032] The nitrate nitrogen processing module is used to process the characteristic spectrum of nitrate nitrogen and fuse the characteristic spectrum of nitrate nitrogen with its first-order derivative spectrum.

[0033] The present invention solves the defects existing in the background technology and has the following beneficial effects:

[0034] (1) The present invention proposes a water quality multi-parameter detection method based on ultraviolet-visible light spectroscopy, which can efficiently and conveniently detect turbidity, chemical oxygen demand and nitrate nitrogen at the same time, and significantly improve the detection accuracy. The method is simple to operate and low in cost, providing a more effective technical means for water quality monitoring and water environment protection.

[0035] (2) The present invention proposes a multi-parameter water quality detection method based on ultraviolet-visible light spectroscopy. By calculating the influence of each component on the spectrum of the mixed water sample, the spectral range is divided into multiple characteristic spectral segments, which correspond to turbidity, chemical oxygen demand and nitrate nitrogen respectively, and multiple parameters can be detected simultaneously. The traditional chemical analysis method requires separate detection of turbidity, chemical oxygen demand and nitrate nitrogen, which is complicated to operate. However, the present invention does not require separate detection, simplifies the operation steps, improves detection efficiency, and can effectively reduce detection costs.

[0036] (3) The present invention proposes a multi-parameter water quality detection method based on ultraviolet-visible light spectroscopy. When analyzing nitrate nitrogen, the first-order derivative spectrum is integrated to effectively enhance the details of the spectral signal and highlight the slope change of the spectrum. When the components of the water sample are complex, the traditional single-wavelength method or dual-wavelength method is easily interfered by the complex background. By integrating the first-order derivative spectrum, the clarity of the nitrate nitrogen signal can be improved, making its characteristic absorption peak more prominent in the complex background, effectively reducing background interference, improving detection accuracy and reliability of detection results, and providing more accurate data support for water quality safety assessment.

[0037] (4) The present invention divides the spectral range into multiple characteristic spectral segments by segmented spectrum analysis, which correspond to turbidity, chemical oxygen demand and nitrate nitrogen, respectively, and realizes the simultaneous detection of multiple parameters. At the same time, when analyzing nitrate nitrogen, the first-order derivative spectrum is integrated to effectively enhance the details of the spectral signal, reduce background interference, and significantly improve the detection accuracy of nitrate nitrogen with weak spectral signals, providing a more effective technical means for water quality monitoring and water environment protection, and has significant advantages in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

[0039] Figure 1 This is a flow chart of a water quality multi-parameter detection method based on ultraviolet-visible light spectroscopy provided by the present invention;

[0040] Figure 2 is a schematic diagram of turbidity prediction results in a preferred embodiment of the present invention;

[0041] Figure 3 is a schematic diagram of chemical oxygen demand prediction results of a preferred embodiment of the present invention;

[0042] Figure 4 It is a schematic diagram of nitrate nitrogen prediction results according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0045] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the scope of protection of the present application. In addition, the terms "first", "second", etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the invention, unless otherwise specified, "multiple" means two or more.

[0046] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0047] Application Overview:

[0048] Water quality testing methods can be mainly divided into three categories: chemical methods, physical methods, and biosensor methods.

[0049] There are two main types of chemical analysis methods: chemical analysis and electrochemical analysis. Chemical analysis is based on the principle of chemical reaction, and quantitatively or qualitatively analyzes the components in water through color changes, precipitation, gas release, electrolyte solubility changes, etc. produced by chemical reactions. Electrochemical analysis detects water quality based on the electrochemical reaction between electrodes and solutions, and quantitatively or qualitatively analyzes the components in water through changes in electrode potential.

[0050] Biosensing uses the specific interaction between biomolecules and pollutants to detect water quality. It combines biological recognition function with signal conversion function to achieve the detection of pollutants.

[0051] The main physical methods are chromatography, mass spectrometry and direct spectroscopy. Mass spectrometry is a technique for analyzing the molecular mass and structure of a substance by ionizing the sample molecules, separating them according to their molecular mass in an electric or magnetic field, and finally detecting and analyzing these ions. Chromatography is a technique for separating and detecting different components in a mixture based on the difference in the distribution coefficients between the stationary phase and the mobile phase. Direct spectroscopy includes ultraviolet-visible spectroscopy, near-infrared spectroscopy and Raman spectroscopy, etc., which analyze samples based on the absorption, emission or scattering properties of light by the substance.

[0052] At present, there are several main methods for analyzing water quality parameters by UV-visible spectroscopy: single wavelength analysis method, dual wavelength compensation method, and spectral difference method.

[0053] Single wavelength analysis method: select a characteristic wavelength of the substance to be tested, analyze and calculate;

[0054] Dual wavelength compensation method: select another wavelength that is less affected by the substance to be tested to compensate the spectrum at the characteristic wavelength of the substance to be tested;

[0055] Spectral difference method: In the spectrum of a mixture system, in order to understand the structure of one of the components, the spectrum of another component can be obtained by subtracting the absorbance spectrum and deducting the spectrum of one component from the spectrum of the mixture.

[0056] However, the chemical analysis method has the disadvantages of complex operation, reagent consumption, secondary pollution, long measurement cycle, and difficulty in online detection. The electrochemical analysis method has the disadvantages of short life, single detection index, and multiple sensors and detection circuits for multiple parameters. The biosensor method has the problem of sensor regeneration. The chromatography method has the disadvantages of complex and expensive instruments, high maintenance costs, sample pretreatment, and long test cycle. The mass spectrometry method also has the disadvantages of expensive equipment and complex analysis. When the components of the water sample are more complex, due to the selective absorption of different wavelengths by different components, single-wavelength or dual-wavelength modeling cannot achieve the ideal effect. In practical applications, absorbance does not strictly meet the additivity, and the spectral difference method needs to obtain the compensation spectra of all different concentrations of the solution composed of other parameters in addition to the parameter to be measured in advance. The workload is large and not convenient and fast enough.

[0057] Exemplary methods:

[0058] like Figure 1 As shown, a water quality multi-parameter detection method based on ultraviolet-visible light spectroscopy includes the following steps:

[0059] S1, obtaining the spectrum data of water samples and converting the spectrum data into absorbance;

[0060] S2. Calculate the influence factors of turbidity, chemical oxygen demand and nitrate nitrogen on the spectrum of mixed samples;

[0061] S3. According to the influencing factors, the spectral range is divided into three characteristic spectral segments, corresponding to turbidity, chemical oxygen demand and nitrate nitrogen respectively;

[0062] S4. Establish a multi-parameter detection model to predict the concentrations of turbidity, chemical oxygen demand, and nitrate nitrogen.

[0063] Below, each step will be described in detail.

[0064] Step S1 includes:

[0065] S11. Scan the water sample using an ultraviolet-visible spectrometer to obtain absorbance data of the water sample in the wavelength range of 190nm-720nm.

[0066] S12. Use the Beer-Lambert law to convert the spectral data into absorbance using the formula: .in, is absorbance; is the outgoing light intensity; is the incident light intensity; is the transmittance (transmittance), which is the intensity of the emitted light ( ) and the incident light intensity ( ) ratio.

[0067] The most sensitive wavelength range is selected for analysis for different water quality parameters, spectral data is acquired and converted into absorbance to ensure data quality.

[0068] Step S2 includes:

[0069] The influence factors of turbidity, chemical oxygen demand, and nitrate nitrogen on the spectrum of the mixed solution were calculated respectively, that is, the contribution of each component to the spectrum of the mixed sample at different wavelengths was calculated.

[0070] S21. Calculate the influencing factor of turbidity using the formula: ,in, is the influence factor of turbidity on the spectrum of mixed solution; is the spectrum of mixed solution; It is the spectrum of a mixture containing equal concentrations of chemical oxygen demand and nitrate nitrogen and without turbidity.

[0071] S22. Calculate the influencing factor of chemical oxygen demand using the formula: ,in is the factor affecting the spectrum of mixed solution by chemical oxygen demand; is the spectrum of mixed solution; It is the spectrum of a mixture containing equal concentrations of turbidity and nitrate nitrogen and no chemical oxygen demand.

[0072] S23. Calculate the influence factor of nitrate nitrogen using the formula: ,in is the influence factor of nitrate nitrogen on the spectrum of mixed solution; is the spectrum of the mixed solution, It is the spectrum of a mixture containing equal concentrations of turbidity and chemical oxygen demand and without nitrate nitrogen.

[0073] By calculating the influence factor of each parameter on the mixed sample spectrum, the contribution of each parameter in the mixed sample can be distinguished more accurately, reducing the possibility of cross-interference.

[0074] Step S3 includes:

[0075] S31. Analyze the changing trend of the influencing factor and look for the wavelength region with a larger influencing factor, that is, the absorption of this parameter in this region is more obvious. At the same time, the influence of other parameters in this region should also be considered, and try to choose the region with less influence of other parameters.

[0076] S32. Divide the characteristic spectral segments of turbidity, chemical oxygen demand and nitrate nitrogen according to the changing trends of the influencing factors.

[0077] Turbidity: 310-720nm is selected as the characteristic spectral band because within this band, turbidity has a greater impact on the spectrum, while the impact of chemical oxygen demand and nitrate nitrogen is relatively small.

[0078] Chemical oxygen demand: 250-310nm is selected as the characteristic spectral band because within this band, chemical oxygen demand has a greater impact on the spectrum, while turbidity and nitrate nitrogen have relatively smaller effects.

[0079] Nitrate nitrogen: 190-250nm is selected as the characteristic spectral band because within this band, nitrate nitrogen has a greater impact on the spectrum, while the impacts of turbidity and chemical oxygen demand are relatively small.

[0080] It should be noted that these characteristic spectral segments are determined based on the results of the influencing factor analysis within the entire spectral range (190nm-720nm) to identify which wavelength range is most relevant for a certain parameter; the absorbance data of the corresponding characteristic spectral segments are selected, that is, only the characteristic spectral segment data of the corresponding component are used for subsequent analysis.

[0081] Step S4 includes:

[0082] The multi-parameter detection model is established using the partial least squares PLS method. The PLS method is a regression analysis method used to process multivariate data and can effectively process the relationship between multiple independent variables and multiple dependent variables. The specific method is as follows:

[0083] Set up the independent variable matrix , contains the absorbance data of each water sample within a specific wavelength range, each row represents a water sample, and each column represents the absorbance value at a wavelength.

[0084] Set up the dependent variable matrix , contains the concentration values ​​of the parameters to be measured (turbidity, chemical oxygen demand, nitrate nitrogen) in each water sample. Each row represents a water sample, and each column represents the concentration of a parameter to be measured.

[0085] (1) Find the matrix The eigenvector corresponding to the largest eigenvalue Get the component score vector , and the residual matrix ,in .

[0086] (2) Find the matrix The eigenvector corresponding to the largest eigenvalue Get the component score vector , and the residual matrix ,in .

[0087] (3)…….

[0088] (4) Find the matrix The eigenvector corresponding to the largest eigenvalue , find the component score vector .

[0089] According to the cross validity, it is determined that a total of r components are extracted , we get a prediction model, then exist The ordinary least squares regression equation on is: .in, , ... is the regression coefficient matrix.

[0090] Will Substitution , that is, the partial least squares regression equation of p dependent variables is obtained, and .in, satisfy , ; Fr is the residual matrix, which represents the part of the dependent variable that cannot be explained by the component score vector.

[0091] Furthermore, when establishing a multi-parameter detection model, data in a specific wavelength range are selected for partial least squares (PLS) modeling: when using the partial least squares PLS method to analyze turbidity, the absorbance of water samples in the 250-400nm band is referenced; when using the partial least squares PLS method to analyze chemical oxygen demand, the absorbance of water samples in the 220-275nm band is referenced; when using the partial least squares PLS method to analyze nitrate nitrogen, the absorbance of water samples in the 195-230nm band is referenced. The selection of the above bands is based on further refinement of the experimental data and theoretical analysis of each parameter, and is determined as the wavelength range that best represents the change of the parameter to ensure that the spectral data used in the modeling process can best reflect the information of the target parameter, and the selection of more specific bands can reduce the interference of other factors and improve the accuracy and reliability of the model.

[0092] Furthermore, when using the partial least squares PLS method to analyze nitrate nitrogen, the characteristic spectrum of nitrate nitrogen is processed, and the specific processing method is to fuse the first-order derivative spectrum. This is because the characteristic absorption peak of nitrate nitrogen in the UV-visible spectrum is weak and easily affected by the complex background, especially when other substances are present in the water sample, its spectral signal may be masked, resulting in reduced detection accuracy. The first-order derivative spectrum processing can extract and highlight subtle features or signal changes in the spectrum. The derivative of the spectrum can be obtained by calculating the difference between consecutive spectral data points. First-order derivative processing can highlight the slope change or peak information in the spectrum, reveal subtle features, edges and change trends in the spectrum, and provide information about sample composition, concentration and reaction. By inputting the spectral information together with the first-order derivative information into the independent variable matrix and the dependent variable matrix In this method, the recognition accuracy can be effectively improved, and since the first-order derivative spectrum is directly calculated through the spectrum to be measured, there is no need to use the spectral information of other samples that do not contain a certain component to be measured, which is convenient and fast. Among them, the background refers to all other signals and noises except the target signal (i.e., the characteristic absorption peak of nitrate nitrogen) in the spectral analysis. These background signals include but are not limited to the following: absorption of solvents and other components, scattering effects, instrument noise, environmental factors, and sample inhomogeneity. In the present invention, the specific method of fusing the first-order derivative spectrum is direct fusion, which directly adds the first-order derivative spectrum data to the original spectrum data.

[0093] In a specific embodiment, seven groups of experimental samples are set, and each group of experimental samples is shown as the sample true value in Table 1:

[0094] Table 1

[0095]

[0096] The detection method provided by the present invention is used to obtain seven groups of experimental sample prediction value data. The prediction value data of each group of experimental samples are shown in Table 2. Figures 2 to 4 As shown:

[0097] Table 2

[0098]

[0099] Based on the predicted data, the model determination coefficient of the multi-parameter detection model in the present invention is shown in Table 3:

[0100] Table 3

[0101]

[0102] R-Square indicates the proportion of the dependent variable variation that the model can explain, and its value ranges from 0 to 1.

[0103] The closer R-Square is to 1, the better the model fit is and the stronger the model prediction ability is.

[0104] The closer R-Square is to 0, the worse the model fit is and the weaker the model prediction ability is.

[0105] The calculation formula of the model determination coefficient R-Square is: Among them, SSR is the residual sum of squares, which represents the sum of squares of the difference between the predicted value and the true value; SST is the total sum of squares, which represents the sum of squares of the difference between the true value and the mean of the true value.

[0106] In this embodiment, the model determination coefficient R-Square is 0.99, 0.99, and 0.98, respectively, indicating that the model fits turbidity, chemical oxygen demand, and nitrate nitrogen well, and the model prediction results are accurate. This is because the spectral range is divided into three characteristic spectral segments, corresponding to turbidity (250-400nm), chemical oxygen demand (220-275nm), and nitrate nitrogen (195-230nm), respectively, to avoid interference caused by severe overlap of multi-parameter spectral lines. Each parameter has its own specific wavelength region, in which the parameter has a greater impact on the spectrum, while the impact of other parameters is smaller, thereby improving the detection accuracy. Secondly, for the detection of nitrate nitrogen, a method of fusing the first-order derivative spectrum is particularly used. In the case of weak nitrate nitrogen absorption peaks, the details of the spectral signal are enhanced, further improving the detection accuracy of nitrate nitrogen.

[0107] Example systems:

[0108] A water quality multi-parameter detection system based on ultraviolet-visible light spectroscopy is used to implement the above detection method, including the following modules:

[0109] Data acquisition and preprocessing module, used to scan water samples, obtain spectral data, and convert spectral data into absorbance data;

[0110] Impact factor calculation module, used to calculate the impact factors of turbidity, chemical oxygen demand and nitrate nitrogen on the spectrum of mixed solutions;

[0111] Characteristic spectrum segment identification module, used to analyze the changing trend of influencing factors and divide characteristic spectrum segments;

[0112] Multi-parameter detection model module, used to train the model for prediction using characteristic spectra of turbidity, chemical oxygen demand and nitrate nitrogen, as well as standard solutions of known concentrations;

[0113] The nitrate nitrogen processing module is used to process the characteristic spectrum of nitrate nitrogen and fuse the characteristic spectrum of nitrate nitrogen with its first-order derivative spectrum.

[0114] The above is based on the ideal embodiment of the present invention. Through the above description, relevant personnel can make various changes and modifications without departing from the technical concept of the present invention. The technical scope of the present invention is not limited to the content in the specification, and the technical scope must be determined according to the scope of the claims.

Claims

1. A water quality multi-parameter detection method based on ultraviolet-visible light spectroscopy, characterized in that: The following steps are involved: S1, obtaining spectral data of water samples and converting the spectral data into absorbance; S2. Calculate the influence factors of turbidity, chemical oxygen demand and nitrate nitrogen on the spectrum of mixed samples; S3. According to the influencing factors, the spectral range is divided into three characteristic spectral segments, corresponding to turbidity, chemical oxygen demand and nitrate nitrogen respectively; S4. Based on the characteristic spectra of turbidity, chemical oxygen demand and nitrate nitrogen, a multi-parameter detection model was established using the partial least squares PLS method to predict the concentrations of turbidity, chemical oxygen demand and nitrate nitrogen; In step S2: Calculate the influence factor of turbidity using the formula: ,in, is the influence factor of turbidity on the spectrum of mixed solution; is the spectrum of mixed solution; It is the spectrum of a mixture containing equal concentrations of chemical oxygen demand and nitrate nitrogen and without turbidity; To calculate the influencing factor of chemical oxygen demand, use the formula: ,in is the factor affecting the spectrum of mixed solution by chemical oxygen demand; is the spectrum of mixed solution; It is the spectrum of a mixture containing equal concentrations of turbidity and nitrate nitrogen and no chemical oxygen demand; To calculate the impact factor of nitrate nitrogen, use the formula: ,in is the influence factor of nitrate nitrogen on the spectrum of mixed solution; is the spectrum of the mixed solution, It is the spectrum of a mixture containing equal concentrations of turbidity and chemical oxygen demand and without nitrate nitrogen; In step S4, when the partial least squares PLS method is used to analyze nitrate nitrogen, the characteristic spectrum of nitrate nitrogen is fused with the first-order derivative spectrum.

2. The method for multi-parameter water quality detection based on ultraviolet-visible light spectroscopy according to claim 1, characterized in that: Step S1 includes: S11, scanning the water sample using an ultraviolet-visible spectrometer to obtain absorbance data of the water sample in the wavelength range of 190nm-720nm; S12. Use the Beer-Lambert law to convert the spectral data into absorbance using the formula: ;in, is absorbance; is the outgoing light intensity; is the incident light intensity; is the transmittance, which is the ratio of the outgoing light intensity to the incident light intensity.

3. The method for multi-parameter water quality detection based on ultraviolet-visible light spectroscopy according to claim 1, characterized in that: Step S3 includes: S31. Analyze the changing trend of impact factors; S32. Divide the characteristic spectral bands, and use 310-720 nm as the characteristic spectral band of turbidity, 250-310 nm as the characteristic spectral band of chemical oxygen demand, and 190-250 nm as the characteristic spectral band of nitrate nitrogen.

4. The method for multi-parameter water quality detection based on ultraviolet-visible light spectroscopy according to claim 1, characterized in that: When analyzing turbidity using the partial least squares PLS method, the absorbance of the water sample in the 250-400 nm band is referenced; When using the partial least squares PLS method to analyze chemical oxygen demand, the absorbance of the water sample in the 220-275 nm band is referenced; When using the partial least squares PLS method to analyze nitrate nitrogen, the absorbance of the water sample in the 195-230 nm band is used as a reference.

5. The method for multi-parameter water quality detection based on ultraviolet-visible light spectroscopy according to claim 1, characterized in that: The method of fusing the first-order derivative spectrum is direct fusion, which adds the first-order derivative spectrum data to the original spectrum data.

6. A water quality multi-parameter detection system based on ultraviolet-visible light spectroscopy, used to implement the detection method described in any one of claims 1 to 5, characterized in that: Includes the following modules: Data acquisition and preprocessing module, used to scan water samples, obtain spectral data, and convert spectral data into absorbance data; Impact factor calculation module, used to calculate the impact factors of turbidity, chemical oxygen demand and nitrate nitrogen on the spectrum of mixed solutions; Characteristic spectrum segment recognition module, used to analyze the changing trend of influencing factors, Divide the characteristic spectrum into segments; Multi-parameter detection model module, used to train the model for prediction using characteristic spectra of turbidity, chemical oxygen demand and nitrate nitrogen, as well as standard solutions of known concentrations; The nitrate nitrogen processing module is used to process the characteristic spectrum of nitrate nitrogen and fuse the characteristic spectrum of nitrate nitrogen with its first-order derivative spectrum.

Images