A method for determining the endpoint of optical titration

Abstract

This invention discloses a method for determining the endpoint of optical titration, comprising: using a turbidity detector to detect the turbidity of different water sources, and obtaining... N × I This study uses a dataset of actual absorbance, water source turbidity, and incident light wavelength. A predictive model based on absorbance compensation during the turbidity-predictive titration process is constructed. Water samples are collected from the test water source, and its turbidity is detected using a turbidity meter. The absorbance of the test water source under different titration times and incident light wavelengths is calculated. Absorbance compensation under different wavelengths of incident light is output to correct the absorbance of the test water source. Absorbance-titer volume relationship curves under different wavelengths are plotted, and the average value of the jump points in the absorbance-titer volume relationship curves is taken to obtain the titration endpoint for water quality testing. This method solves the problem of single-wavelength optical signals being easily interfered with by turbidity and avoids the defects of traditional titration inflection point detection, such as sensitivity to noise and judgment errors.

Description

Technical Field

[0001] This invention relates to the field of water quality testing, and specifically to a method for determining the endpoint of optical titration. Background Technology

[0002] The permanganate index is the amount of oxidant consumed when treating water samples with potassium permanganate under acidic or alkaline conditions. It is expressed in milligrams per liter (mg / L) of oxygen and is mainly used to assess the water quality of drinking water, surface water, and domestic sewage. This index reflects the degree of pollution of organic and inorganic oxidizable substances in water bodies.

[0003] In the process of permanganate index detection, misjudgment of the titration endpoint caused by turbidity interference has long been a core pain point that has plagued the industry, specifically manifested in the following three aspects:

[0004] Light scattering interference from suspended matter: Turbidity in water (such as suspended particulate matter in rivers during the flood season) can cause non-specific attenuation of transmitted light intensity. This attenuation is unrelated to the colorimetric reaction itself, but it can easily be misjudged as the endpoint signal of the reaction, causing judgment bias.

[0005] Limitations of traditional single-optical-path titrators: Existing titrators rely solely on the absorbance change of a single wavelength to determine the endpoint. In high-turbidity water, the absorbance curve is affected by scattering, resulting in a distortion rate of over 40%, which seriously affects the accuracy of the test. Summary of the Invention

[0006] To address the aforementioned shortcomings of existing technologies, this invention provides an optical titration endpoint determination method. By fitting the influence of the turbidity of the water source under different wavelength incident light conditions on the absorbance error, the titration endpoint is determined using incident light of different wavelengths.

[0007] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows:

[0008] A method for determining the endpoint of optical titration is provided, comprising the following steps:

[0009] S1: Set up the target light source for the optical titration experiment, use a turbidity meter to detect the turbidity of different water sources, and collect the transmitted light intensity of incident light of different wavelengths passing through water sources with different turbidities. Calculate the actual absorbance to obtain... N×I A set of data on actual absorbance, water turbidity, and incident light wavelength;

[0010] S2: Construct a prediction model based on absorbance compensation during turbidity prediction titration, and... N×I The system inputs a set of data into the prediction model, fits the prediction model, and outputs the fitted prediction model.

[0011] S3: Collect the water source to be tested and use a turbidity meter to detect the turbidity of the water source. Set the incident light wavelength range for the titration test during water quality testing and calculate the water source at different titration times. Absorbance under incident light conditions of different wavelengths;

[0012] S4: Denoise and reconstruct the absorbance of the water source to be tested, input the turbidity of the water source to be tested into the fitted prediction model, output absorbance compensation under different wavelengths of incident light, and correct the absorbance of the water source to be tested.

[0013] S5: Plot the absorbance-titer volume relationship curves under different wavelength conditions, and take the average value of the jump points in the absorbance-titer volume relationship curves to obtain the titration endpoint when the water source to be tested is subjected to water quality testing.

[0014] Further, step S1 includes:

[0015] S11: Set up the target light source for the optical titration experiment, take water sources with different turbidities and inject them into a transparent titration container, and use a turbidity detector to detect the turbidity of the water source;

[0016] S12: Set the wavelength range of the incident light , This is the minimum value within the wavelength range. The maximum value within the wavelength range, and from the wavelength range Several wavelengths are uniformly selected within the interior as the wavelengths of different incident lights;

[0017] S13: Collect the intensity of incident light transmitted through the water source at different wavelengths and calculate the actual absorbance of the water source under different wavelengths of incident light.

[0018] ;

[0019] in, For the first i Incident light of a certain wavelength in turbidity Actual absorbance under the conditions For the first i The intensity of transmitted light at certain wavelengths. For the first i The intensity of incident light at a certain wavelength, i For the type of wavelength;

[0020] S14: Obtaining different turbidities Water sources at different incident light wavelengths Actual absorbance under the conditions Depending on the type of water source with different turbidity N and the types of incident light wavelengths I , build N×I Data sets .

[0021] Further, step S2 includes:

[0022] S21: Construct a prediction model based on absorbance compensation during turbidity prediction titration;

[0023] ;

[0024] in, These are the relationship coefficients between turbidity and incident light wavelength, respectively. It is an index of turbidity. b For the constant term of the prediction model;

[0025] S22: From N×I From a set of data sets, four data sets are randomly selected with replacement to form a fitted data set. After several rounds of selection, several fitted data sets are obtained. Duplicate fitted data sets are then removed to obtain the desired fitted data set. M One set of fitted data;

[0026] S23: sequentially M Each set of fitted data is input into the prediction model, and the output is... M The parameter sets for group fitting include correlation coefficients within each parameter set. ,index and constant term b ,get M Relation coefficients ,index and constant term , The first M The relationship coefficient between turbidity and incident light wavelength For the first M An index of turbidity, For the first M One constant term;

[0027] S24: Calculate the average value of the relationship coefficients respectively. The average value of the index and the average of constant terms And filter outout values ​​for relationship coefficients, exponents, and constant terms;

[0028] If satisfied Then determine the first m The relationship coefficient of turbidity If it is an outlier, then it is a normal value. For the calculated M The allowable fluctuation range of the relationship coefficient for each turbidity;

[0029] If satisfied Then determine the first m The relationship coefficient of the incident light wavelength If it is an outlier, then it is a normal value. For the calculated M The allowable fluctuation range of the relationship coefficient between the wavelengths of incident light;

[0030] If satisfied Then determine the first m An index of turbidity If it is an outlier, then it is a normal value. For the calculated M The permissible range of fluctuation for an index of turbidity;

[0031] If satisfied Then determine the first m constant terms If it is an outlier, then it is a normal value. For the calculated M The allowable fluctuation range of each constant term;

[0032] S25: Delete the parameter group containing outliers, leaving... Set the parameter set for the fit and calculate it. The average of the relationship coefficients, exponents, and constant terms in the parameter set of the fitted group;

[0033] ;

[0034] in, These are the turbidity correlation coefficient, the incident light wavelength correlation coefficient, the average value of the exponent and the constant term, respectively;

[0035] S26: The average of the turbidity relationship coefficient, the incident light wavelength relationship coefficient, the exponent, and the constant term. Inputting the data into the prediction model yields a fully fitted prediction model:

[0036] .

[0037] Furthermore, step S3 specifically includes:

[0038] Collect water samples from the source to be tested, and use a turbidity meter to measure the turbidity of the water. Set the incident light wavelength range for titration tests during water quality testing. According to the set wavelength interval Within the incident light wavelength range Internal extraction P Calculate at each wavelength point. P The first wavelength point p The wavelength corresponding to each wavelength point Incident light at titration time absorbance;

[0039] ;

[0040] in, Titration time The water source to be tested is affected by wavelength The absorbance of the incident light, Titration time Add the titrant volume from the water source to be tested. wavelength The intensity of the incident light, wavelength The intensity of the emitted light.

[0041] Further, step S4 includes:

[0042] S41: Using the db4 wavelet basis function to measure absorbance Perform a 3-level decomposition;

[0043] ;

[0044] in, v The number of layers in the decomposition. For the first v Layer approximation coefficient, For the first v Layer detail factor;

[0045] S42: Threshold the decomposed detail coefficients;

[0046] ;

[0047] in, The detail coefficients after thresholding. For indicator functions, when the condition is met At that time, indicator function Select 1, otherwise, the indicator function. Take 0, W wavelength The number of titrations under the given conditions, with each titration corresponding to one absorbance reading. The noise standard deviation of the detail factor. For the first v Threshold for layer detail factor;

[0048] S43: Utilizing the detail coefficients after thresholding Reconstructed absorbance after noise reduction;

[0049] ;

[0050] in, The absorbance after noise reduction;

[0051] S44: The turbidity of the water source to be tested and wavelength Input the fitted prediction model into the actual absorbance of the water source to be tested. As absorbance compensation, the absorbance after noise reduction is corrected to obtain the corrected absorbance;

[0052] .

[0053] The beneficial effects of this invention are as follows: This invention utilizes absorbance experiments to fit the absorbance relationship of water sources with different turbidities under different wavelengths of incident light, constructing a predictive model to predict absorbance compensation during the titration process, thereby improving the accuracy of titration endpoint determination. The predictive model fits data obtained from several sets of experiments, exhibiting good predictive performance. It effectively improves the reliability of absorbance compensation in subsequent optical titration processes for water quality testing, enhances the accuracy of water quality assessment during permanganate index detection, solves the problem of single-wavelength optical signals being easily interfered with by turbidity, and avoids the defects of traditional titration inflection point detection, such as sensitivity to noise and judgment errors. Attached Figure Description

[0054] Figure 1 This is a flowchart of the method for determining the endpoint of optical titration.

[0055] Figure 2 This is a schematic diagram of the absorbance-titer volume relationship curve. Detailed Implementation

[0056] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0057] like Figure 1 As shown, a method for determining the endpoint of optical titration includes the following steps:

[0058] S1: Set up the target light source for the optical titration experiment, use a turbidity meter to detect the turbidity of different water sources, and collect the transmitted light intensity of incident light of different wavelengths passing through water sources with different turbidities. Calculate the actual absorbance to obtain... N×I A set of data regarding actual absorbance, water turbidity, and incident light wavelength. Step S1 specifically includes the following steps:

[0059] S11: Set up the target light source for the optical titration experiment, take water sources with different turbidities and inject them into a transparent titration container, and use a turbidity detector to detect the turbidity of the water source;

[0060] S12: Set the wavelength range of the incident light , This is the minimum value within the wavelength range. The maximum value within the wavelength range, and from the wavelength range Several wavelengths are uniformly selected within the interior as the wavelengths of different incident lights;

[0061] S13: Collect the intensity of incident light transmitted through the water source at different wavelengths and calculate the actual absorbance of the water source under different wavelengths of incident light.

[0062] ;

[0063] in, For the first i Incident light of a certain wavelength in turbidity Actual absorbance under the conditions For the first i The intensity of transmitted light at certain wavelengths. For the first i The intensity of incident light at a certain wavelength, i For the type of wavelength;

[0064] S14: Obtaining different turbidities Water sources at different incident light wavelengths Actual absorbance under the conditions Depending on the type of water source with different turbidity N and the types of incident light wavelengths I , build N×I Data sets .

[0065] S2: Construct a prediction model based on absorbance compensation during turbidity prediction titration, and... N×I Each set of data is input into the prediction model, the model is fitted, and the fitted prediction model is output. Step S2 specifically includes the following steps:

[0066] S21: Construct a prediction model based on absorbance compensation during turbidity prediction titration;

[0067] ;

[0068] in, These are the relationship coefficients between turbidity and incident light wavelength, respectively. It is an index of turbidity. b For the constant term of the prediction model;

[0069] This invention constructs a predictive model to accurately predict the interference of nonlinear scattering of suspended particles in water on absorbance under different incident light wavelengths. This model is used for absorbance value compensation during subsequent titration, improving the accuracy of titration endpoint determination. According to experimental simulations, the absorbance of the water source is exponentially correlated with turbidity and positively correlated with the incident light wavelength. By fitting the exponential coefficient between turbidity and absorbance, and the correlation coefficient between incident light wavelength and absorbance, an accurate predictive model can be obtained. As a weight of turbidity-related terms.

[0070] In this embodiment, to facilitate fitting a prediction model with a certain accuracy, at least eight data sets are used to fit the prediction model. Naturally, the more data sets, the higher the fitting accuracy. For example, four water sources with different turbidities are used, and at wavelengths... At least 10 wavelengths of incident light were selected within the range, resulting in a total of 4 × 10 = 40 data sets are used to fit the relationship coefficients, exponents, and constant terms.

[0071] S22: From N×I From a set of data sets, four data sets are randomly selected with replacement to form a fitted data set. After several rounds of selection, several fitted data sets are obtained. Duplicate fitted data sets are then removed to obtain the desired fitted data set. M One set of fitted data;

[0072] In this embodiment, by using a set of fitted data to fit the relationship coefficients, exponents, and constant terms, a set of associated relationship coefficients can be obtained. ,index and constant term b ,use M A fitted set of data can be obtained M Group association coefficient ,index and constant term b。

[0073] S23: sequentially M Each set of fitted data is input into the prediction model, and the output is... M The parameter sets for group fitting include correlation coefficients within each parameter set. ,index and constant term b ,get M Relation coefficients ,index and constant term , The first M The relationship coefficient between turbidity and incident light wavelength For the first M An index of turbidity, For the first M One constant term;

[0074] S24: Calculate the average value of the relationship coefficients respectively. The average value of the index and the average of constant terms And filter outout values ​​for relationship coefficients, exponents, and constant terms;

[0075] If satisfied Then determine the first m The relationship coefficient of turbidity If it is an outlier, then it is a normal value. For the calculated M The allowable fluctuation range of the relationship coefficient for each turbidity;

[0076] If satisfied Then determine the first m The relationship coefficient of the incident light wavelength If it is an outlier, then it is a normal value. For the calculated M The allowable fluctuation range of the relationship coefficient between the wavelengths of incident light;

[0077] If satisfied Then determine the first m An index of turbidity If it is an outlier, then it is a normal value. For the calculated M The permissible range of fluctuation for an index of turbidity;

[0078] If satisfied Then determine the first m constant terms If it is an outlier, then it is a normal value. For the calculated M The allowable fluctuation range of each constant term;

[0079] S25: Delete parameter groups containing outliers. If a parameter group has even one outlier, the entire group must be deleted, indicating a significant error in the fitting process. Set the parameter set for the fit and calculate it. The average of the relationship coefficients, exponents, and constant terms in the parameter set of the fitted group;

[0080] ;

[0081] in, These are the turbidity correlation coefficient, the incident light wavelength correlation coefficient, the average value of the exponent and the constant term, respectively;

[0082] S26: The average of the turbidity relationship coefficient, the incident light wavelength relationship coefficient, the exponent, and the constant term. Inputting the data into the prediction model yields a fully fitted prediction model:

[0083] .

[0084] S3: Collect the water source to be tested and use a turbidity meter to detect the turbidity of the water source. Set the incident light wavelength range for the titration test during water quality testing and calculate the water source at different titration times. Absorbance under incident light conditions of different wavelengths. Step S3 specifically involves:

[0085] Collect water samples from the source to be tested, and use a turbidity meter to measure the turbidity of the water. Set the incident light wavelength range for titration tests during water quality testing. According to the set wavelength interval Within the incident light wavelength range Internal extraction P Each wavelength point, in this embodiment, is still taken as a wavelength. 61 wavelength points within the range, wavelength interval ,calculate P The first wavelength point p The wavelength corresponding to each wavelength point Incident light at titration time absorbance;

[0086] ;

[0087] in, Titration time The water source to be tested is affected by wavelength The absorbance of the incident light, Titration time Add the titrant volume from the water source to be tested. wavelength The intensity of the incident light, wavelength The intensity of the emitted light.

[0088] S4: Denoise and reconstruct the absorbance of the water source to be tested, and input the turbidity of the water source to be tested into the fitted prediction model, outputting absorbance compensation under different wavelengths of incident light to correct the absorbance of the water source to be tested. Step S4 specifically includes the following steps:

[0089] S41: Using the db4 wavelet basis function to measure absorbance Perform a 3-level decomposition;

[0090] ;

[0091] in, v The number of layers in the decomposition. For the first v Layer approximation coefficient, For the first v Layer detail factor;

[0092] S42: Threshold the decomposed detail coefficients;

[0093] ;

[0094] in, The detail coefficients after thresholding. For indicator functions, when the condition is met At that time, indicator function Select 1, otherwise, the indicator function. Take 0, W wavelength The number of titrations under the given conditions, with each titration corresponding to one absorbance reading. The noise standard deviation of the detail factor. For the first v Threshold for layer detail factor;

[0095] S43: Utilizing the detail coefficients after thresholding Reconstructed absorbance after noise reduction;

[0096] ;

[0097] in, The absorbance after noise reduction;

[0098] S44: The turbidity of the water source to be tested and wavelength Input the fitted prediction model into the actual absorbance of the water source to be tested. As absorbance compensation, the absorbance after noise reduction is corrected to obtain the corrected absorbance;

[0099] .

[0100] S5: Plot absorbance-titer volume relationship curves under different wavelength conditions, and take the average value of the jump points in the absorbance-titer volume relationship curves to obtain the titration endpoint for water quality testing of the water source to be tested. Specifically:

[0101] Plotting different wavelengths The absorbance-titer volume relationship curve under the given conditions was obtained. P The absorbance-titer volume relationship curve is shown below. Figure 2 As shown, and from PThe average value of the jump points in the absorbance-titer volume relationship curve is taken to obtain the titration endpoint for water quality testing of the water source to be tested.

[0102] This invention utilizes absorbance experiments to fit the absorbance relationship of water sources with different turbidities under different wavelengths of incident light, constructing a predictive model to predict absorbance compensation during titration, thereby improving the accuracy of titration endpoint determination. The predictive model fits data obtained from several sets of experiments, exhibiting excellent predictive performance. It effectively improves the reliability of absorbance compensation in subsequent optical titration processes for water quality testing, enhances the accuracy of water quality assessment during permanganate index detection, solves the problem of single-wavelength optical signals being easily interfered with by turbidity, and avoids the shortcomings of traditional titration inflection point detection, such as sensitivity to noise and judgment errors.

Claims

1. A method for determining the endpoint of optical titration, characterized in that, Includes the following steps: S1: Set up the target light source for the optical titration experiment, use a turbidity meter to detect the turbidity of different water sources, and collect the transmitted light intensity of incident light of different wavelengths passing through water sources with different turbidities. Calculate the actual absorbance to obtain... N×I A set of data on actual absorbance, water turbidity, and incident light wavelength; S2: Construct a prediction model based on absorbance compensation during turbidity prediction titration, and... N×I Input a set of data into the prediction model, fit the prediction model, and output the fitted prediction model. S3: Collect the water source to be tested, and use a turbidity detector to detect the turbidity of the water source to be tested. Set the incident light wavelength range for the titration test during water quality testing, and calculate the absorbance of the water source to be tested under different titration times and different wavelengths of incident light. S4: Denoise and reconstruct the absorbance of the water source to be tested, input the turbidity of the water source to be tested into the fitted prediction model, output absorbance compensation under different wavelengths of incident light, and correct the absorbance of the water source to be tested. S5: Plot the absorbance-titer volume relationship curves under different wavelength conditions, and take the average value of the jump points in the absorbance-titer volume relationship curves to obtain the titration endpoint when the water source to be tested is subjected to water quality testing.

2. The method for determining the endpoint of optical titration according to claim 1, characterized in that, Step S1 includes: S11: Set up the target light source for the optical titration experiment, take water sources with different turbidities and inject them into a transparent titration container, and use a turbidity detector to detect the turbidity of the water source; S12: Set the wavelength range of the incident light , This is the minimum value within the wavelength range. The maximum value within the wavelength range, and from the wavelength range Several wavelengths are uniformly selected within the interior as the wavelengths of different incident lights; S13: Collect the intensity of incident light transmitted through the water source at different wavelengths and calculate the actual absorbance of the water source under different wavelengths of incident light. ； in, For the first i Incident light of a certain wavelength in turbidity Actual absorbance under the conditions For the first i The intensity of transmitted light at certain wavelengths. For the first i The intensity of incident light at a certain wavelength, i For the type of wavelength; S14: Obtaining different turbidities Water sources at different incident light wavelengths Actual absorbance under the conditions Depending on the type of water source with different turbidity N and the types of incident light wavelengths I , build N×I Data sets .

3. The method for determining the endpoint of optical titration according to claim 2, characterized in that, Step S2 includes: S21: Construct a prediction model based on absorbance compensation during turbidity prediction titration; ； in, These are the relationship coefficients between turbidity and incident light wavelength, respectively. It is an index of turbidity. b For the constant term of the prediction model; S22: From N×I From a set of data sets, four data sets are randomly selected with replacement to form a fitted data set. After several rounds of selection, several fitted data sets are obtained. Duplicate fitted data sets are then removed to obtain the desired fitted data set. M One set of fitted data; S23: sequentially M Each set of fitted data is input into the prediction model, and the output is... M The parameter sets for group fitting include correlation coefficients within each parameter set. ,index and constant term b ,get M Relation coefficients ,index and constant term , The first M The relationship coefficient between turbidity and incident light wavelength For the first M An index of turbidity, For the first M One constant term; S24: Calculate the average value of the relationship coefficients respectively. The average value of the index and the average of constant terms And filter outout values ​​for the relationship coefficients, exponents, and constant terms; If satisfied Then determine the first m The relationship coefficient of turbidity If it is an outlier, then it is a normal value. For the calculated M The allowable fluctuation range of the relationship coefficient for each turbidity; If satisfied Then determine the first m The relationship coefficient of the incident light wavelength If it is an outlier, then it is a normal value. For the calculated M The allowable fluctuation range of the relationship coefficient between the wavelengths of incident light; If satisfied Then determine the first m An index of turbidity If it is an outlier, then it is a normal value. For the calculated M The permissible range of fluctuation for an index of turbidity; If satisfied Then determine the first m constant terms If it is an outlier, then it is a normal value. For the calculated M The allowable fluctuation range of each constant term; S25: Delete the parameter group containing outliers, leaving... Set the parameter set for the fit and calculate it. The average of the relationship coefficients, exponents, and constant terms in the parameter set of the fitted group; ； in, These are the turbidity correlation coefficient, the incident light wavelength correlation coefficient, the constant term, and the average value of the exponent, respectively. S26: The average value of the turbidity relationship coefficient, the incident light wavelength relationship coefficient, the constant term, and the exponent. Inputting the data into the prediction model yields a fully fitted prediction model: 。 4. The method for determining the endpoint of optical titration according to claim 3, characterized in that, Step S3 specifically involves: Collect water samples from the source to be tested, and use a turbidity meter to measure the turbidity of the water. Set the incident light wavelength range for titration tests during water quality testing. According to the set wavelength interval Within the incident light wavelength range Internal extraction P Calculate at each wavelength point. P The first wavelength point p The wavelength corresponding to each wavelength point Incident light at titration time absorbance; ； in, Titration time The water source to be tested is affected by wavelength The absorbance of the incident light, Titration time Add the titrant volume of the water source to be tested. wavelength The intensity of the incident light, wavelength The intensity of the emitted light.

5. The method for determining the endpoint of optical titration according to claim 4, characterized in that, Step S4 includes: S41: Using the db4 wavelet basis function to measure absorbance Perform a 3-level decomposition; ； in, v The number of layers in the decomposition. For the first v Layer approximation coefficient, For the first v Layer detail factor; S42: Threshold the decomposed detail coefficients; ； in, The detail coefficients after thresholding. For indicator functions, when the condition is met At that time, indicator function Select 1, otherwise, the indicator function. Take 0, W wavelength The number of titrations under the given conditions, with each titration corresponding to one absorbance reading. The noise standard deviation of the detail factor. For the first v Threshold for layer detail factor; S43: Utilizing the detail coefficients after thresholding Reconstructed absorbance after noise reduction; ； in, The absorbance after noise reduction; S44: The turbidity of the water source to be tested and wavelength Input the fitted prediction model into the actual absorbance of the water source to be tested. As absorbance compensation, the absorbance after noise reduction is corrected to obtain the corrected absorbance; 。

Images