Automatic noise reduction titration end point judgment method and system

Abstract

The invention provides an automatic noise reduction titration end point judgment method and system, and relates to the technical field of photometric titration analysis. The method comprises the following steps: performing slicing and smooth data preprocessing on luminosity data to be detected, gradually searching a first end point based on peak positions of second-order difference and first-order difference, determining the position of the first end point of titration, and correcting; judging the condition that a plurality of end points exist, and determining the position of a second end point by utilizing a specific judgment condition; through the steps, the interference of end point change caused by noise is eliminated, the stability and the accuracy of an instrument are improved, the accuracy and the consistency of data are improved, specific parameters of differential and smooth operation can be flexibly adjusted and optimized according to actual conditions, and the application range and the practicability of the method are improved; according to the method, the second-order differential end point is firstly calculated, then the first-order differential end point is calculated, and correction is carried out, so that the titration end points of various luminosity curves are relatively stably determined, and the accuracy and efficiency of titration end point judgment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of photometric titration analysis, and in particular to a method and system for judging titration end points with automatic noise reduction. Background Art

[0002] Photometric titration is a volumetric analysis method based on Lambert-Beer's law. It uses a non-contact photometric sensing probe to record the change in absorbance during the titration process, thereby obtaining the titration end point. It is applicable to various titration objects that identify the end point through color change. Due to its highly accurate analysis results and low cost, it is widely used in chemical industries such as metallurgy and monitoring. To meet the needs of batch detection, a method for judging titration end points with automatic noise reduction has also emerged.

[0003] In the prior art, on the one hand, due to the wide variety of detection objects, such as detecting the permanganate index by titrating oxalate with permanganate, controlling the content of sulfur trioxide to control the wet-process phosphoric acid production process, and detecting zinc content by EDTA complexometric method, etc. Under different reagent conditions and processes, the photometric range and change trend of the photometric curve are different, and it is necessary to optimize the end point judgment for specific reagent types, which increases the debugging burden on on-site workers; on the other hand, in the industrial environment, affected by instrument aging, reagent replacement, and liquid turbidity, etc., obvious noise is generated in the absorbance curve, which affects the end point judgment. Currently, the publicly available methods for judging titration end points with automatic noise reduction usually only optimize for a certain process and rarely involve the scheme of automatic noise reduction optimization. Summary of the Invention

[0004] The main purpose of the present invention is to provide a method for judging titration end points with automatic noise reduction, which solves the technical problems in the prior art that the applicable range of the method for judging titration end points with automatic noise reduction is narrow, the need for corresponding debugging when changing reagents increases the manual burden, and it is easily affected by noise.

[0005] To solve the above technical problems, the technical solution adopted by the present invention is: A method for judging titration end points with automatic noise reduction, comprising the following steps: S1. Data preprocessing: Obtain the photometric data to be detected and perform data preprocessing, including slicing and smoothing operations; S2. Finding the first end point: Based on the peak positions of the second-order difference and the first-order difference of the data after data preprocessing, gradually find the position where the first titration end point is located and perform correction; S3. Finding the second end point: When there may be multiple titration end points, judge based on the maximum value, minimum value and their sign relationship between the first end point and the second end point to determine the position of the second end point.

[0006] In the preferred solution, in step S1, Gaussian filtering with a variable window length is used for the smoothing operation, specifically: S11: Perform variable-window long Gaussian filtering on the photometric data x for n times. The smoothing window length w = 5 + 2 * n, and the Gaussian kernel calculation formula is: ; In the formula, is the standard deviation, and μ is the mean; S12: After performing edge compensation, calculate the noise using the photometric data after each smoothing. The formula is: ; ; In the formula, i is the smoothing times, j is the data point position, j_max represents the number of photometric data points, and x i,j represents the intensity of the j-th photometric data point after the i-th smoothing. abs is to take the absolute value, is the intermediate noise, is the final noise; S13: Judge the noise. If the current noise is within the preset noise threshold range, stop the judgment; S14: Average the smoothing times of each-order difference data and round up to an integer to obtain the actual smoothing times used in the application .

[0007] In the preferred solution, in step S2, gradually search based on the peak position of the second-order difference and the peak position of the first-order difference, that is, first find the end point of x2, and then find the end point of x1 near the end point of x2 as the first end point. Specifically, it includes: S21. Find the end point of x2: Determine the end point of x2 by calculating the baseline intensity and peak threshold of x2; S22. Find the end point of x1: Near the end point of x2, determine the end point of x1, that is, the titration end point, by calculating the baseline intensity and peak threshold of x1; Among them, x2 represents the second-order difference, and x1 represents the first-order difference.

[0008] In the preferred solution, step S21 includes calculating the baseline intensity of x2, the peak threshold, the baseline intensity of x1, and the peak threshold, and then obtaining the titration end point. Specifically: S211: Calculate the average value mean2 and standard deviation value std2 of x2 and make a judgment; S212: Eliminate the data points that satisfy x2 > mean2 + t 1 * std2 or x2 < mean2 - t 1 * std2, and then calculate mean2 and std2, and perform iterative elimination loops until no points satisfy the condition and stop; where t 1 is the baseline threshold coefficient; S213: Calculate the mean2 value at this time as the baseline intensity value. The x2 peak threshold calculation formula is; ; In the formula, is the peak threshold of x2, is the peak threshold coefficient of x2, is to take the maximum value, is to take the minimum value; S214: According to Find the x2 end point: If the current x2[j] < mean2 - c3 or x2[j] > mean2 + c3, and x2[j + 2] < mean2 - c3 or x2[j + 2] > mean2 + c3, then the current data point is the end point of x2; S215: Obtain the x1 end point: If the absolute value maximum point of x1 is less than 0, then x1 = -x1; Similar to the process of finding the baseline intensity of x1, obtain mean1 and std1. The x1 peak threshold calculation formula is: ; In the formula, is the x1 peak threshold coefficient, is the x1 noise threshold coefficient; Similar to the process of finding the end point of x2, find the x1 end points on the left and right sides of the x2 end point respectively, and take the end point closer to the x2 end point .

[0009] In the preferred solution, in step S2, a correction formula is used to correct the first end point position to make it closer to the end point of the original photometric data. The formula is: ; In the formula, is the corrected first end point position, is the first end point position before correction, is the number of smoothing times.

[0010] In the preferred solution, when there are multiple titration end points in step S3, the specific judgment includes: If all x2 are greater than zero or less than zero, then there is 1 end point; If the maximum and minimum values of x1 have different signs, and the maximum and minimum values of x2 have the same sign, then there is 1 end point; If the maximum and minimum values of x1 have different signs, and the maximum and minimum values of x2 have different signs, and the minimum value of x2 does not exceed 0.3 times the maximum value, then there is 1 inflection point; Otherwise, there are 2 end points. At this time, starting from the first end point, traverse x1 to the right and apply the process of S2 to find the end point again to determine the second end point.

[0011] An automatic noise reduction titration endpoint judgment system, comprising: A data preprocessing module, configured to obtain photometric data to be detected and perform data preprocessing, including slicing and smoothing operations; A first endpoint searching module, configured to gradually search for the data after data preprocessing based on the peak positions of the second-order difference and the first-order difference, obtain the position where the first titration endpoint is located, and perform correction; A second endpoint searching module, configured to, when there may be multiple titration endpoints, make a judgment based on the maximum value, minimum value and their sign relationship between the first endpoint and the second endpoint, and determine the position of the second endpoint.

[0012] An electronic device, comprising: a processor, a memory, and a computer program stored on the memory and executable on the processor, where when the computer program is executed by the processor, the automatic noise reduction titration endpoint judgment method is implemented.

[0013] A computer-readable storage medium, comprising: a computer program stored on the computer-readable storage medium, where when the computer program is executed by a processor, the automatic noise reduction titration endpoint judgment method is implemented.

[0014] The present invention provides an automatic noise reduction titration endpoint judgment method and system. By performing slicing and smoothing data preprocessing on the photometric data to be detected, and then gradually searching for the first endpoint based on the peak positions of the second-order difference and the first-order difference, the position of the first titration endpoint is determined and corrected; then, according to the characteristics of the titration curve and the analysis requirements, the situation of multiple endpoints is judged, and the specific judgment conditions are used to determine the position of the second endpoint; through the above steps, the interference of endpoint variation caused by noise is eliminated, the instrument stability and accuracy are improved, the accuracy and consistency of the data are improved, the specific parameters of the difference and smoothing operations can be flexibly adjusted and optimized according to the actual situation, and the applicable range and practicability of the method are improved; first calculate the second-order difference endpoint, then calculate the first-order difference endpoint, and perform correction, relatively stably determine the titration endpoints of various photometric curves, and improve the accuracy and efficiency of titration endpoint judgment. Description of the Drawings

[0015] The following further describes the present invention with reference to the drawings and embodiments: Figure 1 is the overall flow schematic diagram of the present invention; Figure 2 is the flowchart of the titration endpoint judgment method of the present invention; Figure 3 is the original photometric data curve diagram of the present invention; Figure 4 is the curve diagram after smoothing and differentiating the photometric data of the present invention; Figure 5 It is a schematic diagram for the present invention to obtain the first end point and the second end point; Figure 6 It is a schematic diagram of the end point for detecting phosphorus pentoxide in the phosphoric chemical industry for comparison in the present invention; Figure 7 It is a schematic diagram of the end point for detecting the total rare earth content in the rare earth industry for comparison in the present invention; Figure 8 It is a schematic diagram of the end point for detecting copper ions in copper plating solution in the electroplating industry for comparison in the present invention. Detailed implementation manners

[0016] Example 1 As Figure 1-8 shown, an automatic noise reduction method for titration end point judgment includes the following steps: S1. Data preprocessing: Obtain the photometric data to be detected and perform data preprocessing, including slicing and smoothing operations.

[0017] S2. Search for the first end point: For the data after data preprocessing, gradually search based on the peak positions of the second-order difference x2 and the first-order difference x1, obtain the position where the first titration end point is located, and perform correction.

[0018] S3. Search for the second end point: When there may be multiple titration end points, judge based on the maximum value, minimum value and their sign relationship of the first end point and the second end point to determine the position of the second end point.

[0019] As Figure 1 shown, it is a schematic diagram of the overall process of this embodiment, Figure 2 and it is a flowchart of the method provided by this embodiment.

[0020] In this embodiment, through slicing and smoothing data preprocessing on the photometric data to be detected, and then gradually searching for the first end point based on the peak positions of the second-order difference and the first-order difference, gradually narrowing the range, determining the position of the first titration end point and performing correction; then according to the characteristics of the titration curve and the analysis requirements, judge the situation where there are multiple end points, and use specific judgment conditions to determine the position of the second end point; through the above steps, the accuracy and consistency of the data are improved, the specific parameters of the difference and smoothing operations can be flexibly adjusted and optimized according to the actual situation, the applicable range of the method is improved, and the comprehensive consideration and calculation of the first end point and the second end point improve the accuracy and efficiency of titration end point judgment.

[0021] Due to the long-term operation of the instrument, noise with inconsistent intensity may be generated, so an automatic noise reduction method needs to be adopted. When the noise reduction reaches a certain level, stop the noise reduction when it is judged that the current data noise meets the requirements. Gaussian filtering may identify noise as peaks during the smoothing process, so a variable window length method is adopted in this embodiment.

[0022] The various method parameters adopted in this embodiment are shown in Table 1 as follows: Table 1 Method Parameter Table

[0023] Table 1 shows the descriptions of the various parameters adopted in this embodiment and their value ranges. The following content will all be represented by the parameters in Table 1.

[0024] As Figure 3 shown, it is the original photometric data curve graph adopted in this embodiment. The vertical axis represents absorbance, and the horizontal axis represents the time axis.

[0025] In the preferred solution, in step S1, variable window length Gaussian filtering is used for smoothing operation. Specifically: S11: Perform variable window length Gaussian filtering on the photometric data x for n times. The smoothing window length w = 5 + 2 * n, and the Gaussian kernel calculation formula is: ; In the formula, is the standard deviation, and μ is the mean.

[0026] S12: After performing edge compensation, calculate the noise using the difference and second-order difference of each smoothed data. The formula is: The formula is: ; ; In the formula, i is the smoothing times, j is the data point position, j_max represents the number of photometric data points, x i,j represents the intensity of the j-th photometric data point after the i-th smoothing, abs is to take the absolute value, is the intermediate noise, is the final noise.

[0027] Among them, linear compensation is adopted for edge compensation. Use the linear fitting of the n + 1 points at the edge, and additionally fit 11 points of data to be added to each edge of the data, and calculate the difference x1 and the second-order difference x2 of each smoothed data x0 respectively.

[0028] S13: Judge the noise. If the current noise is within the preset noise threshold range, stop the judgment.

[0029] That is, if the current noise satisfies or , where, is the minimum noise, is the maximum noise, then stop the judgment. Average the smoothing times of each order of differential data and round up to an integer to obtain the actual smoothing times ii for practical application.

[0030] S14: Average the number of smoothing times for each order of difference data and round up to an integer to obtain the actual smoothing times for application. .

[0031] In this embodiment, in step S11, Gaussian filtering with a variable window length is used to smooth the photometric data to reduce noise and fluctuations; in step S12, for the edge effect generated by Gaussian filtering when processing data edges, the linear fitting is performed using the n + 1 points at the edges, and then the data obtained by fitting is additionally added to each edge of the data to compensate for the edge effect; in step S13, if the current noise is within the preset noise threshold range, it is considered that the smoothing process has met the requirements and further smoothing is stopped; in step S14, by calculating the smoothing results of each order of difference data under different smoothing times, then taking the average and rounding up, the final actual smoothing times for application are obtained to ensure the best smoothing effect.

[0032] Through the above steps, the Figure 3 photometric data in is automatically denoised, improving the accuracy and stability of the data, and obtaining a noise table and a smoothing curve, as shown in Table 2 below and Figure 4 shown.

[0033] Table 2 Noise Table

[0034] As Figure 4 shown, the curves after smoothing and differencing are shown. The orange line is the smoothed x0, the green line is x1, and the red line is x2.

[0035] When it is obtained that x0 is 6 times, x1 is 6, and x2 is 5 respectively to meet the relevant threshold conditions, the appropriate smoothing times are 6 times. For the Figure 4 data, subsequent steps can be performed to obtain the end point.

[0036] Theoretically, the titration end point is the point where the slope of the photometric data changes greatly. The change in slope will cause a peak to form during the differencing process, and this point corresponds to the extreme value point of the peak of the second-order difference. In fact, the first-order difference of the photometric data is not always monotonically changing, resulting in the deviation of the extreme value point of the second-order difference from the titration end point, and the method of finding the extreme value point makes the end point non-adjustable and cannot meet various requirements.

[0037] To address this defect, in this embodiment, the position of a certain height of the peak of the first-order difference is found to avoid the situation where the extreme value point deviates from the end point, and the relative position of the end point is adjustable. Specifically: first, find the x2 end point, and then find the x1 end point near the x2 end point, that is, the titration end point, and the influence caused by the baseline drift of x1 is avoided through an incremental method.

[0038] In the preferred solution, in step S2, a step-by-step search is performed based on the peak positions of the second-order difference x2 and the first-order difference x1, that is, first find the end point of x2, and then find the end point of x1 near the end point of x2 as the first end point, specifically including: S21. Find the end point of x2: Determine the end point of x2 by calculating the baseline intensity and peak threshold of x2.

[0039] S22. Find the end point of x1: Near the end point of x2, determine the end point of x1, that is, the titration end point, by calculating the baseline intensity and peak threshold of x1.

[0040] In the preferred solution, step S21 includes calculating the baseline intensity of x2, the peak threshold, the baseline intensity of x1, and the peak threshold, and then obtaining the titration end point, specifically: S211: Calculate the mean value mean2 and the standard deviation value std2 of x2, and make a judgment.

[0041] S212: Eliminate the data points that satisfy x2 > mean2 + t 1 *std2 or x2 < mean2 - t 1 *std2, and then calculate mean2 and std2, and perform an iterative elimination loop until no points satisfy the condition and stop; where t 1 is the baseline threshold coefficient.

[0042] S213: Calculate the current mean2 value as the baseline intensity value, and the x2 peak threshold calculation formula is; ; In the formula, is the peak threshold of x2, is the x2 peak threshold coefficient, is to take the maximum value, is to take the minimum value.

[0043] S214: According to Find the end point of x2: If the current x2[j] < mean2 - c3 or x2[j] > mean2 + c3, and x2[j + 2] < mean2 - c3 or x2[j + 2] > mean2 + c3, then the current data point is the end point of x2.

[0044] S215: Obtain the end point of x1: If the absolute value maximum point of x1 is less than 0, then x1 = -x1; similar to the process of finding the baseline intensity of x1, obtain mean1 and std1, and the x1 peak threshold calculation formula is: ; In the formula, is the x1 peak threshold coefficient, It is the x1 noise threshold coefficient. Similar to the process of finding the end point for x2, the x1 end points are obtained on the left and right sides of the x2 end point respectively, and the end point closer to the x2 end point is selected. .

[0045] As Figure 5 shown, the abscissa is the time axis and the ordinate is the intensity axis, and the unit has no obvious meaning. The blue line is the smoothed x0, the orange line is x1, the green line is x2, the red line indicates the end point, and the purple line indicates the second end point. Steps 1, 2, 3, 4, 5, 6 indicate the path of finding the end point: first find the extreme point of x2, then find a point at a certain height of this peak, then map it to the x1 peak, find a point at a certain height of the x1 peak on the x1 peak, and map it to the x end point. Find a point at a certain height on the other side of the x1 peak and map it to the second end point on x.

[0046] In this embodiment, the obtained end point is affected by the smoothing and difference processes and will deviate from the end point of the original photometric data, and the end point position needs to be corrected.

[0047] In the preferred solution, in step S2, a correction formula is used to correct the first end point position to make it closer to the end point of the original photometric data. The formula is: ; In the formula, is the corrected first end point position, is the first end point position before correction, is the number of smoothing times.

[0048] Furthermore, when selecting the end points of x2 and x1, in addition to considering the peak positions, it is also necessary to make a judgment in combination with the actual characteristics of the titration reaction (such as color change, absorbance change, etc.).

[0049] In the preferred solution, when there are multiple titration end points in step S3, the specific judgment includes: If x2 is all greater than zero or less than zero, then there is 1 end point; If the maximum and minimum values of x1 have different signs, and the maximum and minimum values of x2 have the same sign, then there is 1 end point; If the maximum and minimum values of x1 have different signs, and the maximum and minimum values of x2 have different signs, and the minimum value of x2 does not exceed 0.3 times the maximum value, then there is 1 inflection point; Otherwise, there are 2 end points. At this time, starting from the first end point, traverse x1 to the right and apply the process of finding the end point in S2 again to determine the second end point.

[0050] In this embodiment, it is set that the minimum value of x2 does not exceed 0.3 times the maximum value. According to the specific situation, it may be necessary to make an adaptive adjustment according to the specific photometric data and the characteristics of the titration reaction.

[0051] In this embodiment, different data are used to obtain the titration end point and verification is carried out. For example, Figure 6-8 As shown, it is a schematic diagram of the end points of other comparison curves. The vertical lines in the figure respectively indicate the first end point and the second end point. Figure 6 It is the original curve graph of the absorbance change for the detection of the phosphorus pentoxide index in the phosphochemical industry. In this method, the indicator is thymolphthalein, the titrant is hydroxide, and the chemical change process is that the solution changes from colorless to blue, which is the titration end point. Figure 7 It is the curve graph of the absorbance change during the chemical reaction process of the total rare earth in the rare earth industry. The reaction end point is that when titrating with 0.3mol / L EDTA standard solution until the solution changes from purple to yellow. Figure 8 It is the curve graph of the absorbance change during the detection process of the copper ion index in the electroplating solution. The indicator for the chemical reaction is 0.1mLPAN, and then titrate with 0.1mol / L EDTA. When the solution color changes from blue-violet to green, it is the titration end point. It can be seen that the end points all point to the vicinity of the point with a large slope change, which all meet the requirements of the titration end point.

[0052] This embodiment adopts automatic noise reduction, which can partially eliminate the interference of the end point variation caused by noise, improve the stability and accuracy of the instrument, and is suitable for the long-term operation detection of absorbance method instruments.

[0053] First, calculate the second-order difference x2 end point, then calculate the first-order difference x1 end point, and perform a certain end point correction, so as to relatively stably determine the titration end points of multiple photometric curves.

[0054] This embodiment gives two end points at the same time, meeting various detection requirements and improving the practicality of the method.

[0055] Embodiment 2 Combined with Embodiment 1 for further illustration, a titration end point judgment system with automatic noise reduction is provided, including: A data preprocessing module, which is used to obtain the photometric data to be detected and perform data preprocessing, including slicing and smoothing operations; A first end point searching module, which is used to gradually search for the position of the titration first end point based on the peak positions of the second-order difference x2 and the first-order difference x1 for the data after data preprocessing, and perform correction; A second end point searching module, which is used to judge based on the maximum value, minimum value and their sign relationship of the first end point and the second end point when there may be multiple titration end points, and determine the position of the second end point.

[0056] An electronic device, including: a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements a titration end point judgment method with automatic noise reduction as in Embodiment 1.

[0057] A computer-readable storage medium, comprising: a computer program stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements an automatic noise reduction titration end point judgment method as in Embodiment 1.

[0058] The above embodiments are only the preferred technical solutions of the present invention and should not be regarded as limitations on the present invention. The protection scope of the present invention should be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A titration endpoint determination method with automatic noise reduction, characterized in that: It includes the following steps: S1. Data preprocessing: Obtain the photometric data to be detected and perform data preprocessing, including slicing and smoothing operations; S2. Finding the first endpoint: For the data after data preprocessing, gradually search based on the peak positions of the second-order difference and the first-order difference to obtain the position of the first titration endpoint and perform correction; S3. Finding the second endpoint: When there may be multiple titration endpoints, judge based on the maximum values, minimum values and their sign relationships of the first endpoint and the second endpoint to determine the position of the second endpoint.

2. The titration endpoint determination method with automatic noise reduction according to claim 1, characterized in that: In step S1, Gaussian filtering with variable window length is used for the smoothing operation. Specifically: S11: Perform Gaussian filtering with variable window length on the photometric data x for n times. The smoothing window length w = 5 + 2 * n, and the Gaussian kernel calculation formula is: ； In the formula, is the standard deviation, μ is the mean; S12: After performing edge compensation, calculate the noise using the photometric data after each smoothing. The formula is: ； ； In the formula, i is the number of smoothing times, j is the data point position, j_max represents the number of photometric data points, and x i,j represents the intensity of the jth photometric data point at the i-th smoothing, and abs is the absolute value. is the intermediate noise, is the final noise; S13: Judge the noise. If the current noise is within the preset noise threshold range, stop the judgment; S14: average the smoothing times of each order difference data and round it up to an integer to obtain the smoothing times for actual application .

3. The titration endpoint determination method with automatic noise reduction according to claim 1, characterized in that: In step S2, gradually search based on the peak positions of the second-order difference and the first-order difference, that is, first find the x2 endpoint, and then find the x1 endpoint near the x2 endpoint as the first endpoint. Specifically include: S21. Finding the x2 endpoint: Determine the endpoint of x2 by calculating the baseline intensity and peak threshold of x2; S22. Finding the x1 endpoint: Near the x2 endpoint, determine the endpoint of x1 by calculating the baseline intensity and peak threshold of x1, that is, the titration endpoint; Among them, x2 represents the second-order difference, and x1 represents the first-order difference.

4. The method for judging the titration endpoint with automatic noise reduction according to claim 3, characterized in that In step S21, it includes calculating the x2 baseline intensity, peak threshold, x1 baseline intensity and peak threshold, and then obtaining the titration endpoint. Specifically: S211: Calculate the average value mean2 and standard deviation value std2 of x2 and make a judgment; S212: Eliminate the data points that satisfy x2 > mean2 + t1 * std2 or x2 < mean2 - t1 * std2, and then calculate mean2 and std2, and perform iterative elimination loops until no points satisfy the condition and stop; where t1 is the baseline threshold coefficient; S213: Calculate the mean2 value at this time as the baseline intensity value, and the x2 peak threshold calculation formula is; ； In the formula, is the peak threshold of x2, is the peak threshold coefficient of x2, To obtain the maximum value, To take the minimum value; S214: According to Find the end point of x2: If the current x2[j] < mean2 - c3 or x2[j] > mean2 + c3, and x2[j + 2] < mean2 - c3 or x2[j + 2] > mean2 + c3, then the current data point is the end point of x2; S215: Obtain the x1 endpoint: If the absolute value maximum point of x1 is less than 0, then x1 = -x1; in the same process of finding the x1 baseline intensity, obtain mean1 and std1, and the x1 peak threshold calculation formula is: ； In the formula, is the x1 peak threshold coefficient, is the x1 noise threshold coefficient; The same process as x2 is used to find the end point. The end point of x1 is found on the left and right sides of the end point of x2. The end point that is closer to the end point of x2 is selected. .

5. The titration endpoint determination method with automatic noise reduction according to claim 1, characterized in that: In step S2, a correction formula is used to correct the position of the first endpoint to make it closer to the endpoint of the original photometric data. The formula is: ； In the formula, is the first end point position after correction, is the first end point position before correction, is the number of smoothing times.

6. The method for judging the titration endpoint with automatic noise reduction according to claim 3, characterized in that When there are multiple titration endpoints in step S3, the specific judgment includes: If x2 is all greater than zero or less than zero, there is 1 endpoint; If the maximum value and minimum value of x1 have different signs, and the maximum value and minimum value of x2 have the same sign, there is 1 endpoint; If the maximum value and minimum value of x1 have different signs, and the maximum value and minimum value of x2 have different signs, and the minimum value of x2 does not exceed 0.3 times the maximum value, there is 1 inflection point; Otherwise, there are two end points. In this case, starting from the first end point, traverse x1 to the right and apply the process of S2 to find the end point again to determine the second end point.

7. An automatic noise reduction titration endpoint determination system, characterized in that: include: A data preprocessing module is used to obtain the photometric data to be detected and perform data preprocessing, including slicing and smoothing operations; A first endpoint search module is used to search the pre-processed data step by step based on the peak position of the second-order difference and the peak position of the first-order difference, obtain the position of the first endpoint of the titration, and perform correction; The module for finding the second endpoint is used to determine the position of the second endpoint based on the maximum value, minimum value and sign relationship of the first endpoint and the second endpoint when there may be multiple titration endpoints.

8. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the titration endpoint determination method with automatic noise reduction as described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: include: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the titration endpoint determination method with automatic noise reduction according to any one of claims 1 to 6 is implemented.

Images