Method and device for measuring content of exchangeable fluorine ions in soil

Abstract

The invention discloses a method and device for measuring the content of exchangeable fluorine ions in soil, and relates to the technical field of chemical analysis of soil, and the method comprises the following steps: analyzing the soil to be measured, obtaining the physical and chemical characteristics of the soil, and selecting extracting agents with different concentrations to react with the soil in an oscillating manner; obtaining a conductivity curve according to the solution after the oscillation reaction, and correcting the conductivity curve based on the established temperature compensation model through regression analysis; screening an extracting agent with the optimal concentration according to the corrected conductivity curve in combination with soil physical and chemical characteristics; performing oscillation reaction with soil at different oscillation rates according to the extracting agent with the optimal concentration, and analyzing and screening the optimal oscillation rate; according to the method, the accuracy, the stability and the adaptability of a measurement result are effectively improved by integrating soil physical and chemical characteristic analysis, conductivity temperature compensation correction, extraction agent concentration optimization screening and oscillation rate intelligent matching.

Description

Technical Field

[0001] The present invention relates to the technical field of soil chemical analysis, and more specifically, the present invention relates to a method and device for determining the content of exchangeable fluoride ions in soil. Background Art

[0002] With the continuous improvement of people's attention to ecological environment and farmland safety issues, the detection of fluoride ion content in soil has become an important research direction in multiple fields such as agriculture, environmental science, and public health. Fluorine is widely present in nature. Appropriate amounts of fluorine are beneficial to organisms, but excessive fluorine can have adverse effects on crop growth, groundwater safety, and human health. Especially in high-fluoride areas, long-term intake of high-fluoride foods or drinking high-fluoride water may lead to health risks such as fluorosis and bone lesions. Therefore, it is of great practical significance to establish an efficient and reliable method for determining the content of exchangeable fluoride ions in soil.

[0003] Existing methods for determining fluoride ions in soil mainly include ion selective electrode method, ultraviolet spectrophotometry, ion chromatography, etc. Although these methods have certain accuracy under laboratory conditions, they have problems such as complex operation, high requirements for experimental environment, cumbersome sample pretreatment, and strong instrument dependence, and it is difficult to meet the actual needs of on-site rapid and high-throughput detection. At the same time, existing methods often ignore the influence of external variables such as extractant concentration, oscillation conditions, and temperature on the extraction efficiency and detection results when extracting fluoride ions, resulting in poor repeatability and stability of detection data and making it difficult to accurately reflect the true content of exchangeable fluoride ions in soil.

[0004] For example, a method for determining the bioavailability of soil fluoride disclosed in the invention patent with publication number: CN111189893A includes: soil sample preparation step; extraction step; blank test solution preparation step; standard curve drawing step; sample determination step: transfer 10 mL of the extracted sample to a volumetric flask, add 10 mL of total ion strength buffer solution, dilute to volume with pure water and mix well, then transfer to a beaker, add a polyethylene magnetic stir bar, place it on a magnetic stirrer, add a fluoride ion selective electrode and a saturated calomel electrode for determination. Wait for the instrument reading to stabilize and record the potential response value; sequentially measure the potential response values of the sample and the blank test solution under the same conditions as when drawing the standard curve, and compare with the standard curve to convert to the corresponding fluoride content; effective concentration calculation step. By the above steps, the determination of fluoride content in soil is simplified, the test process is more convenient, and at the same time, the determination of fluoride content by establishing a standard curve makes the test results more accurate.

[0005] For example, a method for sequential extraction and detection of fluorine in soil disclosed in the invention patent with the publication number of CN117433863A belongs to the technical field of soil characteristic analysis. It includes: S1. Mix the soil with the water-soluble fluorine extraction solution at a certain solid-liquid ratio, ultrasonicate, centrifuge, and measure the content of water-soluble fluorine in the supernatant; S2. When the soil is acidic or neutral, adopt the extraction method of weak base first and then weak acid; when the soil is alkaline, adopt the extraction method of weak acid first and then weak base, and measure the content of adsorbed fluorine; S3. Mix the extraction residue after being extracted in step S2 with the oxidatively bound fluorine extraction solution at a certain solid-liquid ratio, ultrasonicate, centrifuge, and measure the content of oxidatively bound fluorine in the supernatant; S4. Dry and grind the extraction residue after being extracted in step S3, and measure the content of fixed fluorine therein through alkali fusion method. The method of the present invention does not change the soil properties during the extraction of fluorine to determine the content of various forms, and can most truly simulate the ecological environment of fluorine in the soil.

[0006] In the above-disclosed technical solution, there are at least the following technical problems: Traditional methods often select extraction agents with fixed concentrations for fluoride ion extraction, ignoring the influence of soil types on the extraction efficiency, and different oscillation rates have a significant impact on the release effect of fluoride ions. However, the existing technologies generally use fixed oscillation parameters and it is difficult to take into account the differences in various soil structures. In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0007] In order to overcome the above defects of the prior art, an embodiment of the present invention provides a method and device for determining the content of exchangeable fluoride ions in soil. By integrating methods of soil physical and chemical characteristic analysis, conductivity temperature compensation correction, optimized screening of extraction agent concentration, and intelligent matching of oscillation rate, it solves the problems existing in the traditional process of determining the content of exchangeable fluoride ions in soil, such as blind selection of extraction agents, large temperature interference, inaccurate conductivity curve, insufficient optimization of oscillation parameters, etc., and greatly improves the accuracy, stability and adaptability of the measurement results.

[0008] To achieve the above object, the present invention provides the following technical solutions: A method for determining the content of exchangeable fluoride ions in soil, including: analyzing the soil to be measured to obtain the physical and chemical characteristics of the soil, and selecting extraction agents with different concentrations to oscillate and react with the soil; obtaining the conductivity curve of the solution after the oscillation reaction, and correcting the conductivity curve, the correction is based on establishing a temperature compensation model through regression analysis for correction; screening the extraction agent with the best concentration according to the corrected conductivity curve combined with the physical and chemical characteristics of the soil; performing oscillation reactions at different oscillation rates with the soil using the extraction agent with the best concentration, and analyzing and screening the best oscillation rate.

[0009] In a preferred embodiment, the physical and chemical characteristics of the soil are obtained, and different concentrations of extractants are selected to react with the soil by oscillation, specifically as follows: collect the soil samples to be tested and determine the physical and chemical characteristics of the soil; select different concentrations of extractants according to the physical and chemical characteristics of the soil; add the selected concentration of extractant to the soil samples and mix the soil and the extractant through an oscillator.

[0010] In a preferred embodiment, the conductivity curve of the solution after the oscillation reaction is obtained, and the conductivity curve is corrected. The correction is based on establishing a temperature compensation model through regression analysis, specifically as follows: after mixing the soil and the extractant through an oscillator, obtain the supernatant; measure the conductivity of the supernatant in real time through a conductivity meter, record the conductivity values at different time points, and generate a conductivity curve; respectively prepare fluoride ion standard solutions with known concentrations, measure their conductivity values, obtain a temperature-conductivity data set, and perform linear regression analysis through the least squares method to construct a temperature compensation model; based on the temperature compensation model, correct the collected original conductivity data to generate a corrected conductivity curve standardized by temperature.

[0011] In a preferred embodiment, the correction of the conductivity curve is specifically as follows: obtain the conductivity value to be corrected, and select the temperature value where the conductivity value is located as the regression point; obtain the first data of the regression point, and obtain the neighborhood range of the regression point according to the first data. The first data includes the temperature difference and the number of data points; obtain the weight of each data point in the neighborhood range based on the Gaussian weight function; perform locally weighted regression fitting on the data points in the neighborhood range based on the weight of each data point to obtain a local fitting model; output the corrected conductivity value of the regression point according to the local fitting model to obtain the corrected conductivity curve.

[0012] In a preferred embodiment, the obtaining of the neighborhood range of the regression point according to the first data is specifically as follows: obtain the temperature gradient between the regression point to be corrected and its adjacent data points; obtain the fitting accuracy data under different neighborhood ranges according to the temperature gradient, and obtain the initial neighborhood range corresponding to the temperature gradient according to the accuracy data; obtain the actual number of data points in the initial neighborhood range near the regression point, and obtain the variance of the data points in the initial neighborhood range; correct according to the variance of the data points based on a preset neighborhood range correction formula to obtain the neighborhood range.

[0013] In a preferred embodiment, the best concentration of the extractant is screened according to the corrected conductivity curve in combination with the physical and chemical characteristics of the soil, as follows: according to the conductivity curve, the conductivity characteristics are extracted; according to the conductivity characteristics, the extraction effect of the extractant on fluoride ions at different concentrations is evaluated; according to the evaluation results of the extraction effect and the physical and chemical characteristics of the soil, based on the fuzzy comprehensive evaluation method, the adaptation relationship between the soil type and the extractant concentration is constructed; according to the adaptation relationship, the best concentration of the extractant is screened, and based on the BP neural network, a prediction model between the physical and chemical characteristics of the soil and the best extractant concentration is established.

[0014] In a preferred embodiment, the prediction model between the physical and chemical characteristics of the soil and the best extractant concentration is established based on the BP neural network, as follows: the physical and chemical characteristic data are extracted from multiple tested soil samples; the corrected conductivity curve characteristics of each sample at different extractant concentrations are obtained; an input data set is constructed with the soil physical and chemical parameters as the feature vectors, and the best extractant concentration corresponding to the current sample at different concentrations is used as the label to form a supervised learning sample; the supervised learning sample is divided into a training set and a validation set and input into the BP neural network model for training; according to the new soil sample data, the physical and chemical characteristics of the soil are input, and based on the trained BP neural network model, the best extractant concentration is predicted.

[0015] In a preferred embodiment, the oscillating reaction between the best concentration extractant and the soil is carried out at different oscillation rates, and the best oscillation rate is analyzed and screened, as follows: according to the best concentration extractant, several different oscillation rate levels are set to form an oscillation rate gradient group; equal mass soil samples and equal volume of the best concentration extractant are mixed and reacted under different oscillation rate conditions respectively to obtain the supernatant; the conductivity of the supernatant under each oscillation rate is monitored in real time to obtain the conductivity curve, and the conductivity curve is corrected; the data characteristics of the corrected conductivity curve under different oscillation rate conditions are extracted to construct an oscillation rate data set; according to the oscillation rate data set, based on the grey relational analysis method, the extraction effect under each oscillation rate condition is evaluated; the best oscillation rate is screened according to the evaluation results of the extraction effect under each oscillation rate condition.

[0016] Device for method of measuring content of exchangeable fluoride ions in soil, characterized by comprising a data acquisition module, a correction module, an extractant screening module and an oscillation rate screening module, with connections among the modules; the data acquisition module is used for analyzing the soil to be measured, obtaining the physical and chemical characteristics of the soil, and selecting extractants with different concentrations to react with the soil by oscillation; the correction module is used for obtaining the conductivity curve according to the solution after the oscillation reaction, and performing regression analysis to establish a temperature compensation model for correcting the conductivity curve; the extractant screening module is used for screening the optimal concentration extractant according to the corrected conductivity curve in combination with the physical and chemical characteristics of the soil; the oscillation rate screening module is used for performing oscillation reactions at different oscillation rates between the optimal concentration extractant and the soil, and analyzing and screening the optimal oscillation rate.

[0017] Technical effects and advantages of the method and device for measuring the content of exchangeable fluoride ions in soil according to the present invention: 1. The present invention fully considers the diversity and complexity of soil, and introduces a soil physical and chemical characteristic extraction link in the initial stage of the measurement method, such as key parameters such as soil pH value, organic matter content, and particle size distribution, providing basic data support for the personalized optimization of subsequent extractant concentration and reaction parameters. Through this preliminary analysis, the measurement deviation problem in the "fixed extractant - fixed condition" mode in the traditional method is effectively avoided, and the applicability and accuracy of the method are improved.

[0018] 2. The present invention proposes to use the conductivity curve as the core index to reflect the release dynamics of fluoride ions in real time during the extraction reaction process, and for the first time introduces a temperature compensation mechanism based on regression modeling in this type of method. By configuring a standard fluoride ion solution sample and constructing a temperature - conductivity regression model, dynamic correction of conductivity data under different reaction environments is realized, solving the result error problem caused by temperature fluctuations in the traditional method, and significantly improving the stability and repeatability of the measurement results. In the extractant screening link, the present invention adopts a fuzzy comprehensive evaluation method, combines conductivity characteristics with soil physical and chemical properties, and establishes an adaptation relationship between extractant concentration and soil type, avoiding the limitation of single - factor - dominated judgment. At the same time, a prediction model is further constructed based on the BP neural network to realize the intelligent matching of extractant concentration. This method has strong generalization ability, can meet the rapid measurement requirements of various soils, and significantly improves work efficiency. Description of the Drawings

[0019] Figure 1 It is a schematic flow chart of the method for measuring the content of exchangeable fluoride ions in soil according to the present invention.

[0020] Figure 2 It is a schematic structural diagram of the device for the method of measuring the content of exchangeable fluoride ions in soil according to the present invention.

[0021] Figure 3This is the data graph for evaluating the extraction effect under different oscillation rate conditions of the present invention.

[0022] Figure 4 This is the comparison graph of conductivity curves of the present invention under the condition of the extractant being 0.15 mol / L. Detailed implementation manners

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] Embodiment 1 Figure 1 A method for measuring the content of exchangeable fluoride ions in soil according to the present invention is provided, including the following steps: S1. Obtain the physical and chemical characteristics of the soil, and select extractants with different concentrations for oscillating reaction with the soil.

[0025] In this embodiment, obtaining the physical and chemical characteristics of the soil and selecting extractants with different concentrations for oscillating reaction with the soil are specifically as follows: Collect soil samples to be measured, select representative positions of the soil to ensure the uniformity and representativeness of the samples; Use instrument equipment (such as a pH meter, a conductivity meter, a soil moisture meter, etc.) to measure the physical and chemical characteristics of the soil. The physical and chemical characteristics include pH value, conductivity, organic matter content, and soil humidity. For example, the pH value of the soil sample collected in this embodiment is 6.5, the conductivity is 315 μS / cm, the organic matter content is 3.2%, and the moisture content is 22%; According to the physical and chemical characteristics of the soil, select extractants with different concentrations. For example, feasible extractants include ammonium acetate solutions with concentrations set at 0.05 mol / L, 0.10 mol / L, 0.15 mol / L, 0.20 mol / L, and 0.25 mol / L; Add the selected concentration of extractant to the soil sample, and use an oscillator to fully mix the soil and the extractant according to a certain ratio (such as a soil-to-extractant ratio of 1:1).

[0026] S2. Obtain the conductivity curve of the solution after the oscillating reaction, and correct the conductivity curve. The correction is based on establishing a temperature compensation model through regression analysis for correction.

[0027] In this embodiment, obtaining the conductivity curve of the solution after the oscillating reaction and correcting the conductivity curve. The correction is based on establishing a temperature compensation model through regression analysis for correction, specifically as follows: Complete the extraction reaction under constant temperature conditions according to the preset ratio of extractant concentration to soil, and use an oscillation device to fully react the soil with the extractant. The oscillation time is controlled within 30 minutes. After the reaction, centrifuge the mixed solution to obtain the supernatant for detection; Use a conductivity meter to measure the conductivity of the supernatant in real time, record the conductivity values at different time points, and the measurement time interval can be set to once every 30 seconds or once every 1 minute to generate the original curve of conductivity varying with time. Taking the extractant concentration of 0.15 mol / L as an example, the initial conductivity obtained is 350 μS / cm, which rises to 720 μS / cm after 10 minutes, tends to be stable after 20 minutes, and the final value is 760 μS / cm; Use a temperature sensor to monitor the temperature of the extract in real time to ensure that the conductivity measurement and the temperature at the corresponding time point are synchronously recorded for subsequent temperature compensation calculations; Respectively configure fluoride ion standard solutions with known concentrations, measure their conductivity values, obtain the temperature-conductivity data sets, and perform linear regression analysis by the least squares method to construct a temperature compensation model; Based on the constructed temperature compensation model, correct the collected original conductivity data, uniformly correct the conductivity value at each time point to the equivalent conductivity value under the condition of 25 °C, and generate a corrected conductivity curve after temperature standardization.

[0028] The temperature compensation model is specifically as follows:

[0029] In the formula: represents the conductivity at temperature T (unit: μS / cm), represents the conductivity at the standard temperature of 25 °C, is the temperature coefficient, is the actual temperature during measurement (unit: °C).

[0030] For example: when = 30 °C, = 800 μS / cm, Substitute 0.02 / °C into the temperature compensation model to obtain the compensated conductivity of 727.27 μS / cm.

[0031] In this embodiment, the correction of the conductivity curve is specifically as follows: Obtain the conductivity value to be corrected, and select the temperature value where the conductivity value is located as the regression point; Obtain the first data of the regression point, and obtain the neighborhood range of the regression point according to the first data. The first data includes temperature difference and the number of data points. For example, taking the temperature of 30°C as the regression point, the temperatures are 28°C, 29°C, 31°C, 32°C, the number of data points is 4 (excluding the regression point), the corresponding local variance is 0.015, the maximum variance is 0.036, and the neighborhood range is ±3°C; Obtain the weights of each data point within the neighborhood range based on the Gaussian weight function; Perform local weighted regression fitting on the data points within the neighborhood range based on the weights of each data point to obtain a local fitting model; Output the corrected conductivity value of the regression point according to the local fitting model to obtain a corrected conductivity curve.

[0032] In this embodiment, obtaining the neighborhood range of the regression point according to the first data is specifically as follows: Obtain the temperature gradient between the regression point to be corrected and its adjacent data points; Obtain the fitting accuracy data under different neighborhood ranges according to the temperature gradient, and obtain the initial neighborhood range under the corresponding temperature gradient according to the accuracy data; Obtain the actual number of data points within the initial neighborhood range near the regression point, and obtain the variance of the data points within the initial neighborhood range; Perform correction based on the variance of the data points according to a preset neighborhood range correction formula to obtain the neighborhood range.

[0033] The neighborhood range correction formula is specifically as follows:

[0034] In the formula: is the corrected neighborhood range (unit: °C), is the initial neighborhood range (unit: °C), is the adjustment coefficient, is the local variance, is the maximum variance, that is, the maximum variance in the dataset.

[0035] For example: Taking 30°C as the regression point, the temperatures of the data points within the neighborhood are selected as 28°C, 29°C, 31°C, and 32°C. The number of data points is 4 (excluding the regression point), the corresponding local variance is 0.015, and the maximum variance is 0.036. The initial neighborhood range is ±3°C, and the adjustment coefficient β is 0.85. According to the formula, the calculated neighborhood range is ±4.06°C; taking 27°C as the regression point, the temperatures of the data points within the neighborhood are selected as 25°C, 26°C, 28°C, and 29°C. The number of data points is 4 (excluding the regression point), the corresponding local variance is 0.004, and the maximum variance is 0.020. The initial neighborhood range is ±4°C, and the adjustment coefficient β is 0.75. According to the formula, the calculated neighborhood range is ±4.6°C.

[0036] Exemplarily, a comparison graph of the initial conductivity curve and the corrected conductivity curve is as Figure 4 shown: At 0 minute, the initial conductivity is high because the extractant itself has strong conductivity. As time goes by, the conductivity gradually increases; After data correction, the curve is smoother and the trend is clearer. The corrected conductivity is lower than the initial value as a whole, and shows a more stable change trend in a specific time period (such as after 20 minutes), indicating that the correction is effective; The corrected data is closer to the real reaction process, removing the influence of temperature fluctuations. If the difference between the two curves is significant, it indicates that the correction measures have an important impact on the experimental results. If the difference is small, the reliability of the original data is relatively high.

[0037] It should be noted that a temperature compensation model is constructed through regression analysis, which can accurately correct the conductivity curve according to the real-time measured temperature value. Traditional conductivity measurements are usually greatly affected by temperature changes, and the conductivity values at different temperatures vary significantly, which easily leads to inconsistent test results. By using regression analysis and the temperature compensation model, the conductivity values at different temperatures can be effectively standardized to the equivalent value at 25°C, eliminating the interference of temperature fluctuations on the test results and effectively improving the accuracy and reliability of the measurement results.

[0038] Furthermore, a mixed reaction under different temperature conditions is adopted to generate multiple conductivity curves, and similarity analysis is carried out based on the Pearson correlation coefficient method and the conductivity curve at the standard temperature. Through the Pearson correlation coefficient analysis method, the relative deviation between the conductivity at different temperatures and the conductivity at the standard temperature is obtained, providing a scientific basis for correcting the conductivity curve. This data processing method under multiple temperature conditions makes the temperature compensation more detailed and comprehensive, and can adapt to the measurement requirements in different environments.

[0039] S3. According to the corrected conductivity curve and combined with the soil physical and chemical characteristics, the optimal concentration extractant is screened.

[0040] In this embodiment, according to the corrected conductivity curve and combined with the physical and chemical characteristics of the soil, the optimal concentration of the extractant is screened as follows: According to the conductivity curve, the conductivity characteristics are extracted, and the conductivity characteristics include the maximum conductivity value, the conductivity value in the steady stage, and the conductivity rising slope in the early stage of the reaction; According to the conductivity characteristics, the extraction effect of the extractant on fluoride ions at different concentrations is evaluated; According to the extraction effect and the physical and chemical characteristics of the soil, based on the fuzzy comprehensive evaluation method, the adaptation relationship between the soil type and the extractant concentration is constructed; According to the screening, the extractant concentration with the optimal extraction efficiency and the minimum matrix interference under specific soil conditions is used as the target optimal use concentration; Furthermore, the BP neural network is used to train the historical experimental data, and a prediction model between the physical and chemical characteristics of the soil and the optimal extractant concentration is established to realize the rapid recommendation and automatic matching of subsequent samples.

[0041] The evaluation formula for the extraction effect of fluoride ions at different concentrations is as follows:

[0042] In the formula: is the evaluation result of the extraction effect of fluoride ions at different concentrations, is the maximum conductivity value under the i-th concentration condition, is the conductivity value in the steady stage (μS / cm) under the i-th concentration condition, is the conductivity rising slope in the early stage of the reaction under the i-th concentration condition, , and are the weight coefficients, and satisfy .

[0043] For example: taking the extractant concentration of 0.15 mol / L as the condition, recording the maximum conductivity value as 800 μS / cm, the conductivity value in the steady stage as 720 μS / cm, and the conductivity rising slope in the early stage of the reaction as 60 μS / cm / min, setting the weight coefficients as w1 = 0.4, w2 = 0.4, w3 = 0.2, and calculating the extraction effect evaluation result as 620 according to the extraction effect evaluation formula; taking the extractant concentration of 0.20 mol / L as the condition, recording the maximum conductivity value as 900 μS / cm, the conductivity value in the steady stage as 850 μS / cm, and the conductivity rising slope in the early stage of the reaction as 40 μS / cm / min, setting the weight coefficients as w1 = 0.4, w2 = 0.4, w3 = 0.2, and calculating the extraction effect evaluation result as 703 according to the extraction effect evaluation formula.

[0044] In this embodiment, a prediction model is established between the soil physical and chemical characteristics and the optimal extractant concentration to achieve rapid recommendation and automatic matching of subsequent samples, as follows: For multiple tested soil samples, collect their physical and chemical characteristic data, including but not limited to pH value, organic matter content, cation exchange capacity (CEC), particle size composition, moisture content, etc.; At the same time, record the characteristics of the corrected conductivity curve of each sample at different extractant concentrations, including indicators such as maximum conductivity, stable conductivity value, and initial reaction rate; clean and process the above data, remove outliers and missing values, and perform linear interpolation or mean filling if necessary. Subsequently, perform standardization processing on all characteristic data to ensure the consistency of features with different dimensions. Commonly used methods include Z-score standardization or Min-Max normalization; Construct an input data set with soil physical and chemical parameters as feature vectors, and form a supervised learning sample with the optimal extractant concentration corresponding to the current sample at different concentrations as labels; In the BP neural network modeling, construct a three-layer neural network. The input layer corresponds to the dimension of soil parameters, the output layer is the optimal concentration value, and the number of neurons in the hidden layer is set according to the empirical formula. During the training process, use the backpropagation algorithm and introduce the Adam optimizer to update the weights, improving the convergence speed and prediction accuracy; Divide the original sample set into a training set and a validation set, and use the mean squared error, mean absolute error, and coefficient of determination as evaluation indicators to judge the fitting ability and generalization ability of the model. If the error indicators meet the accuracy requirements, the model establishment is completed; According to the new soil sample data, input the soil physical and chemical characteristics, and predict the optimal extractant concentration based on the trained BP neural network model.

[0045] The application process of the BP neural network model is as follows: Extract representative physical and chemical parameters from 80 groups of tested soil samples, including: pH value, conductivity, organic matter content, and soil moisture, a total of 4-dimensional features. At the same time, record the characteristics of the corrected conductivity curve of each group of samples at various extractant concentrations, and extract the maximum conductivity, conductivity at the stable stage, and slope at the initial reaction stage as the basis for extraction efficiency; Complete the missing values of all sample data (using mean filling), and after removing outliers, use Z-score standardization (mean is 0, standard deviation is 1) to ensure the scale consistency between input parameters; Construct a three-layer BP neural network model with a structure of 4-10-1, that is: the input layer has 4 nodes corresponding to 4 physical and chemical characteristics, the hidden layer is set with 10 neurons, the activation function is ReLU, and the output layer has 1 node, corresponding to the predicted optimal extractant concentration (unit: mol / L); 80 groups of samples were divided into a training set and a validation set at a ratio of 8:2. During the training process, the loss decline curve was monitored in real time to avoid overfitting. The model finally achieved a mean absolute error (MAE) of 0.012 mol / L on the validation set, and the final prediction accuracy was 0.93, showing good generalization ability.

[0046] It should be noted that in the data acquisition stage, this method fully utilizes the temperature-compensated conductivity curve to extract representative conductivity characteristic indexes, including the maximum conductivity value, the conductivity value in the stable stage, and the rising slope in the initial stage of the reaction, etc. These characteristics reflect the extraction efficiency and reaction dynamics of the extractant for fluoride ions at different concentrations, which are more comprehensive than the traditional method relying on single-point conductivity measurement, and can accurately reflect the overall performance of the extraction process, providing a key reference for subsequent analysis.

[0047] S4. Perform oscillation reactions between the extractant with the optimal concentration and the soil at different oscillation rates, and analyze and screen the optimal oscillation rate.

[0048] In this embodiment, perform oscillation reactions between the extractant with the optimal concentration and the soil at different oscillation rates, and analyze and screen the optimal oscillation rate as follows: According to the extractant with the optimal concentration, set several different oscillation rate levels to form an oscillation rate gradient group. For example, set 5 different oscillation speeds: 100 rpm, 150 rpm, 200 rpm, 250 rpm, and 300 rpm. And at each group of oscillation rates, keep the reaction time, temperature, and extractant concentration consistent to ensure single-variable control; Mix soil samples of equal mass with an equal volume of the extractant with the optimal concentration, and react them under the above different oscillation rate conditions respectively. The reaction time is uniformly fixed, and the experimental environment temperature is controlled at a constant level (such as 25°C); Perform real-time monitoring of the conductivity of the extract obtained at each oscillation rate to obtain a conductivity curve, and correct the conductivity curve through a temperature compensation model to obtain a corrected conductivity change curve; Extract the data characteristics of the corrected conductivity curves corresponding to different oscillation rate conditions. The data characteristics include characteristic indexes such as the maximum conductivity value, the time required for the conductivity to reach the stable stage, and the initial conductivity growth rate, etc., to construct an oscillation rate data set; Based on the oscillation rate data set, evaluate the extraction effects under each oscillation rate condition by using the grey relational analysis method; Select the rate with high extraction efficiency, low energy consumption required, and good reaction stability as the optimal oscillation rate of the target soil according to the evaluation results.

[0049] The specific evaluation formula for the extraction effects under each oscillation rate condition is as follows:

[0050] Standard sequence:

[0051] Where: is the evaluation result of the extraction effect under each oscillation rate condition, is the total number of data features, represents the grey correlation coefficient between the i-th oscillation rate and the standard sequence under the j-th data feature, is the standard sequence.

[0052] For example: Taking the extraction experiment with an oscillation rate of 200 rpm as an example, three representative characteristic parameters are selected: the maximum conductivity of 0.95, the time required to reach equilibrium of 0.90, and the rising slope of the conductivity at the initial stage of the reaction of 0.91. The grey correlation coefficients are 1.00, 0.75, and 0.789 respectively. The standard sequence is set as , the number of data features is m = 3, and according to the extraction effect evaluation formula, 0.846 is output. The data graph of the extraction effect evaluation result is as shown in Figure 3 shown: The extraction effect is the best at 250 rpm, and the extraction effect evaluation is 0.92. After exceeding 250 rpm, the extraction effect gradually decreases, probably because high-speed oscillation leads to an increase in foam, uneven dispersion of the solution, or mechanical damage to the active ingredients.

[0053] Example 2, Figure 2 The device for the method for determining the content of exchangeable fluoride ions in soil according to the present invention is provided, which is characterized in that it includes a data acquisition module, a correction module, an extractant screening module, and an oscillation rate screening module, and there are connections between the modules; The data acquisition module is used to analyze the soil to be measured, obtain the physical and chemical characteristics of the soil, and select extractants with different concentrations to react with the soil by oscillation; The correction module is used to obtain the conductivity curve based on the solution after the oscillation reaction, and perform regression analysis to establish a temperature compensation model to correct the conductivity curve; The extractant screening module is used to screen the optimal concentration extractant according to the corrected conductivity curve in combination with the physical and chemical characteristics of the soil; The oscillation rate screening module is used to perform oscillation reactions at different oscillation rates between the optimal concentration extractant and the soil, and analyze and screen the optimal oscillation rate.

[0054] The above formulas are all dimensionless and take their numerical calculations. The formula is obtained by collecting a large amount of data and performing software simulation to obtain a formula closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0055] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0056] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0057] In addition, the functional modules in the various embodiments of this application can be integrated into a processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0058] As described above, this is only a specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0059] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for determining the exchangeable fluoride ion content in soil, characterized in that: include: Obtain the physical and chemical characteristics of the soil, and select extractants of different concentrations to react with the soil in an oscillation reaction; Obtaining a conductivity curve of the solution after the oscillation reaction, and correcting the conductivity curve, wherein the correction is based on a temperature compensation model established through regression analysis; The optimal concentration of extractant was screened based on the corrected conductivity curve combined with soil physical and chemical characteristics; According to the optimum concentration of extractant, oscillation reaction is carried out with soil at different oscillation rates, and the optimum oscillation rate is analyzed and screened.

2. The method for determining the exchangeable fluoride ion content in soil according to claim 1, characterized in that: The physical and chemical characteristics of the soil are obtained, and different concentrations of extractants are selected to react with the soil in an oscillating manner, as follows: Collect soil samples to be tested and determine the physical and chemical characteristics of the soil; According to the physical and chemical characteristics of the soil, different concentrations of extractants are selected; The selected concentration of extractant was added to the soil sample and the soil and extractant were mixed by a shaker.

3. The method for determining the exchangeable fluoride ion content in soil according to claim 2, characterized in that: The conductivity curve of the solution after the oscillation reaction is obtained, and the conductivity curve is corrected. The correction is based on the temperature compensation model established by regression analysis, which is specifically as follows: After mixing the soil and the extractant by an oscillator, a supernatant is obtained; The conductivity of the supernatant is measured in real time by a conductivity meter, the conductivity values ​​at different time points are recorded, and a conductivity curve is generated; Preparing fluoride ion solutions of known different concentrations respectively, and measuring the conductivity values ​​of the solutions at multiple temperature gradients to obtain a temperature-conductivity data set; A temperature compensation model was constructed by performing linear regression analysis using the least squares method based on the temperature-conductivity data set; Based on the temperature compensation model, the collected original conductivity data is corrected to generate a temperature-standardized corrected conductivity curve.

4. The method for determining the exchangeable fluoride ion content in soil according to claim 3, characterized in that: The conductivity curve correction is as follows: Obtain the conductivity value to be corrected, and select the temperature value where the conductivity value is located as the regression point; Acquire first data of the regression point, and obtain a neighborhood range of the regression point according to the first data, wherein the first data includes a temperature difference and a number of data points; Obtain the weight of each data point in the neighborhood based on the Gaussian weight function; Based on the weight of each data point, local weighted regression fitting is performed on the data points within the neighborhood to obtain a local fitting model; According to the corrected conductivity value of the regression point output by the local fitting model, a corrected conductivity curve is obtained.

5. The method for determining the exchangeable fluoride ion content in soil according to claim 4, characterized in that: The neighborhood range of the regression point obtained according to the first data is specifically as follows: Obtaining the temperature gradient between the regression point to be corrected and its adjacent data points; According to the temperature gradient, the fitting accuracy data under different neighborhood ranges are obtained, and according to the accuracy data, the initial neighborhood range under the corresponding temperature gradient is obtained; Get the actual number of data points in the initial neighborhood around the regression point, and get the variance of the data points in the initial neighborhood; According to the variance of the data points, correction is performed based on the preset neighborhood range correction formula to obtain the neighborhood range.

6. The method for determining the exchangeable fluoride ion content in soil according to claim 5, characterized in that: The optimal concentration of extractant is screened according to the modified conductivity curve combined with the soil physical and chemical characteristics, as follows: Extracting conductivity features according to the conductivity curve; The extraction effect of the extractant on fluoride ions at different concentrations was evaluated based on the conductivity characteristics; According to the extraction effect evaluation results and soil physical and chemical characteristics, the matching relationship between soil type and extractant concentration was constructed based on the fuzzy comprehensive evaluation method. The optimal concentration of extractant was screened out according to the matching relationship, and a prediction model between soil physical and chemical characteristics and the optimal concentration of extractant was established based on the BP neural network.

7. The method for determining the exchangeable fluoride ion content in soil according to claim 6, characterized in that: The prediction model between soil physicochemical characteristics and optimal extractant concentration is established based on the BP neural network, as follows: Extract physical and chemical characteristics data from multiple tested soil samples; Obtain the corrected conductivity curve characteristics of each sample at different extractant concentrations; The soil physical and chemical parameters are used as feature vectors to construct input data sets, and the optimal extractant concentration corresponding to the current sample at different concentrations is used as a label to form supervised learning samples; Divide supervised learning samples into training set and validation set, and input them into BP neural network model for training; According to the new soil sample data, the soil physical and chemical characteristics are input, and the optimal extractant concentration is predicted based on the trained BP neural network model.

8. The method for determining the exchangeable fluoride ion content in soil according to claim 7, characterized in that: The extractant and soil are subjected to oscillation reactions at different oscillation rates according to the optimal concentration, and the optimal oscillation rate is analyzed and screened, as follows: According to the optimal concentration of the extractant, a number of different oscillation rate levels are set to form an oscillation rate gradient group; The soil sample of equal weight was mixed with the extractant of optimal concentration of equal volume, and the mixture was reacted under different shaking rate conditions to obtain the supernatant; Conducting real-time monitoring of the conductivity of the supernatant at each set of oscillation rates to obtain a conductivity curve, and then correcting the conductivity curve; Extract the data features of the modified conductivity curve under different oscillation rate conditions and construct an oscillation rate data set; According to the oscillation rate data set, the extraction effect under each oscillation rate condition is evaluated based on the grey correlation analysis method; The optimal oscillation rate was screened out based on the extraction effect evaluation results under various oscillation rate conditions.

9. The method for determining the exchangeable fluoride ion content in soil according to claim 8, characterized in that: The neighborhood range correction formula is as follows: Where: is the corrected neighborhood range, is the initial neighborhood range, is the adjustment coefficient, is the local variance, is the maximum variance, i.e., the largest variance in the data set.

10. A device for measuring the exchangeable fluoride ion content in soil using the method according to any one of claims 1 to 9, characterized in that: It includes a data acquisition module, a correction module, an extractant screening module and an oscillation rate screening module, and there are connections between the modules; The data acquisition module is used to analyze the soil to be tested, obtain the physical and chemical characteristics of the soil, and select different concentrations of extractants to oscillate with the soil; A correction module is used to obtain a conductivity curve according to the solution after the oscillation reaction, and to correct the conductivity curve based on the established temperature compensation model through regression analysis; Extraction agent screening module, used to screen the optimal concentration of extraction agent based on the corrected conductivity curve combined with soil physical and chemical characteristics; The oscillation rate screening module is used to perform oscillation reactions between the extractant and the soil at different oscillation rates according to the optimal concentration, and analyze and screen the optimal oscillation rate.

Images