A method and device for measuring the content of exchangeable fluoride ions in soil

Abstract

The present invention discloses a method and device for measuring the content of exchangeable fluoride ions in soil, which relates to the technical field of soil chemical analysis, and includes the following steps: analyzing the soil to be measured to obtain the physical and chemical characteristics of the soil, and selecting extractants with different concentrations to react with the soil by oscillation; obtaining the conductivity curve according to the solution after the oscillation reaction, and performing regression analysis to establish a temperature compensation model to correct the conductivity curve; screening the optimal concentration of the extractant 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 optimal concentration of the extractant, and analyzing and screening the optimal oscillation rate; The present application effectively improves the accuracy, stability and adaptability of the measurement results by integrating methods such as soil physical and chemical characteristics analysis, conductivity temperature compensation correction, extractant 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. An appropriate amount of fluorine is beneficial to organisms, but excessive fluorine will have an adverse impact 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] The existing methods for determining fluoride ions in soil mainly include ion selective electrode method, ultraviolet spectrophotometry, ion chromatography, etc. Although these methods have a certain degree of accuracy under laboratory conditions, they have problems such as complex operation, high requirements for experimental environment, cumbersome sample pretreatment, and strong instrument dependence, making it difficult to meet the actual needs of on-site rapid and high-throughput detection. At the same time, the 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 the 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 the publication number of 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, make up the volume with pure water and mix well, then transfer it 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; measure the potential response values of the sample and the blank test solution in turn 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 extractant for water-soluble fluorine according to 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 first weak base and then weak acid; when the soil is alkaline, adopt the extraction method of first weak acid and then weak base, and measure the content of adsorbed fluorine; S3. Mix the extraction residue after extraction in step S2 with the extractant for oxidatively bound fluorine according to 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 extraction 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:

[0007] Traditional methods often use extractants with fixed concentrations for fluoride ion extraction, ignoring the influence of soil types on 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

[0008] In order to overcome the above defects of the prior art, an embodiment of the present invention provides a method and device for measuring 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 extractant concentration, and intelligent matching of oscillation rate, it solves the problems existing in the traditional process of measuring the content of exchangeable fluoride ions in soil, such as blind selection of extractants, large temperature interference, inaccurate conductivity curve, insufficient optimization of oscillation parameters, etc., and greatly improves the accuracy, stability and adaptability of the measurement results.

[0009] To achieve the above object, the present invention provides the following technical solution:

[0010] A method for measuring the content of exchangeable fluoride ions in soil includes: analyzing the soil to be measured to obtain the physical and chemical characteristics of the soil, and selecting extractants 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; screening the optimal concentration extractant 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 optimal concentration extractant, and analyzing and screening the optimal oscillation rate.

[0011] In a preferred embodiment, obtaining the physical and chemical characteristics of the soil and selecting extraction agents with different concentrations for oscillating reaction with the soil is as follows: collecting soil samples to be measured and determining the physical and chemical characteristics of the soil; selecting extraction agents with different concentrations according to the physical and chemical characteristics of the soil; adding the selected concentration of extraction agent to the soil sample and mixing the soil and the extraction agent through an oscillator.

[0012] In a preferred embodiment, obtaining the conductivity curve of the solution after the oscillating reaction and correcting the conductivity curve, where the correction is based on establishing a temperature compensation model through regression analysis, is as follows: after mixing the soil and the extraction agent through an oscillator, obtaining the supernatant; measuring the conductivity of the supernatant in real time through a conductivity meter, recording the conductivity values at different time points, and generating a conductivity curve; respectively preparing fluoride ion standard solutions with known concentrations, measuring their conductivity values, obtaining a temperature-conductivity data set, and performing linear regression analysis through the least squares method to construct a temperature compensation model; based on the temperature compensation model, correcting the collected original conductivity data to generate a corrected conductivity curve after temperature standardization.

[0013] In a preferred embodiment, correcting the conductivity curve is as follows: obtaining the conductivity value to be corrected and selecting the temperature value where the conductivity value is located as the regression point; obtaining the first data of the regression point, and obtaining the neighborhood range of the regression point according to the first data, where the first data includes temperature difference and the number of data points; obtaining the weight of each data point in the neighborhood range based on the Gaussian weight function; performing 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; outputting the corrected conductivity value of the regression point according to the local fitting model to obtain the corrected conductivity curve.

[0014] In a preferred embodiment, obtaining the neighborhood range of the regression point according to the first data is as follows: obtaining the temperature gradient between the regression point to be corrected and its adjacent data points; obtaining the fitting accuracy data under different neighborhood ranges according to the temperature gradient, and obtaining the initial neighborhood range corresponding to the temperature gradient according to the accuracy data; obtaining the actual number of data points within the initial neighborhood range near the regression point, and obtaining the variance of the data points within the initial neighborhood range; correcting according to the variance of the data points based on a preset neighborhood range correction formula to obtain the neighborhood range.

[0015] 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, specifically 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.

[0016] 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, specifically 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.

[0017] In a preferred embodiment, the oscillatory reaction of the best concentration extractant with the soil at different oscillation rates is carried out, and the best oscillation rate is analyzed and screened, specifically 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 best concentration extractants are mixed and reacted under different oscillation rate conditions respectively to obtain supernatant liquids; the conductivity of the supernatant liquids under each oscillation rate is monitored in real time to obtain conductivity curves and the conductivity curves are corrected; the data characteristics of the corrected conductivity curves 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 effects under each oscillation rate condition are evaluated; the best oscillation rate is screened according to the evaluation results of the extraction effects under each oscillation rate condition.

[0018] Device for method of measuring content of exchangeable fluoride ions in soil, 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 different concentrations of extractants 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 combined 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.

[0019] Technical effects and advantages of a method and device for measuring the content of exchangeable fluoride ions in soil according to the present invention:

[0020] 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 problem of measurement deviation in the "fixed extractant - fixed condition" mode in traditional methods is effectively avoided, and the applicability and accuracy of the method are improved.

[0021] 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 standard fluoride ion solution samples and constructing a temperature - conductivity regression model, dynamic correction of conductivity data under different reaction environments is realized, solving the problem of result errors caused by temperature fluctuations in traditional methods, and significantly improving the stability and repeatability of 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 types of soil, and significantly improves work efficiency. Description of the Drawings

[0022] Figure 1 It is a schematic flow chart of a method for measuring the content of exchangeable fluoride ions in soil according to the present invention.

[0023] Figure 2 It is a schematic structural diagram of a device for a method for measuring the content of exchangeable fluoride ions in soil according to the present invention.

[0024] Figure 3 This is a data graph for evaluating the extraction effect under different oscillation rate conditions of the present invention.

[0025] Figure 4 This is a comparison graph of conductivity curves of the present invention under the condition of 0.15 mol / L extractant. Detailed implementation manners

[0026] 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.

[0027] 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:

[0028] S1. Obtain the physical and chemical characteristics of the soil, and select extractants with different concentrations to react with the soil by oscillation.

[0029] In this embodiment, obtaining the physical and chemical characteristics of the soil and selecting extractants with different concentrations to react with the soil by oscillation are specifically as follows:

[0030] Collect soil samples to be measured, select representative positions of the soil to ensure the uniformity and representativeness of the samples;

[0031] Use instrument equipment (such as pH meter, conductivity meter, 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%;

[0032] According to the physical and chemical characteristics of the soil, select extractants with different concentrations. For example, feasible extractants include ammonium acetate solution, and the concentrations are set to 0.05 mol / L, 0.10 mol / L, 0.15 mol / L, 0.20 mol / L, and 0.25 mol / L;

[0033] Add the selected concentration of extractant to the soil sample, and according to a certain ratio (such as a soil-to-extractant ratio of 1:1), use an oscillator to fully mix the soil and the extractant.

[0034] S2. Obtain the conductivity curve of the solution after the oscillation reaction, and correct the conductivity curve. The correction is based on establishing a temperature compensation model by regression analysis for correction.

[0035] In this embodiment, the conductivity curve of the solution after the oscillating 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:

[0036] According to the preset ratio of extractant concentration to soil, the extraction reaction is completed under constant temperature conditions, and the soil and the extractant are fully reacted through an oscillating device. The oscillation time is controlled within 30 minutes. After the reaction ends, the mixed solution is centrifuged to obtain the supernatant for detection;

[0037] An electrical conductivity meter is used to measure the conductivity of the supernatant in real time, and the conductivity values at different time points are recorded. The measurement time interval can be set to once every 30 seconds or once every 1 minute to generate an original curve of conductivity changing 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;

[0038] A temperature sensor is used 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 calculation;

[0039] Standard fluoride ion solutions with known concentrations are respectively prepared, their conductivity values are measured to obtain a temperature-conductivity data set, and a temperature compensation model is constructed through linear regression analysis by the least squares method;

[0040] Based on the constructed temperature compensation model, the collected original conductivity data is corrected, and the conductivity value at each time point is uniformly corrected to the equivalent conductivity value under the condition of 25 °C to generate a corrected conductivity curve after temperature standardization.

[0041] The temperature compensation model is specifically as follows:

[0042]

[0043] 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).

[0044] For example: when = 30 °C, = 800 μS / cm, Taking 0.02 / °C and substituting it into the temperature compensation model, the compensated conductivity is 727.27 μS / cm.

[0045] In this embodiment, the conductivity curve is corrected as follows:

[0046] Obtain the conductivity value to be corrected, and select the temperature value where the conductivity value is located as the regression point;

[0047] 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;

[0048] Obtain the weight of each data point within the neighborhood range based on the Gaussian weight function;

[0049] Perform local weighted regression fitting on the data points within the neighborhood range based on the weight of each data point to obtain a local fitting model;

[0050] Output the corrected conductivity value of the regression point according to the local fitting model to obtain the corrected conductivity curve.

[0051] In this embodiment, obtaining the neighborhood range of the regression point according to the first data is as follows:

[0052] Obtain the temperature gradient between the regression point to be corrected and its adjacent data points;

[0053] 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;

[0054] 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;

[0055] Correct according to the variance of the data points based on the preset neighborhood range correction formula to obtain the neighborhood range.

[0056] The neighborhood range correction formula is specifically as follows:

[0057]

[0058] 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.

[0059] 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.

[0060] Exemplarily, a comparison graph of the initial conductivity curve and the corrected conductivity curve is as Figure 4 shown:

[0061] The initial conductivity is high at 0 minutes because the extractant itself has strong conductivity. As time goes by, the conductivity gradually increases;

[0062] 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;

[0063] 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.

[0064] 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.

[0065] Furthermore, mixed reactions under different temperature conditions are 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 meet the measurement requirements in different environments.

[0066] S3. According to the corrected conductivity curve and combined with the soil physical and chemical characteristics, screen the extraction agent with the optimal concentration.

[0067] In this embodiment, according to the corrected conductivity curve and combined with the soil physical and chemical characteristics, screen the extraction agent with the optimal concentration, which is specifically as follows:

[0068] According to the conductivity curve, extract the conductivity characteristics, where the conductivity characteristics include the maximum conductivity value, the conductivity value in the stable stage, and the conductivity rising slope in the early stage of the reaction;

[0069] Evaluate the extraction effect of the extraction agent on fluoride ions at different concentrations;

[0070] Based on the extraction effect and soil physical and chemical characteristics, and using the fuzzy comprehensive evaluation method, construct the adaptation relationship between the soil type and the extraction agent concentration;

[0071] According to the screening, select the extraction agent concentration with the optimal extraction efficiency and the minimum matrix interference under specific soil conditions as the target optimal usage concentration;

[0072] Furthermore, use the BP neural network to train the historical experimental data, establish a prediction model between the soil physical and chemical characteristics and the optimal extraction agent concentration, and realize the rapid recommendation and automatic matching of subsequent samples.

[0073] The evaluation formula for the extraction effect of fluoride ions at different concentrations is specifically as follows:

[0074]

[0075] 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 stable 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 .

[0076] For example: with an extractant concentration of 0.15 mol / L as the condition, the maximum conductivity value recorded is 800μS / cm, the conductivity value in the stable stage is 720μS / cm, the conductivity rising slope in the early stage of the reaction is 60μS / cm / min, the weight coefficients are set to w1=0.4, w2=0.4, w3=0.2, and the extraction effect evaluation result calculated according to the extraction effect evaluation formula is 620; with an extractant concentration of 0.20 mol / L as the condition, the maximum conductivity value recorded is 900μS / cm, the conductivity value in the stable stage is 850μS / cm, the conductivity rising slope in the early stage of the reaction is 40μS / cm / min, the weight coefficients are set to w1=0.4, w2=0.4, w3=0.2, and the extraction effect evaluation result calculated according to the extraction effect evaluation formula is 703.

[0077] In this embodiment, a prediction model between soil physicochemical characteristics and optimal extractant concentration is established to achieve rapid recommendation and automatic matching of subsequent samples, as follows:

[0078] For multiple tested soil samples, collect their physical and chemical characteristics data, including but not limited to pH value, organic matter content, cation exchange capacity (CEC), particle size composition, moisture content, etc.;

[0079] At the same time, the corrected conductivity curve characteristics of each sample at different extractant concentrations are recorded, including indicators such as maximum conductivity, stable conductivity value and initial reaction rate; the above data are cleaned, outliers and missing values are removed, and linear interpolation or mean filling is performed when necessary. Then all characteristic data are standardized to ensure the consistency of characteristics of different dimensions. Z-score standardization or Min-Max normalization method is often used;

[0080] 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;

[0081] In the BP neural network modeling, a three-layer neural network is constructed. The input layer corresponds to the soil parameter dimension, the output layer is the optimal concentration value, and the number of neurons in the hidden layer is set according to the empirical formula. The back propagation algorithm is used in the training process and the Adam optimizer is introduced to update the weights to improve the convergence speed and prediction accuracy.

[0082] The original sample set is divided into a training set and a validation set. The mean square error, mean absolute error and coefficient of determination are used as evaluation indicators to judge the model's fitting ability and generalization ability. If the error index meets the accuracy requirements, the model is established.

[0083] According to the new soil sample data and the input of soil physical and chemical characteristics, the optimal extractant concentration is predicted based on the trained BP neural network model.

[0084] The application process of the BP neural network model is as follows:

[0085] Extract representative physical and chemical parameters from 80 groups of tested soil samples, including: pH value, conductivity, organic matter content, and soil humidity, a total of 4-dimensional features. At the same time, record the characteristics of the corrected conductivity curve of each group of samples under various extractant concentrations, and extract the maximum conductivity, conductivity at the stable stage, and slope at the initial stage of the reaction as the basis for extraction efficiency;

[0086] 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;

[0087] Construct a three-layer BP neural network model with a structure of 4-10-1, that is: 4 nodes in the input layer correspond to 4 physical and chemical characteristics, 10 neurons are set in the hidden layer, the activation function is ReLU, and 1 node in the output layer corresponds to the predicted optimal extractant concentration (unit: mol / L);

[0088] Divide the 80 groups of samples into a training set and a validation set according to 8:2. During the training process, monitor the loss decline curve in real time to avoid overfitting. The model finally obtains a mean absolute error (MAE) of 0.012 mol / L on the validation set, and the final prediction accuracy is 0.93, with good generalization ability.

[0089] It should be noted that in the data acquisition stage, this method makes full use of the conductivity curve after temperature compensation to extract representative conductivity characteristic indexes, including the maximum conductivity value, conductivity value at the stable stage, and rising slope at the initial stage of the reaction. These characteristics reflect the extraction efficiency and reaction dynamics of the extractant for fluoride ions at different concentrations, which is 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.

[0090] S4. Conduct an oscillation reaction between the soil and the extractant with the optimal concentration at different oscillation rates, and analyze and screen the optimal oscillation rate.

[0091] In this embodiment, conduct an oscillation reaction between the soil and the extractant with the optimal concentration at different oscillation rates, and analyze and screen the optimal oscillation rate, specifically as follows:

[0092] 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 under each group of oscillation rates, keep the reaction time, temperature, and extractant concentration consistent to ensure single-variable control;

[0093] Mix equal - mass soil samples with an equal - volume extractant at the optimal concentration, and react them under the above - mentioned different oscillation rate conditions. The reaction time is uniformly fixed, and the experimental environmental temperature is controlled at a constant level (such as 25°C).

[0094] Monitor the conductivity of the extract obtained under each oscillation rate in real - time to obtain a conductivity curve, and correct the conductivity curve through a temperature compensation model to obtain a corrected conductivity change curve.

[0095] Extract the data characteristics of the corrected conductivity curves under 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, and construct an oscillation rate data set.

[0096] According to the oscillation rate data set, based on the grey relational analysis method, evaluate the extraction effect under each oscillation rate condition.

[0097] Select the rate with high extraction efficiency, low energy consumption, and good reaction stability as the optimal oscillation rate of the target soil according to the evaluation results.

[0098] The evaluation formula for the extraction effect under each oscillation rate condition is as follows:

[0099]

[0100] Standard sequence:

[0101]

[0102] In the formula: is the evaluation result of the extraction effect under each oscillation rate condition, is the total number of data characteristics, represents the grey correlation coefficient between the i - th oscillation rate and the standard sequence under the j - th data characteristic, is the standard sequence.

[0103] For example: Taking the extraction experiment with an oscillation rate of 200 rpm as an example, select three representative characteristic parameters: the maximum conductivity of 0.95, the time required to reach stability of 0.90, and the slope of the conductivity increase in the initial stage of the reaction of 0.91. The grey correlation coefficients are 1.00, 0.75, and 0.789 respectively. Set the standard sequence as , the number of data characteristics is m = 3, and according to the extraction effect evaluation formula, output 0.846. The data graph of the obtained extraction effect evaluation result is as Figure 3 shown:

[0104] The extraction effect is optimal at 250 rpm, and the extraction effect is evaluated as 0.92. After exceeding 250 rpm, the extraction effect gradually decreases, probably because high-speed oscillation causes an increase in foam, uneven dispersion of the solution, or mechanical damage to the active ingredients.

[0105] Example 2 Figure 2 The device for the method for measuring the content of exchangeable fluoride ions in soil according to the present invention is provided, which is characterized by including a data acquisition module, a correction module, an extractant screening module, and an oscillation rate screening module, and there are connections between the modules;

[0106] The data acquisition module is used to analyze the soil to be measured, obtain the physical and chemical characteristics of the soil, and select different concentrations of extractants to react with the soil by oscillation;

[0107] 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;

[0108] The extractant screening module is used to screen the extractant with the optimal concentration according to the corrected conductivity curve combined with the physical and chemical characteristics of the soil;

[0109] The oscillation rate screening module is used to perform oscillation reactions at different oscillation rates between the extractant with the optimal concentration and the soil, and analyze and screen the optimal oscillation rate.

[0110] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get 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.

[0111] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0112] 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. Professional technicians 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.

[0113] In addition, the functional modules in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0114] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the said claims.

[0115] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, 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 content of exchangeable fluoride ions in soil, characterized in that, Including: Obtain the physical and chemical characteristics of the soil, and select extractants with different concentrations to react with the soil by oscillation; Obtain the conductivity curve of the solution after the oscillation reaction, and correct the conductivity curve. The correction is based on establishing a temperature compensation model through regression analysis for correction; According to the corrected conductivity curve combined with the physical and chemical characteristics of the soil, screen the extractant with the best concentration. The extractant with the best concentration is obtained by evaluating the extraction effect of fluoride ions; React the soil with the extractant with the best concentration under different oscillation rates, and analyze and screen the best oscillation rate. The best oscillation rate is obtained by evaluating the extraction effect under each oscillation rate condition; The correction of the conductivity curve includes: Obtain the conductivity value to be corrected, and select the temperature value where the conductivity value is located as the regression point; 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 number of actual data points and the variance of the data points within the initial neighborhood range near the regression point, and perform correction based on the preset neighborhood range correction formula to obtain the neighborhood range; Obtain the weights of each data point within the neighborhood range based on the Gaussian weight function, and perform locally weighted regression fitting on the data points within the neighborhood range 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; The temperature compensation model is as follows: Wherein: is the conductivity at temperature T, is the conductivity at the standard temperature of 25°C, is the temperature coefficient, is the actual temperature during measurement; The evaluation formula for the extraction effect of fluoride ions is as follows: Wherein: 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 under the i-th concentration condition, is the rising slope of the conductivity in the early stage of the reaction under the i-th concentration condition, , and are weighting coefficients; The evaluation formula for the extraction effect under each oscillation rate condition is as follows: In the formula: is the evaluation result of the extraction effect under each oscillation rate condition, is the total number of data features, is the grey correlation coefficient between the i-th oscillation rate and the standard sequence under the j-th data feature; The neighborhood range correction is as follows: Wherein: is the corrected neighborhood range, is the initial neighborhood range, is the adjustment coefficient, is the local variance, is the maximum variance.

2. The method for determining the content of exchangeable fluoride ions in soil according to claim 1, characterized in that, The obtaining of the physical and chemical characteristics of the soil, and the selection of extractants with different concentrations to react with the soil by oscillation are specifically as follows: Collect soil samples to be tested, and measure the physical and chemical characteristics of the soil; Select extractants with different concentrations according to the physical and chemical characteristics of the soil; Add the selected concentration of extractant to the soil sample, and mix the soil and the extractant through an oscillator.

3. The method for determining the content of exchangeable fluoride ions in soil according to claim 2, wherein The obtaining of the conductivity curve of the solution after the oscillation reaction, and the correction of the conductivity curve. The correction is based on establishing a temperature compensation model through regression analysis for correction, and is 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; Configure fluoride ion solutions with known different concentrations respectively, and measure the conductivity values of the solutions at multiple temperature gradients to obtain a temperature-conductivity data set; Perform linear regression analysis through the least squares method according to the temperature-conductivity data set to construct a temperature compensation model; Based on the temperature compensation model, correct the collected original conductivity data to generate a corrected conductivity curve after temperature standardization.

4. The method for determining the content of exchangeable fluoride ions in soil according to claim 3, wherein The screening of the extractant with the best concentration according to the corrected conductivity curve combined with the physical and chemical characteristics of the soil is specifically as follows: Extract conductivity characteristics according to the conductivity curve; Evaluate the extraction effect of the extractant on fluoride ions at different concentrations according to the conductivity characteristics; Based on the evaluation results of the extraction effect and the physical and chemical characteristics of the soil, construct an adaptation relationship between the soil type and the extractant concentration based on the fuzzy comprehensive evaluation method; The best concentration extractant is screened according to the adaptation relationship, and a prediction model between soil physical and chemical characteristics and the best extractant concentration is established based on the BP neural network.

5. The method for determining the content of exchangeable fluoride ions in soil according to claim 4, characterized in that, The prediction model between soil physical and chemical characteristics and the best extractant concentration is established based on the BP neural network, which is specifically as follows: Extract physical and chemical characteristic data from multiple tested soil samples; Obtain the corrected conductivity curve characteristics of each sample at different extractant concentrations; Construct an input data set with soil physical and chemical parameters as feature vectors, and form a supervised learning sample with the best extractant concentration corresponding to the current sample at different concentrations as the label; Divide the supervised learning sample into a training set and a validation set, and input it into the BP neural network model for training; According to the new soil sample data, input the soil physical and chemical characteristics, and predict the best extractant concentration based on the trained BP neural network model.

6. The method for determining the content of exchangeable fluoride ions in soil according to claim 5, wherein The oscillating reaction between the best concentration extractant and the soil is carried out at different oscillating rates, and the best oscillating rate is analyzed and screened, which is specifically as follows: According to the best concentration extractant, set several different oscillating rate levels to form an oscillating rate gradient group; Mix soil samples of equal mass with an equal volume of the best concentration extractant, and react them under different oscillating rate conditions respectively to obtain the supernatant; Carry out real-time monitoring of the conductivity of the supernatant under each oscillating rate to obtain the conductivity curve, and correct the conductivity curve; Extract the data characteristics of the corrected conductivity curve under different oscillating rate conditions to construct an oscillating rate data set; According to the oscillating rate data set, evaluate the extraction effect under each oscillating rate condition based on the grey correlation analysis method; Screen out the best oscillating rate according to the evaluation results of the extraction effect under each oscillating rate condition.

7. An apparatus for using the method for measuring the content of exchangeable fluoride ions in soil according to any one of claims 1-6, characterized in that, It includes a data acquisition module, a correction module, an extractant screening module and an oscillating 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 concentration extractants to react with the soil by oscillation; The correction module is used to obtain the conductivity curve according to the solution after the oscillating reaction, and perform regression analysis to correct the conductivity curve based on the established temperature compensation model; The extractant screening module is used to screen the best concentration extractant according to the corrected conductivity curve combined with the soil physical and chemical characteristics; The oscillating rate screening module is used to carry out the oscillating reaction between the best concentration extractant and the soil at different oscillating rates, and analyze and screen the best oscillating rate.

Images