Method for determining content of iron oxide in ore by using spectrophotometer

Abstract

The invention provides a method for determining the content of iron oxide in ore by using a spectrophotometer, which relates to the technical field of chemical component analysis and comprises five steps of pretreatment of ore samples, construction and training of a dynamic correction model, preparation of a standard solution, establishment of a spectrophotometer measurement standard and correction of the dynamic correction model. According to the method, the dynamic correction model is constructed based on the XGBoost regression model, so that an intelligent algorithm and a spectrophotometer measurement method are combined, measurement interference caused by environment temperature and humidity changes and ore component differences can be effectively eliminated through the dynamic correction model, the problem of large measurement errors in the prior art is solved, and the measurement accuracy is improved. The method not only improves the accuracy and stability of iron content detection of raw materials in industries such as photovoltaic glass, but also solves the interference of environmental factors and ore matrixes on measurement results, and meets the requirements of photovoltaic glass production on strict control of the iron content of the raw materials.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of chemical component analysis, and particularly relates to a method for measuring the content of iron oxide in ore by using a spectrophotometer. BACKGROUND

[0002] Iron is one of the common impurities in ore, especially in limestone, dolomite and other ores. The content of iron not only affects the whiteness of the ore, but also directly affects the industrial value of the ore. For limestone and dolomite, the content of iron directly determines its applicability as a raw material for photovoltaic glass. Photovoltaic glass requires extremely high purity to ensure the high efficiency of photovoltaic cells, therefore, the application of ore with too high iron content in this field is strictly limited.

[0003] In the production of photovoltaic glass, the content of iron is a strictly controlled factor. Photovoltaic glass requires extremely high light transmittance, and the content of iron directly affects the optical performance of the glass. In particular, when manufacturing photovoltaic glass, if the content of iron exceeds the specified standard, it will lead to a decrease in the optical transmittance of the glass, thereby affecting the efficiency of the photovoltaic cell. Therefore, photovoltaic glass manufacturers are very strict in controlling the content of iron in the raw materials, which is one of the reasons for measuring and controlling the content of iron in the prior art.

[0004] At present, common methods for detecting the content of iron oxide in ore include gravimetric method, chemical titration method, atomic absorption spectrometry (AAS) and spectrophotometric method, etc. Among them, the spectrophotometric method is a relatively simple, economical and efficient detection method, and has been widely used in ore analysis. It calculates the content of iron oxide in the sample by measuring the absorbance of the solution. However, in actual use, the spectrophotometric method still faces some challenges.

[0005] For example, factors such as environmental temperature, humidity and air pressure may affect the measurement results of the spectrophotometer; factors such as the instrument accuracy of the spectrophotometer, the purity of the solvent and the preparation of the solution may introduce measurement errors, such as trace impurities in the solvent or inaccurate preparation of the solution may cause changes in absorbance; other components in the ore (such as calcium, magnesium and other minerals) may interact with iron oxide, affecting the accuracy of the absorbance, especially the mineral matrix of different ores is quite different, which makes it difficult to accurately reflect the actual concentration of the measured results of iron content, therefore, the present application proposes a method for measuring the content of iron oxide in ore by using a spectrophotometer to solve the problems existing in the prior art. SUMMARY

[0006] In view of the above problems, the purpose of the present application is to provide a method for measuring the content of iron oxide in ore by using a spectrophotometer, which can solve the problems in the prior art by combining the spectrophotometer with a dynamic correction model.

[0007] To achieve the purpose of the present application, the present application realizes the following technical solutions: a method for measuring the content of iron oxide in ore by using a spectrophotometer, comprising the following steps:

[0008] Step one, pretreatment of ore sample

[0009] Prepare the ore sample to be measured, and give it a unique digital ID according to the type of the ore sample to be measured, record it, then grind the ore sample to be measured to obtain a ground ore sample, and then post-treat the ground ore sample to obtain a sample solution;

[0010] Step two, construction and training of dynamic correction model

[0011] A dynamic correction model is constructed based on an XGBoost regression model, and is trained based on historical data, so that the input data is apparent concentration, environmental temperature, environmental humidity and sample information, and the output data is a correction value;

[0012] Step three, preparation of standard solution

[0013] High-purity iron trioxide powder is used as a standard sample, pretreated, dissolved with hydrochloric acid, and then diluted with deionized water to a predetermined volume to prepare a standard iron solution, and then ascorbic acid solution, buffer and phenanthroline chromogenic solution are added in turn, mixed uniformly to obtain a chromogenic standard solution;

[0014] Step four, establishment of spectrophotometer measurement standard

[0015] Several groups of standard solutions with different concentrations are taken, and the spectrophotometer is used to measure at a wavelength of 510 nm, and the environmental temperature, humidity and measurement timestamp of each measurement are recorded, and then a linear standard curve equation is fitted according to the measured absorbance;

[0016] Step five, correction of dynamic correction model

[0017] Take a sample solution with the same volume as the standard solution, and follow the standard solution coloration process, then use the spectrophotometer to measure and record the information, then substitute the measured absorbance value of the sample solution into the standard curve equation in step four to obtain the apparent concentration of the sample solution, input it into the dynamic correction model, output the correction value, and then calculate the corrected iron oxide concentration, combine the corrected iron oxide concentration with the mass of the ore sample to be measured to obtain the iron oxide content, and complete the measurement.

[0018] Further improvement lies in that the specific way of the grinding treatment in the step one is that the coarse grinding is performed by using a jaw crusher to make the particle size of the ore sample to be measured less than or equal to 5 mm, and then the fine grinding is performed by using a planetary ball mill to make the particle size of the ore sample to be measured less than or equal to 0.15 mm.

[0019] Further improvement lies in that the specific way of the post-treatment in the step one is that the grinding ore sample is dissolved by using dilute hydrochloric acid at a volume ratio of 1:1, and the heating and stirring are performed by using a heating and stirring device at a temperature of 80-90 DEG C, the heating is stopped when the light transmittance is greater than 90% and the light transmittance change rate of the solution is less than 1% per minute, and the vacuum filtration method is used to filter to obtain a sample solution, and finally the deionized water is used to set the volume to a preset volume.

[0020] Further improvement lies in that the specific steps of the training based on the historical data in the step two are as follows:

[0021] S1: Collect historical experimental data as a data set, the historical experimental data including spectrophotometer measurement values, environmental temperature and humidity, sample information and known iron oxide content;

[0022] S2: The data in the data set is standardized and then cleaned;

[0023] S3: The data set is divided into a training set and a validation set at a ratio of 7:3, the XGBoost regression model is trained by using the training set, and after the training is completed, the trained model is evaluated by using the validation set;

[0024] S4: The stability of the model is verified by using a five-fold cross-validation method, and thus a dynamic correction model is obtained.

[0025] Further improvement lies in that the specific way of the pretreatment in the step three is that the high-purity iron trioxide powder is placed in a muffle furnace, and is calcined at 550 DEG C ± 25 DEG C for 50-60 min, and then cooled to room temperature, and then hydrochloric acid is added at a ratio of 1 g:20 ml, and is heated and stirred in a constant-temperature water bath at 80-90 DEG C until the solid is completely dissolved to obtain a standard iron solution.

[0026] Further improvement lies in that the specific color developing process of the color developing standard solution in the step three is as follows:

[0027] SS1: The total volume of the standard iron solution to be prepared is set, and then V ml of the standard iron solution is taken, wherein V is greater than or equal to 0;

[0028] SS2: 10% of the total volume of ascorbic acid solution is added, and after stirring uniformly, it is placed for 5 min;

[0029] SS3: Then 20% of total capacity of buffer solution is added, and stirring is carried out;

[0030] SS4: 10% of total capacity of phenanthroline color developing solution is added, and stirring is carried out;

[0031] SS5: Deionized water is used to make up to total capacity, and color development is carried out after standing for 15 min.

[0032] Further improvement lies in that in the step four, a linear standard curve equation is fitted as follows:

[0033] C a =k·A 标 +b

[0034] In the formula, C a is a display concentration, A 标 is an absorbance of a standard solution, k is a slope, and b is an intercept.

[0035] Further improvement lies in that in the step five, a calculation formula of output calculation is as follows:

[0036] C 修 =C 样 +ΔC

[0037] In the formula, C 修 is a corrected iron oxide concentration, C 样 is an apparent concentration of a sample solution, and ΔC is a correction value obtained by a dynamic correction model.

[0038] Further improvement lies in that in the step five, a calculation formula of iron oxide content is as follows:

[0039]

[0040] In the formula, V is a volume of a sample solution, and m is a sample mass.

[0041] The present application has the following beneficial effects:

[0042] The present application is based on the XGBoost regression model to construct a dynamic correction model, thereby combining the intelligent algorithm with the spectrophotometric method, measuring the absorbance of the sample by the spectrophotometer, and calculating the apparent concentration of the sample by using the standard curve, then, through the XGBoost dynamic correction model, combining real-time environmental data (such as temperature, humidity) and ore type information, data correction is carried out to obtain more accurate iron oxide content. Thus, through the dynamic correction model, the measurement interference caused by the change of environmental temperature and humidity and the difference of ore composition can be effectively eliminated, the problem of large measurement error in the prior art is overcome, not only the accuracy and stability of the detection of the iron content of the raw material in the photovoltaic glass industry and the like are improved, but also the interference of environmental factors and ore matrix on the measurement result is solved, and the demand of the photovoltaic glass production for strict control of the iron content of the raw material is met. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 is a step flowchart of the present application. DETAILED DESCRIPTION

[0044] In order to deepen the understanding of the present application, the present application will be further described in combination with examples, and the present examples are only used to explain the present application and do not constitute a limitation on the protection scope of the present application.

[0045] The whiteness of the ore is an important indicator of its purity. In the production process, iron impurities will reduce the natural whiteness of the ore. Iron exists in the form of iron oxide in the ore, especially during the high-temperature calcination process of the ore, the iron oxide will further affect the color of the ore, resulting in a decrease in its whiteness. Ore with too high iron content not only has poor appearance color, but also its physical and chemical properties will be affected. For example, when the concentration of iron in limestone is too high, it may cause the desulfurization performance to decrease in the production of cement, affecting the quality of the final product.

[0046] For limestone, dolomite and other ores, the content of iron is a crucial quality indicator. The iron content in the ore not only reduces its natural whiteness, but also has a significant impact on its application in industry. Taking dolomite as an example, it is widely used in the construction, glass manufacturing and steel production industries. The quality of dolomite directly affects the efficiency of the reaction in the production process and the quality of the final product. Dolomite with too high iron content has increased solubility when calcined at high temperatures, which may result in low reaction efficiency and increased processing costs.

[0047] Therefore, it is necessary to control the iron content in the ore through measurement. The spectrophotometric method is a relatively simple, economical and efficient detection method, which has been widely used in ore analysis. It calculates the iron oxide content in the sample by measuring the absorbance of the solution. However, in actual use, the spectrophotometric method still faces some challenges, mainly in the following aspects:

[0048] (1) Environmental factors such as temperature, humidity and air pressure may affect the measurement results of the spectrophotometer. Fluctuations in temperature and humidity will change the absorbance of the solution, resulting in measurement errors.

[0049] (2) Other components in the ore (such as calcium, magnesium and other minerals) may interact with iron oxide, affecting the accuracy of absorbance. Especially the mineral matrix of different ores is quite different, which makes it difficult to accurately reflect the actual concentration of iron content.

[0050] (3) The instrument accuracy of the spectrophotometer, the purity of the solvent and the preparation of the solution will introduce measurement errors. For example, trace impurities in the solvent or inaccurate preparation of the solution may cause changes in absorbance.

[0051] Therefore, according to Figure 1 It is shown that the embodiment proposes a method for determining the content of iron oxide in ore by using a spectrophotometer, which includes the following steps:

[0052] Step one, pretreatment of ore sample

[0053] Prepare the ore sample to be tested. According to the type of ore sample to be tested (such as limestone, dolomite, etc.), assign a unique digital ID to each ore sample, and provide the ore type characteristics for the subsequent intelligent model.

[0054] Specifically, use dictionary mapping, limestone corresponds to code "1", dolomite corresponds to code "2", and other ore types are similarly coded. Then record the detailed information of each ore sample, including ore type, source, collection time, collection place, etc., to ensure the completeness and traceability of the sample information.

[0055] Then, the ore sample to be tested is ground. First, use a jaw crusher for coarse grinding to make the particle size of the ore sample to be tested ≤5mm, which is convenient for subsequent fine grinding operation. Then use a planetary ball mill for fine grinding to make the particle size of the ore sample to be tested ≤0.15mm, and obtain the ground ore sample

[0056] Then the post-processing of the ground ore sample is carried out, the ground ore sample is placed in a container, 1:1 volume ratio of dilute hydrochloric acid is added (for example, 3g of ground ore sample, the amount of dilute hydrochloric acid added is 30ml), and heating and stirring are carried out by a heating and stirring device. The heating temperature is set between 80-90℃ (80℃ in this embodiment), to ensure that the iron oxide in the ore is completely dissolved into iron ions. At the same time during the dissolution process, the light transmittance of the solution is monitored in real time by a turbidity sensor, and the heating is stopped when the light transmittance is >90% and the light transmittance change rate of the solution is less than 1% / min. After cooling to room temperature, the solid impurities in the solution are filtered out by using a vacuum filtration method through a 0.45μm micron filter membrane, to obtain a clear sample solution, and then deionized water is used for constant volume to ensure that the volume of the solution reaches the preset capacity (for example, based on 3g of ground ore sample, the capacity is 100mL). After constant volume, the sample solution is ready for subsequent spectrophotometer measurement;

[0057] Step two, construction and training of dynamic correction model

[0058] The dynamic correction model is constructed based on the XGBoost regression model, and is trained based on historical data, so that the input data is apparent concentration, environmental temperature, environmental humidity and sample information, and the output data is correction value. The goal of this step is to correct the spectrophotometer measurement result in real time through the XGBoost regression model, and to improve the accuracy of the determination of iron oxide content. The XGBoost regression model is a regression model based on gradient boosting decision tree (GBDT), which has high prediction accuracy and strong processing capability, can handle nonlinear relationships, and can predict by constructing multiple decision trees (basic learners of the model) and gradually correcting the errors of the previous trees. Correspondingly, after constructing the dynamic correction model, it is deployed to a computing device.

[0059] Specifically, the specific steps of training based on historical data are as follows:

[0060] S1: Collect historical experimental data as a data set, which includes apparent concentration (calculated from absorbance value), environmental temperature and humidity (environmental temperature at each measurement, accuracy requirement is ±0.5℃; environmental humidity, accuracy requirement is ±3%RH), sample information (type of ore sample, grinding particle size, sample dissolution temperature and light transmittance of solution, wherein the ore type code is consistent with the sample ID assigned in the pretreatment stage), and known iron oxide content (true iron oxide content measured by atomic absorption method, the requirement for expanded uncertainty is ≤2%, as the target value of the training model);

[0061] S2: Standardize the data within the dataset to adapt its range and distribution to the model training of XGBoost, then clean the data to remove outliers and noise data, specifically: eliminate the dimensional differences between features by Z-score standardization method, so that the mean of each feature is 0 and the standard deviation is 1, so that after standardization, all features will be in a similar numerical range, avoiding model bias caused by different dimensions, then clean the data, delete the samples with missing iron oxide true value, and mark the ore type unknown strip as special code 0, finally ensure that the retention rate of the cleaned data is not less than 95%;

[0062] S3: Divide the standardized and cleaned dataset into training set and validation set with a ratio of 7:3, train the XGBoost regression model using the training set, and set the key parameters during training: the regression objective function is the minimization of mean square error, generate 1200 decision trees with a maximum depth limit of 6 layers (to prevent overfitting), control the learning rate at 0.05 (to avoid over updating) and configure the subsampling rate at 0.75 (to reduce the risk of overfitting by random sampling). Early stopping mechanism is introduced during training, which automatically terminates optimization when the performance of the validation set does not improve for 50 consecutive rounds. Immediately after training, evaluate the model performance using the validation set, requiring that the three core indicators be met simultaneously: average absolute error ≤0.5 ppm, determination coefficient R 2 ≥0.98, maximum residual error ≤1.2 ppm, and any indicator not meeting the requirements needs to be re-adjusted.

[0063] S4: Use five-fold cross-validation method to verify the stability of the model, specifically, divide the entire dataset into five equal parts, and use four parts as training data and one part as test data, repeat the training and evaluation five times, finally require that the MAE fluctuation range of the five test rounds be less than 0.15 ppm, and the R 2 value of all test rounds be higher than 0.95, confirming that the model can maintain stable and reliable prediction performance under different data distributions, and obtaining the dynamic correction model.

[0064] Step three, preparation of standard solution

[0065] High-purity iron trioxide powder (purity ≥99.9%) is used as a standard sample, pretreated, dissolved with hydrochloric acid, and then diluted with deionized water to a predetermined volume to prepare a standard iron solution. Then add ascorbic acid solution, buffer and phenanthroline chromogenic solution in turn, mix well, and obtain the standard solution.

[0066] Specifically, the specific way of pretreatment is: put high-purity iron trioxide powder into a muffle furnace, and burn at 550℃±25℃ for 50-60min. At present, the water, impurities and crystal water in the iron trioxide are removed to ensure the accuracy of the standard solution. Then cool to room temperature (±25℃) in a desiccator, then add hydrochloric acid (6mol / L) according to the ratio of 1g (iron trioxide powder) : 20ml (hydrochloric acid), ensure that the concentration of hydrochloric acid is sufficient to dissolve iron trioxide, then place in a constant temperature water bath at 80-90℃ and heat and stir until the solid is completely dissolved, to obtain a standard iron solution.

[0067] According to the obtained standard iron solution, a standard solution is prepared and colored. The specific preparation process of coloring is:

[0068] SS1: First, set the total capacity of the standard iron solution to be prepared. In this embodiment, 100ml is taken as an example, then take Vml standard iron solution, where V≥0;

[0069] SS2: First, add 20% of the total capacity of ascorbic acid solution (total capacity is 100mL, then take 20mL of ascorbic acid solution), stir evenly, and stand for 5min to ensure that ascorbic acid and iron ions react fully. The role of ascorbic acid is to prevent further oxidation of iron oxide in the solution;

[0070] SS3: Then add 20% of the total capacity of the buffer solution (acetic acid-sodium acetate buffer solution, pH4.5), stir evenly. The addition of buffer solution helps to control the pH value of the solution within a suitable range;

[0071] SS4: Then add 10% of the total capacity of the phenanthroline color developing solution (0.1% concentration), shake vigorously to ensure that the color developing agent and iron ions react fully to form a stable red complex, enhance the absorbance of the solution, and facilitate subsequent spectrophotometer measurement

[0072] SS5: Use deionized water to make up to the total capacity, and stand for 15min for color development;

[0073] Step four, spectrophotometer measurement standard establishment

[0074] Take several groups of standard solutions with different concentrations (the concentration refers to the standard solution prepared by Vml standard iron solution), for example, five groups are taken in this embodiment, which are 0ml, 1ml, 2ml, 3ml and 4ml. Take 0ml as an example, which refers to the standard solution prepared by 0ml standard iron solution. In this way, measure the absorbance at 510nm wavelength by spectrophotometer, and record the environmental temperature, humidity and measurement time stamp of each measurement. According to the measured absorbance, the linear standard curve equation is fitted as follows:

[0075] Ca = k · A 标 + b

[0076] wherein C a is the apparent concentration, A 标 is the absorbance of the standard solution, k is the slope, and b is the intercept;

[0077] Step five, correction of the dynamic correction model

[0078] An equal volume of the sample solution is taken as the standard solution, and the color development procedure of the standard solution is followed. Then, the spectrophotometer is used to measure (with the same parameters as in step four), and the information is recorded. The absorbance value of the sample solution is substituted into the standard curve equation in step four to obtain the apparent concentration of the sample solution, which is input into the dynamic correction model to output the correction value. The corrected iron oxide concentration is obtained through output calculation, and the calculation formula of the output calculation is:

[0079] C 修 = C 样 + ΔC

[0080] wherein C 修 is the corrected iron oxide concentration, C 样 is the apparent concentration of the sample solution, and ΔC is the correction value obtained by the dynamic correction model. The corrected iron oxide concentration is combined with the mass of the ore sample to be measured to obtain the iron oxide content, and the measurement is completed. The calculation formula of the iron oxide content is:

[0081]

[0082] wherein V is the volume of the sample solution, and m is the sample mass. The present method is compared with the traditional spectrophotometric measurement method, and the brief steps of the traditional spectrophotometric measurement method are as follows:

[0083] A1: The sample to be measured is treated with dilute hydrochloric acid to form a transparent solution;

[0084] A2: Select the characteristic absorption wavelength of the component to be measured on the spectrophotometer;

[0085] A3: Use a blank solution (without the measured substance) to adjust the absorbance of the instrument to zero;

[0086] A4: Place the sample solution in the instrument, and the light source emits light of a specific wavelength through the sample. The detector measures the transmitted light intensity, and the instrument automatically calculates and displays the absorbance value (A);

[0087] A5: According to the absorbance value (A) of the sample, find or substitute into the formula on the pre-established standard curve (concentration vs. absorbance) to obtain the concentration of the component to be measured.

[0088] The comparison results are shown in Table 1 below:

[0089] Table I

[0090] Indicator Conventional spectrophotometric measurement method This embodiment Mean absolute error 0.82 ppm 0.28 ppm Root mean square error 1.15 ppm 0.35 ppm Error range when temperature and humidity fluctuate ± 1.5 ppm ± 0.4 ppm Error difference for different ore matrix ≤ 2.0 ppm ≤ 0.5 ppm

[0091] As shown in the above table, by introducing the dynamic correction model (XGBoost regression model), the method can correct the measurement results according to real-time environmental data and ore type information, significantly reduce the error, and the average absolute error is reduced to 0.28 ppm, improve the measurement accuracy, at the same time, can eliminate the error caused by temperature and humidity fluctuation and ore matrix difference, RMSE is significantly reduced to 0.35 ppm, provide more accurate measurement results, thus the present application not only performs excellently in improving the detection accuracy and stability, but also effectively solves the problems of temperature and humidity fluctuation, ore matrix difference and the like in the traditional method, provides an efficient and stable solution for the accurate detection of the content of oxidized iron in ore.

[0092] The basic principles, main features and advantages of the present application are shown and described above. It should be understood by those skilled in the art that the present application is not limited by the above examples, and the above examples and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the framework and scope of application of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for determining the content of oxidized iron in an ore by means of a spectrophotometer, characterized in that: The method comprises the following steps: Step one, pretreatment of ore sample Prepare the ore sample to be tested, assign a unique digital ID according to the type of the ore sample to be tested, record it, then grind the ore sample to be tested to obtain a ground ore sample, and then post-treat the ground ore sample to obtain a sample solution; Step two, construction and training of dynamic correction model A dynamic correction model is constructed based on an XGBoost regression model, and is trained based on historical data, with the input data being apparent concentration, environmental temperature, environmental humidity and sample information, and the output data being a correction value; Step three, preparation of standard solution High-purity iron trioxide powder is used as a standard sample, pretreated, dissolved with hydrochloric acid, and then diluted with deionized water to a predetermined volume to prepare a standard iron solution, and then ascorbic acid solution, buffer and o-diazenium phenothiazine color developing solution are sequentially added and mixed to obtain a color developing standard solution; Step four, establishment of spectrophotometer measurement standard Several groups of standard solutions with different concentrations are taken, measured by a spectrophotometer at a wavelength of 510 nm, and the environmental temperature, humidity and measurement timestamp of each measurement are recorded, and a linear standard curve equation is fitted according to the measured absorbance; Step five, correction of dynamic correction model An equal volume of sample solution is taken as the standard solution, and the color developing process of the standard solution is followed, then the sample solution is measured by a spectrophotometer, and the information is recorded, then the absorbance value of the sample solution is substituted into the standard curve equation in step four to obtain the apparent concentration of the sample solution, which is input into the dynamic correction model to output a correction value, and the corrected iron oxide concentration is calculated to obtain the iron oxide content combined with the mass of the ore sample to be tested, and the measurement is completed.

2. A method for determining the content of iron oxide in an ore using a spectrophotometer according to claim 1, characterized in that: In step one, the specific way of grinding treatment is: first, use a jaw crusher for coarse grinding to make the particle size of the ore sample to be tested ≤5mm, and then use a planetary ball mill for fine grinding to make the particle size of the ore sample to be tested ≤0.15mm.

3. A method for determining the content of iron oxide in an ore using a spectrophotometer according to claim 1, characterized in that: In step one, the specific way of post-treatment is: dissolve the ground ore sample with dilute hydrochloric acid at a volume ratio of 1:1, heat and stir the solution with a heating and stirring device while dissolving, the temperature is 80-90℃, stop heating when the transmittance is >90% and the transmittance change rate of the solution is less than 1% / min, filter the solution with a vacuum filtration method to obtain a sample solution, and finally dilute the sample solution to a predetermined volume with deionized water.

4. A method for determining the content of iron oxide in an ore using a spectrophotometer according to claim 1, characterized in that: In step two, the specific steps for training based on historical data are: S1: collect historical experimental data as a data set, which includes spectrophotometer measurement values, environmental temperature and humidity, sample information and known iron oxide content; S2: standardize the data in the data set, and then clean the data; S3: divide the data set into a training set and a validation set at a ratio of 7:3, train the XGBoost regression model using the training set, and evaluate the trained model using the validation set after training is completed; S4: Reuse five-fold cross-validation method to verify the stability of the model, thus obtaining the dynamic correction model.

5. A method for determining the content of iron oxide in an ore using a spectrophotometer according to claim 1, characterized in that: The specific way of the third step is: put high-purity iron trioxide powder into a muffle furnace, calcine at 550℃±25℃ for 50-60min, then cool to room temperature, add hydrochloric acid according to the ratio of 1g:20ml, and place in a constant temperature water bath at 80-90℃ to heat and stir until the solid is completely dissolved to obtain a standard iron solution.

6. A method for determining the content of iron oxide in an ore using a spectrophotometer according to claim 1, characterized in that: The specific coloration process of the coloration standard solution in the third step is: SS1: First, set the total capacity of the standard iron solution to be prepared, then take Vml standard iron solution, wherein V≥0; SS2: First, add 10% of the total capacity of ascorbic acid solution, stir evenly, and stand for 5min; SS3: Then add 20% of the total capacity of buffer solution, stir evenly; SS4: Then add 10% of the total capacity of phenanthroline coloration solution, shake vigorously; SS5: Use deionized water to make up to the total capacity, stand for 15min for coloration.

7. A method for determining the content of iron oxide in an ore using a spectrophotometer according to claim 1, characterized in that: The linear standard curve equation in the fourth step is: C a = k · A 标 + b where C is the concentration of the analyte, A is the absorbance of the sample, k is the slope, and b is the intercept. a where C is the concentration of the analyte, A is the absorbance of the sample, k is the slope, and b is the intercept. 标 where C is the concentration of the analyte, A is the absorbance of the sample 8. A method for determining the content of iron oxide in an ore using a spectrophotometer according to claim 1, characterized in that: The calculation formula of the output calculation in the fifth step is: C 修 = C 样 + ΔC where C is the concentration of iron oxide, C is the concentration of iron oxide after correction, C is the apparent concentration of the sample solution, and ΔC is the correction value obtained by the dynamic correction model. 修 where C is the concentration of iron oxide, C is the concentration of iron oxide after correction, C is the apparent concentration of the sample solution, and ΔC is the correction value obtained by the dynamic correction model. 样 where C is the concentration of iron oxide, C is the concentration of 9. A method for determining the content of iron oxide in an ore using a spectrophotometer according to claim 1, characterized in that: The calculation formula of the iron oxide content in the fifth step is: In the formula, V is the volume of the sample solution, and m is the sample mass.

Images