Method for judging whether element grade prediction result is abnormal or not based on LIBS

Abstract

The invention discloses a method for judging whether an element grade prediction result is abnormal or not based on LIBS. The method comprises the following steps: S1, obtaining average LIBS spectral data of a to-be-detected ore pulp sample; s2, respectively inputting the average LIBS spectral data of the ore pulp sample to be detected into the element grade prediction model of each selected element to obtain an element grade prediction result corresponding to each selected element; inputting the average LIBS spectral data of the to-be-detected ore pulp sample into a moisture content prediction model to obtain a moisture content prediction result of the to-be-detected ore pulp sample; and S3, judging whether the water content prediction result of the to-be-detected ore pulp sample is within a preset range or not, and if not, judging that the element grade prediction result corresponding to each selected element is abnormal. According to the method, whether the element grade prediction result is abnormal or not can be automatically, quickly and accurately judged.

Description

Technical Field

[0001] The present invention relates to the technical field of slurry element grade analysis, and more specifically to a method for judging whether an element grade prediction result is abnormal based on LIBS. Background Art

[0002] At present, element grade analysis of ore pulp is mostly done offline, that is, a batch of samples is manually intercepted at a designated sampling point. After a series of pre-treatment processes such as laboratory filtration, drying, and sample preparation, the ore pulp element grade is obtained by chemical titration or X-ray fluorescence spectrometry. The entire analysis process takes about an hour. If the number of batches is large, the analysis time will be longer, which leads to a serious lag in the acquisition of ore pulp element grade. There is no real-time guidance for the adjustment of the front-end flotation process and the addition of flotation reagents, resulting in poor quality stability of concentrate products in the production process, which has a great impact on the production cost and product quality of the enterprise.

[0003] Laser induced breakdown spectroscopy (LIBS) is a plasma emission spectroscopy analysis technology with the advantages of simultaneous measurement of multiple elements, simple sample pretreatment, fast analysis speed, and little damage to the sample. It is very suitable for full-element detection. Currently, this technology has been studied and applied in the fields of mining industry process monitoring, agriculture, food, environment and medicine.

[0004] One of the difficulties in analyzing the element grade of ore pulp using LIBS technology is that the intensity and stability of the LIBS spectral signal are easily affected by the water content in the ore pulp, which in turn affects the accuracy of the ore pulp element grade analysis.

[0005] Therefore, how to provide a method for automatically, quickly and accurately determining whether the element grade prediction result is abnormal is an urgent problem that technicians in this field need to solve. Summary of the invention

[0006] In view of this, the present invention provides a method for judging whether the element grade prediction result is abnormal based on LIBS, and the technical solution is as follows: A method for judging whether an element grade prediction result is abnormal based on LIBS comprises the following steps: S1: Obtain the average LIBS spectrum data of the slurry sample to be tested; S2: inputting the average LIBS spectrum data of the slurry sample to be tested into the element grade prediction model of each selected element respectively to obtain the element grade prediction result corresponding to each selected element; The average LIBS spectrum data of the ore pulp sample to be tested is input into the moisture content prediction model to obtain the moisture content prediction result of the ore pulp sample to be tested; S3: Determine whether the moisture content prediction result of the slurry sample to be tested is within a preset range: if the moisture content prediction result is within the preset range, the element grade prediction result is determined to be normal; if the moisture content prediction result is not within the preset range, the element grade prediction result is determined to be abnormal; The preset range is 65~80%.

[0007] The method for obtaining the element grade prediction model is: S21: Obtain average LIBS spectrum data of slurry samples; Obtain measured elemental grades of selected elements in slurry samples; S22: using the average LIBS spectrum data in S21 as the input of the partial least squares method; using the measured element grade in S21 as the output of the partial least squares method; and obtaining an element grade prediction model for the selected element.

[0008] The method for obtaining the moisture content prediction model is: S21': obtain the average LIBS spectrum data of the pulp sample; Obtain the measured moisture content of the slurry sample; S22': using the average LIBS spectrum data in S21' as the input of the partial least squares method; using the measured moisture content in S21' as the output of the partial least squares method; and obtaining the moisture content prediction model.

[0009] The method for obtaining the average LIBS spectrum data is: (1) Use the slurry sampler to intercept the slurry in the main slurry pipe; (2) Using a multi-channel reducer to reduce the intercepted slurry; (3) The slurry after reduction is input into the LIBS analyzer to obtain the average LIBS spectrum data.

[0010] The method for obtaining the measured element grade is: A pulp sample is intercepted from the sample port of the multi-channel reducer; the intercepted pulp sample is filtered and dried to prepare a sample into a mineral powder; and the measured element grade of the selected element in the pulp sample is obtained by chemical titration or X-ray fluorescence spectrometry.

[0011] The method for obtaining the measured moisture content is: A slurry sample is intercepted from the sample port of the multi-channel reducer; the intercepted slurry sample is weighed to obtain a weighing value A; the slurry sample is filtered and dried to prepare a mineral powder, and the mineral powder is weighed to obtain a weighing value B; based on the weighing value, the measured moisture content of the slurry sample is calculated, and the measured moisture content = (AB) / A.

[0012] The average LIBS spectrum data includes the characteristic spectral line intensities of all elements in the ore pulp sample to be tested.

[0013] The beneficial effects of the present invention are: The present invention provides a method for judging whether an element grade prediction result is abnormal based on LIBS. By establishing an element grade prediction model and a moisture content prediction model, the element grade prediction result and the moisture content prediction result are obtained according to LIBS spectral data, and whether the element grade prediction result is abnormal is judged according to the moisture content prediction result. Whether the element grade prediction result is abnormal can be judged automatically, quickly and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Schematic diagram of the process of the present invention; Figure 2 A diagram showing the relationship between the training data, validation data, and prediction data for phosphorus in slurry samples. DETAILED DESCRIPTION

[0015] The embodiments of the present invention will be described in detail below with reference to examples. The following examples are only used to illustrate the present invention and should not be construed as limiting the scope of the present invention.

[0016] Example 1 Prediction model building: 1) Select magnesium concentrate slurry with a moisture content of 65% to 80%, and use a magnesium concentrate sampler to intercept the slurry every two hours; Specifically: the present invention obtains a total of 560 magnesium concentrate slurry samples; 2) The slurry samples intercepted each time are input to the multi-channel divider through a pipeline (the multi-channel divider enables a part of the slurry to be input to the LIBS analyzer). After being divided, they flow into the LIBS analyzer for spectral measurement, and the average LIBS spectral data of the slurry samples are obtained and saved; Specifically: In actual measurement, each slurry sample is excited with 1000 pulses to obtain 1000 LIBS spectral data. The first 200 LIBS spectral data with large volatility are removed, and then the average value of the remaining 800 LIBS spectral data is calculated; this average value is the average LIBS spectral data of the slurry sample. The reason for taking the average value after multiple measurements is that the particles in the slurry are uneven, which leads to large volatility in the obtained spectrum. Only by taking the average value after multiple measurements can part of the measurement error be eliminated, making the spectral data more representative.

[0017] While the LIBS analyzer is measuring, a slurry sample is obtained at the sample port of the multi-channel reducer; the slurry sample is filtered and dried to prepare ore powder; Measured elemental grades of four selected elements, namely phosphorus, magnesium, iron and aluminum, are obtained by chemical titration or X-ray fluorescence spectrometry; Then, weigh the mineral powder before and after filtration, calculate the difference between the weight B of the mineral powder after drying and the weight A of the mineral powder before filtration, and obtain the measured moisture content of the slurry sample = (AB) / A; Based on this, the present invention obtained a total of 560 average LIBS spectral data, 560 measured elemental grades of phosphorus, 560 measured elemental grades of magnesium, 560 measured elemental grades of iron, 560 measured elemental grades of aluminum, and 560 measured moisture contents; 3) The first 500 average LIBS spectral data, 500 measured elemental grades of phosphorus, 500 measured elemental grades of magnesium, 500 measured elemental grades of iron, 500 measured elemental grades of aluminum and 500 measured water contents were used as training data.

[0018] The first 500 average LIBS spectral data were used as the input of the partial least squares method, and the corresponding first 500 measured elemental grades of phosphorus were used as the output of the partial least squares method to obtain the elemental grade prediction model of phosphorus; The first 500 average LIBS spectral data were used as the input of the partial least squares method, and the corresponding first 500 measured elemental grades of magnesium were used as the output of the partial least squares method to obtain the elemental grade prediction model of magnesium. The first 500 average LIBS spectral data were used as the input of the partial least squares method, and the corresponding first 500 measured elemental grades of iron were used as the output of the partial least squares method to obtain the elemental grade prediction model of iron; The first 500 average LIBS spectral data were used as the input of the partial least squares method, and the corresponding first 500 measured elemental grades of aluminum were used as the output of the partial least squares method to obtain the elemental grade prediction model of aluminum. The first 500 average LIBS spectral data were used as the input of the partial least squares method, and the corresponding first 500 measured moisture contents were used as the output of the partial least squares method to obtain the moisture content prediction model.

[0019] Example 2 Prediction model validation: On the basis of Example 1, the remaining 60 average LIBS spectral data, the remaining 60 measured elemental grades of magnesium, the remaining 60 measured elemental grades of iron, the remaining 60 measured elemental grades of iron and the remaining 60 measured water contents were used as verification data and prediction data to verify and predict the above five models; Specifically: the remaining 60 average LIBS spectral data are input into the above 5 prediction models to obtain prediction data; the remaining 60 measured element grades of phosphorus, the remaining 60 measured element grades of magnesium, the remaining 60 measured element grades of iron, the remaining 60 measured element grades of aluminum and the remaining 60 measured moisture contents are verification data.

[0020] Figure 2 is the relevant data of the element grade prediction model of phosphorus element. 2 The value of P2P is 0.984 and the value of RMSEP is 0.465.

[0021] In addition, the R of the moisture content prediction model 2 The R value of the element grade prediction model of magnesium is 0.988. 2 The R value of the element grade prediction model of iron is 0.983. 2 The R value of the element grade prediction model of aluminum is 0.982. 2 It is 0.983.

[0022] Example 3 Formal production application: 1) Importing the five prediction models of Example 1 into the LIBS analyzer; 2) using a magnesium concentrate sampler to obtain a magnesium concentrate slurry sample to be tested, and inputting the sample into a phosphorus element grade prediction model, a magnesium element grade prediction model, an iron element grade prediction model, an aluminum element grade prediction model and a moisture content prediction model, respectively, to obtain the phosphorus element grade, the magnesium element grade, the iron element grade, the aluminum element grade and the moisture content of the magnesium concentrate slurry; 3) Determine whether the moisture content of magnesium concentrate slurry is between 65% and 80%: If the moisture content is not within the range of 65% to 80%, the predicted results of the elemental grades of phosphorus, magnesium, iron and aluminum are judged to be abnormal, and the abnormal information is uploaded to the data center, prompting the flotation workshop to pay attention to the slurry concentration and adjust the process in time.

[0023] If the water content is within the range of 65% to 80%, it is judged that the predicted results of the elemental grade of phosphorus, the elemental grade of magnesium, the elemental grade of iron, and the elemental grade of aluminum are normal.

Claims

1. A method for judging whether an element grade prediction result is abnormal based on LIBS, characterized in that: The following steps are involved: S1: Obtain the average LIBS spectrum data of the slurry sample to be tested; S2: inputting the average LIBS spectrum data of the slurry sample to be tested into the element grade prediction model of each selected element respectively to obtain the element grade prediction result corresponding to each selected element; The average LIBS spectrum data of the ore pulp sample to be tested is input into the moisture content prediction model to obtain the moisture content prediction result of the ore pulp sample to be tested; S3: Determine whether the predicted result of the moisture content of the slurry sample to be tested is within a preset range: if the predicted result of the moisture content is within the preset range, then determine that the predicted result of the element grade is normal; If the water content prediction result is not within the preset range, it is determined that the element grade prediction result is abnormal.

2. The method for judging whether the element grade prediction result is abnormal based on LIBS according to claim 1, characterized in that: The method for obtaining the element grade prediction model is: S21: Obtain average LIBS spectrum data of slurry samples; Obtain measured elemental grades of selected elements in slurry samples; S22: using the average LIBS spectrum data in S21 as the input of the partial least squares method; using the measured element grade in S21 as the output of the partial least squares method; and obtaining an element grade prediction model for the selected element.

3. The method for judging whether the element grade prediction result is abnormal based on LIBS according to claim 2, characterized in that: The method for obtaining the moisture content prediction model is: S21': obtain the average LIBS spectrum data of the pulp sample; Obtain the measured moisture content of the slurry sample; S22': using the average LIBS spectrum data in S21' as the input of the partial least squares method; using the measured moisture content in S21' as the output of the partial least squares method; and obtaining the moisture content prediction model.

4. The method for judging whether the element grade prediction result is abnormal based on LIBS according to claim 3, characterized in that: The method for obtaining the average LIBS spectrum data is: (1) Use the slurry sampler to intercept the slurry in the main slurry pipe; (2) Using a multi-channel reducer to reduce the intercepted slurry; (3) The slurry after reduction is input into the LIBS analyzer to obtain the average LIBS spectrum data.

5. The method for judging whether the element grade prediction result is abnormal based on LIBS according to claim 4, characterized in that: The method for obtaining the measured element grade is: A pulp sample is intercepted from the sample port of the multi-channel reducer; the intercepted pulp sample is filtered and dried to prepare a sample into a mineral powder; and the measured element grade of the selected element in the pulp sample is obtained by chemical titration or X-ray fluorescence spectrometry.

6. The method for judging whether the element grade prediction result is abnormal based on LIBS according to claim 5, characterized in that: The method for obtaining the measured moisture content is: A slurry sample is intercepted from the sample port of the multi-channel reducer; the intercepted slurry sample is weighed to obtain a weighing value A; the slurry sample is filtered and dried to prepare a mineral powder, and the mineral powder is weighed to obtain a weighing value B; based on the weighing value, the measured moisture content of the slurry sample is calculated, and the measured moisture content = (AB) / A.

7. The method for judging whether the element grade prediction result is abnormal based on LIBS according to claim 4, characterized in that: The average LIBS spectrum data includes the characteristic spectral line intensities of all elements in the ore pulp sample to be tested.

8. The method for judging whether the element grade prediction result is abnormal based on LIBS according to claim 1, characterized in that: The preset range is 65~80%.

Images