Method for on-line prediction of iron content in copper flash smelting copper slag and model construction

Abstract

The invention relates to the technical field of non-ferrous metal smelting, in particular to a method for predicting the iron content in copper flash smelting copper slag on line and model construction, starting from the actual incidence relation of copper slag and production input, model training-verification-testing is carried out through input data and output data of historical copper slag production, and the iron content in the copper flash smelting copper slag is predicted. And the input data of the copper flash smelting copper slag in a certain time period are obtained in real time and serve as model input to be substituted into the prediction model for on-line operation, the predicted value of the iron content in the copper slag discharged at the corresponding time point is obtained in real time, the problem that input data and output data of the prediction model constructed in the prior art are not matched is solved, and the prediction accuracy is improved. And online accurate prediction of the iron content in the copper flash smelting copper slag is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of nonferrous metal smelting, and in particular to an online method for predicting the iron content in copper slag from copper flash smelting and model construction. Background Art

[0002] Copper flash smelting is a process that combines roasting and smelting in one device to carry out copper smelting production. It is the most widely used sulfide ore copper smelting method in the existing technology. Copper slag is the main product discharged from the copper flash smelting process. The ratio of the iron content in the copper slag to the silicon dioxide in the copper slag is an important factor affecting the copper content in the copper slag. During production time, it is necessary to adjust the copper flash smelting process by monitoring the real-time value of the iron content in the copper slag to minimize the copper content in the copper slag. In the existing technology, the online iron content detection equipment in copper slag is immature and cannot achieve real-time prediction or detection of the iron content in the copper slag. Therefore, the use of predictive model judgment has become a feasible solution currently selected.

[0003] Currently, research has been conducted on predictive models to monitor the content of certain chemical components in chemical products.

[0004] For example: Patent No. 202410489618.2 discloses collecting actual production data (including mineral unit oxygen consumption, solvent rate, oxygen enrichment concentration, copper content, sulfur content, iron content, silica content and magnesium oxide content) during the stable production cycle of a copper smelter, normalizing the collected data, using 80% of the processed data as training data and 20% as test data, and then improving the genetic algorithm GA based on the existing GA-BP model to obtain an improved genetic algorithm GA. The specific improvement is to add an adaptive crossover probability adjustment mechanism to the genetic algorithm GA. and an adaptive mutation probability adjustment mechanism, wherein the adaptive crossover probability adjustment mechanism is used to adjust the population update speed in the early genetic stage and the stability in the late genetic stage of the genetic algorithm GA, and the adaptive mutation probability adjustment mechanism is used to improve the population diversity; the BP neural network model is optimized by the improved genetic algorithm GA, and the actual production data of the mineral unit oxygen consumption, solvent rate, oxygen enrichment concentration, copper content, sulfur content, iron content, silica content and magnesium oxide content are used as inputs of the BP neural network model to finally obtain the copper matte grade value; and the prediction accuracy of the improved GA-BP model is verified.

[0005] Another example: Patent application number 202411808546.X discloses an improved genetic algorithm core mechanism to enhance optimization efficiency and model prediction accuracy. The method includes collecting actual production data during the stable production cycle of a copper smelter; improving the existing GA-BP neural network by adding an adaptive crossover probability adjustment mechanism and an adaptive mutation probability adjustment mechanism to the genetic algorithm. The adaptive crossover probability adjustment mechanism not only relies on individual fitness, but also combines global and local information to enhance global search capabilities and local optimization capabilities. The adaptive mutation probability adjustment mechanism not only combines individual fitness, but also introduces global entropy and dynamic noise effects, while allowing the mutation probability to increase randomly in the later stages of evolution to escape the local optimum; using the improved GA-BP neural network, the actual production data is used as input to predict the grade of oxygen-enriched bottom-blown copper matte and output the copper matte grade value.

[0006] However, after searching and analyzing, no technical literature was found that used a prediction model to predict the iron content in copper slag from copper flash smelting in real time. Moreover, in the existing technology, statistical and machine learning algorithms were used, and the product quality index inspection data collected during the actual production process were used as the model output annotation value, and the historical production input data synchronized with the output product were used as the model input data to jointly construct the model training and test data sets. Through model training and testing, a prediction model for the quality index of the product to be tested was obtained. However, the copper flash smelting process has the following conditions: (1) The flash smelting pool is long, and the copper discharge port at the tail of the flash smelting pool is more than ten meters away from the flash smelting tower molten pool area, resulting in a long time delay from feeding to the discharge of copper slag; (2) There are complex physical and chemical reactions and heat transfer and mass transfer processes in the copper flash smelting pool, so that the iron content in the copper slag discharged at a certain point in time is actually the result of the combined effect of all feeding and production conditions in a long period of time before that point in time. However, the prediction model in the existing technology can only take into account the production conditions synchronized with the discharge time of the predicted product and the impact of the feeding on the content, resulting in poor accuracy in the prediction results of the iron content in copper slag. Summary of the Invention

[0007] Based on the above technical problems, the present invention provides a method and model construction for online prediction of the iron content in copper slag from copper flash smelting, which takes into account the feeding and production conditions during the period from feeding to the residence time in the flash smelting pool before the time point of copper slag production, so as to construct an online prediction model for the iron content in copper slag, solve the technical problem of mismatch between input data and output data in the existing technology, and thus improve the accuracy of the prediction of the iron content in copper slag from copper flash smelting.

[0008] The specific technical solutions are:

[0009] One of the purposes of the present invention is to provide a method for online prediction of the iron content in copper flash smelting slag, comprising the following steps:

[0010] S1: Set the data sampling time points of the copper flash smelting production process at equal intervals n;

[0011] S2: Collect input data according to the data sampling time, and collect and calculate the average residence time A from feeding to discharging copper slag according to the following formula:

[0012] A=(S×H)÷(Q×D)

[0013] Among them, S is the effective area of the flash smelting pool, unit is m2; H is the average thickness of the copper slag layer in the flash smelting pool, unit is m; Q is the average amount of copper concentrate fed per unit time, unit is m 3 / h; D is the average copper slag yield, in %;

[0014] S3: Collecting copper slag output data at the data sampling time point, and constructing an input time period by tracing back A from the data sampling time point, and intercepting the input data within the input time period; marking and matching the input data and the copper slag output data one by one into a data set;

[0015] S4: Divide the dataset into a training dataset: validation dataset: test dataset ratio of 7:2:1, construct a primary model using an artificial intelligence algorithm, use the dataset for model training, validation, and testing, and select the optimal model T for predicting the iron content in copper slag based on the test results;

[0016] S5: real-time acquisition of input data and copper slag discharge time point to input into the copper slag iron content prediction model T, and the model calculation to obtain the predicted value of the copper slag iron content at the copper slag discharge time point.

[0017] The input data created by the present invention includes but is not limited to feeding data and equipment operation data; the feeding data includes but is not limited to feeding copper content, feeding sulfur content, feeding iron content and feeding silica content; the equipment operation data includes but is not limited to feeding amount per unit time, air flow rate and oxygen flow rate; the copper slag output data is the iron content detection data in the copper slag.

[0018] The feed rate per unit time of the present invention includes but is not limited to the feed rate of copper concentrate, the feed rate of slag forming agent and the feed rate of fly ash; the air flow rate includes but is not limited to the process air flow rate and the dispersed air flow rate; the oxygen flow rate includes but is not limited to the central oxygen flow rate, the regulating oxygen flow rate and the secondary oxygen lance oxygen flow rate.

[0019] The present invention creates the step S5, deploying the prediction model T for the iron content in the copper slag in a computer, which is connected to the batching and control system of the copper flash smelting pool and can read the feeding data of the copper flash smelting process.

[0020] The artificial intelligence algorithm created by the present invention is capable of supporting vector machines, logistic regression, random forests and / or deep learning.

[0021] A second object of the present invention is to provide a method for constructing a model for online prediction of the iron content in copper flash smelting slag, comprising the following steps:

[0022] (1) Determine the prediction target: the iron content in copper flash smelting slag;

[0023] (2) Setting model evaluation criteria: using mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) data to evaluate model prediction accuracy;

[0024] (3) Data collection and cleaning: Collect input data and copper slag output data; process missing data, abnormal data, and data standardization;

[0025] (4) Through feature selection or feature extraction, the original data is converted into valid input and the corresponding prediction target data is constructed into a data set;

[0026] (5) Select several artificial intelligence regression algorithms to construct primary models respectively, and then select the best one based on the actual prediction effect of each primary model;

[0027] (6) The data set is divided into a training set, a validation set, and a test set. The training set is used for model training, the validation set is used for model parameter adjustment and optimization, and the test set is used to test the model prediction effect.

[0028] The present invention creates a method for converting the original data into an effective input:

[0029] a. Set the sampling time point of the copper slag as t, then trace back an average residence time A from time t, construct an input time period from tA to time t, and intercept the input data in the input time period to form a set of input data; at the set sampling time point t, sample the discharged copper slag, number it, and send it for inspection, and record the iron content in the sampled copper slag to construct a copper slag output data;

[0030] b. Assemble the model input and output data, set the number of input data to E, and calculate the number of input data collection times L corresponding to a piece of copper slag output data according to L = A ÷ n; suppose that the kth sampling of the output copper slag is carried out at the sampling time point t, and the iron content in the copper slag is analyzed and the copper slag output data is Q k , which corresponds to a set of input data I k i,j(i=1,2,3……E;j=1,2,3……L);the input data collection times corresponding to j=1,2,3……L are tA, t-A+n, t-A+2n,……,t respectively;

[0031] c. Record the input data and copper slag output data collected in step b and organize them into input data I k 1,1 , I k 1,2 , I k 1,3 ...I k 1,E , I k 2,1 , I k 2,2 , I k 2,3 ...I k 2,E , I k 3,1 , I k 3,2 , I k 3,3 ...I k 3,E ,……,I k L,1 , I k L,2 , I k L,3 ...I k L,E and copper slag output data Q k One-to-one corresponding data sets.

[0032] Compared with the prior art, the technical effects created by the present invention are embodied in:

[0033] The present invention utilizes an artificial intelligence algorithm to construct a prediction model, uses historical data of a copper flash smelting production process to perform model training, verification, and testing, and obtains a prediction model for the iron content in copper slag from copper flash smelting. The prediction model is deployed in a computer, and input data within a certain period of time is obtained in real time by the computer. The data is substituted into the prediction model for the iron content in copper slag from copper flash smelting as model input for online calculation. Based on the actual correlation between the copper slag and the production input, the predicted value of the iron content in the copper slag released at the corresponding time point is predicted in real time, thereby solving the problem of mismatch between input data and output data in the prediction model constructed in the prior art and realizing accurate online prediction of the iron content in copper slag from copper flash smelting. DETAILED DESCRIPTION

[0034] In order to facilitate those skilled in the art to correctly understand the present invention and enable those skilled in the art to fully understand the technical content of the present invention, the technical solution of the present invention is further explained below in conjunction with specific implementation methods, but this explanation does not limit the scope of protection required for the present invention. The scope of protection of the present invention by those skilled in the art cannot be limited to the following explanations. Any equivalent replacement or change made by any those skilled in the art or persons familiar with the technology in this field, on the basis of the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0035] Example

[0036] The method for online prediction of iron content in copper flash smelting slag includes:

[0037] The feed copper concentrate samples were taken from the feed belt every 5 minutes and sent to the testing laboratory for testing of feed composition. The copper content, sulfur content, iron content, silica content and other impurity contents in the copper concentrate fed during this period were recorded to obtain the feeding data (as shown in Table 1).

[0038] Table 1 Sample table for collecting and recording feeding data

[0039]

[0040]

[0041] At the same time period as the sampling of the feed copper concentrate, the equipment operating data such as the feed amount per unit time (total amount of copper concentrate, slagging agent and fly ash), air flow (total flow of process air and dispersed air) and oxygen flow (total flow of central oxygen, regulating oxygen and secondary oxygen lance oxygen) are read from the copper flash smelting pool control system (as shown in the sample table in Table 2).

[0042] Table 2 Sample table for recording data of equipment operation

[0043] Sampling time Feeding amount, t / hour <![CDATA[Air flow rate, m 3 / hour]]> <![CDATA[Oxygen flow rate, m 3 / hour]]> 01:00 136.536 5523 15528 01:05 135.326 5548 15206 ... ... ... ... 15:00 135.843 5462 15439 15:05 135.408 5516 15302 ... ... ... ...

[0044] The feeding data and equipment operation data collected in Tables 1 and 2 are sorted according to the sampling time to obtain the input data (as shown in the sample table in Table 3).

[0045] Table 3

[0046]

[0047] When the copper slag is discharged, samples are taken, numbered, and sent for inspection every 5 minutes to determine the iron content in the copper slag, and the output data of the copper flash smelting copper slag is obtained (as shown in Table 4 for samples).

[0048] Table 4 Sample record of copper slag output data

[0049] Copper slag sampling batch Copper slag sampling time Iron content in copper slag, % 001 15:00 57.79 002 15:05 58.15 ... ... ...

[0050] The effective area S of the copper flash smelting pool was collected and tested to be 150 m2, the average thickness H of the copper slag in the pool was 1.82 m, and the average feed volume Q per unit time was 30 m 3 / h, the average copper slag yield C is 65%, and the average residence time of the feed components in the copper flash smelting pool from feeding to discharging copper slag is 14h according to the formula of average residence time A = (S×H) / (Q×D).

[0051] For each copper slag batch, the starting time point is obtained by tracing back the average residence time of the feed (the total amount of copper concentrate, slag-making agent, and fly ash) in the copper flash smelting pool from the time of copper slag discharge and sampling. From the starting time point to the time of copper slag sampling, data records (A ÷ n) = 14h ÷ 5min = (14 × 60) ÷ 5 = 168 are intercepted from the input data (as shown in the sample table of Table 3) to construct a set of input data. From the copper flash smelting copper slag output data (as shown in the sample of Table 4), a copper slag grade data corresponding to the copper slag batch is selected as a copper slag output data. The input data and copper slag output data are matched one by one to construct a data set of copper flash smelting copper slag, and the data are recorded (as shown in the sample table of Table 5).

[0052] Table 5 Data set of copper flash smelting slag - sample record table

[0053]

[0054] Note: In Table 5, E is 7; the sampling batch is k; j is the number of samplings within the same sampling batch; i is the number of the corresponding input factor E, such as copper content is 1, sulfur content is 2, iron content is 3, silica content is 4, feed amount is 5, air flow rate is 6, and oxygen flow rate is 7.

[0055] The dataset of copper flash smelting slag was divided into three sub-datasets: training set, validation set and test set according to the ratio of 7:2:1; artificial intelligence machines such as support vector, random forest and deep learning were used for modeling algorithms respectively, the training set was used for model training, the validation set was used to verify the trained model and adjust the parameters, and the test set was used to test the model after verification and parameter adjustment. According to the test results, the random forest model was selected as the prediction model for iron content in copper flash smelting slag.

[0056] The prediction model for the iron content in copper slag from copper flash smelting is deployed in a computer at the copper flash smelting copper slag production site, and the computer is connected to the copper flash smelting pool batching system and control system through an intranet. The computer can read the copper concentrate in the batching system, including but not limited to the copper content, sulfur content, iron content, silica content and other impurity contents; at the same time, the computer can read the control system, including but not limited to the feed amount per unit time (copper concentrate feed amount, slagging agent feed amount and fly ash feed amount), air flow (process air flow and dispersed air flow), oxygen flow (central oxygen flow, regulating oxygen flow and secondary oxygen lance oxygen flow); and can trace back an average residence time A (17 hours) from the current time point, and intercept all sampled input data in the time period from t-17 hours to t, as model input and substituted into the prediction model for the iron content in copper slag from copper flash smelting. The model operation outputs the result, that is, the predicted value of the iron content in the copper slag released at the time point t.

[0057] This prediction method starts from the actual correlation between copper slag and production inputs, uses the input data and output data of historical copper slag production to conduct model training, verification and testing, and then obtains the copper slag input data of copper flash smelting within a certain time period in real time. It uses this data as model input and substitutes it into the prediction model for online calculation, and obtains the predicted value of the iron content in the copper slag released at the corresponding time point in real time. This solves the problem of mismatch between the input data and output data of the prediction model constructed by the existing technology, and realizes the online and accurate prediction of the iron content in copper slag from copper flash smelting.

[0058] Tables 1 to 5 in the present invention are sample tables, which are intended to accurately express the process of constructing the prediction model and processing the data to form a data set in the present invention, so that those skilled in the art can make an accurate understanding. Moreover, the actual model training requires a large amount of data, and the data set is divided by random sampling. It is neither necessary nor meaningful to list all of them in the table. Therefore, ellipsis is used in the table instead.

[0059] In addition, the data set division ratio involved in the invention is a reference. Those skilled in the art may adopt other data set division ratios, which may be specifically adjusted according to the common sense basic requirements of those skilled in the art and artificial intelligence algorithm model training.

Claims

1. A method for online prediction of iron content in copper flash smelting slag, characterized in that: The following steps are involved: S1: Set the data sampling time points of the copper flash smelting production process at equal intervals n; S2: Collect input data at the data sampling time points, and collect and calculate the average residence time A from feeding to discharging copper slag according to the following formula: A=(S×H)÷(Q×D) Among them, S is the effective area of the flash smelting pool, unit is m2; H is the average thickness of the copper slag layer in the flash smelting pool, unit is m; Q is the average amount of copper concentrate fed per unit time, unit is m 3 / h; D is the average copper slag yield, in %; S3: Collecting copper slag output data at the data sampling time point, and constructing an input time period by tracing back A from the data sampling time point, and intercepting the input data within the input time period; The input data and copper slag output data are marked and matched one by one to form a data set; S4: Divide the dataset into a training dataset: validation dataset: test dataset ratio of 7:2:1, construct a primary model using an artificial intelligence algorithm, use the dataset for model training, validation, and testing, and select the optimal model T for predicting the iron content in copper slag based on the test results; S5: real-time acquisition of input data and copper slag discharge time point to input into the copper slag iron content prediction model T, and obtain the predicted value of the copper slag iron content at the copper slag discharge time point through model calculation.

2. The method for online prediction of iron content in copper flash smelting slag according to claim 1, characterized in that: The input data includes feeding data and equipment operation data; the feeding data includes feeding copper content, feeding sulfur content, feeding iron content and feeding silica content; the equipment operation data includes feeding amount per unit time, wind flow rate and oxygen flow rate; the copper slag output data is the iron content detection data in the copper slag.

3. The method for online prediction of iron content in copper flash smelting slag according to claim 2, characterized in that: The feed rate per unit time includes the copper concentrate feed rate, the slag forming agent feed rate and the fly ash feed rate; the air flow rate includes the process air flow rate and the dispersed air flow rate; the oxygen flow rate includes the central oxygen flow rate, the regulating oxygen flow rate and the secondary oxygen lance oxygen flow rate.

4. The method for online prediction of iron content in copper flash smelting slag according to claim 1, characterized in that: In step S5, the prediction model T for the iron content in the copper slag is deployed in a computer, which is connected to the batching and control system of the copper flash smelting pool and can read the feeding data of the copper flash smelting process.

5. The method for online prediction of iron content in copper flash smelting slag according to claim 1, characterized in that: The artificial intelligence algorithm can be support vector machine, random forest and / or deep learning.

6. The method for online prediction of iron content in copper flash smelting slag according to claim 1, characterized in that: The step S4, the steps of constructing a model using an artificial intelligence logistic regression algorithm are as follows: (1) Determine the prediction target: the iron content in copper flash smelting slag; (2) Setting model evaluation criteria; (3) Data collection and cleaning: Collect input data and copper slag output data; process missing data, abnormal data, and data standardization; (4) Through feature selection or feature extraction, the original data is converted into valid input and the corresponding prediction target data is constructed into a data set; (5) Select several artificial intelligence logistic regression algorithms to construct primary models respectively, and then select the best one based on the actual prediction effect of each primary model; (6) The data set is divided into a training set, a validation set, and a test set. The training set is used for model training, the validation set is used for model parameter adjustment and optimization, and the test set is used to test the model prediction effect.

7. The method for online prediction of iron content in copper flash smelting slag according to claim 6, characterized in that: The original data is converted into a valid input method: a. Set the sampling time point of the copper slag as t, then trace back an average residence time A from time t, construct an input time period from tA to time t, and intercept the input data in the input time period to form a set of input data; at the set sampling time point t, sample the discharged copper slag, number it, and send it for inspection, and record the iron content in the sampled copper slag to construct a copper slag output data; b. Assemble the model input and output data, set the number of input data to E, and calculate the number of input data collection times L corresponding to a piece of copper slag output data according to L = A ÷ n; suppose that the kth sampling of the output copper slag is carried out at the sampling time point t, and the iron content in the copper slag is analyzed and the copper slag output data is Q k , which corresponds to a set of input data I k i,j (i=1,2,3……E;j=1,2,3……L);the input data collection times corresponding to j=1,2,3……L are tA, t-A+n, t-A+2n,……,t respectively; c. Record the input data and copper slag output data collected in step b and organize them into input data I k 1,1 , I k 1,2 , I k 1,3 ...I k 1,E , I k 2,1 , I k 2,2 , I k 2,3 ...I k 2,E , I k 3,1 , I k 3,2 , I k 3,3 ...I k 3,E ,……,I k L,1 , I k L,2 , I k L,3 ...I k L,E and copper slag output data Q k One-to-one corresponding data sets.

8. The method for online prediction of iron content in copper flash smelting slag according to claim 6, characterized in that: In step (2), mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) data are used to evaluate the model prediction accuracy.

9. A method for constructing a model for online prediction of the iron content in copper flash smelting slag, characterized in that: The following steps are involved: (1) Determine the prediction target: the iron content in copper flash smelting slag; (2) Setting model evaluation criteria: using mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) data to evaluate model prediction accuracy; (3) Data collection and cleaning: Collect input data and copper slag output data; process missing data, abnormal data, and data standardization; (4) Through feature selection or feature extraction, the original data is converted into valid input and the corresponding prediction target data is constructed into a data set; (5) Select several artificial intelligence logistic regression algorithms to construct primary models respectively, and then select the best one based on the actual prediction effect of each primary model; (6) The data set is divided into a training set, a validation set, and a test set. The training set is used for model training, the validation set is used for model parameter adjustment and optimization, and the test set is used to test the model prediction effect.

10. The model construction method for online prediction of iron content in copper flash smelting slag according to claim 9, characterized in that: The original data is converted into a valid input method: a. Set the sampling time point of the copper slag as t, then trace back an average residence time A from time t, construct an input time period from tA to time t, and intercept the input data in the input time period to form a set of input data; at the set sampling time point t, sample the discharged copper slag, number it, and send it for inspection, and record the iron content in the sampled copper slag to construct a copper slag output data; b. Assemble the model input and output data, set the number of input data to E, and calculate the number of input data collection times L corresponding to a piece of copper slag output data according to L = A ÷ n; suppose that the kth sampling of the output copper slag is carried out at the sampling time point t, and the iron content in the copper slag is analyzed and the copper slag output data is Q k , which corresponds to a set of input data I k i,j (i=1,2,3……E;j=1,2,3……L);the input data collection times corresponding to j=1,2,3……L are tA, t-A+n, t-A+2n,……,t respectively; c. Record the input data and copper slag output data collected in step b and organize them into input data I k 1,1 , I k 1,2 , I k 1,3 ...I k 1,E , I k 2,1 , I k 2,2 , I k 2,3 ...I k 2,E , I k 3,1 , I k 3,2 , I k 3,3 ...I k 3,E ,……,I k L,1 , I k L,2 , I k L,3 ...I k L,E and copper slag output data Q k One-to-one corresponding data sets.

Images