Development of a human-machine interface for cascade control mills in obtaining nepheline ore charge

Alumina (Al 2 O 3 ) is the main source material used in the production of primary aluminum, which, in turn, is the main material for space engineering, both in the production of vehicle parts and in the organization of power supply and electronics. In most cases, bauxite acts as an ore base when creating alumina, but nepheline and other ores containing alumina are also used. One of the most common methods of alumina production is the alkaline method by sintering. The essence of this method consists in binding Al 2 O 3 to sodium aluminate by sintering and subsequent leaching of the resulting sinter [1-7]. This method of alumina production is used at the Achinsk Alumina Plant (AGP), the largest enterprise in the aluminum industry in Russia [8].

The preparation of raw materials for sintering is handled by the charge preparation department. The process of preparing the charge of nepheline ore is one of the most important stages of obtaining alumina, and with the improvement of the control scheme, a significant economic effect can be achieved by reducing energy and material costs [9; 10].

This effect can be achieved by solving some problems of the automated control system of technological process (ACS TP):

• –    lack of automatic monitoring and indication of the operation of the main technological equipment (on/off mode, emergency status indication);

• –    lack of output to the ACS TP of important parameters of operational control of equipment operation (bearing temperature, oil pressure, etc.);

• –    monitoring of parameters only in place;

• –    the absence of an automatic control system for lime and alkali modules of the raw charge;

• –    implementation of the output to the target parameters of the charge in manual mode based on the results of chemical laboratory analysis.

Description of the technological process

The charge preparation department is designed for the preparation of a charge from nepheline ore and limestone with specified characteristics. The composition of the charge is characterized by:

• –    limestone modulus – the molar ratio of the sum of calcium oxide to silicon dioxide ( M izv = 1.07 - (CaO/SiO 2 ));

• –    alkaline modulus – molar ratio of the sum of sodium and potassium oxides in terms of Na 2 O to aluminum oxide ( M gap = 1.645 -[ (Na 2 O + 0.659K 2 O) / Al 2 O 3 ] );

• –    fineness of grinding, fraction content +0.08 mm, %;

• –    humidity ( W ), %;

• –    consistency of the composition.

The grinding of nepheline ore and limestone with the addition of a circulating solution is carried out in tubular ball mills.

There are four stages of charge preparation at AGP (Fig. 1):

Stage I – grinding of nepheline ore with the addition of circulating solution (mills No. 15-22);

Stage II – re-grinding of nepheline pulp (mills No. 11-14);

Stage III – limestone grinding together with nepheline pulp (mills No. 1-8);

Stage IV – re-grinding of nepheline charge, limestone (mills No. 9-10, 23-28, 31-35).

Fig. 1. Stage of the mixing processes

Рис. 1. Стадии приготовления шихты

After the re-grinding mills of the fourth stage, the charge is collected in a prefabricated agitator and moved by pumps into flow averaging schemes of the charge in the correction pools.

From the flow averaging schemes, the charge is pumped into the flow pools, from where it is pumped to the sintering furnaces. The excess charge through the ring pipelines is returned to the flow pools.

Cascade of mills as an object of automation

In Fig. 2 shows a cascade of mills as an automation object, where X ( t ) is a vector of control actions; F ( t ) and M ( t ) are vectors of controlled and uncontrolled disturbances, respectively; Y ( t ) is a vector of output parameters. The table contains vector variables of the cascade of mills.

After the fourth stage, the charge enters the correction pools, where the main output parameters of the alkaline and limestone modulus cascade are measured by laboratory analysis:

M gap - alkaline modulus, 1.12 ± 0.02 units.;

M izv - limestone module, 2.04 ± 0.02 units.

The adjustment of these parameters consists in regulating the input parameters of the mills in the cascade (changes in the consumption of ore, mortar, etc.). This gives additional complexity for automated control, since the modules depend not only on the consumption of input indicators, but also on the composition of the components. Therefore, in order to achieve the necessary indicators of the alkaline and limestone module, it is possible to have variations in the humidity index in the range of 30-35%.

Vector variables of a four-stage cascade

Fig. 2. Cascade as automation object

Рис. 2. Каскад как объект автоматизации

СЛ о ед Й	Vector variables
	Control actions X ( t )	Output parameters Y(t)	Controlled disturbances F ( t )	Uncontrolled disturbances M ( t )
о ед й	x 1.1 – feed of nepheline ore, 0-160 t/h; x 1.2 – feed of circulating solution, 0-63 m3/h	y 1.1 – charge humidity, 28-35 %; y 1.2 – charge level in the agitator, 1.5–4.5 m; y 1.3 – pulp at the outlet of the agitator, 0-200 t/h	f 1 – overheating of the bearings of the drive and gearbox of the mill, up to 75 ° ; f 2 – overheating of mill axle bearings, up to 65 ° ;	m 1 – sensor failure; m 2 – equipment wear and tear
сч о ед л й	x 2 – pulp feed from the first stage of grinding, 0200 t/h	y 2.1 – charge humidity, 30-35 %; y 2.2 – charge level in the mixer, 1.5–4.5 m; y 2.3 – pulp at the outlet of the agitator, 0-125 t/h

ед Й	x 3.1 – supply of nepheline pulp, 0-125 t/h; x 3.2 – limestone feed, 0160 t/h; x 3.3 – feed of circulating solution, 0-63 m3/h	y 3.1 – charge humidity, 26.5–27 %; y 3.2 – charge level in agitators, 1.5–4.5 m; y 3.3 – pulp at the outlet of the agitator, 0-200 t/h	f 3 – clogging of the mill loading funnel, up to 3 m; f 4 – lack of material in the ore bunker; f 5 – oil filter clogging (pressure increase)
ед й	x 4 – pulp feed from the third stage of grinding, 0200 t/h;	y 4.1 – charge humidity, 30-35 %; y 4.2 – charge level in agitators, 1.5–4.5 m; y 4.3 – pulp at the outlet of the agitator, 0-200 t/h

In order to meet the requirements of the process for several indicators at once and at the same time do it automatically, it is necessary to develop a control program containing a circuit for automatic regulation of the chemical composition of the charge, and a mnemonic circuit based on modern software. WinCC was chosen from a number of different SCADA systems, as it has a powerful set of tools for any automation tasks. The simulation capabilities of WinCC software allow real-time monitoring of changes in the technological process described using software and functional blocks, as well as influencing output variables during the process [11; 12].

Development of a mnemonic circuit for cascade control

A virtual control project has been developed, which includes several screens.

The screen "Mill 1" (Fig. 3) is the main control screen of the mill of the first stage of charge preparation. The upper part of the screen is the general control panel. It includes toggle switches for controlling mills, selecting the operating modes of mills of stage 1 and 3, and setting the consumption of materials at the entrance to the mills and exit from the agitators [13].

Fig. 3. Screen «Mill 1»

Рис. 3. Экран «Мельница 1»

On the left side of the screen "Mill 1" shows a mill, an ore hopper, a conveyor, a pipeline with a solution, an agitator and a pump at the exit of the mill. The current sensor readings are also located here. At the bottom there is a navigation bar. On the right are the trends showing the current and preset material costs, as well as the trend of the pulp level in the agitator.

Imitation of the mill operation

The mill cascade is controlled in several stages, which can be repeated during operation. Figure 4 shows the order of operation of the control program blocks. Input indicators on the flowchart include control actions and the composition of raw materials.

Simulation of sensor readings is implemented using scripts in the VBS language.

Changing the values of the flow of materials into the mill can also change the humidity of the incoming mixture. To show this on the mnemonic scheme, a script was developed. The operation of the script is that the equality between the current humidity in the agitator and the humidity of the incoming mixture is continuously checked, and if the equality is violated, then the current humidity value decreases or increases to the required value [14].

Fig. 4. Block-diagram of the cascade control

Рис. 4. Блок-схема управления каскадом

The automatic mode of operation of the mill assumes that the deviation of the current humidity value from the set one will lead to a change in the values of the consumption of mate- rials entering the mill. For example, if the value of the current humidity is less than the set value, it is necessary to increase the consumption of the solution, but at the same time reduce the consumption of ore so that the overall productivity of the mill does not change. The trend bar below shows such a case (Fig. 5) [15].

Fig. 5. Bar of the flow trends

Рис. 5. Панель трендов расхода

Figure 6 shows the bar specifying the composition of raw materials and calculating modules. You can go to it through the navigation panel – the "Analysis" tab. To simulate the operation of laboratory analyzers, a program for calculating alkaline and limestone modules in FBD was developed. Thanks to the program, you can change the composition of raw materials and see the change in the value of modules in real time. Limestone and alkaline modules are calculated on the basis of calculation formulas according to the technological regulations described above.

	Состав руды
Текущий, % Регламент, %
AI2O3	+26	25,5-27,5
SiO2	+ 40	39-41
СаО	+ 8	7-9
Na2O	+ 10	10-11,5
К2О	+ 3	2,5-3
Модуль известняковый
	|+2.379

Состав известняка Текущий, % Регламент, %
AI2O3	+ 1	0,1-1
SiO2	+3	1-5
СаО	+53	50-55
Состав раствора Текущий, % Регламент, %
Na2O	+ 9	8-9
	Модуль щелочной
	1 + 1.705

Fig. 6. Bar for specifying the raw material composition and module calculation

Рис. 6. Панель задания состава входного сырья и расчета модулей

Conclusion

Automation of the charge preparation process in the production of alumina has some problems that were described above. The developed mnemonic scheme and programs in the WinCC SCADA system allow to partially solve these problems, namely: to monitor technological parameters, output their values to the automated process control system and display the operation of equipment; to implement automatic cascade control based on data collection in the automated process control system; to calculate the output charge indicators based on data on the composition of input raw materials and the volume of input flows to the cascade of mills. The latter allows you to set and adjust the ratio of the supply of ore, limestone and water to ensure the desired quality of the charge.