JPH01136912A - Automatic control system of furnace heat in blast furnace - Google Patents

Abstract

PURPOSE:To standardize operational control, to prevent miss-judgement of an operator and to save the man-hour by executing the prescribed inference as an artificial knowledge from processed data generated from each kind of data setting in a blast furnace and knowledge base based on experience, etc. CONSTITUTION:Data of each kind of sensor are read by collecting means 12 at the prescribed timing in order and stored 11 into filing means with the lapse of time. Successively, this data are data-processed with a sensor-data- preprocessing means in an arithmetic processing means 16 to execute the data- treatment on loading-down, temp., gas utilizing ratio, tapping iron and slag, etc. Further, by comparing this data with the prescribed reference value, the prescribed processed data are generated and stored in an interface buffer 19 to forward this data to the common data buffer 24. Then, the inference engine means 26 infer the condition in a blast furnace based on the knowledge data prestoring in the knowledge base 22 and the processed data in the buffer 24.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an automatic blast furnace heat control system that automatically controls the temperature of hot metal tapped from a blast furnace.

[Prior Art] Conventionally, as a method for estimating, managing and controlling the temperature of hot metal in a blast furnace, a blast furnace operator qualitatively judges information from various sensors installed in the blast furnace. The method used is to evaluate the situation and make optimal adjustments to operational factors.

However, there are individual differences in the evaluation results depending on the ability and experience of the operator, which makes it difficult to standardize operational actions, and the evaluation is not quantitative, making it difficult to estimate the hot metal temperature. there were.

For this reason, a process control method for a blast furnace has been proposed, for example, as disclosed in Japanese Patent Publication No. 51-30007. This process control method maintains the blast temperature constant, calculates the amount of direct reduction in the furnace from measurements that can be obtained continuously during operation, and calculates the target St content in the pig iron and an exponentially smoothed value that represents the actual value. The blast humidity is determined by an equation with a correction term added to prevent long-term fluctuations in the Si content in the pig iron, and the heat balance in the furnace is controlled using this blast moisture determined value. ing.

For this reason, accurate operation has been achieved by correcting the large wave changes (long-period changes) that occur when the blast furnace status is controlled by calculation.

[Problems to be solved by the invention] In the conventional process control method disclosed in the above-mentioned Japanese Patent Publication No. 51-30007, information from sensors is analyzed and input into a model to perform predetermined calculations. I have to. For this reason, the computer that executes the calculation uses, for example, Fortran as the language, but the calculation capacity is extremely large. Furthermore, as blast furnaces change over time, the analysis model itself must be changed for maintenance, but there is a problem in that the analysis modem itself is complex, and changing the conditions of the analysis model is an extremely troublesome task.

In addition, as a means to solve the above-mentioned problems, it is possible to improve maintainability by using a computer system that uses an artificial intelligence language such as LISP, but here, sensor information (authenticity data, various sensor data) ,
In using the knowledge base to reason about the furnace thermal situation,
When using production rules, for example, ε IF xi -x THEN A-A C
F=Y.

IF xl -x THEN A-A C
As shown in F-Y2, there was a problem in that creating a program for writing or inputting numerical values described in the rules became complicated.

The present invention was made to solve these problems, and it is possible to control the furnace heat of the blast furnace at a constant level.
The purpose is to improve the calculation capacity and calculation speed when realized on a computer, and to obtain an automatic blast furnace thermal control system that can easily add and modify rules in response to new situations such as aging of the blast furnace. shall be.

[Means for Solving the Problems] The blast furnace furnace thermal automatic control system according to the present invention includes a data input means that takes in data at a predetermined timing from various sensors installed in the blast furnace, and a system that, based on the data from the sensors, It has means for creating various data indicating the status of the blast furnace, such as tuyere embedding temperature, unloading speed, pressure loss, furnace top temperature, gas utilization rate, solution loss amount, etc.

Furthermore, means for comparing the various data with reference data to create processed data (data indicating authenticity and relationship with furnace heat), storage means for temporarily storing the processed data, and information on blast furnace operation. a knowledge base means in which various knowledge bases based on experience, achievements, mathematical models, etc. are stored;
Inferring the furnace heat level and furnace heat transition based on the processed data of the storage means and the knowledge base of the knowledge base means,
It has an inference means for determining the amount of action for the blast furnace, and a means for automatically controlling the amount of action.

According to a preferred embodiment of the invention, said reasoning means:
A three-dimensional function with three axes of sensor information, furnace heat level or furnace heat transition, and reliability is used.

[Operations] In the present invention, various data indicating the conditions of the blast furnace are created from the blast furnace data from the data input means, and then processing data is created based on the data.

Based on the processed data and knowledge base, artificial intelligence performs inference calculations and determines the amount of action for the blast furnace. Then, the blast furnace furnace heat is automatically controlled by adjusting the operating factors based on the action.

[Example] Embodiments of the present invention will be described below based on the drawings.

FIG. 1 is a conceptual diagram of an automatic blast furnace thermal control system according to an embodiment of the present invention, which includes a process computer with a blast furnace operation support function, an AI dedicated processor for inference processing with AI roots, and a digital instrumentation device. Main composition.

In the figure, (1) is the blast furnace to be controlled, (1o) is the process computer with blast furnace operation support function, sensor data collection means (12), file means (4), calculation means (1B), blast furnace control It includes a system (18) and an interface buffer (19).

The sensor data collection means (12) collects data from various sensors such as temperature sensors, pressure sensors, gas sensors, etc. to data scanners (lla), (llb), (lll).
c), the data is manually processed in chronological order.

The file means (14) includes a time series file means (for normal control) and a time series file means (for artificial intelligence), and performs a data banking function. The arithmetic processing means (16) includes a system processing means, a time series processing means (for artificial intelligence), and a sensor data preprocessing means.

In this embodiment, the file means (13
) incorporates a time series file means (for artificial intelligence),
The difference from conventional process computers is that a time series processing means (for artificial intelligence) and a sensor data preprocessing means are incorporated into the calculation means (16), and an interface buffer (19) is also incorporated. These devices preprocess sensor data for calculation on a small computer as an AI dedicated processor.

(20) is a small computer as an AI-dedicated processor, which includes a knowledge base (22), a common data buffer (24), and an inference engine (26) shown in FIG. (The structure of the items indicated by the symbols (10) and (20) above is based on the patent application filed in 1986-11, which was already filed by the same applicant
It is basically the same as that related to No. 3795. )(3
0) is a CRT, on which the results of the inference of the inference engine (25) are transmitted and displayed via the interface-buffer (19) and file means (14). (32) is a digital instrumentation device that controls the temperature control of the blast furnace based on the inference results of the inference engine (2B). It adjusts the amount of operation, and the interface buffer (
19) and the blast furnace control system (18).

(40) is a hot air stove, (42), (44), (
46) are control valves.

The operation of this embodiment having the above configuration will be explained.

(1) First, data from various sensors is sequentially read at a predetermined timing by the sensor data collection means (12) via the data scanner (11),
14) is stored in the time series file means (for normal control).

(2) The data stored in the time series file means (for normal control) of the file means (14) is stored in the arithmetic processing means (16).
) The data is processed by the sensor data preprocessing means.

This sensor data preprocessing means processes data regarding unloading, temperature, gas utilization rate, tap slag, etc. Each of these data is compared with a predetermined reference value to create predetermined processing data, and the interface buffer (1
9).

(3) Next, the processed data stored in the interface buffer (19) is transferred to the common data buffer (24).

(4) The inference engine means (2B) has a knowledge base (22
) and the processed data in the common data buffer (24).

Here, the knowledge base (22) is composed of knowledge units for making furnace heat judgment, furnace heat transition, action judgment, and action correction, as shown in FIG. 2, taking into account the efficiency of reasoning and the like. These are IFs based on the operator's knowledge and experience.
-THEN~ type production rules or frames, and the CF value (Certainty Factor) is used as an index to express the ambiguity of each rule.
; confidence) is introduced to increase the reliability of the inference.

Furthermore, by using a three-dimensional function, which will be described later, as a means for determining the CF value, the number of rules is reduced and maintainability is improved.

As shown in FIG. 2, the inference engine means (2B) first determines the furnace heat level and the furnace heat transition, and then determines the amount of action based on the results of these determinations.

This action amount is subjected to a predetermined correction, and the result is sent to the digital instrumentation device (3) via the blast furnace control system (18).
2).

Then, the control valve (42) is controlled by the digital instrumentation device (32).
), (44), and (4B) are appropriately controlled,
An action is taken to control the temperature of the blast furnace (1) to a desired value.

Next, the structure of the knowledge base and its specific inference will be explained based on FIG. 2.

(A) Furnace heat level determination K S (Knowl
knowledge source) group; This is a knowledge base that determines the state of the furnace heat at the inference start time. It consists of KS groups divided into sensor attributes and functions such as "hot metal temperature,""blowpressure," and "tuyere embedded temperature," and each KS group is divided into seven levels from high to low. The CF value distribution is determined for the furnace heat level determined by the method described later, and the level of maximum certainty is taken as the furnace heat level at the current time. In addition, if it is determined that the furnace heat level is extremely low,
It has a ``human judgment rule'' that allows you to input information such as the tuyere status and the tapping status at any time in an interactive format, and to incorporate information that cannot be obtained using sensor-related rules.

In addition, it also includes knowledge bases that can deal with various conditions such as pig iron slag storage, hot metal composition, temperature deviation between tap holes, and tap type.

(B) Furnace heat transition judgment KS group: The furnace heat transition is classified into 5 levels from rapid rise to steady steady decline depending on the degree of change from the past to the present, and the confidence level is calculated for each rank. Let the step position of the value be the furnace heat transition state at the current time.

The knowledge base consists of various sensors such as "hot metal temperature" and "hot metal components", and "short-term trends" for "hot metal temperature" and "hot metal components".
The rules are divided into "long-term trends" and "long-term trends."

(C) Action judgment KS group: Find the current furnace heat state on a matrix centered on the furnace heat transition and furnace heat level, and decide what action to take.

FIG. 3 shows an example of this, and this example shows that the peak (maximum value) of the CF value is at a furnace heat level of -4 and a furnace heat transition of -3.

Note that the action type and amount of action at each position on the matrix are stored in advance as knowledge in the frame.

The action type is configured as shown in FIG. 4, for example, and the amount of action is quantified as shown in FIG.

Note that when adopting an action-type action at a predetermined position, it is necessary that the CF value reaches a predetermined value. In principle, all action amounts are automatically controlled, but it is also possible to manually control some of them (for example, G in FIG. 5).

(D) Action correction amount determination KS group: Determines actions taken in the past or disturbances, and determines the correction action amount by considering their shadow 'fl fa at the current time. The contents include air humidity, air temperature,
It consists of rules for detecting and responding to changes in manipulated variables such as liquid fuel and coke ratio, as well as disturbances such as coke moisture and falling off deposits.

For example, when the ventilation humidity is changed, the change time and amount are automatically detected by the "operation amount change detection" rule, and the subsequent influence amount is considered as a function of time by the "ventilation humidity" rule. In addition, when something attached to the furnace wall falls off, the "wall fall" rule automatically determines the falling place, the amount of influence on the furnace heat, and the time for the tuyere tip to fall, and the operation time and amount of the preliminary action are determined and correction calculations are made. be incorporated into.

(E) Overall judgment KS.

The amount of action to be taken is comprehensively determined based on the determination results of (A) to (D) above. Then, the judgment results are manually inputted to the operating state judgment KS along with the wind break judgment KS to determine the operating state, and are sent to the CRT (30) via the screen display KS.
to instruct the operator on the amount of action to take.
At the same time as providing guidance, feedback is provided to the digital instrumentation device (32) for fully automatic control (see Figure 1).

Next, the three-dimensional functions used to obtain the CF value in each of the above knowledge bases (A) and (B) will be explained.

In normal production rules that do not introduce this function, for all related sensors, for example: (1) IF (the temperature of sensor 1 is in the range of T1 to T2).
THEN (CF value, which is a high fever level, is CI),...
..., (CF value, which is a low fever level, is Cn); ■IF (
The temperature of sensor i is in the range T1 to T3. )THEN
(CF value C'l of high fever level).

・・・・・・(CF value of low fever level is C'n);■・・・
. . . It is necessary to express the rules as follows, which results in a huge number of rules, which increases the inference time and has the disadvantage that the adjustment of the CF value becomes extremely complicated.

FIG. 6 shows a three-dimensional function of the furnace heat transition for one piece of sensor information. In the figure (.) is an input value given as a boundary condition of this function.

In the figure, X-Z is the X-axis secondary processing data (for example, the amount of solution loss carbon), the Y-axis: the furnace heat transition status (the degree of change is classified into 1 to 5 levels), and the Z-axis: confidence level (CF value). means. Here, it is shown that when the value of a certain sensor is X on the X-axis, the reliability Zj (X) for each stage (j-1 to j-5) on the Y-axis shown in FIG. 6 is determined. After finding the confidence distribution for each piece of information in this way,
Each sensor is weighted and cumulatively added to determine the overall confidence distribution of the furnace heat transition.

Next, a method for creating a three-dimensional function of the hot metal temperature level will be described in detail. Figure 7 shows the distribution of hot metal temperature changes during tapping, and each axis represents the following: There is. Next, FIG. 8 shows the relationship between the hot metal temperature Ti and the tap maximum hot metal temperature Tmi at the tapping time X-X1. Based on this figure, the three-dimensional function shown in FIG. 9 is obtained by adjusting the relationship between the tap maximum hot metal temperature and the furnace heat, and the relationship between the appearance frequency and the CF value. In addition,
Since tapping temperature is measurement information that depends on the elapsed time from the start of tapping and operating conditions, more than 30 types are available so that you can use it properly, and it is automatically selected according to the conditions.

The above three-dimensional functions are written in a frame, and the calculation procedure is given by a LISP function.

Figure 10 shows the results of fully automatic control operation using blown water ratio control. For example, action Di (water ratio -3g/
Nm3) is due to the decrease in the furnace heat level. Also, the action at 15:00 (water ratio -2g/Nm3) is type C (no action) at present, but it was taken in consideration of the fact that the coke ratio -2kg/T would later fall to the tip of the tuyere. It is.

Comparing the operational results according to this embodiment with the operational results by conventional operators, it can be seen that the accuracy of controlling the hot metal temperature has been improved as shown in FIG. 11. This means that operational management has been standardized.

[Effects of the invention] As described above, according to the present invention, processed data is created from various data installed in a blast furnace, and predetermined inferences as artificial intelligence are made using the processed data and a knowledge base based on experience. As a result, conventional experience has been fully utilized and the following effects have been obtained.

(1) Standardization of operational management (2) Prevention of human misjudgments (3) Stable supply of high-quality hot metal with little temperature and composition fluctuations to the next process (4) Avoidance of furnace cooling (5) Labor saving (6) When the device is implemented using a computer, its calculation capacity and calculation speed are improved, and rules can be easily added and modified in response to new situations such as aging of the blast furnace.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an automatic blast furnace thermal control system according to an embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of a knowledge base, and FIG. 3 is an explanatory diagram showing an example of an action matrix. Fig. 4 is an explanatory diagram showing an example of the type of action matrix, Fig. 5 is an explanatory diagram showing an example of the amount of action, Fig. 6 is an explanatory diagram of a three-dimensional function of furnace heat transition, and Figs. Figure 9 is an explanatory diagram showing how to create a three-dimensional function of the furnace heat level, Figure 10 is a characteristic diagram showing the results of a fully automatic control operation example using injection water ratio side control, and Figure 11 is an explanatory diagram showing the method of creating a three-dimensional function of the furnace heat level. FIG. 3 is a characteristic diagram showing the error occurrence rate. Agent: Patent Attorney Souji Sasaki

Claims (2)

[Claims] (1) A data input means that takes in data from various sensors installed in the blast furnace at a predetermined timing; Means for creating various data indicating the status of the blast furnace, such as utilization rate, solution loss amount, etc., and data indicating the authenticity and relationship with furnace heat by comparing the various data with standard data (hereinafter referred to as processed data) a storage means for temporarily storing the processed data; a knowledge base means storing various knowledge bases based on experience, results, mathematical models, etc. regarding blast furnace operation; and a storage means for temporarily storing the processed data. and inferring the furnace heat level and furnace heat transition based on the knowledge base of the knowledge base means,
An automatic blast furnace thermal control system comprising: inference means for determining an action amount for a blast furnace; and means for automatically controlling the action amount. (2) The blast furnace according to claim 1, wherein the inference means makes inferences using a three-dimensional function with three axes of sensor information, furnace heat level or furnace heat transition, and confidence level. Thermal automatic control system.