JPH0730368B2 - Blast furnace furnace thermal controller - Google Patents

Description

TECHNICAL FIELD The present invention relates to a blast furnace thermal control device for controlling the temperature of hot metal tapped from a blast furnace.

[Prior Art] As a method of estimating the temperature of hot metal in a conventional blast furnace, and managing / controlling the temperature, generally, a blast furnace operator qualitatively determines information from various sensors installed in the blast furnace. A method is adopted in which the condition of the blast furnace is evaluated and the operating factors are optimally adjusted. However, there are individual differences in the evaluation results due to the ability and experience of operators, and therefore it is difficult to standardize the operation actions and it is difficult to estimate the hot metal temperature because the evaluation is not quantitative. there were.

Under such circumstances, a blast furnace process control method as disclosed in Japanese Patent Publication No. 51-30007 has been proposed. This process control method keeps the blast temperature constant and obtains the amount of direct reduction in the furnace from the measured values that can be obtained continuously during operation, and the exponential smoothed value that represents the target value of Si content in pig iron and its actual value. The blast moisture content is determined by an equation that adds a correction term to prevent long-term fluctuations in the Si content in the pig iron, and the blast moisture content determination value controls the heat balance in the furnace. ing.

Therefore, the large wave change (change of long cycle) that occurs when the blast furnace condition is controlled under calculation is corrected to realize an accurate operation.

[Problems to be Solved by the Invention]

In the conventional process control method disclosed in Japanese Patent Publication No. 51-30007, the information from the sensor is analyzed and input into a model to perform a predetermined calculation. For this reason, a computer that executes the calculation uses, for example, Fortran as a language, but the calculation capacity is extremely large. Furthermore, since the blast furnace changes over time, it is necessary to change and maintain the analysis model itself, but since the analysis model itself is complicated, changing the conditions of the analysis model is a very troublesome task.

Further, as a means for solving the above problems, a maintainability can be improved by a computer system using an artificial intelligence language, for example, LISP.
When inferring the furnace heat situation using sensor information (true / false data, various sensor data) and the knowledge base, if a production rule is used, for example, IF ( The temperature of sensor i is in the range of T1 to T2.) THEN
(CF value with high heat level is C1) ……, (CF value with low heat level is Cn); IF (Temperature of sensor i is in the range of T1 to T3.) THEN
(CF value at high heat level C'1), ... (CF value at low heat level is C'n); ... It is necessary to express the rules as follows, and the inference time increases because of the huge number of rules. Moreover, the adjustment of the CF value is extremely complicated.

Therefore, there has been a problem that the amount of numerical value indicating the certainty factor described in the rule program or in the rule becomes enormous, and the input work becomes complicated.

Also, if the deposits on the furnace wall of the blast furnace fall off, it takes some time to reach the tuyere, and the effect on the furnace heat during that time is not uniform, but it was not possible to grasp that in the past. From this point as well, there is a problem that the furnace heat control cannot be sufficiently performed.

The present invention has been made in order to solve such a problem, it is possible to control the furnace heat of the blast furnace with high accuracy, and when it is realized by a computer, its calculation capacity and calculation speed are improved, Moreover, it is an object of the present invention to obtain a blast furnace heat control device in which rules can be easily added and modified even in a new situation such as aging of the blast furnace, and the effect of falling of deposits on the furnace wall is taken into consideration.

[Means for Solving the Problem] FIG. 1 is a block diagram showing the concept of a blast furnace heat control device according to the present invention. The blast furnace heat control device according to the present invention,
Data input means for capturing data from various sensors installed in the blast furnace at a predetermined timing, and based on the data from the sensor, tuyere embedding temperature, unloading speed, pressure loss, furnace top temperature, gas utilization rate, And processing data creating means for creating various data indicating the state of the blast furnace, including the amount of solution loss, comparing the various data with the reference data, and creating the difference data thereof. Storage means for temporarily storing data and difference data (hereinafter referred to as processing data); knowledge base storage means for storing various knowledge bases based on experience, results, and mathematical model of blast furnace operation; Determining the furnace heat level and furnace heat transition based on the processing data and the knowledge base of the knowledge base storage means to determine the action amount for the blast furnace It has an inference means and a means for outputting the action amount.

The knowledge base storage means includes a "furnace heat level determination knowledge base" used for estimating a furnace heat level, a "furnace heat transition determination knowledge base" used for estimating a furnace heat transition, and a furnace. "Action judgment knowledge base" used to determine action amount from heat level and furnace heat transition
And "action correction knowledge base" that determines the correction value of the action amount based on past actions and disturbances
And an "integrated judgment knowledge base" for determining an actual action amount from the action amount and the action correction amount.

The "reactor heat level judgment knowledge base" and the "furnace heat transition judgment knowledge base" have the processing data and the furnace heat level as independent variables, and the probability of occurrence of a combination of these (hereinafter referred to as confidence) as a dependent variable. A confidence function and a group of rules for determining how to apply this confidence function, respectively, the "action determination knowledge base" has a rule to determine the amount of action from the furnace heat level and furnace heat transition,
The “total judgment knowledge base” has a rule for determining an actual action amount from the action amount and the action correction amount.

Furthermore, the above-mentioned "action correction knowledge base" is for detecting the falling of the deposit on the furnace wall, determining the falling position and the influence amount, and inferring the thermal compensation amount and the timing based on the determination result. Contains the wall drop rules used.

[Operation] In the present invention, after the processing data creating means creates various data indicating the state of the blast furnace based on the blast furnace data from the data input means, the processing data is created based on the data. An inference operation as artificial intelligence is performed based on the processing data and the knowledge base stored in the knowledge base storage unit to determine the action amount for the blast furnace.

In the inference calculation, the furnace heat level determination knowledge base is used to estimate the furnace heat level, and the furnace heat transition determination knowledge base is used to estimate the furnace heat transition. When inferring the furnace heat level and the furnace heat transition, a certainty factor function and a group of rules for determining the application method are used to obtain certainty factor for the inference result.

Next, the above-mentioned furnace heat level and furnace heat transition are applied to the action determination knowledge base to determine the action amount.

When the wall fall rule is detected using the wall fall rule of the action correction knowledge base, the amount of heat compensation and the timing of the heat compensation are calculated and used as the action correction amount.

Next, an actual action amount is determined from the action amount and the action correction amount using the comprehensive judgment knowledge base. Then, the blast furnace heat is controlled by adjusting the operation factor based on the action amount.

[Embodiment] An embodiment of the present invention will be described below with reference to the drawings. FIG. 2 is a block diagram showing a blast furnace heat control device and related equipment according to an embodiment of the present invention.

In the figure, (1) is a blast furnace to be controlled, (10) is a blast furnace thermal control device according to the present invention, and a data scanner (1
1), sensor data collection means (12), file means (1
4), computing means (16), and output interface (1
8), knowledge base (22), inference engine (26), common data buffer (24).

The sensor data collecting means (12) inputs data from various sensors, for example, a temperature sensor, a pressure sensor, a gas sensor, etc. in time series through the data scanners (11a), (11b), (11c).

The file means (14) performs a file data banking function. The calculation means (16) is a file means (1
The data from the sensor data collecting means (12) stored in 4) is exponentially smoothed and then stored again in the file means (14). Then, at a predetermined time, for example, every 20 minutes, the average value and the difference data between the average value and the reference value are used as the processed data, and the common data buffer (2
4) Send to.

The inference engine (26) has its data and knowledge base (22)
A predetermined inference operation is performed on the basis of the knowledge of (1), the amount of action to be taken next is calculated, and the result is stored again in the common data buffer (24) and also in the file means (14).

(30) is a CRT, and the inference result of the inference engine (26) is transmitted through the file means (14) and displayed.

(32) is a digital instrumentation device, which is a blast furnace heat control device (10)
The temperature of the blast furnace is controlled based on the above command, and the operation amount of the operation type such as blast humidity, blast temperature, and liquid fuel such as heavy oil that can adjust the furnace heat is adjusted. At this time, the blast furnace heat control device (1
The command 0) is sent to the digital instrumentation device (32) via the output means (18).

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

An outline of the operation of this embodiment having the above configuration will be described.

(1) First, data of various sensors is sequentially read by the sensor data collecting means (12) via the data scanner (11) at a predetermined timing, for example, at 1 minute intervals, and stored in the file means (14). .

(2) The data stored in the file means (14) is exponentially smoothed by the arithmetic means (16). Here, data relating to unloading, temperature, gas utilization rate, slag, etc. is processed. The various kinds of data that have been subjected to arithmetic processing are stored again in the file means (14). Next, for these various data, the average value and the difference between the average value and the predetermined reference value are obtained at a predetermined time, for example, at intervals of 20 minutes, and are transferred to the common data buffer (24) as processed data. To do.

(3) The inference engine means (26) infers and calculates the situation in the blast furnace based on the knowledge data stored in advance in the knowledge base (22) and the processing data in the common data buffer (24).

Here, the knowledge base is, as shown in FIG. 3, furnace heat level judgment KS group, furnace heat transition judgment KS group (KS; knowledge source), action judgment KS, action correction judgment KS, comprehensive judgment KS group, operation status judgment. It is formed from units of each knowledge base of KS.

Reactor heat level determination The KS group is a knowledge base used by the inference engine (26) to determine what level the furnace heat level of the blast furnace is, and determines the furnace heat level with the hot metal temperature as the main determining factor. "Hot metal temperature-furnace heat level
KS ”and“ sensor-furnace heat level KS ”that determine the furnace heat level with the measured quantities of other sensors as the main judgment factors.

Each of these KS has a certainty factor function (hereinafter referred to as a CF function) in which each measured amount and furnace heat level are independent variables, and a probability (confidence factor) of the combination of them is a dependent variable. It consists of a set of rules that determine the usage procedure.

Reactor heat transition judgment The KS group uses the inference engine (2) to determine the level of the furnace heat transition of the blast furnace.
This is the knowledge base used by 6), and the main decision factor is the "hot metal temperature-furnace heat transition KS" that determines the transition of the furnace heat with the transition of the hot metal temperature as the main determination factor, and the transition of the measured amount of other sensors. Judgment of furnace heat transition "Sensor-furnace heat transition
KS ”etc. are included.

These KS also have CF functions with each measured quantity and furnace heat transition level as an independent variable, and the probability (confidence) of the combination of them as a dependent variable, and a set of rules that determine the procedure for using that CF function. It is established from.

The ansion determination KS is made up of a group of rules for determining the action amount based on the combination of the furnace heat level and the furnace heat transition level. The action correction amount KS group is made up of a group of rules for correcting the current action amount based on the information of the action taken in the past and the disturbance that occurred in the past. The comprehensive judgment KS is made up of a group of rules for determining the final action amount based on the result of the action judgment and the result of the action correction amount judgment.

Then, the operation state determination KS determines whether or not the operation of the furnace is normally performed based on, for example, the furnace heat level, and if it is normal, the above action amount is sent to the control system as it is, and if it is abnormal, For example, it is composed of a group of rules that display that fact and give guidance to the operator.

The inference engine means (26) executes each knowledge base. As shown in the flowchart of FIG. 3, first, the furnace heat level and the furnace heat transition are judged, and then the action amount is determined based on these judgment results. To judge. The action amount is subjected to a predetermined correction, and the result is temporarily stored in the common data buffer (24) and then stored in the digital instrumentation device (32) via the file means (14) and the output means (18). Sent.

And the control valve (42) by the digital instrumentation device (32),
The opening degrees of (44) and (46) are appropriately controlled, an action operation is performed, the temperature of the blast furnace (1) is controlled, and as a result, the hot metal temperature is controlled to a desired value.

Next, the structure of the knowledge base and the outline of its specific inference will be described with reference to FIG.

(A) Reactor heat level judgment KS (KS (Knowlege Source); knowledge source) group; This furnace heat level judgment KS group is a knowledge base for judging the state of the furnace heat at the inference start time, and as described above, Temperature-furnace heat level KS "," sensor-furnace heat level K
It consists of KS groups such as “S”. As shown below, for each furnace heat level divided into 7 levels from high to low level, the CF value distribution is obtained by the method described below, and the maximum The confidence level is the furnace heat level at the current time.

Furnace heat level evaluation 7 Large heat 6 Medium heat 5 Normal 4 Small cooling 3 Medium cooling 2 Large cooling 1 Extra large cooling Here, an example of the hot metal temperature-furnace heat level KS will be described. KS
Consists of an IF section that sets conditions and a THEN section that sets the contents of instructions when the conditions are satisfied. An example is as follows.

Rule No.1 [IF part] Pan order = 1 NOT (a large amount of residue) NOT (elapsed time after completion of wind reduction ≤ 180 minutes) Si, S judgment is "low" [THEN part] Melted by normal 3D function Obtain the CF value of pig temperature-heat level.

This rule No.1 is that when the operation state of the blast furnace is in a steady state, the normal hot metal temperature-furnace heat level
It indicates that the CF value is to be obtained.

That is, the CF value function of hot metal temperature-furnace heat level is usually
There are three types: f _MTN , somewhat high ~ f _MTH , high ~ f _MTEH ,
The inference engine (26) selects which function to use according to the above logic.

In other words, the hot metal temperature-furnace heat level KS is the order of the pots (the number indicating the number of the ladle used to measure the hot metal temperature from the start of tapping) and the amount of residue and after the reduction of wind. A rule for deciding which of the above-mentioned three types of functions to be adopted is stored in correspondence with the elapsed time from. Another example of this rule is as follows: Rule No. 2 [IF part] Pan order = 1 NOT (many residues) NOT (elapsed time after completion of wind reduction ≤ 180 minutes) Si, S judgment is "slightly high" [THEN part] The CF value at the hot metal level is calculated by the three-dimensional function when it is "slightly high".

Rule No. 3 [IF part] Pan order = 1 NOT (a large amount of residue) NOT (elapsed time after wind reduction ≤ 180 minutes) Si, S judgment is "high" [THEN part] 3D when "high" The CF value at the hot metal level is obtained by a function.

These rules No.1 to No.3 indicate that the type of function is selected by the judgment of Si and S in the hot metal when there is not much residue and sufficient time has passed since the end of the wind reduction. .

Also, all three CF functions of hot metal temperature-furnace heat level are functions of pan order LN, hot metal temperature MT, and furnace heat level FHL.

That is, f _MTN = f _MTN (LN, MT, FHL) f _MTH = f _MTH (LN, MT, FHL) f _MTEH = f _MTEH (LN, MT, FHL) The inference engine (26) was selected by the above rule. CF
Regarding the value function, LN and MT are applied to the actual measurement values, and FHL is applied to the numerical values from 1 to 7 described above, and the CF value corresponding to each furnace heat level is obtained.

Figure 4 shows the CF value function for one pan order, and since the pan order is fixed, the CF value is a function of the hot metal temperature and the furnace heat level.

For example, when fMTN is selected according to the rules, when the hot metal temperature is 1400 ° C, the CF value is the highest when the furnace heat level is 4 and the CF value is
= 0.2, when the hot metal temperature is 1480 ° C, the maximum CF value is 0.4 when the furnace heat level is 7.

Next, the sensor-furnace heat level (tuyere embedding temperature-furnace heat level K
S, solution loss C amount-furnace heat level KS) will be explained.

This KS is a rule that determines whether to use the tuyere embedding temperature-furnace heat level KS and solution loss C amount-furnace heat level KS, and the tuyere embedding temperature, furnace heat level, and solution loss C, respectively. the amount and CF value function Ronetsu level two independent variables f _{HT 'consists f SL f HT = f HT (} HT, FHL) f SL = f SL (SL, FHL) to compute this function As for the tuyere embedding temperature HT and the solution loss C amount SL used for, the actual measurement value (the exponential smoothed value or moving average value) is used.

An example of a rule that decides whether or not to use these functions is shown below.

Rule No. 1 [IF part] NOT (there is a lot of residue) [THEN part] (1) Depending on the tuyere embedding temperature, the tuyere embedding temperature-furnace heat level
Calculate the CF value.

(2) Obtain the CF value of the solution loss C amount-furnace heat level based on the solution loss C amount.

In this case, the inference engine (26) uses both functions.

Rule No. 2 [IF part] (there is a lot of residue) [THEN part] (1) Tuyere embedding temperature-CF value of furnace heat level is set to "0".

(2) Solution loss C amount-CF value of furnace heat level is set to "0".

In this case, the tuyere embedded temperature-furnace heat level CF value and the solution loss C amount-furnace heat level CF value are constant values regardless of the furnace heat level.

This means that these factors do not contribute to the determination of the furnace heat level, which will be described later, and the furnace heat level is the above-mentioned hot metal temperature-furnace heat level.
Means to be determined only by KS.

FIG. 5 is an explanatory diagram showing the operation of the inference engine (26), in which the CF value for each furnace heat level is obtained based on the hot metal temperature-furnace heat level KS, and the tuyere embedded temperature-furnace heat CF value for each furnace heat level is calculated based on the level KS and the solution loss C amount-furnace heat level KS. And the CF value and solution loss C amount at each level depending on the tuyere embedding temperature KS
Add the CF value of each level by KS. The CF value of the sensor level thus obtained and the CF value of each level based on the hot metal level KS are added. In this way, the CF value of each furnace heat level (7-1) is obtained.

(B) Furnace heat transition determination KS group; This furnace heat transition determination KS group includes hot metal temperature-furnace heat transition
KS and sensor-furnace heat transition KS is included, and according to the degree of change from the past to the present, the furnace heat transition is divided into 5 stages from sudden rise to constant to sudden fall, and for each rank. The CF value is calculated, and the stage position with the maximum value is taken as the furnace heat transition state at the current time.

Level Contents 5 Rapid rise 4 Upward rise 3 Sideways 2 Lower down 1 Rapid fall Here, the hot metal temperature-furnace heat transition KS is explained.

This KS is a CF function f _ΔMT = f _ΔMT (ΔMT, VFHL) that has two independent numbers of the hot metal temperature difference ΔMT between the front and back taps (one tapping) and the furnace heat transition level VFHL, and this CF Pre-processing and post-processing rules for functions are stored.

The usage of this KS will be described below.

Rule No. 0 [IF section] (Initial setting) [THEN section] Set the following values as CF values for the hot metal transition.

Level 1 2 3 4 5 CF value 0 0 0 0 0 0 Rule No. 1 [IF part] (1) NOT (Si, S judgment is "somewhat high") (2) NOT (Si, S judgment is "high" (3) Stability flag is ON (state in which the furnace condition is stable) [THEN part] (1) ΔMT = (hot metal temperature of current tap-hot metal temperature of previous tap) for each furnace heat transition level Obtain the CF value and use it as the "current hot metal temperature-furnace heat transition CF value".

(2) Next, the CF value is calculated for each furnace heat transition level as ΔMT = (hot metal temperature of front tap-hot metal temperature of front tap),
The value is "the previous hot metal temperature-furnace heat transfer CF value".

(3) Multiply the "current hot metal temperature-furnace heat transition CF value" and the "previous hot metal temperature-furnace heat transition CF value" by weighting factors and add them to the "hot metal temperature-smooth furnace heat transition CF value".

(These calculations are performed separately for each level of the furnace heat transition.) In other words, in the stable state of the furnace condition, up to the latest data are used for estimating the furnace heat transition.

Rule No.2 [IF part] (1) (Si, S judgment is "high") (2) NOT (Si, S judgment is "high") (3) Stability flag is OFF [THEN part] ( 1) Calculate the CF value for each furnace heat transition level as ΔMT = (pre-tap hot metal temperature-pre-previous tap hot metal temperature) and set it as "previous hot metal temperature-furnace heat transition CF value".

(2) Multiply the "previous hot metal temperature-furnace heat transition" by a weighting factor and add it to the "hot metal temperature-smooth furnace heat transition CF value" and set again as "hot metal temperature-smooth furnace heat transition CF value" To do.

Since the furnace conditions are not stable, "current hot metal temperature-furnace heat transition" is not considered here. Next, the sensor-furnace heat transition KS will be described. The sensor-furnace heat transition KS stores the CF value function that uses the transitions of the measured values of many sensors and the furnace heat transition as two independent variables, and the rules that determine how to use it. It is provided in.

That is, f _ΔSi = f _ΔSi (ΔSi, VFHL) (i = 1 to n: for each corresponding sensor) An example will be described in which the sensor has the tuyere embedded temperature.

Rule No. 1 [IF part] (1) NOT (there is a large amount of residue) [THEN part] ΔSi = (tuyere embedding temperature-60 minutes before tuyere embedding temperature) as the tuyere embedding temperature-furnace heat transition Calculate the CF value.

Rule No.2 [IF part] A large amount of residue CF value of tuyere transition level 4> 0 [THEN part] Overwrite setting of "0" to CF value of tuyere transition level 4.

Level 1 2 3 4 5 CF value * * * 0 * (* indicates that the value is the same as before) Rule No. 3 A large amount of residue CF value of tuyere transition level 5> 0 [THEN part] Overwrite the CF value of the tuyere transition level 5 with "0".

Level 1 2 3 4 5 5 CF value ＊＊＊＊ 0 ……………………………… Fig. 6 shows the wing where the tuyere embedding temperature (difference from the reference value) and the furnace heat transition level are independent variables. The mouth filling temperature-furnace heat CF function is shown.

Note that the hot metal temperature-furnace heat transition KS can be divided into “short-term transition” and “long-term transition” and set as rules.
In addition to the tuyere-embedded KS, the sensor transition KS also includes other KS, such as unloading, blast pressure, gas utilization rate, solution loss, and other KS, and this information is also taken into consideration.

The inference engine (26) calculates the CF value for each transition based on each rule of the hot metal temperature-furnace heat transition KS and each sensor based on each sensor-furnace heat transition KS rule. Calculate the CF value for each transition. And
The CF values of these KS are added for each furnace heat transition level to obtain the CF value for each furnace heat transition level.

(C) Action judgment KS; This action judgment KS is a knowledge base for determining the action to be taken by obtaining the furnace heat state at the current time on a matrix with the furnace heat transition and the furnace heat level as axes.

The inference engine (26) obtains the product of the CF value of the furnace heat level and the CF value of the furnace heat transition based on the above action determination KS and writes it in the matrix.

FIG. 7 shows an example thereof, and in this example, the peak (maximum value) of the CF value is that the furnace heat level = 4 and the furnace heat transition = 3. The action type and the amount of action at each position on the matrix are stored as knowledge in the frame in advance.

FIG. 8 is a diagram showing an example of action type, and FIG. 9 is a diagram showing an example of action amount.

When adopting an action-type action at a predetermined position, the CF value needs to reach a predetermined value. Further, in principle, all of the action amount is automatically controlled, but a part of the action amount can be manually controlled (for example, the action amount G in FIG. 9).

(D) Action correction amount determination KS group: In this action correction amount determination KS group, the action or disturbance taken in the past is determined, and the correction action amount is determined in consideration of the influence amount at the present time. Contains various KS to do. The content is
It is composed of rules for responding to changes in manipulated variables such as blast humidity, blast temperature, liquid fuel, coke ratio, etc., and disturbances such as coke moisture and falling off of adhering substances.

For example, when the blast humidity is changed, the change time and the changed amount are automatically detected by the "manipulation amount change detection" rule, and the subsequent influence amount is considered as a function of time by the "blast humidity" rule. In addition, when the deposits on the furnace wall fall off, the "falling wall" rule automatically determines the drop-off point, the amount of influence on the furnace heat and the tuyere tip descent time, and determines the operation time and amount of the preliminary action to make a correction calculation. Incorporated into.

The inference engine (26) executes each of the above rules to obtain the necessary correction action amount and operation time.

(E) Comprehensive judgment KS; This comprehensive judgment KS is a knowledge base for comprehensively judging the amount of action to be taken based on the judgment results of (C) and (D). When the inference engine (26) infers this KS and obtains a determination result, the determination result is input to the operating state determination KS to determine the operating state, and the CR
The amount of action to be displayed is displayed on T (30) to give an instruction to the operator to give guidance, and at the same time, it is fed back to the digital instrumentation device (32) to perform predetermined automatic control (see FIG. 2).

Next, a method of creating a multidimensional function used for obtaining a CF value in each of the above knowledge bases (A) and (B), for example, a CF function of hot metal temperature-furnace heat level will be described.

FIG. 10 is a characteristic diagram showing the transition of the hot metal temperature in the tap (results of continuous measurement). As shown in the figure, even if the thermal condition in the furnace is stable, the effect of cooling the bottom of the furnace during the stay in the lower part of the furnace and the cooling of the hot metal in the hot metal tap, which is the hot metal flow path Initially, the hot metal having a relatively low temperature is discharged. Over time, the temperature of the hot metal discharged rises and stabilizes because the temperature of the hot metal is increased due to the sensible heat of the hot metal and the effect of residence in the furnace is reduced. The information will be well representative of the furnace heat that is controlled by the system.

Furthermore, since the tendency of this temperature rise is not always constant and changes depending on the operating conditions, the information on the hot metal temperature is ambiguous because its eligibility as information representative of the system control target changes moment by moment. Becomes Therefore,
In order to observe the hot metal temperature and link it to the estimation of the state of the furnace heat, it is necessary to consider the elapsed time from the start of tapping, and also to consider the ambiguity peculiar to the process. Will be needed.

From the above viewpoint, if we grasp the hot metal temperature during tapping,
The temperature distribution is represented as shown in FIG. In the figure, each axis means the X-axis; the dimensionless time Y-axis; the hot metal temperature Z-axis; the appearance frequency (frequency of occurrence). Figure 12 shows the tapping time X in Figure 11 = Hot metal temperature Ti in Xi
And the maximum hot metal temperature Tmi of the tap.
Based on this figure, the relationship between the tap hot metal temperature and the furnace heat,
By adjusting the relationship between the appearance frequency and the CF value, the three-dimensional function shown in FIG. 13 is obtained. Since the tapping temperature is measurement information that depends on the elapsed time from the tapping start and the operating conditions, more than 30 types are prepared so that they can be used properly, and are automatically selected according to the conditions.

The characteristic diagram of FIG. 13 is shown in FIG. 14 when the X axis is replaced with sensor data. In the figure, point A1
The CF value of the straight line connecting points A4 and A4 and the straight line connecting points A3 and A6 are "0", and the CF values increase toward the middle of both lines. The CF value shows the maximum value “1” on the straight line connecting the points A2 and A5, which indicates that the reliability is highest.

As a result of controlling the furnace temperature based on such a three-dimensional function,
As shown in FIG. 15, it can be seen that the frequency of error occurrence is reduced as compared with the conventional operator method.

The three-dimensional function of FIG. 14 is formed by connecting a minimum value and a maximum value in a predetermined range of sensor data with a straight line.
In Fig. 16, the area of sensor data is expanded, and it is divided into multiple parts according to the size of the sensor data (in this example, it is divided into 3), and the three-dimensional function is formed by connecting the broken lines so that each divided area is continuous. I am configuring. Such a three-dimensional function in FIG. 16 shows a more actual relationship between sensor data, furnace heat level and CF value (confidence) than the three-dimensional function in FIG. The maximum value of the CF value is "1" in each of the divided areas, but it goes without saying that the maximum value may be different for each area.

The reason why the sensor data of the above three-dimensional function is difference data (= measured value-target value) from the reference value is as follows.

B) The target value of the hot metal temperature changes due to component adjustment and other factors.

B) Regarding the sensor data, the standard is changed daily depending on the operation policy, for example, whether or not the low fuel ratio is aimed.

C) Further, the temperature of the thermometer or the like differs depending on the degree of wear of the installation position of the brick or the like, even if the conditions inside the furnace are the same.

For the above reasons, the three-dimensional function uses the difference data as a reference and can cope with various changes.

Figure 17 shows the relationship between action instructions, furnace heat level and hot metal temperature. The action instruction by the three-dimensional function in FIG. 14 is made at the timing shown by the solid line in the figure at the tap of, and the action instruction (dashed line) by the three-dimensional function in FIG. 16 is made at the timing shown by A and B in the figure. . As a result, in the case of using the three-dimensional function in FIG. 16, the hot metal temperature becomes the target value at the next tap. On the other hand, in the case of using the three-dimensional function shown in FIG. 14, it is slightly deviated from the target value.

FIG. 18 shows the actual results for the hot metal temperature, Si and S when the three-dimensional function of FIGS. 14 and 16 is used, and the one when the three-dimensional function of FIG. 16 is used. Excellent characteristics are obtained as compared with the case of FIG.

As a result, in the case of using the three-dimensional function in FIG. 16, the hot metal temperature becomes the target value at the next tap. 14th compared to this
In the case of using the three-dimensional function in the figure, it is slightly deviated from the target value.

FIG. 18 shows the actual results for the hot metal temperature, Si and S when the three-dimensional function of FIGS. 14 and 16 is used, and the one when the three-dimensional function of FIG. 16 is used. Excellent characteristics are obtained as compared with the case of FIG.

By the way, the operation of the above action determination KSC is explained when there is no wall drop, but in the present embodiment, the "wall drop rule" is in the action correction determination KS group D, and when there is a wall drop, "wall drop" is performed. Inference operations are performed based on the "falling rule", and both the action determination KSC determination and the action correction determination KS group determination are considered in the overall determination KS. The details will be described below.

As shown in FIG. 19, when the deposit (51) on the wall of the blast furnace (1) falls off, the sensor (52) installed in the blast furnace catches this as a rapid change in temperature and detects the drop. To do.

As shown in Fig. 20, the blast furnace (1) is equipped with sensors on the shaft and cooling board (hereinafter referred to as "CS"). There are 8 shafts in the circumferential direction of 5 to 1 stages, and 3 in CS.
16 pieces are attached in the circumferential direction in two steps.

For the output of each sensor, the average temperature (▲
▼) is calculated and a certain threshold value ΔT ° C / min, here 5
When there was a change of 0 ° C / min or more, it was judged that there was a wall drop. However, if the same sensor determines that a wall has fallen within a predetermined time, it is not added to the strength rank described below.

Here, by introducing the concept of strength rank,
Rank 3.

Strength 1; ▲ ▼ <T1j 2; T1j ≤ ▲ ▼ <T2j 3; T2j ≤ ▲ However, i; 1 to 8 (shaft); 1 to 16 (CS) j; 1 to 7 (height direction) CS 3 steps Taking the case of as an example, it becomes as follows (16th
See the right side of the figure).

Tij; 100 ° C T2j; 300 ° C ΔT; 50 ° C / min In Fig. 20, the correction coefficient corresponding to the number of wall drop judgments of each sensor is clearly shown. The reason for the large value (rank 1 has a larger correction coefficient than rank 2) is that detachment of deposits at low temperatures has a large effect. Furthermore, the relationship between the number of detections of dropouts and the CF value is as shown in FIG. 21, and it can be seen that the certainty factor is saturated at each predetermined number.

Next, the wall drop time T2 is calculated. This wall drop time T2 is the time from when the wall drop occurs (after detection) until it reaches the tuyere, and is calculated by the following equation.

ch; Charges per hour (times / hour) O / C; ore / coke weight ratio CB; 1 charge coke amount (T / ch) ρ _o , ρ _c ; ore, coke bulk density (T / m ³ ) α: Compression constant Vj; Volume between sensor and tuyere (m ³ ) Also, the effect of this wall drop becomes smaller as the adhering material gets closer to the tuyere, and the time T1 when the effect begins to decrease is past experience. It is determined from the following equation.

T1 = T2-αα; 60 minutes After the times T1 and T2 are obtained in this way, a correction action is obtained for each stage, taking into consideration the tuyere tip descent time.

Step correction action = (action amount) × {Rank 1 correction coefficient + Rank 2 correction coefficient + Rank 3 correction coefficient} The step correction actions described above are aggregated and grasped as the influence amount of wall fall. The action amount in the above equation is a constant obtained in consideration of theoretical calculation and empirical value.

For example, the influence amount of wall drop is 2.0g / Nm ^{3 in} terms of water injection ratio Moi.
(If it is considered that the amount of heat compensation to prevent the drop in hot metal temperature due to falling walls corresponds to a coke ratio of 2 kg / T, the water injection ratio input from the tuyere
Moi is equivalent to 2.0 g / Nm ³ . ), As shown in FIG. 22, the correction action amount is −2.0 g / Nm ³ , that is, the water injection ratio Moi
To obtain a correction amount that reduces 2.0 g / Nm ³ . Next, when it reaches the time T1 when the effect of wall fall begins to decrease (here, 60 minutes before T2), it is assumed that the effect level is 1.0g / Nm ³ assuming that the effect level is 0.5. . Therefore, −1.0 g / Nm ³ is obtained as the correction action amount.

In addition, in FIG. 22, in order to simplify the calculation, the time
Although it is assumed that the influence degree from time T1 to time T2 is uniformly 0.5, the influence degree is reduced according to a straight line connecting the influence degree “1” at time T1 and the influence degree “0” at time T2. It goes without saying that it is good.

The correction action amount obtained in the action correction determination KS group as described above is added to the action amount of the action determination KS group in the inference calculation based on the comprehensive determination KSE to obtain the operating value. At this time, since the correction action amount is a negative value, as shown in FIG.
The absolute value of the corrected action amount is subtracted from the action amount. The blast furnace is operated based on the operation values obtained in this way.

Figure 23 shows the relationship between the hot metal temperature and the water injection ratio, the solid line shows the control performance without consideration of wall drop, and the broken line shows the control performance with wall drop taken into consideration. Is. When the hot metal temperature is 1459 ° C, the effect of wall fall occurs, but as a measure against it, if the wall drop is not taken into consideration, Δt time (= 6 hours) after the effect appears. In addition, the water injection ratio was reduced from 44g / Nm ³ to 40g / Nm ³ . When considering wall drop, the water injection ratio is 44g / Nm ^{3 at} T0 hours (= 3 hours) after the wall drop occurs.
From 35g / Nm ³

In this way, when the wall drop is taken into consideration, it can be dealt with before the hot metal temperature changes, so the change in the hot metal temperature after the wall drop can be relatively small, and the deviation from the target temperature is small.

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

[Effects of the Invention] As described above, according to the present invention, processing data is created from various data installed in a blast furnace, and predetermined inference as artificial intelligence is made based on the processing data and a knowledge base based on experience and the like. Therefore, the conventional experience is fully utilized, standardization of operation management, prevention of human misjudgment, stable supply of high quality hot metal with little temperature and component fluctuations to the next process, avoidance of furnace cooling, labor saving Have been realized.

In addition, when the control device is realized by a computer, the knowledge base is used as a reference, so the calculation capacity and calculation speed are improved, and rules can be easily added and modified even in new situations such as aging of the blast furnace. It has become.

Since the belief function is used to calculate each furnace heat level and each furnace heat transition CF value, the number of rules and the work of inputting the numerical values are reduced.

Furthermore, since the next action amount is determined in consideration of the influence of the wall fall of the blast furnace, the control corresponding to the actual state of the furnace can be performed, which enables extremely accurate furnace temperature control. .

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of the present invention, FIG. 2 is a block diagram of a blast furnace heat control device according to an embodiment of the present invention,
FIG. 3 is a flow chart showing the knowledge base and its inference sequence, FIG. 4 is a characteristic diagram showing a three-dimensional function of the furnace heat level, and FIG. 5 is an explanatory diagram explaining the inference operation. FIG. 6 is a characteristic diagram showing a three-dimensional function of the tuyere transition level,
FIG. 7 is an explanatory view showing an example of the action matrix, FIG. 8 is an explanatory view showing an example of the type of the action matrix, and FIG. 9 is an explanatory view showing an example of the action amount. Fig. 10 is a characteristic diagram showing the transition of hot metal temperature in the tap.
FIG. 13 to FIG. 13 are explanatory views showing a method of creating a three-dimensional function of the furnace heat level. FIG. 14 is an explanatory diagram showing an example of a three-dimensional function, and FIG.
Fig. 14 is a characteristic diagram showing the results by the three-dimensional function, Fig. 16 is an explanatory diagram showing another example of the three-dimensional function, Fig. 17 is the three-dimensional function of Fig. 14 and the three-dimensional function of Fig. 16. The characteristic diagram showing the relationship between the control action instruction, the furnace heat level and the hot metal temperature by Fig. 18, Fig. 18 is the temperature difference from the target temperature showing the control results by the three-dimensional function of Fig. 14 and the three-dimensional function of Fig. 16, It is a characteristic view showing distribution of Si and S. FIG. 19 is an explanatory view showing the fall of the deposits on the wall of the blast furnace, FIG. 20 is an explanatory view showing the sensor attached to the blast furnace and its correction coefficient, and FIG. 21 is the number of detected wall drops and the CF value. Fig. 22 is a characteristic diagram showing the relationship, Fig. 22 is an explanatory diagram showing the relationship between wall drop and its effect, and Fig. 23 is a characteristic diagram showing the relationship between hot metal temperature and water injection when there is wall drop. FIG. 24 is a characteristic diagram showing the error occurrence rate of the hot metal temperature. (11), (12); data input means (16), (14); processed data creation means (19), (24); storage means (22); knowledge base storage means (26); inference means

Claims (1)

[Claims] 1. A data input means for taking in data from various sensors installed in a blast furnace at a predetermined timing, and a tuyere embedding temperature based on the data from the sensor,
Various data indicating the condition of the blast furnace including the unloading speed, pressure loss, furnace top temperature, gas utilization rate, and solution loss amount are created, and the various data are compared with the reference data, and the difference data is calculated. Processing data creating means to create, the various data and difference data (hereinafter referred to as processing data)
A storage means for temporarily storing, a knowledge base storage means for storing various knowledge bases based on experience, achievements, and mathematical model about blast furnace operation; processing data of the storage means and knowledge of the knowledge base storage means The knowledge base storing means estimates the furnace heat level, and has an inference means for inferring the furnace heat level and the furnace heat transition based on the base, and determining an action amount for the blast furnace, and a means for outputting the action amount. And a furnace heat level determination knowledge base used for estimating the furnace heat transition, and a furnace heat transition determination knowledge base used for estimating the furnace heat transition,
Action judgment knowledge base used to determine action amount from furnace heat level and furnace heat transition, action correction knowledge base that determines action amount correction value based on past actions and disturbances, action amount and action Including a comprehensive determination knowledge base for determining the actual action amount from the correction amount, the furnace heat level determination knowledge base and the furnace heat transition determination knowledge base, the processing data and the furnace heat level as independent variables,
The action determination knowledge base has a certainty function having a certainty of occurrence of these combinations (hereinafter referred to as certainty factor) as a dependent variable, and a rule group for determining an application method of the certainty function. There is a rule that determines the action amount from the level and the furnace heat transition, the comprehensive judgment knowledge base has a rule that determines the actual action amount from the action amount and the action correction amount, and further, the action correction knowledge base, It has a wall drop judgment rule that is used to detect the drop-off of deposits on the furnace wall, determine the drop-off position and influence amount, and infer the thermal compensation amount and timing based on the decision result. Blast furnace furnace heat control device.