JPH0663005B2 - Blast furnace operation method - Google Patents

Description

The present invention relates to a method of operating a blast furnace using a knowledge engineering system.

[Conventional technology]

Blast furnace operation consists of a large number of operation factors that are mutually related, and it is difficult to directly visually monitor the inside of the furnace from the equipment conditions, etc. In order to achieve this, it is necessary to comprehensively judge the information from the sensors installed in the blast furnace and control it appropriately.
For this reason, the experience and knowledge of operators are still important for the daily operation management of blast furnaces.

Since the knowledge engineering system can take in such human know-how into a computer and process it, it is not possible to use the knowledge engineering system.
The introduction of a knowledge engineering system for blast furnace operation management as shown in Japanese Patent Laid-Open No. 708 and Japanese Patent Laid-Open No. 62-270712 is being promoted. By systematizing operation management, problems such as oversight of information and erroneous judgments will be eliminated, and operation management will be optimized and standardized.

[Problems to be Solved by the Invention]

In the knowledge engineering system disclosed in JP-A-62-270708 and JP-A-62-270712, as a result obtained by inference, in a blast furnace such as predicting blow-through and slip and controlling furnace heat. A knowledge base for partial items of the phenomenon is established, and inferences are made independently of each other. However, in-furnace phenomena such as aeration, load drop, furnace heat, etc. occur as one process in the blast furnace, and these individual phenomena are comprehensively considered in the operation management system of the blast furnace. It is necessary to judge and link to action. Furthermore, as the action of actual operation, there is a retreat action (escape action) such as increase of fuel ratio or reduction of air flow to avoid abnormal reactor conditions, and escape action such as reduction of fuel ratio due to operational stability. Although there is a return action or an operation level improvement action (aggressive action) for improving the operation level, the conventional system has a problem that it cannot cover all of the daily operation management.

Then, this invention is made in order to solve such a problem, and an object of this invention is to obtain the operating method of the blast furnace based on the daily operation management which made all the daily operations into the protection range by the knowledge engineering system. .

[Means for Solving the Problems]

The operation method of the blast furnace according to the present invention is (1) inferring the in-furnace condition of the blast furnace using knowledge engineering, and performing operation action based on the result, in the first step, the operation fluctuation of various blast furnaces is analyzed. Prediction is performed to judge the situation inside the reactor by the defense rule group that derives the backward action such as increase in fuel ratio and decrease in air flow rate, and when the situation requiring action occurs in the defense rule group, the action instruction is given to infer If no action occurs in the defensive rule group, the situation inside the reactor is judged by the attack rule group that leads to the operation level improvement action such as fuel ratio reduction due to the stability of operation as the second step. , If an action-needed situation occurs in the attack rule group, instruct action is given and inference is stopped, and if the action does not occur in the attack rule group, the current state is maintained Infer furnace conditions by construction and the knowledge base as a method of operating a blast furnace, characterized in that to perform the operation action based on the results and
(2) A distribution improvement rule that determines the appropriate charge distribution action by determining the possibility of improving reaction efficiency and reducing heat dissipation from the furnace body as the third step when action does not occur in the attack rule group The situation of the reactor is judged by the group, and when the situation requiring action occurs in the distribution improvement rule group, the action is instructed and the inference is stopped, and the current state is maintained when the action does not occur in the distribution improvement rule group. The method according to the above item 1 is configured as described above.

[Action]

In the present invention, after making an escape action judgment for avoiding a reactor condition abnormality by the defense rule group, making an offensive action decision for returning the escape action and improving the operation level by the attack rule group However, it is desirable to make a distribution improvement action judgment based on a distribution improvement rule group so as to obtain an operation state in which an aggressive action can be performed, so that it is possible to cover all of the daily operations.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings. First
The figure is a schematic flow diagram of inference by a knowledge base of the present invention. When the inference start 1 starts, the operation variation hypothesis inference 2 is first executed. The presence / absence 3 of the action in the inference result is determined, and if there is an action, the backward action instruction 4 is given and the inference is stopped 12. If there is no action based on the determination of action presence / absence 3 in the inference result, the operating margin hypothesis inference 5 is executed. The presence / absence 6 of the action in the inference result is determined, and if there is an action, the attack action instruction 7 is given, and the inference is stopped 12. If there is no action, it is possible to return to the original state by returning to the original state by judging the action presence / absence 6 based on the inference result.
To execute. The presence / absence 9 of the action in the inference result is determined, and if there is an action, the distribution improvement action instruction 10 is given and the inference is stopped 12. If there is no action based on the determination of action presence / absence 9 in the inference result, the current state maintaining instruction 11 is given and the inference is stopped 12. The retreat action instruction 4, the attack action instruction 7, the distribution improvement action instruction 10, and the current state maintenance instruction 11 may be a message output to the display terminal or may be control data transmission to the process computer.

Each inference will be described in more detail below. FIG. 2 is an explanatory diagram of the contents of the operational fluctuation hypothesis inference 2 of FIG. When the central gas flow is weak, the ventilation resistance in the furnace rises, and the furnace brick temperature rises, the central flow shortage operation fluctuation 15 occurs,
Based on the operation variation hypothesis 14 and the data 13 such as the detection end, which represent the phenomenon that the peripheral flow shortage operation variation 16 occurs when the ambient gas flow is weak, the heat level is reduced, and the furnace brick temperature is reduced, The inference 17 on whether or not the in-reactor situation corresponds to the above hypothesis is executed, and the retreat final determination 18 is performed.

FIG. 3 is an explanatory diagram of the contents of the operating margin hypothesis inference 5 of FIG. Gas flow distribution is proper, heat level is high, ventilation resistance is low and stable, furnace brick temperature is high, load drop is stable, and there is a margin in operating level. Based on the operating margin hypothesis 19 that expresses knowledge and the data 13 such as the detection end, the current state is inferred 17 and the final attack determination 20 is performed.

FIG. 4 is an explanatory diagram of the contents of the distribution improvement hypothesis reasoning 8 of FIG. When the central gas flow is strong, it is judged that the central gas flow can be suppressed by increasing the amount of ore charged to the central part 22 in order to improve the reaction efficiency in the central part, and similarly, the intermediate gas flow is strong. When the intermediate gas flow can be suppressed23, when the surrounding gas flow is strong,
Based on the distribution improvement hypothesis 21 configured to determine that the peripheral gas flow can be suppressed 24 and the data 13 such as the detection end, the inference 17 is executed and the distribution improvement final determination 25 is performed.

An example of blast furnace operation by the system configured as described above is shown in FIG. The distribution situation has become short of peripheral flow from 9 ° (9:00) of the day, and the actual reactor action was carried out in accordance with the instruction of peripheral fluidization distribution adjustment by swinging the ore to shift the charging position of the ore to the center side. . Since then, the gas flow distribution has become appropriate from 13 °. The temperature of the belly bricks has dropped from around 10 ° due to a temporary shortage of peripheral flow, but since the peripheral fluidization distribution adjustment has been implemented, the decrease has been stopped at 13 °, and has since turned to an upward direction. Around 13 °, it is predicted that the reactor heat will fall due to the group of defense rules, and actions are being taken in accordance with instructions to increase heat by increasing the fuel ratio by 5 kg / tp. As a result, the drop in hot metal temperature stopped at 15 ° and recovered thereafter. Due to the recovery of the furnace heat condition and the stabilization of other operating conditions, the 23:00 attack rule group issued an instruction to reduce heat by reducing the fuel ratio by 5 kg / tp in order to return the heat increase action, and the action is being implemented. 2 more
At 8 o'clock on the day, it was judged by the attack rule group that there was a margin in the heat condition of the reactor, and actions were taken in accordance with the instruction to reduce heat by reducing the fuel ratio by 2 kg / tp. At 17 °, the ventilation condition and furnace heat condition are not in the state of performing receding and attack co-action, but it is judged by the distribution improvement rule group to be in the peripheral flow strength state, and the surrounding by the ore swing An instruction is given to improve the flow restriction distribution and actions are being taken. As a result, the gas flow distribution status has returned to the proper level again.

In this embodiment, inference based on the knowledge base of the present invention is carried out at a cycle of 10 minutes, actions in the figures 1 to 3 lead to action occurrence, and all other inference results are the current state.

〔The invention's effect〕

When operating a blast furnace with a knowledge engineering system, the knowledge base has a hierarchical structure that executes inference in the order of defense rule group, attack rule group, distribution improvement rule group, not only abnormal prediction but also retreat action Later return,
Most of the daily operation judgments such as actions to improve the operation level can be incorporated into the system with a structure that is easy to read, and the use of this system improves the operation level by speeding up and making accurate operation judgments. Is planned.

[Brief description of drawings]

FIG. 1 is a schematic flow chart of inference by a knowledge base of the present invention, FIG. 2 is an explanatory diagram of contents of inference of operation fluctuation hypothesis in FIG. 1, and FIG. 3 is contents of inference of operation margin hypothesis in FIG. FIG. 4 is an explanatory view of the contents of the distribution improvement hypothesis inference in FIG. 1, and FIG. 5 is an explanatory view showing an embodiment of blast furnace operation by the system of the present invention.

Claims (2)

[Claims] Claim: What is claimed is: 1. In a method of inferring the internal condition of a blast furnace by using knowledge engineering and performing an operational action based on the result, the first step is to predict the operational fluctuation of various blast furnaces and increase the fuel ratio or The in-reactor situation is judged by a group of defense rules that induces a backward action such as a reduction in the amount of blown air, and when a situation that requires action occurs in the group of defense rules, an action instruction is given to stop the inference and the rules for defense. If action does not occur in the group, as a second step, the in-reactor situation is judged by the attack rule group that derives the action level improvement action such as the fuel ratio reduction due to the stability of the operation, and the action is performed by the attack rule group. When a necessary situation occurs, an action instruction is given and inference is stopped, and if the action does not occur in the attack rule group, the current state is maintained. Knowledge base by infers furnace conditions, a method of operating a blast furnace, characterized in that to perform the operation action based on this result. 2. A distribution for deriving an appropriate charge distribution action by judging possibilities such as improvement of reaction efficiency and reduction of heat dissipation from the furnace body as a third step when action does not occur in the attack rule group. Judgment of the in-reactor situation based on the improvement rule group, when an action-needed situation occurs in the distribution improvement rule group, the action is instructed to stop reasoning, and the current state is maintained when no action occurs in the distribution improvement rule group The method of operating a blast furnace according to claim 1, wherein the method is as follows.