KR19980019613A - Furnace level diagnosis device using expert system and its method - Google Patents

Abstract

The present invention relates to an apparatus and method for diagnosing a furnace level using an expert system, and more particularly, to diagnose a current furnace level of a blast furnace, to predict future furnace trends, and to use the expert system to infer an appropriate action of an operator based on this. The present invention relates to an apparatus and a method for diagnosing a current level.

The present invention identifies the most important current furnace level diagnosis in a furnace control expert system that guides an operator's action to maintain an appropriate furnace by using the current furnace level determined by the present invention and the furnace trend deduced by the knowledge base of the expert system. By the method of the present invention, the operator judges that the furnace level is expressed in quantitative numerical value by the method of the present invention, which not only improves the degree of the furnace control expert system but also judges the state of the furnace vaguely based on the experience of a conventional skilled worker. It is possible to prevent misoperation of workers and to manage the furnace by relying on the system, so that anyone can understand the current state of blast furnace. It has a special effect.

Description

Furnace level diagnosis device using expert system and its method

The present invention relates to an apparatus and method for diagnosing a furnace level using an expert system, and more particularly, to diagnose a current furnace level of a blast furnace, to predict future furnace trends, and to use the expert system to infer an appropriate action of an operator based on this. The present invention relates to an apparatus and a method for diagnosing a current level.

In general, a blast furnace is a huge reaction in which raw iron ore and fuel coke are charged from the top of the furnace into the furnace, hot air flows into the tuyere to combust the coke and heat, reduce and dissolve the charges by the reducing gas generated therefrom. As a container, pig iron and slag are discharged through the outlet and the reaction gas is recovered to the top. The most important factor for stable operation management of this blast furnace is furnace management, that is, maintaining the thermal balance in the furnace. The operator judges the furnace heat by using the data provided at regular intervals from the sensor installed in the blast furnace and the short-term data provided by the process computer in order to maintain proper furnace heat, and based on the judgment, moisture, wind temperature, pulverized coal, coke sbi The same actions have been taken, but due to the complexity of the blast furnace process, the use of short-term data, unstructured formulas, and the blast furnace process are not enough for long-term data analysis to diagnose the level of heat. In some cases, maintenance failures often occur, causing a problem of lowering productivity.

The present invention has been made to solve the above problems and to achieve an improvement.

Accordingly, an object of the present invention is to accurately diagnose the current heat level, which is the most important element of the heat control expert system, and to prevent misoperation due to the simple judgment of the operators, thereby maintaining the proper heat. To provide a level diagnostic apparatus and method thereof.

As a technical means for achieving the above object of the present invention, the furnace level diagnostic apparatus using the expert system of the present invention is a sensor data including the molten iron temperature, the moisture addition amount, the blowing temperature, the pulverized coal injection amount and the coke charge amount from the blast furnace Instrumentation system for receiving the input to the process computer; A data collection unit for collecting sensor data from the instrumentation system, a data preprocessing unit for preprocessing the collected data, a data storage unit for storing the preprocessed data, and a computer dedicated to an expert system for transmitting and receiving data. A process computer including an interface portion; An interface unit for receiving and transmitting data from the processor computer, an inference engine for inferring an evaluation chart temperature and a heat balance according to a knowledge base of the data received from the processor computer, and a blast furnace operator's experience and actual business analysis. Based on the knowledge base that stores various knowledge data and belonging function and fuzzy rules for fuzzy control, and the belonging function for evaluation chart temperature and heat balance, fuzzy theory is performed to determine the heat level. An expert system dedicated computer (4) comprising a fuzzy controller for controlling the output of the screen; And a guide screen for outputting a result from the expert system dedicated computer on a screen.

In addition, as another technical means for achieving the object of the present invention, the furnace level diagnostic method using the expert system of the present invention is the melting temperature of the blast furnace, the moisture addition amount, the cowung temperature, fine coal injection amount and coke loading amount through the instrumentation system A first step of collecting, by the process computer, sensor data including the data, and transmitting the sensor data to the expert system dedicated computer 4 after the preprocessing process; A computer dedicated to expert systems for receiving data from the process computer corresponds to one of dozens of chartered use conditions, which are measured in topedoca units, and then to a chartered evaluator according to the method for determining the chartered temperature. Determining a temperature; The expert system dedicated computer includes a third step of calculating a thermal balance within a furnace by adding a singular information including a coke ratio converted per unit action amount, a couss charge amount, a difference between a current action value taken by an operator and a previous action value; And a fourth step of calculating and displaying a aging level by applying the fuzzy inference rule in the knowledge base using the belonging function of the evaluation chart temperature and the heat balance in the second and third steps as input variables. Characterized in that made.

1 is a configuration diagram of a conventional furnace level diagnostic apparatus.

2 is a configuration diagram of a furnace level diagnostic apparatus according to the present invention.

3 is a flowchart showing a method for diagnosing a furnace level using an expert system according to the present invention.

4 is a classification diagram for the molten iron temperature pattern in accordance with the present invention.

5 is a view showing the function of belonging to the charter temperature and heat balance and the heat level according to the present invention.

Explanation of symbols on the main parts of the drawings

1: blast furnace 2: instrumentation system

3: Process computer 4: Computer for expert system

5: data collection unit 6: data preprocessing unit

7: Data storage unit 8: Interface unit

9: Reasoning Engine 10: Knowledge Base

11: Purge Controller 12: Guide Screen

Hereinafter, with reference to the accompanying drawings a preferred embodiment of the furnace level diagnostic apparatus using an expert system according to the present invention. In the drawings referred to in the present invention, components having substantially the same configuration and function will use the same reference numerals.

2 is a configuration diagram of a furnace level diagnosis apparatus according to the present invention. Referring to FIG. 2, a furnace level level apparatus using the expert system of the present invention is a molten iron temperature, a moisture addition amount, a blowing temperature, and pulverized coal from the blast furnace 1. The instrumentation system 2 which receives sensor data including the blowing amount and the coke charging amount and transmits the sensor data to the process term computer, the data collection unit 5 which collects sensor data from the instrumentation system 2 and collected data. A process computer 3 comprising a data preprocessor 6 for preprocessing, a data storage 7 for storing the preprocessed data, a computer dedicated to expert systems, and an interface 8 for transmitting and receiving data; The interface unit 8 for receiving and transmitting data from the processor computer 3 and the data received from the processor computer 3 according to the knowledge base 10 and the evaluation chart temperature; Inference engine (9) for inferring balance, knowledge base (10) which stores various instructional data based on the experience and actual industry analysis of blast furnace operators, fuzzy control function and fuzzy rules for fuzzy control, and the above-mentioned evaluation chart temperature and heat Dedicated expert system computer 4 and fuzzy controller 11 which performs fuzzy theory based on the belonging function for balance to determine the heat level and controls the screen output for the determined heat level result. The guide screen 12 which outputs the result from the computer 4 to a screen is comprised.

FIG. 3 is a flowchart showing a method for diagnosing furnace level using an expert system according to the present invention. FIG. 4 is a classification diagram for an evaluation chart temperature pattern according to the present invention, and FIG. 5 is an evaluation chart temperature and heat according to the present invention. The figure shows the membership function for balance and aging level.

Operation according to the device of the present invention configured as described above will be described in detail below based on the accompanying drawings.

Referring to FIG. 2 to FIG. 5, the present invention will be described firstly, in the first step, the molten iron temperature, the moisture addition amount, the bowing temperature, the pulverized coal injection amount, and the coke in the blast furnace 1 through the instrumentation system 2. The sensor computer including the charge amount is collected by the process computer 3 in the data collection unit 5.

The data sensed by the accounting system 2 includes about 700 pieces of data such as temperature, pressure, gas components, etc. by the data collection unit 5 of the process computer 3. 240 data were selected through correlation analysis of historical performance data from operation index calculations such as aeration index and sensitive data of thermal changes among collected data, and persistence, circumferential balance, manual balance, standard deviation, and correlation The preprocessing of the data is performed in the data preprocessing section 6 and stored in the data storage section 7 and transmitted to the expert system dedicated computer 4 via the interface section 8.

In the second step 32, the expert system dedicated computer 4, which receives data from the process computer 3, corresponds to the molten iron use state divided into dozens of forms of the molten iron use measured in topedoca units. In this case, the evaluation charter temperature is determined according to the method for determining the charter temperature.

Specifically, the expert system dedicated computer 4 uses this data to determine the current heat level, heat trend, and proper action every 10 minutes, and guide the result to the operator. An appropriate axer is taken into the blast furnace. To accurately diagnose the current furnace level of the blast furnace, the evaluated coolant temperature and operator action thermal balance are calculated for the fuzzy theory.

First, the molten iron temperature is calculated using the molten iron temperature measured in units of topedoca (container for repairing and moving molten iron). The difference in temperature of the chartered ship is generated according to the blast furnace furnace situation and the exit port. In order to minimize the change in the molten iron temperature due to the change of the external conditions such as the change in the situation of each tapping line and to use the change of the molten iron temperature according to the change of the actual furnace condition as the evaluation molten iron temperature as shown in FIG. After the repair of the pioneer, it is classified into 24 types in consideration of whether or not there is a deviation of the molten iron temperature by the molten iron, whether or not the recently measured molten iron temperature is low, and the molten iron temperature is measured.

If the molten iron temperature is determined to be one of the 24 types shown in FIG. 3, the molten iron temperature is calculated in accordance with one of the methods shown in Table 1 below.

In the case of methods 4 and 5 in Table 1, Ttap (1) is the first, that is, the highest molten iron temperature in descending order of the last nine outgoing average temperatures, and Ttap (9) is the last temperature, that is, the lowest. Temperature. Ttlc is the latest molten iron temperature measured in topedoca.

Next, in the third step 33, the expert system dedicated computer 4 sums up a plurality of pieces of information including the difference between the annular cokes ratio, the coss loading amount, the current action value taken by the operator and the previous action value per unit action amount. Calculate the thermal balance in the furnace.

On the other hand, the furnace heat balance by the operator action is calculated by the following equation (1).

Thermal Balance = ∑ (αi × △ Xi × Ti) ------------- (1)

Here, α is the coke ratio converted per unit action amount of the furnace control action particles, and i is an operator's action that affects the furnace heat, and the coke loading amount, wind temperature, pulverized coal injection amount, and moisture addition amount are added. ΔX is the difference between the current action value and the previous action value taken by the operator as the amount of deviation of the heat control action factor.

T contained in said Formula (1) is calculated | required by following formula (2) as reaction efficiency of a heat control action factor.

Here, Td is obtained as the elapsed time of the recent action taken by the operator (current time-current action time), Tb is the time when the action taken by the worker starts to affect more than 10% of the heat, and Tf is the action of the worker. It is time to affect 100% of this furnace.

Finally, in the fourth step (34, 35) using the fuzzy inference rule in the knowledge base (10) by using the belonging function of the evaluation chart temperature and heat balance in the second and third steps as input variables The furnace level is calculated and displayed on the guide screen 12.

Because the reaction in the blast furnace (1) is complicated, it is impossible to quantitatively calculate the level of heat by using a simple arithmetic equation for the relationship between the molten iron temperature and the heat balance in the furnace. The current heating level is diagnosed by using a fuzzy technique, which is calculated by evaluating the molten iron temperature (standard molten iron temperature) and the heat balance by the two-man action as the input variable, and the current heating level is the output variable. Fuzzy inference of two inputs and one output is performed.

As shown in Figs. 5 (a), (b), and (c), the six temperature-dependency functions are defined for the molten iron temperature from very low to very high, and the heat balance is defined by five membership functions from low to high. Define. In addition, the output level of frost level is stored in the knowledge base by defining five membership functions ranging from very low to very high. For example, fuzzy inference is carried out using fuzzy rules such as the following fuzzy rule example. will be.

Fuzzy Rule Example:

If the melt temperature is very frequent and the heat balance is low, the furnace level is very low.

According to the present invention as described above, a furnace control expert system for guiding an operator's action to maintain proper furnace heat using the current furnace level determined by the method of the present invention and the furnace trend deduced by the knowledge base of the expert system. In the present invention, by performing the diagnosis of the current heating level which is the most important in the method of the present invention, it is possible not only to improve the degree of the thermal control expert system but also to quantitatively determine the heating state by the experience of a conventional skilled worker. By using the furnace level expressed by the operator in the furnace judgment, it is possible to prevent the operator's misjudgment and implement the furnace management based on the system so that anyone can understand the current furnace condition of the blast furnace. Standardization of operation management, There are special effects.

The above description is only a description of one embodiment of the present invention, the present invention is capable of various changes and modifications within the scope of the configuration. In addition, it will be apparent to those skilled in the art that the apparatus and method for diagnosing a furnace level using the expert system are not limited to the above-described embodiment.

Claims (2)

An instrumentation system (2) which receives sensor data including the molten iron temperature, the moisture content, the blowing temperature, the fine coal injection amount and the coke charge amount from the blast furnace 1 and transmits it to the process computer; A data collection unit 5 for collecting sensor data from the instrumentation system 2, a data preprocessing unit 6 for preprocessing the collected data, a data storage unit 7 for storing the preprocessed data, A process computer 3 comprising an expert system dedicated computer and an interface section 8 for transmitting and receiving data; Interface unit 8 for receiving and transmitting data from the processor computer 3, and an inference engine for inferring the temperature of the charter temperature and heat balance from the data received from the processor computer 3 according to the knowledge base 10 ( 9), the knowledge base (10) which stores various knowledge data based on the blast furnace operator's experience and the analysis of unemployment, and the fuzzy control function and fuzzy rules for fuzzy control, and the membership function for the evaluation chart temperature and heat balance. An expert system dedicated computer (4) comprising a fuzzy controller (11) for performing a fuzzy theory to determine the heat level and to control the screen output for the determined heat level result; And a guide screen (12) for outputting the results from the expert system dedicated computer (4) to a screen. Through the instrumentation system (2), the process computer (3) collects the sensor data including the molten iron temperature, the moist moisture addition amount, the bovine temperature, the fine coal injection amount and the coke charge amount of the blast furnace 1, and after the pretreatment process, the expert system A first step 31 of transmitting to the dedicated computer 4; The expert system dedicated computer 4, which receives data from the process computer 3, corresponds to the molten iron use state measured in topedoca units in one of the dozens of different molten iron use states. A second step 32 of determining an evaluation molten iron temperature according to a determination method; The expert system dedicated computer 4 includes a third step of calculating a thermal balance within the furnace by summing a plurality of pieces of information including a converted coke ratio per unit action amount, a coss charge amount, a difference between the current action value taken by the operator and the previous action value ( 33); A fourth step of calculating and displaying a frost level by applying a fuzzy inference rule in the knowledge base (10) using the belonging function of the evaluation chart temperature and the heat balance in the second and third steps as input variables; (34, 35); Depth level diagnosis method using an expert system, characterized in that consisting of.