Modern lines of development of metallurgy are characterized by working out, introduction and wide use of information systems and technologies. A basis of information technologies and systems are computers and the computer networks having the richest software, and also control systems of databases, computer decision support systems which methodological basis on theory of systems and the system analysis, the theory of data modeling, technological processes and knowledge.
The hot stove, is a very large batch type of heat exchange, and generates the hot blast for the blast furnace. The purpose of hot stove combustion control lies in supplying blast furnace with a hot blast of the prescribed temperature and the highest thermal efficiency, and also in maintaining the temperature of the refractor bricks of the hot stove within a certain temperature range for optimum durability. Operating conditions which satisfy such demands can be theoretically obtained by calculating in-and-out heat quantities. In reality, however, process data that indicate the characteristics of a hot stove such as the blast volume passing through cannot be accurately measured, and the consequent difficulty in correctly defining the operating condition prevents a model from producing satisfactory results. In contrast, a skilled operator can provide satisfactory control by inferring the heat condition of the hot stove from certain process data. In analogy, a control model that expresses the know-how of an operator by a type of fuzzy rule has recently been applied, to several other blast furnaces. The hot stove combustion control system that is used for furnace is a hybrid type consisting for a physical model applying the heat balance calculation, and an expert system applying fuzzy inference.
The physical model is activated when the blast volume and blast temperature change significantly. The optimum new operating conditions for the hot stove are then calculated to provide the highest thermal efficiency and lowest operating coast. If the characteristics of the hot stove are the same as those represented by the physical model, operations according to the conditions obtained from the physical model should achieve the best result. In practice, however, the operation conditions are changed. The control model using fuzzy inference is then activated to meet these unplanned situations before the stove commence combustion after completing blasting, and checks the heat condition of the stove and the brick temperature, thereby correcting control parameters to maintain the most suitable operating conditions.
The configuration of the fuzzy rules in the present control model is shown in Fig.1 in the first-stage fuzzy group, an index is obtained that evaluates the heating condition of the stove (the heat index), using as inputs, the opening of the mixed blast butterfly valve when single blowing of the hot stove occurs at change-over and the opening of the cold blast butterfly valve just before blasting ends. For vale opening, the heating states of two stoves are shown in a synthesized shape, but sing the present rule group is required to judge the heating condition of a single stove, four items of information about valve opening are input to form a rule that can infer the heating condition of the stove being considered. In the second-stage fuzzy rule group, the following inputs are used: the heat index obtained by the first-stage inference, the brick temperature of the object stove at the end of blowing, and the change of brick temperature during times of cycles in the past. The required adjustment to the control parameters is then obtained by fuzzy inference. The output from this rule group are: the change of input heat the target colorific value of the mixed gas, the dome temperature, and the change of opening of the mixed blast butterfly valve at the time of two stoves blowing.
Figure 1 - The configuration of the fuzzy rules
Conclusion In this work, we have designed and implemented a Development expert system of Decision Support System to control of blast-furnace air in blast-furnace production by using a fuzzy logic.
1. Спирин Н. А., Лавров В.В. Информационные системы в металлургии: Конспект лекций (отдельные главы из учебника для вузов). - Екатеринбург: УПИ, 2004. – 495 с.
2. Паклин Н. В. Нечеткая логика - математические основы [Internet resource] : http://www.basegroup.ru/library/analysis/fuzzylogic/math/
3. Лобатый А. А. Современные методы синтеза систем управлени [Internet resource] : http://abitur.bsuir.by/eumk/smssu/work_plan.html
4. American Iron and Steel Institute [Internet resource] : http://www.steel.org
5. Рамм А.Н. Современный доменный процесс. – М.: Металлургия, 1989. – 304 с.
6. Спирин Н.А., Лавров В.В., Шаврин В.С. Оптимизация, идентификация и оценивание теплотехнических процессов в металлургии: Учебное пособие для вузов. – Екатеринбург: УГТУ, 1996. – 188 с.
7. Трахтенгерц Э.А. Компьютерная поддержка принятия решений: Научно-практическое издание. – М.: СИНТЕГ, 1998. – 376 с.
8. Ченцов А.В., Чесноков Ю.А., Шаврин С.В. Балансовая логико-статистическая модель доменного процесса. – М.: Наука, 1991. – 92 с.
Important remark: during the writing this abstract the master's work wasn’t finished yet. The final ending is planning on December, 2009. The complete text of work and all the information about work you can receive from the author or her scientific adviser after the mentioned date.