Methods for developing models and decision making in a fuzzy environment for effective control of the coking process

Abstract

The paper studies the problems of developing fuzzy and linguistic models of poorly formalized production facilities using available fuzzy information and other types of data. Since known methods do not allow us to effectively solve these problems in a fuzzy environment, we propose a systematic method and a heuristic approach to decision making for the effective objects control in a fuzzy environment. The proposed system method makes it possible to develop fuzzy, linguistic models of complex objects, and a heuristic approach to the formulation and solution of the adoption problem - to adequately and effectively solve the problem of fuzzy described objects control. In the proposed heuristic method of fuzzy decision making, effective control of the coking process is ensured through the maximum and effective use of fuzzy information representing the experience, knowledge and intuition of the decision maker, experts. Using the proposed system method, fuzzy and linguistic models of the studied coking reactors of the delayed coking unit of the Atyrau refinery were developed. A computer modeling system has been created that allows, as a result of modeling, to select the optimal operating modes of coking reactors. Based on the developed models using the proposed heuristic decision-making method, modifying and combining the principles of maximin and Pareto optimality, the problem of effective control of the delayed coking process was successfully solved. Comparison of the obtained results with the results of known methods and real data shows the advantages, efficiency and high adequacy of the proposed methods in a fuzzy environment. The practical significance of the results obtained is that they can be adapted and successfully applied in modeling and decision making for the effective complex objects control in a fuzzy environment.