FORMULATION AND SOLUTION TO CONTROL PROBLEMS OF OPERATION MODES IN DELAYED COKING UNIT IN FUZZY ENVIRONMENT BASED ON THE HEURISTIC METHOD

摘要

This paper presents a formulation to the optimal control problems of operation modes for technological systems in the fuzzy environment and proposes a heuristic approach for solutions to the introduced problems. For the purposes of formalization and mathematical problem formulation, we employ methods of fuzzy set theory and various compromise schemes, methods of multi-criteria selection and optimality principles. By modifying and adapting the methods of main criteria, maximin and the principle of Pareto optimality, new formulations to the optimal control problems are obtained and effective heuristic methods for their solutions are developed. The uniqueness and novelty of the proposed approach to solving the control problems of operation modes in the fuzzy environment provides more efficient solutions without prior conversions of a fuzzy problem into an equivalent deterministic problem, i.e. maximizes the use of initial fuzzy information. This is justified by the fact that, in contrast to well-known approaches, problems are set in a fuzzy environment without replacing them with equivalent deterministic tasks, and they are solved using the original fuzzy information. The authors' research contribution refers to the fact that for the first time mathematical formulations of optimization problems are formulated during the control of operation modes for a delayed coking unit (DCU), and heuristic methods for the problems are developed based on various optimality principles (main criterion, maximin and Pareto optimality). Moreover, the developed method based on maximin and Pareto optimality are employed in solving the problem of controlling the operation modes for delayed coking unit of the Atyrau Oil Refinery (AOR). We present and analyze the results of applying the proposed heuristic method-based maxi-min method and Pareto optimality principle, considering possible concessions during solving the problem of decision-making on the optimal control of operating modes for the delayed coking unit in Atyrau Oil Refinery, which works in the fuzzy environment. The obtained results of applying heuristic methods show the effectiveness of the proposed method for solving the control problem of operation modes for DCU in the fuzzy environment. In comparison with the results of known methods, the proposed method gives the best results, as well as calculates the membership functions of fuzzy constraints, i.e. allows controlling the extent to which the fuzzy constraints are met. For verifying the claims and the adequacy of the obtained decisions, we apply scientifically-based methods of multi-criteria optimization, decision-making methods, theories of fuzzy sets, the software package "Delayed coking unit" developed by participation of the author, modern computers with Pentium V processors, as well as real, i.e. experimental production data from the existing delayed coking unit.