Eventos Anais de eventos
COBEM 2019
25th International Congress of Mechanical Engineering
MULTIOBJECTIVE OPTIMIZATION OF HEAT EXCHANGER DESIGN THROUGH HEURISTIC KALMAN ALGORITHM
Submission Author:
Helon Vicente Hultmann Ayala , RJ , Brazil
Co-Authors:
Pedro Henrique Domingues, Leandro dos Santos Coelho, Roberto Zanetti Freire, Helon Vicente Hultmann Ayala
Presenter: Helon Vicente Hultmann Ayala
doi://10.26678/ABCM.COBEM2019.COB2019-1400
Abstract
Due to the constant presence of heat exchangers in a wide range of industries, the design optimization of heat exchangers draws the scientific community’s attention, which has explored the use of metaheuristics in the development of this type of equipment. In order to accomplish the heat exchanger design optimization, effectiveness and cost are generaly used as optimization objectives and owing to their conflicting characteristics, a multiobjective optimization approach is indicated. A new interesting multiobjective metaheuristic for non-convex constrained optimization problems is the Multiobjective heuristic Kalman Algorithm (MOHKA), which is attractive by its ease of implementation and the use of few adjustable parameters, being one of them the value. Therefore, in this work, MOHKA and Nondominated Sorting Genetic Algorithm version II (NSGA-II) were applied to the heat exchanger design optimization problem for a range of values, considering Spacing (S), Euclidean Distance (ED) and Hypervolume (HV) metrics to evaluate the best value and to compare its performance with NSGA-II. The preliminary results showed that NSGA-II presented a better performance than MOHKA, regardless of the value and that the best value for Plate-Fin and Shell-Tube heat exchanger design were 0.25 and 0.30, respectively, considering only the percentage of dominance of each method’s Pareto Front (PF) over the other PFs.
Keywords
