Enhanced Krill Herd Algorithm for Chemical Engineering Problems

บทคัดย่อ

Krill Herd algorithm, Meta-Heuristic optimization, was introduced and wildly applied to many fields. However, Meta-heuristic sometimes got stuck in the local trapped. This article proposes a new technique to enhance the Krill Herd algorithm to apply to various chemical engineering problems. It focuses on balancing between the exploration and exploitation operator for escaping the optimum trapped by using the genetic algorithm named crossover operator to update a new position of the krill and also use the checking method to prevent the unexpectable error from running program. This new technique is the so-called Enhanced Krill Herd algorithm (EKHA). In this research, seven benchmark problems have validated the performance of the developed algorithm. In addition, EKHA also brought to solve the real working field. Four chemical engineering problems with different design conditions were selected to test the performance. The obtained results from the proposed method and standard krill herd algorithm show the achievements reaching the optimum results. This EKHA gave more accuracy comparing with the exact solution from the exact algorithm of the problem. The only weakness of the proposed algorithm is required more computational time for operating because of the checking method. The comparison between the two methods can prove the efficiency and superiority of the proposed algorithm.