JISE


  [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]


Journal of Information Science and Engineering, Vol. 36 No. 6, pp. 1155-1166


Heat Exchanger Design using Differential Evolution-Based ABC


MOSTAFA A. ELHOSSEINI1,2
1College of Computer Science and Engineering in Yanbu
Taibah University
Yanbu, 30012 Saudi Arabia

2Department of Computers Engineering and Control Systems
Mansoura University
Dakahlia Governorate, 35516 Egypt
E-mail: mmaoustafa@taibahu.edu.sa; melhosseini@mans.edu.eg


The main purpose of this work is to develop a cost-effective design of the shell and tube heat exchanger (STHE). The STHE objective function to be minimized is the total cost of STHE, which is a function of the surface area of the heat transfer and pressure drop at both tube and shell side. Artificial Bee Colony (ABC) is a robust population-based swarm optimization algorithm with a few numbers of control parameters. Slow convergence and poor exploitations of ABC may cause solutions to be stuck in local minima. Differential evolution (DE) is arguably one of the most potent stochastic real-parameter optimization algorithms in current use. Compared to most other EAs, DE is much simpler and more straightforward to implement. Despite its simplicity, DE exhibits much better performance in comparison with several others on a wide variety of problems, including unimodal, multimodal, separable, non-separable, and so on. Besides, the number of control parameters in DE is very few, and the space complexity of DE is low as compared to some of the most competitive real parameter optimizers. These features help in extending DE for handling large scale and expensive optimization problems. Hybridizing ABC with DE seems a reasonable suggestion to combine the merits of both resulting in proposed Hybrid ABC DE (HABCDE). The HABCDE is compared against five algorithms using two different cases with a different number of passes, pitch type, and fluid type. The results show that HABCDE gets the minimum total cost. Total cost decreases by a percentage ranging from 22.29% to 0.93%, compared to other algorithms.


Keywords: artificial bee colony (ABC), cost optimization, differential evolution (DE), shell and tube heat exchanger (STHE), swarm optimization, particle swarm optimization, genetic algorithm

  Retrieve PDF document (JISE_202006_01.pdf)