| dc.description.abstract | In this work, a new approach to optimizing the design of a shell and tube heat exchanger is developed. The Grey Wolf Optimizer algorithm (GWO) is applied to minimize the objective total cost function of the proposed heat exchanger. The optimization variables adopted are: number of tubes passes, tube outside diameter, shell inside diameter, baffles spacing and baffle cut. The Bell-Delaware method and the Kern method are used to calculate the heat transfer coefficient and the pressure drop from the shell side. The results of the optimization are compared with the original design and with other optimization algorithms in the literature. In addition, the GWO algorithm is compared with other optimization meta-heuristics in three different test functions. The results of the comparison of the GWO algorithm in the test functions show a competitive performance compared to the Genetic Algorithm, Particle Swarm Optimization, Differential Evolution algorithm and Firefly Algorithm. Already, the results of the optimization of the heat exchanger using the Kern method shows a good performance, with the reduction of capital investment by 11.95%, 7.93%, 6.24%, 2.37% and 0.19% compared to the original project, GA, ICA, PSO and GSA, respectively. In addition, the total discounted operating cost was lower than the original project and the rest of the metaheuristics except for the GSA algorithm where the GWO algorithm obtained 22.33% higher result. Overall, the combined reduction in capital investment and total discounted operating cost obtained by applying the GWO algorithm led to a total cost reduction of 20.80%, 7.28%, 6.07% and 4.06% compared to the original project, GA, ICA and PSO, respectively.
Finally, the results of optimization of the heat exchanger using the Bell-Delaware method compared to the original design show satisfactory performance with reduction of capital investment by 13.32 %. The total discounted operating cost was lower by 32.56%. In this case, the combined reduction in capital investment and discounted total operating cost obtained by applying the GWO algorithm led to a total cost reduction of 17.19%. | en |
