Applications of Hybrid Optimization Algorithms to the Unconstrained and Constrained Problems

Abstract

This chapter presents the applications of four different hybrid algorithms to the unconstrained and constrained benchmark functions and also to the mechanical element design optimization problems of a simple gear train, radial ball bearing, Belleville spring, multi-plate disc clutch brake, robot gripper, hydrostatic thrust bearing, a four stage gear train, pressure vessel, welded beam, tension/compression spring, speed reducer, stiffened cylindrical shell, step-cone pulley, screw jack, C-clamp, hydrodynamic bearing, cone clutch, cantilever support, hydraulic cylinder and a planetary gear train. Four different optimization algorithms such as PSO, BBO, DE and GA are chosen to hybridize them with ABC algorithm. Comparison of the overall performance of hybrid algorithms with the basic algorithms is made and it is observed that hybridization of ABC and PSO is effective than the basic PSO algorithm. For searching the best solutions, hybridization of ABC with PSO and GA is not so effective than the basic ABC. Hybridization of ABC with BBO and DE is effective than the basic ABC in finding the best solutions. Moreover, Hybridization of ABC with DE is very effective than the basic ABC and other algorithms.