Particle Swarm Optimization and Its Application to Structural Optimum Design
LI Wei-ji · Mechanical Science and Technology · 2005
One of the evolutionary computation technology, Particle Swarm Optimization (PSO), is investigated in this paper. The principle and implementation of this method are introduced. We analyse the impact of main parameters in PSO upon the search directions of particles. PSO is compared with Genetic Algorithm(GA) in optimizaiton operations and searching techniques. We solve the optimization of test functions and truss structural design via PSO and GA respectively. The results produced by PSO are compared with those produced by GA. It is concluded that PSO is more effective when both algorithms satisfy the same precision in calculation. We also design the wing structure by PSO effectively. Therefore, PSO is an effective optimization tool and suited for optimum design of large-scale complex structure.