The Heuristic Particle Swarm Optimizer And Its Application To The Structural Optimal Design

Great attention has been paid to the optimal design, due to the fact that it can reduce the cost. Most of the traditional optimal algorithms, which are based on the numerical linear and nonlinear programming methods, require substantial gradient information and usually seek to improve the solution in the neighborhood of a starting point. The traditional optimal algorithms provide a useful strategy to obtain the global optimal solution in a simple model. However, many practical engineering optimal problems are very complex and hard to solve by the traditional optimal algorithms. The Particle Swarm Optimizer, which is a swarm intelligence algorithm, has been proposed. It is a population-based optimization algorithm and requires few parameters to be implemented.Firstly, this thesis presents the particle swarm optimizer (PSO) and the particle swarm optimizer with passive congregation (PSOPC). Then, the thesis proposes the heuristic particle swarm optimizer (HPSO), which is based on the PSOPC. It handles the problem-specific constraints using a 'fly-back mechanism' technique and the variables' constraints using the harmony search scheme.The HPSO is tested by several truss optimization problems, which has been researched by many scholars (Benchmarks). The results show that the HPSO is able to accelerate the convergence rate effectively and has the faster convergence rate than the PSO and PSOPC.Finally, the HPSO is applied to the double layer spherical shell's optimal design. The feasibility is tested for the practical application of the algorithm proposed in this thesis.