Hybrid Evolutionary Algorithms And Engineering Applications

Optimal design is a modern design method which researches on how to get the best solution under constrains. Evolutionary algorithm does not rely on the mathematical model and has nice global searching characteristics. Several branches of evolutionary algorithm were developed one after another. Each branch has its advantages and disadvantages. It will improve the searching characteristics by constructing mixed evolutionary algorithm.Particle Swarm Optimization (PSO) is a branch of evolutionary algorithm with nice searching characteristics. PSO is efficient because the information sharing mechanism made all the individuals tend to converge to the current global best solution quickly in the intelligent evolution. But PSO is much easier to run into the local optimal solution because the diversity of the population losses quickly. In this paper, the antibody density in artificial immune algorithm is introduced into PSO to improve searching characteristics by keeping the diversity of the population. The Competitive Particle Swarm Optimization algorithm (CPSO) is developed. A parameter is also defined to describe the diversity mathematically. The CPSO algorithm is then applied to structural optimal design, packing problems, UAV path planning, trial rendition of link lever mechanism and disk brake optimization. The result shows that the CPSO acted efficiently. In this paper, a new sharing function is also defined to describe the diversity of Pareto set in multi-objective PSO.Topology optimization is the highest phase of structure optimization. The topological form of the structure changes in optimization. In this paper, several simple and common used methods are included. They will show how topology optimization runs.