| Swarm Intelligence (SI) is the property of a system whereby the collective behaviors of (unsophisticated) agents interacting locally with their environment cause coherent functional global patterns to emerge. SI provides a basis with which it is possible to explore collective (or distributed) problem solving without centralized control or the provision of a global model.Some of the swarm intelligent algorithms, such as ant colony system and particle swarm optimization, also fall into the category of evolutionary computation, so that they are similar to evolutionary algorithms in principle and mechanism, and also the flaw: contradiction between the speed by which they converge and the quality of solution to which they converge. On the other hand, lots of ideas of improving derive from the area of evolutionary computation in the process of their development.Beginning with discussion of various classic methods applied in the realm of evolutionary computation, the thesis have a research on ant colony system and particle swarm optimization, including the principle and mechanism, implementation, advantageous and disadvantageous, idea and method of improving. The routing and scheduling mechanism of wasp nests is introduced by the way.Finally, the thesis dives into particle swarm optimization on the analysis of its shortcoming in global and local searching, variety of strategies of improving, the rule and quantity of sensation in psychology. Then, a sentient particle swarm optimization is presented, in which the sensation model is introduced to the exploitation part of particle swarm optimization to enhance the capability of local searching. Comparing of outcome as a result of simulation among genetic algorithm, basic particle swarm optimization and dissipative particle swarm optimization shows the superiority of the sentient particle swarm optimization inits capability of exploiting, cooperating between global and local searching. Some important paraments of the algorithm are also discussed.
|
PDF Full Text Request