论文部分内容阅读
粒子群优化(PSO,Particle Swarm Optimization)算法是继遗传算法、蚁群算法之后的又一种新的群体智能算法,经常用于复杂问题的求解.由于其迭代公式是面向连续空间的,因此更适合解决非网格拓扑的航路规划问题.标准的粒子群优化算法在寻优的过程中容易出现早熟现象,针对这种现象,提出了一种改进的粒子群优化算法.改进算法根据相应的代价函数选择精英粒子和较差粒子,对较差粒子采用了带有动能补偿的速度更新策略,从而避免了寻优过程中的早熟现象;在单个粒子的运动方面引入了最差粒子的失败经验,让群体中粒子有效避开最差解.仿真表明:改进算法在航路规划的应用中具有更强的搜索能力,获得的航路代价在进化代数相同的前提下更小.
Particle swarm optimization (PSO) algorithm is a new swarm intelligence algorithm after genetic algorithm and ant colony algorithm, which is often used to solve complex problems. Because its iterative formula is continuous space-oriented Which is suitable to solve the route planning problem of non-grid topology.The standard Particle Swarm Optimization (PSO) algorithm is prone to premature phenomenon in the process of optimization.An improved particle swarm optimization algorithm is proposed for this phenomenon.An improved algorithm based on the corresponding cost The function selects the elite particles and the poorer particles, and applies the velocity update strategy with kinetic energy compensation to the poorer particles, thus avoiding the premature phenomenon in the optimization process. Introducing the failure experience of the worst particle in the movement of a single particle, The particles in the population are effectively avoided from the worst solution.The simulation results show that the improved algorithm has stronger search ability in the application of route planning and the route cost obtained is smaller under the same evolutionary algebra.