一种新型的带有小生境技术和精英集策略的多目标粒子群算法

A New Multi-objective Particle Swarm Algorithm with a Niche Technology and Elite Set Policies

为提高多目标粒子群算法的有效性和运行效率,利用小生境技术求解适应度,采取轮盘赌的方法根据精英集中各个粒子的适应度选取全局最佳位置,提出一种新型的带有小生境技术和精英集策略的多目标粒子群算法。论文对算法运行的过程作了调整,加入小概率变异方法,采用测试函数验证算法的有效性。结果表明,在相同的实验环境中本文算法的运行时间为2.113 s,比基于粒子群的多目标优化算法(4.157s)缩短近一半,即本算法的运算效率大大提高了。仿真结果还表明本文中的算法不仅有很好的收敛性,所得的解还有较好的均匀性。

In order to improve the effectiveness and efficiency of the multi -objective particle swarm optimization （pso） algorithm, by using niche technology to solve the fitness, this paper puts forward a new kind of multi - objective particle swarm optimization （PSO） algorithm with a niche technology and elite set strategy, the fitness was solved with niche technology and the roulette method was utilized to select global best position with the fitness of each particle of elite set. The adjustment in the process of algorithm running was made in this paper and the small probability variation method was also used. Finally, the test function was adopted to verify the ef- fectiveness of the algorithm. The results show that the algorithm running time is 2.113 s, multi-objective optimization ratio on particle swarm algorithm（4. 147s） was reduced by neraly half, and that the operation efficiency of the algorithm is greatly improved. The simu- lation results also show that the algorithm has better convergence and the solution has better uniformity.