摘要
针对单目标粒子群优化算法局部搜索能力差,不能有效求解高维、复杂工程问题等缺点,提出了一种改进的粒子群优化算法,即单纯形粒子群优化方法的混合算法(SM PSO)。该混合算法,在继承粒子群优化算法原有优点的同时,不但可减少计算规模,且有效地增强了粒子群优化算法的局部搜索能力,提高了算法的鲁棒性能。文中采用30维经典测试函数及齿轮减速器优化问题作为算例,验证了该算法的优越性能。
In this paper, an improved particle swarm optimization algorithm, hybrid algorithm of simplex and particle swarm optimization(SM-PSO)is proposed, because particle swarm optimization has weak local searching abilities and can not be applied to solving complex problems with high dimensions. The hybrid algorithm can adopt lesser population, and decrease calculation scale, thus improves local searching abilities and robustness. Proved by classical testing function with 30 dimensions and gear-driven moderator, the algorithm has excellent performance.
出处
《机械科学与技术》
CSCD
北大核心
2005年第4期415-417,共3页
Mechanical Science and Technology for Aerospace Engineering
基金
国家自然科学基金项目(10377015)资助
