摘要
微粒群算法在处理约束条件时最常采用的方法是约束保持法,但该方法易使粒子在搜索中停滞不前,为了改进传统约束保持法的缺点,将微粒群算法与信赖域算法相结合,从而保持了粒子的多样性并使最优解在可行域内。另外,采用与信赖域搜索技术相结合的随机惯性权重,改善了算法的全局寻优能力,提高了算法的收敛速度和计算精度。实验结果表明:与标准微粒群算法和一些其他优化算法相比,改进算法具有较强的寻优能力和寻优效率。
The most commonly-used method of particle swarm optimization constraints is constraint preserving method, but the method is easy to make the particles standstill in the search process. In order to improve the shortcom- ing, particle swarm optimization is combined with the trust region algorithm to keep the diverse of particles and make the best solution in the feasible region. In addition, a random inertia weight by combining with a trust domain search technology was used to improve the ability of global optimization and increase the convergence speed and computational accuracy. Numerical experiments show that：compared with the standard particle swarm algorithm and some other optimization algorithms, the improved algorithm has strong optimization ability and optimizing efficiency.
出处
《太原科技大学学报》
2012年第5期406-409,共4页
Journal of Taiyuan University of Science and Technology
关键词
约束优化问题
微粒群算法
信赖域算法
随机惯性权重
寻优能力
constrained optimization problem, particle swarm optimization, trust region algorithm, random inertia weight, optimization ability
