中文摘要：

针对粒子群优化算法中出现早熟和不收敛问题,分析了基本PSO算法搜索速度对其优化性能的影响,提出了一种根据速度信息非线性自适应调整参数的粒子群优化算法。在算法迭代过程中,粒子随迭代次数和递减指数确定的非线性变化的理想速度自适应调整参数进行搜索,提高了粒子群算法的性能。提出的算法经过测试函数的模拟实验验证,并与其他已有算法进行了比较。实验结果表明,该算法在搜索精度和收敛速度等方面有明显优势,特别是高维、多峰等复杂非线性优化问题时,算法的优势更明显。

英文摘要：

Due to the prematurity and non-convergence problems in particle swarm optimization algorithm,the influences of the search velocity on optimization performance of standard PSO algorithm was analysed.A nonlinear adaptive parameter adjusting particle swarm optimization algorithm（NAPSO-VI） was proposed.During the iterations,particles,could adaptively adjust parameters to search space,according to the ideal velocity,which usually changes along with the iterations and constant descending exponent.As a result,it improved the performance of PSO algorithm,through empirically tesing and comparing with other published methods on benchmark functions.The experimental results illustrated that the proposed algorithm had evident advantages in the search accuracy and convergence speed,especially in multi-dimension and multi-peak nonlinear optimization problems.