粒子群优化算法的改进研究

发布时间：2018-09-01 13:56

【摘要】：粒子群优化(Particle Swarm Optimization,PSO)算法是通过对鸟群和鱼群群体运动行为的观察研究提出的一种新兴的仿生优化算法。粒子群优化算法的优点是概念较简单、参数较少、收敛速度较快、实现较容易。因此,引起了许多国内外学者广泛的关注,目前已应用于函数优化、神经网络的训练、模糊系统的控制等应用领域并取得了较好的效果。但是其缺点是容易陷入局部极小点,搜索精度不高,这也吸引了许多学者对粒子群优化算法进行大量的改进研究。本文首先构造了一个新的粒子群优化算法更新公式,该更新公式大大的增加了算法的多样性。然后把新的更新公式转化为一个离散线性系统,根据系统稳定性理论,推出了保证改进粒子群系统稳定的参数设置区域,通过数值实验验证了算法的稳定性。另外,将和声优化搜索算法与粒子群优化算法相结合,给出了混合粒子群优化算法,并通过数值实验验证了算法的稳定性和可行性。
[Abstract]:Particle Swarm Optimization (Particle Swarm Optimization,PSO) algorithm is a new bionic optimization algorithm based on the observation of the movement behavior of birds and fish populations. Particle Swarm Optimization (PSO) has the advantages of simple concept, less parameters, fast convergence rate and easy implementation. Therefore, many scholars at home and abroad have paid close attention to it, which has been applied in the fields of function optimization, neural network training, fuzzy system control and so on. However, its shortcoming is that it is easy to fall into local minima, and the search accuracy is not high, which has attracted many scholars to do a lot of improved research on particle swarm optimization (PSO) algorithm. In this paper, a new updating formula of particle swarm optimization algorithm is constructed, which greatly increases the diversity of the algorithm. Then, the new updating formula is transformed into a discrete linear system. According to the stability theory of the system, the parameter setting region is derived to ensure the stability of the improved particle swarm optimization system. The stability of the algorithm is verified by numerical experiments. In addition, the hybrid particle swarm optimization algorithm is proposed by combining the harmonic optimization search algorithm with the particle swarm optimization algorithm, and the stability and feasibility of the algorithm are verified by numerical experiments.
【学位授予单位】：渤海大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18

【相似文献】