线性递减的粒子群优化算法

粒子群优化算法（ PSO）是一种仿生类的全局优化算法，它借助记忆与反馈机制完成了寻优搜索。该算法受到了鸟类觅食活动的启发而得，其基本思想源于对鸟类简化社会模型的研究及行为模拟，其中的每个个体充分利用自身与群体的智能，不断地调整学习，最终得到满意解。该算法常用于求解非线性问题、组合优化问题等。因其具有易理解，易实现，控制参数少，收敛速度快等优点，该算法一经提出就吸引了广泛的关注，逐渐成为一个新的研究热点。然而粒子群优化算法也有些不足，如搜索精度不高，易早熟以及易陷入局部极值等。而且算法在搜索后期也有产生振荡现象的可能，使得算法收敛起来会较慢。所以，文中就粒子群在迭代后期所出现的振荡现象进行了研究，并作出改进，提出了一种飞行时间单调递减的粒子群优化算法。新算法改善了算法的寻优能力，减小了粒子在寻优过程中的振荡现象。

A Particle Swarm Optimization Algorithm of Linear Decreasing

Particle Swarm Optimization （ PSO） algorithm is a global optimization algorithm of bionics,with the help of memory and feed-back mechanism to complete the search for optimum. The algorithm is inspired by the foraging birds. The basic idea is the result of the study on birds simplified social model and behavior simulation,each of these individuals makes full use of their own and the collective in-telligence,constantly adjusts learning,finally gets satisfied solution. The algorithm is often used to solve nonlinear problem,combinatorial optimization problem and so on. Because of the advantages which is easy to understand and implement,with less control parameters and fast convergence speed,the algorithm is attracted widespread attention since proposing,gradually becoming a new research hotspot. How-ever,there exists a premature convergence,particle swarm optimization algorithm is easy to fall into local optimum and search accuracy of inherent defects,and the algorithm may appears oscillation phenomenon in the late iterations, algorithm ＆#39; s convergence speed is slow. Therefore,based on particle swarm in the late iterations of iterative phenomenon is studied and improved,design a flight time linear de-creasing particle swarm optimization algorithm. The new algorithm improves the searching capability,reduces the particle in the oscillation phenomenon in the process of optimization.