摘要
粒子群优化算法是一类基于群智能的优化搜索算法。该算法初期收敛较快,但后期易陷入局部最优点。为了提高粒子群算法的性能,将粒子群算法全局搜索的快速性和混沌算法的一定范围内的遍历性二者结合,利用罚函数的思想把有约束的非线性规划问题转化为无约束最优化问题,并利用了混沌运动遍历性、随机性等特点,对传统粒子群算法进行改进,摆脱了粒子群算法后期易陷入局部极值点的缺点,然后与罚函数方法结合,构造出一个基于罚函数的混沌粒子群优化算法。数值结果表明文中所提出的算法是有效的。
Particle Swarm Optimization （PSO） is a kind of optimization search algorithnm based on swarm intelligence. The algorithm weakens quickly initially,but falls into local extreme value easily. In order to improve the performance of PSOtcombined its rapid global searching ability and chaos ergedicity in certain range,employing penalty function transforms nonlinear programming problems into un- constrained optimization problems. Also,considering the ergodicity and randomness of chaotic motion,the traditional PSO is improved, which avoids falling into local extreme point,with penalty function produces a CPSO based on penalty function. Numerical results show that the proposed algorithm is effective.
出处
《计算机技术与发展》
2012年第10期15-17,共3页
Computer Technology and Development
基金
国家自然科学基金资助项目(70871081)
上海市重点学科建设资助项目(S30504)
关键词
罚函数
混沌粒子群
算法
penalty function
chaos particle swarm
optimization
algorithm
