摘要
分析了一类混沌优化算法所存在的不足,即在二次载波过程中只是在近似最优解的单侧邻域内进行搜索,同时可调参数也不能随着搜索进程的变化而变化,使得寻优结果并不是很理想。对此进行了适当地改进,利用混沌变量对当前点进行扰动,并且通过时变参数逐渐减小搜索进程中的扰动幅度,同时,以一定方式确定了时变参数的初值。用改进后的算法对连续对象的全局优化问题进行优化,仿真结果表明,该方法可以显著提高收敛速度和精度。
The deficiencies of a kind of chaos optimization algorithm are analyzed. This algorithm can only search the optimal values in one side of the approximate values, at the same time the control parameter doesn＇t change adaptively. According to the above deficiencies, an improved chaos optimization algorithm is presented. It disturbs the current state by means of chaotic variable. Because of the intervention of a variable parameter, The number of searching steps is larger, the smaller range of the disturbance. In the meantime, the initial value of the variable parameter is given. The improved chaos optimization algorithm is applied to some continuous functions, the numerical simulation results indicate that the convergence speed and accuracy of global optimization are significantly improved
出处
《科学技术与工程》
2007年第3期307-309,313,共4页
Science Technology and Engineering
关键词
混沌
遍历性
混沌优化
全局最优
chaos ergodic property chaos optimization global optimization
