基于原对偶状态转移算法的分数阶多涡卷混沌系统辨识

未知分数阶混沌系统参数辨识问题可转化为函数优化问题, 是实现分数阶混沌系统同步与控制的关键. 结合正交学习机制和原对偶学习策略, 提出一种原对偶状态转移算法, 用于解决分数阶混沌系统的参数辨识问题. 利用正交学习机制产生较优的初始种群增加算法的收敛能力, 并引入原对偶操作增加状态在空间的搜索能力, 提高算法的寻优性能. 在有噪声和无噪声情况下以分数阶多涡卷混沌系统的参数辨识为研究对象进行仿真. 结果表明了该算法的有效性、鲁棒性和通用性.

Parameter identification for fractional-order multi-scroll chaotic systems based on original dual-state transition algorithm

Parameter estimation for fractional-order chaotic systems is a multi-dimensional optimization problem, which is one of the important issues in fractional-order chaotic control and synchronization. With the orthogonal learning strategies and the original dual learning mechanism, the original dual-state transition algorithm is proposed for solving the problem of parameter estimation in fractional-order chaotic systems. The orthogonal learning strategy is presented which can increase the diversity of initial population and improve the convergence ability. And the original dual learning mechanism is presented which can increase the space ability of states, and also can improve the search capability of the algorithm. In the process of identification, we adopt Radau IIA method to solve the fractional-order differential equation. The simulation of the fractional-order multi-scroll chaotic systems with or without noise is conducted and the results demonstrate the e?ectiveness, robustness, and versatility of the proposed algorithm.