一种多策略改进鲸鱼优化算法的混沌系统参数辨识

Parameter identification of chaotic system based on a multi-strategy improved whale optimization algorithm

针对混沌系统参数辨识精度不高的问题,以鲸鱼优化算法(whale optimization algorithm,WOA)为基础,提出一种多策略改进鲸鱼优化算法(multi-strategy improved whale optimization algorithm,MIWOA)。采用Chebyshev混沌映射选取高质量初始种群,采用非线性收敛因子和自适应权重,提高算法收敛速度,为了避免算法陷入局部最优,动态选择自适应t分布或蚁狮优化算法更新后期位置,提高处理局部极值的能力。通过对10个基准函数和高维测试函数进行仿真试验,表明MIWOA具有良好的稳定性和收敛精度。将MIWOA应用于辨识Rossler和Lu混沌系统参数,仿真结果优于现有成果,表明本文MIWOA辨识混沌系统参数的高效性和实用性。

Aimed at the problem of low parameter identification accuracy of chaotic systems,a multi-strategy improved whale optimization algorithm(MIWOA)is proposed based on the whale optimization algorithm(WOA).MIWOA uses Chebyshev chaotic mapping to select high-quality initial populations,and nonlinear convergence factor and adaptive weight to improve the convergence speed of the algorithm.In order to avoid falling into local optimal solution,MIWOA dynamically selects adaptive t distribution or ant lion optimization algorithm to update the later position and improve the ability to handle local extremum.Through simulation experiments on 10 benchmark functions and high-dimensional test functions,it is shown that MIWOA has good stability and convergence accuracy.Applying MIWOA to identify the parameters of Rossler and Lu chaotic systems,the simulation results are superior to existing achievements,indicating the efficiency and practicality of MIWOA in identifying chaotic system parameters in this paper.