基于高斯混合飞蛾优化算法的Wiener模型辨识
    点此下载全文
引用本文:王迦祺,张 健,吴天慧,晋佳浩.基于高斯混合飞蛾优化算法的Wiener模型辨识[J].计算技术与自动化,2022,(4):103-111
摘要点击次数: 359
全文下载次数: 0
作者单位
王迦祺,张 健,吴天慧,晋佳浩 (营口理工学院 电气工程学院辽宁 营口 115014) 
中文摘要:针对Wiener模型辨识问题,结合函数连接型神经网络(FLANN)和飞蛾优化算法(MFO)的优势,提出了一种新型的辨识方案。利用FLANN来拟合静态非线性模块,通过将辨识问题转化为优化问题来对线性部分和非线性部分的参数同时进行更新。为了提升飞蛾优化算法的辨识性能,将高斯混合分布思想引入飞蛾种群初始化以及位置更新中,提出了一种新型的高斯混合飞蛾优化算法(GMFO),并通过测试函数验证了其寻优性能。最后通过仿真实验结果证明了所提出辨识方案的有效性和鲁棒性。
中文关键词:Wiener模型  系统辨识  飞蛾优化算法  高斯混合分布  测试函数  函数连接型神经网络
 
Wiener Model Identification Based on Gaussian-mixture Moth-Flame Optimization
Abstract:Aiming at the identification problem of Wiener model, this paper proposes a novel identification scheme combining the advantages of FLANN and Moth-Flame Optimization (MFO). The scheme uses FLANN to fit the static nonlinear block and update the parameters of linear part and nonlinear part simultaneously by converting the identification problem to an optimization problem. In order to improve the identification performance of MFO, this paper proposes a novel version named Gaussian-mixture Moth-Flame Optimization (GMFO) by introducing the idea of Gaussian-mixture distribution into the moth population initialization and position adjustment. Test functions are used to verify its optimization performance. Finally, simulation results verify the effectiveness and robustness of the proposed identification scheme.
keywords:wiener model  system identification  moth-flame optimization  Gaussian-mixture distribution  test function  functional link artificial neural network
查看全文   查看/发表评论   下载pdf阅读器