摘要
韦伯分布在非线性寻优问题中具有较好的寻优精度和全局搜索能力,为此提出一种基于韦伯分布的布谷鸟搜索(WCS)算法来解决颗粒粒径分布反演的问题。使用WCS算法对服从Johnson’s S_(B)分布、Rosin-Rammler分布和正态分布的单峰颗粒系和双峰颗粒系进行颗粒粒径分布的反演,并分别与其他传统算法的处理结果进行比较。结果表明,WCS算法的整体效果优于人工鱼群算法和人工蜂群算法,且改进后的4种重尾分布CS算法的标准差比原CS算法提升2~3个数量级。目标函数散射光能加入噪声后,WCS算法比其他三种重尾分布的相对均方根误差值至少可降低1/2。使用小角前向散射测量系统对单峰颗粒系和双峰混合颗粒系进行实验研究,发现WCS算法的相对均方根误差比原CS算法降低约为40%。
Weber distribution has better optimization accuracy and global search ability in nonlinear optimization problems. For this reason, a cuckoo search(WCS) algorithm based on Weber distribution is proposed to solve the problem of particle size distribution inversion. The WCS algorithm is used to invert the particle size distribution of unimodal and bimodal particle systems which follow Johnson’s S_(B) distribution, Rosin-Rammler distribution, and normal distribution, and the results are compared with those of other traditional algorithms. The results show that the overall performance of the WCS algorithm is better than that of the artificial fish swarm algorithm and the artificial bee colony algorithm, and the standard deviation of the improved four heavy-tailed distribution CS algorithm is 2-3 orders of magnitude higher than the original CS algorithm. Compared with the other three heavy-tailed distributions, the relative root mean square error of the WCS algorithm can be reduced by at least 1/2 when the scattering light energy of the objective function is added into the noise. The small angle forward scattering measurement system is used to study the unimodal particle system and bimodal mixed particle system. It is found that the relative root mean square error of the WCS algorithm is about 40% lower than that of the original CS algorithm.
作者
单良
查婷婷
孔明
洪波
Shan Liang;Zha Tingting;Kong Ming;Hong Bo(Key Laboratory of Electromagnetic Wave Information Technology and Metrology of Zhejiang Province,College of Information Engineering,China Jiliang University,Hangzhou,Zhejiang 310018,China;College of Metrology&Measurement Engineering,China Jiliang University,Hangzhou,Zhejiang 310018,China)
出处
《光学学报》
EI
CAS
CSCD
北大核心
2022年第2期242-251,共10页
Acta Optica Sinica
基金
国家自然科学基金(51874264,52076200)。
关键词
散射
小角前向散射
布谷鸟搜索算法
韦伯分布
颗粒粒径反演
scattering
small angel forward scattering
cuckoo search algorithm
Weber distribution
particle size inversion
