摘要
本文提出一种基于改进遗传算法的随机及演方法，该方法具有更强的全局收敛性，同时也具有对粒径分布形状不敏感、抗噪声能力强等优点，适合于工业现场应用．
In this paper an improved inverse genetic algorithm for measuring particle size distribution by light scattering method was developed. A type of new 'migrated' individuals that had been generated by random was introduced in the algorithm. The new 'migrated' individuals were permitted to eliter current population and compete with 'native' individuals.Because the 'migrated' individuals could bring a lot of new gene fEatures in the evolution,the improved inverse method could be considered easier to escape the local optimum. Numerical tests for inverting angular light scattering data showed that the algorithm could be successfully applied to inverse problem of light scattering with high stability in presence of random noise and low susceptibility to the type of distributions.
出处
《工程热物理学报》
EI
CAS
CSCD
北大核心
1999年第5期632-636,共5页
Journal of Engineering Thermophysics
关键词
微粒尺寸分布
光散射法
反演遗传算法
particle size distribution, light scattering method, genetic algorithm
