基于正交设计的反向学习差分进化算法

为了克服反向学习带来的维度退化现象,生成部分反向解增强对反向解空间的勘测能力,提出了一种基于正交设计的反向学习差分进化算法,利用正交设计仅生成若干具有代表性的部分反向解,在增强算法勘测能力与减少函数评价次数上达到了一个良好的平衡.实验结果表明:该算法相对于其他几种反向学习差分进化算法有更好的收敛精度及速度,同时对函数维度变化不敏感,鲁棒性较强.

一种包括三代费米子的统一模型

提出了一种定义在１０维时空上的ＳＯ（１７）统一模型，通过相对于陪集空间ＳＵ（３）／Ｕ（１）×Ｕ（１）×Ｚ２的维度退化，得到了具有三代轻费米子的标准模型．

Relative humidity and temperature dependence of mechanical degradation of natural fiber composites^(?)

In this paper,the mechanical degradation of natural fiber composites is studied with the consideration of the relative humidity and the temperature.A nonlinear constitutive model is established,which employs an internal variable to describe the mechanical degradation related to the energy dissipation during moisture absorption.The existing experimental researches demonstrated that the mechanical degradation is an irreversible thermodynamic process induced by the degradation of fibers and the damages of interfaces between fiber and matrix,both of which depend on the variation of the relative humidity or the temperature.The evolution of the mechanical degradation is obtained through the determination of dissipation rates as a function of the relative humidity and the temperature.The theoretically predicted mechanical degradations are compared with experimental results of sisal fiber reinforced composites subject to different relative humidity and temperatures,and a good agreement is found.