基于克隆选择的小世界优化算法

针对小世界算法在多极值等复杂函数优化中存在算法后期种群多样性退化、全局搜索效率下降等问题,提出一种基于种群克隆选择的小世界优化算法。该算法以小世界现象信息传递的高效性改进克隆过程中体细胞高频变异的随机性,实现克隆增殖、克隆选择以及小世界网络短连接等算子在局部空间的搜索,克隆删除与小世界随机长连接在全局空间的搜索。实验结果表明:各种克隆算子与小世界变异算子相结合,增加了种群的多样性,扩大了搜索范围。与其他算法相比,该算法在收敛速度和多极值点函数搜索能力等方面具有明显改善。     

基于ReliefF-NOSCA-AdakNN的肌肉疲劳识别技术研究

针对竞技体育训练中的肌肉疲劳监测问题,提出了一种基于ReliefF-NOSCA-AdakNN（RNA）的表面肌电信号（sEMG）特征提取和分类算法. 该算法结合了特征和类别之间的相关性分析和启发式搜索算法,对高维特征进行了有效的筛选和分类. 将RNA算法应用于经过滤波处理的肱二头肌肌电信号数据,对不同疲劳状态进行了识别和分类. 实验结果表明,提出的RNA算法在平均分类准确率和标准差方面分别达到了83.88%和0.012 7,均显著优于传统单一算法,体现了较好的分类性能.     

考虑数量折扣的多周期供应商选择决策

针对一条由一个主导制造商、多个上游供应商组成的供应链,为了尽可能降低制造商采购时的总成本,结合产品的价格、质量、交付三大因素,以供应商生产能力、供应商数量折扣、供应商与制造商关系维持和制造商的库存能力等为约束条件,综合考虑制造商的采购成本、质量因素带来的销售损失和交付因素影响的库存成本,为制造商建立了多目标的多周期订单分配整数线性规划模型,通过算例对三个目标进行权重分析,通过Lingo17.0得到模型的最优解,为制造商提供一个多周期订单决策的框架。最后,对模型中几个重要的影响参数进行了灵敏度分析,并为制造商与供应商提出了一些参考建议。     

基于OLAP的副本选择体系结构OLAPRES

针对当前副本选择算法处理时间随副本个数增加而急剧增加的问题, 提出一种基于OLAP技术的副本选择体系结构OLAPRES, 利用OLAP技术中的数据预处理和快速查询功能, 使大量副本中的快速选择成为可能并描述了基于OLAPRES副本选择的工作流程.     

组合动力技术的未来应用

 组合动力将两种或以上单一类型的动力有机结合，结构相互融合，功能相互补充，大大拓展了飞行包线、提高飞行效率，在临近空间高速飞行器、重复使用航天运输系统等领域具有重要应用。介绍了组合动力基本分类及工作原理，综述了组合动力技术在临近空间高速飞行器、重复使用航天运输系统等领域的应用，总结了组合动力飞行器发展趋势及主要关键技术。     

案例结合问题教学在药物分析课程中的应用

药物分析是一门理论与实践并重的学科,不仅需要有效地为学生传授药物分析的基础知识、方法技能,更要注重培养学生解决实际问题的能力.基于药物分析的学科特点,在教学过程中实施案例结合问题教学法,以案例激发学生学习热情、以问题启发学生自主学习,理论联系实际,有效培养学生专业素养并提高其综合素质,取得了良好的教学效果.     

国外组合循环动力技术研究进展

 组合循环动力技术是多种动力技术从热力循环、结构布局等方面的有机融合，具有工作包线宽、综合性能优等特点。综述了国外组合循环发动机的发展历程，介绍了几种典型组合循环发动机，概述了组合循环动力技术的发展趋势、关键技术，提出了组合循环动力技术领域的发展建议。     

An imperfect imago? Post-mating loss of iridescent scales in Cheimas butterflies may change female from attractive to cryptic (Lepidoptera: Nymphalidae: Satyrinae)

Adults of both sexes of Cheimas opalinus (Staudinger), a Satyrinae butterfly occurring in the cloud forests of the Venezuelan Cordillera de Mérida, have simple wing colour patterns, dark brown dorsa marked by one conspicuous element, large greenish-blue patches in the discal area of the hind wing. It was observed that in young, freshly emerged females these patches are frequently absent. They fade away and eventually disappear during mating and nuptial flights when brushed out by lateral movements of the male’s hind wing put between folded female hind wings. The falling off of the scales is made easy because they do not adhere to the wing as firmly as brown background scales due to their particular folded extremities. We speculate that this process is related to the fitness of the females. Prior to mating, blue-green patches are advantageous because they attract the attention of the opposite sex and enhance the chances of successful mating. After mating they lose their sexual role. Females without the blue-green patches become cryptic. Thus, they are less apparent not only for the males, which helps them avoid sexual harassment, but also for potential bird predators. Additionally, the loss of blue reflecting scales may speed up the warming up of the abdomen and egg maturation. In the males, positive role of the patches (signalling), and negative (bird attraction), is balanced during their entire life span, and they fade gradually due to regular usage of wings scaling in flight. Our hypothesis is supported by morphological, optical, experimental and statistical analysis in which we used 509 individuals of both sexes. A similar ratio of young (with undamaged wing) females with or without blue patch was recorded, whereas in the males no individuals without any trace of blue patch were reported, and a correlation of age (wing damage) and gradual fading off of the patch was demonstrated. Such an adaptation involving an active change of the appearance of adults, i.e. colour patterns related to sexual selection, has not been reported previously in Lepidoptera.     

基于EEMD-PSO-LSSVM的中国沿海散货运价指数预测

针对中国沿海散货运价指数(CBFI)预测对精度的要求,从内在波动特性角度出发,提出一种基于集合经验模态分解(EEMD)-粒子群优化算法(PSO)-最小二乘法支持向量机(LSSVM)的组合预测模型.对比LSSVM、PSO-LSSVM、EMD-PSO-LSSVM三种预测模型,EEMD可对CBFI序列中波动较大数据进行降噪分解,保留序列的内在波动特性,且预测精度有一定提升,预测性能更佳.     

Moving average threshold heterogeneous autoregressive (MAT-HAR) models

We propose moving average threshold heterogeneous autoregressive (MAT-HAR) models as a novel combination of heterogeneous autoregression (HAR) and threshold autoregression (TAR). The MAT-HAR has multiple groups of lags of a target series, and a threshold term can appear in each group. The threshold is a moving average of lagged target series, which guarantees time-varying thresholds and simple estimation via least squares. We show via Monte Carlo simulations that the MAT-HAR has sharp in-sample and out-of-sample performance. An empirical application on the industrial production of Japan suggests that significant threshold effects exist, and the MAT-HAR has a higher forecast accuracy than the HAR.