新疆沙滩排球后备力量培养现状分析研究

通过对新疆沙滩排球后备力量培养现状的调查与研究,针对运动员、教练员在培养过程中存在的不足进行深入地剖析,并提出可行性建议,以期促进新疆沙滩排球事业的更快发展。     

考虑响应共变特性的多响应稳健参数设计

针对响应共变特性的稳健参数设计问题,在多任务高斯过程(multi-task Gaussian processes,MTGP)建模框架下,结合质量损失函数和考虑响应不确定性的优化函数构建了一个考虑输出响应不确定性的MTGP(uncertainty of MTGP,UNMTGP)优化模型。首先,利用MTGP模型拟合实验数据,构建考虑响应间共变特性对优化结果影响的多元高斯模型。其次,提出考虑输出响应不确定性的优化目标函数,构建多响应稳健优化模型。最后,结合全局优化方法,获得最优参数设计。此外,结合真实案例,利用质量损失函数的相关评价指标,论证所提方法的有效性。结果表明,所提方法考虑了响应共变特性和输出响应不确定性对优化结果的影响,有效改善了模型的预测质量,提升了输出响应的稳健性。     

The explanatory role of abstraction processes in models: The case of aggregations

Though it is held that some models in science have explanatory value, there is no conclusive agreement on what provides them with this value. One common view is that models have explanatory value vis-à-vis some target systems because they are developed using an abstraction process (i.e., a process which involves omitting features). Though I think this is correct, I believe it is not the whole picture. In this paper, I argue that, in addition to the well-known process of abstraction understood as an omission of features or information, there is also a family of abstraction processes that involve aggregation of features or information and that these processes play an important role in endowing the models they are used to build with explanatory value. After offering a taxonomy of abstraction processes involving aggregation, I show by considering in detail several models drawn from different sciences that the abstraction processes involving aggregation that are used to build these models are responsible (at least partially) for their having explanatory value.     

Exponential convergence of fisher entropy for diffusion processes

In this paper, we introduce the definition of Γ-Fisher entropy. For the stochastic differential equation in Rd and the unique invariant probability measure μ, we obtain the exponential convergence of Γ-Fisher entropy with respect to μ under some condition about Γ and the drift b. Moreover, we revisit the exponential convergence of the usual Fisher entropy under the Bakry Emery condition.     

针对细分方案的快速FBM算法

在对分数布朗运动进行理论研究的基础上，给出了一种针对细分方案的快速ＦＢＭ算法，即对于不同格式的剖分，填加相应的型值点，以实时地生成各种不同ＦＢＭ或者类ＦＢＭ曲线，此法可以有效地用于各种复杂曲线及边界线模拟。     

Modelling the frequency and severity of extreme exchange rate returns

Risk managers are often concerned about tail probabilities of asset return distributions, in particular the frequency and severity of extreme returns. In this article, we propose a model that integrates extreme value theory and point processes to model the frequency and severity of exchange rate returns. The proposed model is applied to daily spot exchange rate series and the parameters of interest, such as the tail index, the mean size and rate of occurrence of extreme returns, are estimated using maximum likelihood estimation. We study the impact of recent currency crises on the frequency and severity of the series and find that, during 1995–9, the frequency of extreme daily Japanese yen–US dollar spot exchange rate returns increases twofold, and the time duration of high volatility persists longer for the Japanese yen series than for the Swiss franc and Danish krone series. Copyright © 2001 John Wiley & Sons, Ltd.     

极端冲击下系统竞争性失效过程的可靠性建模

由于系统运行过程中可靠性的不断降低,对于相同幅值的冲击,系统运行的前期和后期会有不同的冲击结果.为了描述这一现象,考虑了系统自然退化和冲击两个竞争性失效过程间的相关性,一方面冲击造成系统退化量的增加,另一方面系统的自然退化程度对冲击的结果也有一定的影响.文中假设系统因冲击而失效的过程是极端冲击过程,通过系统自然退化过程和冲击过程的分布函数,导出了系统的可靠度函数,建立了系统可靠度模型的一般形式和一种特例的具体形式,最后利用文献中的具体参数进行仿真验证了模型的正确性和有效性.     

独立同分布环境中配对依人口数两性分支过程L~1收敛

考虑了随机环境中配对依人口数两性分枝过程模型,并且得到了独立同分布环境配对依人口数两性分枝过程{Zn}对应的过程{Wn}的L1收敛的充分条件.     

非高斯过程与微小故障的故障检测方法

针对生产过程中数据服从非高斯分布的故障检测问题进行了讨论,给出了一个改进的基于稀疏表示的故障检测方法,并通过构建重构误差控制限和距离控制限区间,提高了基于稀疏表示的故障检测水平;给出了在统一框架下的微小故障检测方法,并通过在微小故障字典矩阵中引入时间常数t来提高针对微小故障的检测水平;通过两例数值仿真验证了方法的有效性,并与经典的基于主元分析的故障检测方法进行了对比.结果表明,所提出的方法在上述2种情况下的故障检测水平均超过基于主元分析的故障检测水平.