不同损失函数下不同无信息先验的Bayes估计及比较

讨论了采用损失函数L2（θ,δ）=aθ^m（δ（x）-x）^2时,在二项分布场合下关于产品合格率的不同无信息先验分布下的Bayes估计,并从Bayes风险的角度,对损失函数为L2（θ,δ）=θ（δ（x）-x）^2。的Bayes估计与[1]中所得的Bayes估计进行了比较。从而得出了采用损失函数L2（θ,δ）=θ（δ（x）-x）^2进行Bayes估计较优的结论。     

二项分布的经验贝叶斯估计及应用

讨论了二项分布的经验贝叶斯估计。对用共轭分布β（ａ,ｂ）作为先验分布可能出现的问题,提出了用另外一种先验分布来解决的方法,并给了其在可靠性增长试验中的应用。     

部分先验信息Г下分布族参数的Г—容许估计

讨论了在某一族先验分布族Г或者一定的先验限制Г下分布参数的Г－容许估计，给出了相应的充分条件，分别推广了文献（１）和（２）的结果，并给出若干例子。     

熵损失函数下几何分布参数的Bayes估计

本文在熵损失下，给出了对于任何先验分布的几何分布参数θ的Bayes估计，同时由参数θ的充分统计量Σni=1Xi，给出了熵损失函数下，不同先验分布时几何分布参数θ的Bayes估计，并且证明了在熵损失函数下，对任一先验分布，几何分布的参数θ的Bayes估计δB（X）是可容许估计．     

多环芳烃对土壤环境影响的调查与思考

以新余钢铁城市为例,对其土壤中多环芳烃（PAHs）含量进行了定量分析,研究其分布特征,初步探讨其污染水平,并对土壤PAHs污染防治,减少对农作物污染提出了建议。对典型区域土壤中采集样品,对8种可能高致癌PAHs：苯并[a]蒽（4）、屈、苯并[b]荧蒽（4）、苯并[k]荧蒽（4）、苯并[a]芘（5）、二苯并[a,h]蒽（5）、苯并[gh i〗]苝（6）、茚并（1,2,3-cd）芘（5）,运用高压液相色普仪方法,进行分析测定,结果显示,样品土壤中8种PAHs总含量范围在32.3 ng.g-1-241 200 ng.g-1,平均含量80 447.4 ng.g-1。     

(x1,x2,x3)～M(n;P1,P2,P3)的BAYES检验

设（x1,x2,x3）服从参数为（P1，P2，P3）的多项分布，在使用Bayes方法时，我们选择先验分布为无信息先验，从而求出每个假设的后验概率．     

一类偶数阶非线性中立型微分方程的振动性

考虑一类带有分布型偏差变元的偶数阶非线性中立型微分方程：d^n/dt^n[a（t）y（t）＋m∑i=1ci（t）y（t-τi）]＋∫a^bf（t,ξ,y[g（t,ξ）]）dσ（ξ）=0,t≥t0,的振动性，得到了这个方程其解振动的充分性条件，推广了PG Wang，WY Shi[J．Appl．Math．Let．，2003，16：1011-1018]中相关结论。     

口味比较与厂家盈利问题

通过考虑两种钙片的口味比较与厂家能否盈利之间的关系时,简化了文献[1]中的模型.设(x1,x2)服从参数为(p1,p2)的二项分布,则厂家能否盈利的问题转为检验H t:(p1,p2)∈Θi,i=1,2,3,三个假设的问题.在使用B ayes方法时,此处选择先验分布为无信息先验,然后求出每种假设的后验概率.     

统筹时间的理想分布

本文提出：统筹时间X服从参数为p．q，r的理想分布β（cm，x）．它较之文献[1],[2]所给分布有精确度高且计算简明的优点．     

Excel计算二元弱酸溶液的分布系数和平衡浓度

本文利用Excel对琥珀酸（二元弱酸）溶液各存在形式的分布系数和平衡浓度进行计算,可以看出利用Excel2000来处理化学计算问题,免除了大量数据的复杂计算,使计算变得简便、快速和准确[1]。     

多级多受灾点连续消耗应急物资调度优化策略

灾后应急救助十分关键,特别是多地同时受灾情况下.为此,考虑将灾民安置与物资运输一体化规划,建立连续消耗多级应急物资调配数学模型.在此基础上,以应急系统施救成本和施救不及时损失为目标函数构建优化模型,并应用遗传算法(GA)和序列线性规划(SLP)算法相结合策略对物资调配进行优化.构造四级救灾物资调配算例模拟救灾过程.数值结果表明,使用GA全局寻优结合SLP局部寻优策略相较单一优化算法能有效找到更优解.所提出的优化策略可为灾后多级多受灾点连续消耗应急物资调度提供可行方案.     

改进系统布置设计法在车间布局设计中的应用

现代车间布局设计时,往往应用基于遗传算法(genetic algorithm, GA)选择最优布局的系统布置设计(systematic layout planning, SLP)方法。为减少GA中轮盘赌模型随机性的影响,融合熵权-逼近理想解排序法(entropy weight-technique for order preference by similarity to an ideal solution, ET)组合评价模型对GA的选择算子进行重新计算,形成ET-GA模式实现SLP方法的改进。最后,利用改进SLP方法对某车间进行布局设计,结合ET-GA模型的改进SLP方法与原方案、基于GA的SLP方法相比,物料搬运成本分别节约了12.50%、5.88%,非物流关系分别提高了3.72%、2.50%,物料搬运时间分别降低了21.02%、14.75%。     

基于动态规划的自动向量化方法

由于SLP自动向量化算法使用的启发式策略会丢失一定的向量化机会,本文提出一种基于动态规划的自动向量化方法DPSLP,该方法采用比SLP更加激进的策略在基本块内寻找候选的SIMD指令分组,依据动态规划方程计算指令分组的代价并从众多指令分组中选择最优的分组进行向量化转换.实验结果显示,DPSLP与SLP相比测试程序的运行时间平均减少了8%,静态指令代价平均减少10%,平均向量宽度增加66.4%.      

解廣義線性互補問題的一個序列線性規劃算法

在適當條件下,給出了廣義線性互補問題的絕對誤差界估計,基于這個誤差界,建立了求解此問題的一個序列線性規劃(SLP)算法,并在不要求存在非退化解的情況下,證明了算法的全局收斂性.     

解垂直线性互补问题的一个序列线性规划算法

建立了求解垂直线性互补问题的一个序列线性规划（SLP）算法,并证明了算法的全局收敛性。     

Spatial teleconnection pattern between Indian Ocean Dipole and the southern high-latitude sea level pressure field

Sea level pressure (SLP) anomalies over the southern high latitudes associated with the Indian Ocean Dipole Mode (IOD) are investigated. Partial correlation and composite analysis depict, for the first time, the seasonal spatial variability in the relationship of SLP field in Southern Ocean with IOD. Results suggest that the IOD signal exists in the southern high latitudes and it is enhanced in boreal autumn, an active season of IOD. On interannual to subdecadal timescales, the spatial teleconnection pattern exhibits a wavenumber-3 pattern around the circumpolar Southern Ocean and the lead-lag correlation analysis shows that there are about 3 months of SLP anomalies in southern high latitudes lagging IOD, which indicates that the response time is almost instantaneous.     

Signals of Antarctic Circum-polar Wave over the Southern Indian Ocean as recorded in an Antarctica ice core

Oxygen stable isotopic and ionic records, covering a period of 1745--1996, are recovered in DT001 ice core drilled in Princess Elizabeth Land, East Antarctica.Using empirical orthogonal function (EOF) analysis of the annually resolved glaciochemical time series, we find the first EOF (EOFI) represents sea-salt aerosols and is the proxy of sea level pressure (SLP) over a quasi-stationary low in the Southern Indian Ocean (SIO). δ^18O represents the sea surface temperature (SST) of the same ocean area. In the past two decades, four climatic waves as represented by SLP and SST proxies are found in the DT001 ice core, which in coincident with four Antarctic Circum-polar Waves (ACW) as revealed by NCEP/NCAR reanalysis. The phase difference between SST and SLP in the ice core is also coincident with that in ACW. Both ice-core record and reanalysis suggestthat there were no signals of ACW during 1958--1980, none during the overall recording period between 1745--1996, as there is no regular phase difference between SST and SLP.The ACW signal after early 1980s is probably attributable to the climate shift occurring over Antarctic Peninsula-Drake Passage region.     

SLP和遗传算法结合在工厂平面布置中的应用

用经典的系统布置设计求得综合相互关系表之后,采用遗传算法求解具体的平面布置方案,以高效率获得可视的、满意的设计结果,弥补传统SLP设计过程中手工操作的繁琐迭代、易受主观影响、结果不稳定等缺点。结合某液压转向器厂的案例对改进的SLP法的具体实施,以及中等规模工厂平面布局问题的有效性、灵敏性作了相关探讨。     

东亚夏季风异常的研究

本文运用经验正交函数分析方法,对10～65°N,90°E～175°W范围内近30年逐年7月和8月平均海平面气压场进行了分析研究.结果表明,第一、第三特征向量与东亚季风异常有关,第二特征向量与西风环流异常有关.讨论了东亚季风异常与大气环流的关系,以及季风异常对华北东部地区降水时空分布的影响.     

Decadal change of the East Asian summer monsoon and its related surface temperature in Asia-Pacific during 1880–2004

The East Asian summer monsoon （EASM） and its related change of surface temperature in the past century were not clearly ad- dressed due to absence of atmospheric reanalysis data before 1948. On the benchmark of station-observed sea level pressure （SLP） in China, we utilized multiple SLP datasets and evaluated their qualities in measuring the SLP-based EASM index （EASMI）. It is found that the EASMI based on the SLP of the Hadley center version 2 （HadSLP2） has shown the best performance on the inter- annual and decadal time scales. Instead of showing a linear weakening trend pointed out by the previous study, the EASMI has likely exhibited the decadal variability, characterized by weakened trends during 1880-1906, 1921-1936, and 1960-2004, and with enhanced trends during 1906-1921 and 1936-1960, respectively. Corresponding to the weakened and enhanced periods of EASMI since the 1920s, the surface air temperature （SAT） index （SATI） averaged in eastern China has likely shown a warming and a cooling trend, respectively. However, the decadal abrupt transitions between the two indices do not occur concurrently, which results in a weak correlation between two indices on the decadal time scale. Further analysis indicates that there are four key regions where the SAT is significantly correlated with the EASMI, suggesting the joint impact of surface temperature in Asia-Pacific on the EASM during 1880-2004. In which, the decadal change of SAT near the Lake Baikal plays an important role in the linear trends of the EASM before and after 1960.