关于系数平方增长的带跳BSDE的解(Ⅰ)

讨论了系数关于q为平方增长,p和-y为指数增长的带跳倒向随机微分方程(BSDE)解的存在性,以及有这种系数的反射BSDE解的存在性.     

常系数非齐次线性微分方程组特解公式的新推导及其应用

首先利用推广的分部积分法导出一阶线性方程组的两个特解公式,然后将有关的结果应用到高阶线性方程(组),得出了特解的一些新公式。     

Atomic-scale computer simulation for early precipitation process of Ni75Al10V15 alloy

The kinetic model for a ternary system is introduced based on the microscopic diffusion form of the phase-field equations for a binary alloy. The equation is solved in the reciprocal space. This model is used to investigate the early precipitation process of Ni75-Al10V15 by simulating the atomic pictures of the two ordered phases and calculating the order parameters of γ′(Ni3Al) phase. Simulation results show that the γ′ ordered phases precipitate from the disordered matrix by a non-classical nucleation mechanism, and the nonstoichiometric γ′ ordered phase appears first and then transforms into the stoichiometric one. Clusters of V atoms appear at the γ′ phase boundaries followed by the formation of the nonstoichiometric θ ordered phase. The farther the location from γ′ phase boundary is, the lower the order degree of θ phase is. There exist two kinds of DO22 ordered domains: a horizontal one and a vertical one, related to their adjacent γ′ phase boundaries. The model could describe the atomic ordering and composition clustering simultaneously, and any a priori assumption about the new phase structure and precipitation mechanism etc. is unnecessary.     

Computer simulation for the precipitation process of Ni75Al7.5V17.5 alloy

The precipitation mechanism of Ni75Al7.5V17.5 alloy above the L12 instability line, between the L12 and D022 instability lines and below the D022 instability line are studied using microscopic phase-field kinetic equation. This paper is aimed at investigating the effect of temperature on precipitation mechanism and morphological evolution of the alloy. Our simulations demonstrate that the precipitation is a mixed mechanism of non-classical nucleation growth and spinodal decomposition above the L12 instability line. It needs certain thermal fluctuations for nucleation and the number of θ phases is small at this temperature. The precipitation mechanism of γ'phase is congruent ordering followed by spinodal decomposition, and θ phase is a mixed mechanism of non-classical nucleation growth and spinodal decomposition between the L12 and D022 instability lines. The mechanism below the D022 instability line is similar to that between the L12 and D022 instability lines. With the decrease of the temperature, ordering and phase separation becomes fast, the dimension of γ'phase becomes small, the shape transforms from equiaxed to block, the dimension of θ phase becomes large and the shape transforms from strip to circle.     

基于质心分析的淮河流域降水时空分布特征

为定量表征淮河流域降水量非均匀性特征,建立年降水质心时间指标,以概化年内降水过程时间分布,从时域、空域、频域3个维度解析流域近55 a降水质心时间的总体格局与演变规律。结果表明:各典型站点近55a降水质心时间序列中全部存在28a的第一主周期;年降水质心时间序列突变年份多发生于1975—1989年,突变主要集中于淮河上中游地区;年降水质心时间序列多呈现不显著的增加趋势,说明流域大部分地区年降水集中期有微弱的后移倾向;年降水质心时间随纬度增加而呈现明显的后推趋势,流域南北地区年降水质心时间可相差20 d左右,经度变化对年降水质心时间没有明显影响;淮河流域暴雨与大雨质心时间普遍迟于其他等级年降水质心时间,流域年降水质心时间对暴雨、大雨较为敏感。     

正火对高Cr马氏体耐热钢组织和性能的影响

通过光学显微镜、透射电镜(TEM)及扫描电镜(SEM)分析,研究了正火处理对高Cr马氏体耐热钢显微组织、力学性能及断裂机理的影响.结果表明,900~970℃正火后,晶粒尺寸十分细小,在10μm以下.1060~1200℃正火晶粒尺寸迅速增长,1060℃正火晶粒尺寸约为33μm.经过1060℃×2h正火+760℃×3h回火热处理后,室温和600℃高温拉伸屈服强度分别达到535MPa和380MPa,综合力学性能优良,而1060℃长时间正火对力学性能并无明显影响.1060℃×2h正火+760℃×3h回火热处理后得到破碎的晶粒细小回火马氏体组织,晶界上200~300nm M₂₃C₆和晶内5~50nm MX型弥散析出有效地阻碍位错运动,进而提高了材料力学性能.     

高铝熔渣物相析出演变规律研究

以改善高铝渣浸出性能为目标,采用DSC实验和SEM-EDS检测方法,研究了高铝渣控温冷却过程中物相析出规律,考察了降温速率、n(CaO)/n(Al₂O₃)和MgO含量对铝酸钙渣物相析出演变规律的影响.结果表明:当降温速率加快时,熔渣凝固过程趋于更大程度的偏析,C₂S相变愈不充分,粒度粗化,劣化了自粉行为和浸出条件;随着调高渣中n(CaO)/n(Al₂O₃),C₁₂A₇的析出温度升高,物相析出活性按照CA→C₁₂A₇→C₃A逐渐增强;当渣中含有MgO时,物相中出现C₂MS₂,不利于C₂S物相转变,更为重要的是MgO富集在含铝物相中形成Q相,显著恶化氧化铝浸出率和收得率.     

格林函数法求解一维系统中两介质的戴逊方程

用格林函数方法推导出了2个介质存在的准一维系统中电子的格林函数戴逊方程,给出了电子处于2个最低标准模时的解及透射系数与反射系数,并得出同一个亚能带之间的格林函数与不同亚能带之间的格林函数都服从透射系数的关系。     

一种新的项目反应理论等值准则——余弦准则

根据同一被试在锚题上的得分概率向量的模式应该十分相似的直观想法,该文提出了一种新的项目反应理论框架下的等值准则——余弦准则.我们采用GPC模型,利用Monte Carlo模拟方法将余弦准则与Haebara等值准则(Hcrit)和Stocking-Lord等值准则(SLcrit)进行比较.研究结果表明,当项目参数估计精度较高或中等而A的取值范围在0.9~2.0之间,新方法往往比Hcrit和SLcrit方法估计的误差小且有显著性差异.     

一阶迎风格式和二阶ENO Local Lax-Friedrich格式修正系数研究

针对一阶迎风格式和二阶ENO Local Lax-Friedrich格式,通过修正系数法对其进行修正,在不增加模板节点数的情况下,将逼近精度提高一阶,分别称为MCupwind格式和MCENOLLF格式。理论与数值模拟实验表明,修正后的格式保持了原有格式的所有优点,在最坏的情况下自动降为原格式,修正系数法若能在间断或激波附近引入间断解思想,将会有更好的发展前景。