带先验知识的波阻抗反演正则化方法研究

针对波阻抗反演中存在的不适定性问题,本文提出了一种带先验知识的正则化重开始共轭梯度法.该方法的内层循环采用修改的共轭梯度法,并使用重开始技巧;外层循环使用Morozov偏差准则作为停机准则.正则参数的选取采用连续几何选取法.克服了传统共轭梯度法迭代不足或迭代过度的缺点,将迭代步数控制在了合适的范围,使算法能够更快速更准确的收敛.同时考虑了用最速下降法计算先验解和对解施加非均一的规范约束.通过理论模型试算和实际资料处理,并与共轭梯度法进行对比,表明该算法具有精度高、抗病态能力强,运算速度快的优点,具有实用性.     

时间域航空电磁2.5维非线性共轭梯度反演

对于时间域航空电磁法二维和三维反演来说,最大的困难在于有效的算法和大的计算量需求.本文利用非线性共轭梯度法实现了时间域航空电磁法2.5维反演方法,着重解决了迭代反演过程中灵敏度矩阵计算、最佳迭代步长计算、初始模型选取等问题.在正演计算中,我们采用有限元法求解拉式傅氏域中的电磁场偏微分方程,再通过逆拉氏和逆傅氏变换高精度数值算法得到时间域电磁响应.在灵敏度矩阵计算中,采用了基于拉式傅氏双变换的伴随方程法,时间消耗只需计算两次正演,从而节约了大量计算时间.对于最佳步长计算,二次插值向后追踪法能够保证反演迭代的稳定性.设计两个理论模型,检验反演算法的有效性,并讨论了选择不同初始模型对反演结果的影响.模型算例表明:非线性共轭梯度方法应用于时间域航空电磁2.5维反演中稳定可靠,反演结果能够有效地反映地下真实电性结构.当选择的初始模型电阻率值与真实背景电阻率值接近时,能得到较好的反演结果,当初始模型电阻率远大于或远小于真实背景电阻率值时反演效果就会变差.     

裂陷盆地基底双界面模式二维重力反演

裂陷盆地基底的起伏表现为非光滑的几何形态,传统的重力反演结果并不能很好地反映这种特点.此外,大多数情况下,重力观测面并不位于盆地上界面,应为单独的起伏观测面,盆地应为上界面和基底组成的双界面模式.基于此,本文研究了起伏观测面上裂陷盆地基底双界面模式二维重力反演方法.研究中假设沉积盆地的沉积层与基底的密度差随深度按双曲线规律变化.将沉积盆地的沉积层剖分成相邻的垂直柱体,其水平尺寸是已知的,顶面与沉积层上界面重合,底面深度代表基底的深度,即为要反演的参数.反演中引入全变差函数作为盆地模型的约束,使得反演结果呈现非光滑形态,符合裂陷盆地基底特征.为减小反演多解性,引入已知深度点作为约束.建立由重力数据拟合、已知深度约束及全变差函数组成的目标函数,采用非线性共轭梯度算法使目标函数最小化.模型试算结果表明该方法可反演裂陷盆地基底起伏,并通过调整正则化参数的值可反演坳陷盆地基底起伏.将该反演方法用于珠江口盆地惠州凹陷和运城-临汾裂陷盆地实际资料处理,其结果较好地反映了裂陷盆地基底起伏特征,为研究盆地构造、油气勘探等提供重要参考.     

极区电离层F2层峰值电子浓度对太阳活动依赖性的共轭研究

本文利用南极中山站(ZHS),以及北极与其地理共轭的Tromso站(TRO)、地磁共轭的Longyearbyen站(LYB)各自约一个太阳活动周的观测数据,对比分析了极区电离层F₂层峰值电子浓度(N_mF₂)对太阳活动的依赖性.结果表明,三个台站N_mF₂月中值随修正太阳10.7 cm通量指数F_10.7P(简称P)增大在总体上呈线性增长,这说明在这三个台站,太阳辐射仍是其F₂层主要电离源.其中TRO站N_mF₂与P线性关系最好,ZHS站的次之,LYB站的最差.在日变化中,TRO站N_mF₂对太阳活动响应最为敏感的时刻出现在地方时中午附近,LYB站出现在磁中午,ZHS站则出现在地方时中午和磁地方时中午之间.这主要是由地理/地磁纬度差异引起的不同强度的光致电离与极区等离子体对流共同作用的结果.在年变化中,TRO站N_mF₂随太阳活动变化上升最快的季节出现在冬季,夏季上升最慢.在ZHS站与LYB站,N_mF₂对太阳活动变化的响应都在两分季最为敏感.这种季节上的差异则是由于三个台站光致电离与中性大气成分R[O/N₂]的不同所致.     

Interpretation of the experimental data provided by Gómez-Rivas and Griera(2012) in terms of the MEM-criterion

The systematical experiments carried out by Gómez-Rivas and Griera(2012) demonstrate that the ductile shear zones initiate at ~55° to ?1 just as predicted by the MEM-criterion. However, the data are explained in terms of dilatancy, which requires many prerequisites and implies that the ±55° angle is only valid for the used material. In contrast, the ~55? predicted by the MEM-criterion is material independent, which makes it widely applicable to explaining the development of ductile shear zones in nature.     

反假频非均匀地震数据重建方法研究

研究基于Fourier变换的数据重建方法,既能进行非均匀采样数据重建,又可以去除空间假频. 将不规则采样数据重建问题归结为信息重建的地球物理反演问题,采用最小二乘方法从观测的稀疏或不规则数据反演模型空间完全信息. 在求解信息重建反演问题时,引入DFT 加权范数规则化策略,采用预条件共轭梯度法（PCG）求解,保证解的稳定性和收敛速度. 处理线性同相轴假频问题时,根据采样定理,引入线性预测方法,采用Yule Walker方程由带限信号的无假频低频功率谱预测高频功率谱,达到反假频目的. 本文研究了均匀采样数据内插,非均匀采样数据重建,非均匀分布高频信息重建等方面问题,数值试验取得较好效果.     

青海鄂拉山断裂带晚第四纪构造活动及其所反映的青藏高原东北缘的变形机制

鄂拉山断裂带是分隔青海乌兰盆地 (柴达木盆地的一部分 )与茶卡—共和盆地的一条重要边界断裂 ,长约 2 0 7km ,由 6条规模较大的主要以右阶或左阶次级断裂段羽列而成 ,阶距约 1～ 3.5km。该断裂右旋走滑的起始时代为第四纪初期 ,约在 1.8～ 3.8MaB .P .期间 ,大的地质体累积断错约 9～12km。断裂新活动形成了一系列山脊、冲沟和阶地等的右旋断错及断层崖、断层陡坎等。晚更新世晚期以来 ,鄂拉山断裂带的平均水平滑动速率为 (4 .1± 0 .9)mm/a ,垂直滑动速率为 (0 .15± 0 .1)mm/a。鄂拉山地区的构造变形受区域NE向构造应力作用下的剪切压扁与鄂拉山断裂的右旋剪切和挤压的共同影响 ,共和—茶卡盆地和乌兰盆地均属于走滑挤压型盆地。青藏高原东北缘地区在区域性北东向挤压的作用之下 ,应变被分解为沿北西西向断裂的左旋走滑和沿北北西向断裂的右旋走滑运动 ,形成一对共轭的剪切断裂。鄂拉山断裂及其他北北西走向断裂的发展演化和变形机制表明青藏高原东北缘向东的挤出和逃逸是非常有限的。     

天文大地网与空间网联合平差中用共轭梯度法解算高阶法方程的某些策略

本文研究了用共轭梯度法解算天文大地网与空间网联合平差中２０万阶法方程的有关问题,提出了“系数调整策略”和“平差与求逆既分又合的策略”。利用的天文大地网（４７０５７点）和空间网（４７６点）在ＰＩＩ／２３３微机上解算１４．１万阶方程组共迭代４９９３次,花费机时２８分４７秒,与未进行系数调整相比,功效提高４０多倍。     

Expression of Mammalian Metallothionein-I Gene in Cyanobacteria to Enhance Heavy Metal Resistance

Metallothionein (MT) from mammals is a low molecular weight (about 6–7 KD), and cystein-rich (20 cystein per molecular) protein (Vallee, 1991). So MT has the strong binding ability of metal such as Cd, Hg and Pd. Wild-typed cyanobacteria have their own metallothionein-like proteins, which are of low cysteine content and have weak ability to bind heavy metal. Therefore, it is imperative to transfer the MT-I gene from mammal into cyanobacteria cell for the absorption and accumulation of heavy metals in aqueous system including marine and lake ( Misra and Gedamu, 1989). We inserted the MT-I cDNA gene after the strong promoter PpsbA or Psmt into the intermediary vector pRL-439, and then pRL-MT was ligated with shuttle vector pKT-210 (a shuttle vector pDC-08 was used at an earlier time) to construct the shuttle expression vector pKT-MT. (The MT-I mutant -KKS- also was selected as aim gene.) Then the constructed vector was introduced into cyanobacterium Anabaena sp. PCC7120, Synechocystis sp. PCC6803 and Synechococcus sp. PCC7942 with triparental conjunctive transfer (Bryant, 1994). The expression efficiency is about 1%. We also used homologous recombination vector pTZ18-8 to introduce MT-I cDNA into chromosomes in cyanobacteria and got higher expression efficiency. Metal tolerance tests show that the transgenic cyanobacteria acquire metal resistance up to 60 μmol/g CdCl₂ and expression efficiency up to 1045 μg MT/g fresh cells according to the data of ELISA.     

Study and application of an improved four-dimensional variational assimilation system based on the physical-space statistical analysis for the South China Sea

The four-dimensional variational assimilation (4D-Var) has been widely used in meteorological and oceanographic data assimilation. This method is usually implemented in the model space, known as primal approach (P4D-Var). Alternatively, physical space analysis system (4D-PSAS) is proposed to reduce the computation cost, in which the 4D-Var problem is solved in physical space (i.e., observation space). In this study, the conjugate gradient (CG) algorithm, implemented in the 4D-PSAS system is evaluated and it is found that the non-monotonic change of the gradient norm of 4D-PSAS cost function causes artificial oscillations of cost function in the iteration process. The reason of non-monotonic variation of gradient norm in 4D-PSAS is then analyzed. In order to overcome the non-monotonic variation of gradient norm, a new algorithm, Minimum Residual (MINRES) algorithm, is implemented in the process of assimilation iteration in this study. Our experimental results show that the improved 4D-PSAS with the MINRES algorithm guarantees the monotonic reduction of gradient norm of cost function, greatly improves the convergence properties of 4D-PSAS as well, and significantly restrains the numerical noises associated with the traditional 4D-PSAS system.