预处理工艺对硬质合金与金刚石膜间粘结力的影响

在两种经不同预处理的硬质合金YG8基底上，采用微波等离子体化学气相沉积法，在微波功率2kW，压强4.0kPa和6.5kPa，CH4和H2流量分别为1.6cm/s和100.0cm/s的条件下生长金刚石薄膜。利用X射线衍射检测了金刚石薄膜是否存在，用拉曼光谱分析了薄膜的质量，用金相显微镜观察了薄膜的洛氏硬度压痕，标定并比较了不同预处理工艺膜与基底的结合力。实验结果表明，不同的预处理方法对于粘结力的影响不大，最主要的因素是钴含量的多少。     

A concise process technology for 3-D suspended radio frequency micro-inductors on silicon substrate

In this letter, a concise process technology is proposed for the first time to enable the fabrication of good quality three-dimensional (3-D) suspended radio frequency (RF) micro-inductors on bulk silicon, without utilizing the lithography process on sidewall and trench-bottom patterning. Samples were fabricated to demonstrate the applicability of the proposed process technology.     

Interfacial segregation of Ti in the brazing of diamond grits onto a steel substrate using a Cu-Sn-Ti brazing alloy

Diamond grits were brazed onto a steel substrate using a prealloyed Cu-10Sn-15Ti (wt pct) brazing alloy at 925 °C and 1050 °C. Due to the relatively high concentration of Ti in the brazing alloy, the braze matrix exhibited a composite structure, composed of β-(Cu,Sn), a Cu-based solid solution, and various intermetallic compounds with different morphologies. The reaction of Ti with diamond yielded a continuous TiC layer on the surfaces of the diamond grits. On top of the TiC growth front, an intermetallic compound, composed of Sn and Ti, nucleated and grew into a randomly interwoven fine lacey structure. An interfacial structure developed as the interwoven fine lacey phase was semicoherently bonded to the TiC layer, with the Cu-based braze matrix filling its interstices. The thickness of such a composite layer was increased linearly with the square root of isothermal holding time at 925 °C, complying with the law of a diffusion-controlled process. However, at 1050 °C, the segregation behavior of Ti and Sn to the interfaces between the TiC layer and the braze matrix diminished, due to the increased solubility of Ti in the Cu-based liquid phase. The enhanced dissolution of Ti in the Cu-based liquid phase at 1050 °C also caused the precipitation of rod-like CuTi with an average diameter of about 0.2 μm during cooling. SnTi₃ was the predominant intermetallic compound and existed in three different forms in the braze matrix. It existed as interconnected grains of large size which either floated to the surface of the braze matrix or grew into faceted grains. It also exhibited a nail-like structure with a mean diameter of about 1 μm for the rod section and a lamellar structure arising from a eutectic reaction during cooling.     

A random switching method for HPWM full-bridge inverter

To overcome the problem of unequal switching loss in power switches, in conventional hybrid pulse width modulation (HPWM) full-bridge inverters, a random switching method for HPWM full-bridge inverters is proposed. The proposed method equalizes switching losses of the four switches, while also providing good output performance     

全IP移动通信网网络层移动性管理

首先介绍了全IP移动通信网网络层移动性管理的概念、框架和最经典的移动IP（Mobile IP）协议，然后讨论了基于Mobile IP的四种主要扩展方案：蜂窝IP、切换敏感无线接入Internet结构（HAWAII)、电信增强移动IP（TeleMIP）和边缘移动性体系结构（EMA）方案，最后结合第三代移动通信发展的状况得出了一些结论。     

深穿透复合射孔技术在中原油田的应用

深穿透复合射孔是一项把射孔与高能气体压裂两种技术有机地结合在一起的综合改造油气层的新技术。重点对它的技术原理，包括设计原理、枪身内火药燃烧气体对射孔孔眼的冲刷作用原理、火药燃烧气体射流的高能气体压裂作用原理等进行了研究与探讨，同时对它的施工工艺、现场应用等方面进行了阐述，经中原油田189口井现场应用证明，其工艺简便，成本低廉，可实现多层跨隔层使用，处理后增产效果显著，投入产出比高，可代替传统的聚能射孔技术，具有广阔的推广和应用价值。     

等偏移距自动初至静校正

茅金根 ,梁秀文 ,杨午阳 ,冯有奎 .等偏移距自动初至静校正 .石油地球物理勘探 ,2 0 0 2 ,37(5 ) :4 6 9～ 4 72　　本文提出一种自动计算静校正量的新方法。它首先在等偏移距道集内对相邻道的初至波采用高精度相关构建一个关于炮、检时差的超大矩阵方程 ,然后采用优化统计迭代算法求解炮点和检波点的静校正量 ,并用概率分布函数确定精确的炮、检静校正量 ,最后用误差校正函数消除炮、检静校正误差。实际资料处理结果表明 ,本方法自动、省时、精度高。     

试论会计电算化系统的内部控制

阐述了会计电算化系统产生和发展对内产控制的影响，以及内部控制的实施方法。     

Improving signal prediction performance of neural networks through multiresolution learning approach.

In this paper, we extend the original work on multiresolution learning for neural networks, and present new developments on the multiresolution learning paradigm. The contributions of this paper include: (1) proposing a new concept and method of adjustable neural activation functions in multiresolution learning to improve neural network learning efficacy and generalization performance for signal predictions; (2) providing new insightful explanations for the multiresolution learning paradigm from a multiresolution optimization perspective; (3) exploring underlying ideas why the multiresolution learning scheme associated with adjustable activation functions would be more appropriate for the multiresolution learning paradigm; and (4) providing rigorous validations to evaluate the multiresolution learning paradigm with adjustable activation functions and comparing it with the schemes of multiresolution learning with fixed activation functions and traditional learning. This paper presents systematically new analytical and experimental results on the multiresolution learning approach for training an individual neural network model, demonstrates our integral solution on neural network learning efficacy, and illustrates the significant improvements on neural networks' generalization performance and robustness for nonlinear signal predictions.