基于半马尔科夫条件随机场的命名体识别及其关系抽取研究

随着互联网和电子化办公的发展，出现了大量的文本资源。信息抽取技术可以帮助人们快速获取大规模文本中的有用信息。命名体识别与关系抽取是信息抽取的两个基本任务。本文在调研当前命名体识别和实体关系抽取中采用的主要方法的基础上，分别给出了解决方案。论文开展的主要工作有：（1）从模型选择和特征选择两个方面总结了命名体识别及实体关系抽取的国内外研究现状，重点介绍用于命名体识别的统计学习方法以及用于实体关系抽取的基于核的方法。（2）针对当前命名体识别中命名体片段边界的确定问题，研究了如何将 Semi-Markov CRFs 模型应用于中文命名体识别。这种模型只要求段间遵循马尔科夫规则，而段内的文本之间则可以被灵活的赋予各种规则。将这种模型用于中文命名体识别任务时，我们可以更有效更自由的设计出各种有利于识别出命名体片段边界的特征。实验表明，加入段相关的特征后，命名体识别的性能提高了 4-5 个百分点。（3）实体关系抽取的任务是判别两个实体之间的语义关系。之前的研究已经表明，待判别关系的两个实体间的语法树结构对于确定二者的关系类别是非常有用的，而相对成熟的基于平面特征的关系抽取方法在充分提取语法树结构特征方面的能力有限，因此，本文研究了基于核的中文实体关系抽取方法。针对中文特点，我们探讨了卷积（Convolution）核中使用不同的语法树对中文实体关系抽取性能的影响，构造了几种基于卷积核的复合核，改进了最短路依赖核。因为核方法开始被用于英文关系抽取时，F1 值也只有40%左右，而我们只使用作用在语法树上的卷积核时，中文关系抽取的F1 值达到了35%，可见核方法对中文关系抽取也是有效的。

Electrochemical impedance spectroscopy analysis for oxygen reduction reaction in 3.5% NaCl solution

The electrochemical impedance spectroscopy (EIS) at different potentials has been used to study the oxygen reduction reaction (ORR) in 3.5% NaCl solution on glassy carbon (GC) electrode in this work. Results show that ORR consists of three two-electron reaction steps and both superoxide ion (O-2(-)) and hydrogen peroxide (H2O2), which are produced by ORR, obstruct the diffusion of oxygen to the surface of the electrode and make the EIS results change into a transmissive finite diffusion process with the real part contraction and a reflective finite diffusion process from a semi-infinite diffusion process. The values of electron transfer resistance (R-t) and diffusion resistance (R-d) were calculated from EIS. O-2(-) influenced strongly on the Rt values and induced a maximum at -0.45 V.

强化学习算法中启发式回报函数的设计及其收敛性分析

回报函数设计的好与坏对学习系统性能有着重要作用,按回报值在状态-动作空间中的分布情况,将回报函数的构建分为两种形式:密集函数和稀疏函数,分析了密集函数和稀疏函数的特点.提出启发式回报函数的基本设计思路,利用基于保守势函数差分形式的附加回报函数,给学习系统提供更多的启发式信息,并对算法的最优策略不变性和迭代收敛性进行了证明.启发式回报函数能够引导学习,加快学习进程,从而可以实现强化学习在实际大型复杂系统应用中的实时控制和调度.

基于MRF模型的油气储层属性随机建模方法研究

Stochastic reservoir modeling is a technique used in reservoir describing. Through this technique, multiple data sources with different scales can be integrated into the reservoir model and its uncertainty can be conveyed to researchers and supervisors. Stochastic reservoir modeling, for its digital models, its changeable scales, its honoring known information and data and its conveying uncertainty in models, provides a mathematical framework or platform for researchers to integrate multiple data sources and information with different scales into their prediction models. As a fresher method, stochastic reservoir modeling is on the upswing. Based on related works, this paper, starting with Markov property in reservoir, illustrates how to constitute spatial models for catalogued variables and continuum variables by use of Markov random fields. In order to explore reservoir properties, researchers should study the properties of rocks embedded in reservoirs. Apart from methods used in laboratories, geophysical means and subsequent interpretations may be the main sources for information and data used in petroleum exploration and exploitation. How to build a model for flow simulations based on incomplete information is to predict the spatial distributions of different reservoir variables. Considering data source, digital extent and methods, reservoir modeling can be catalogued into four sorts: reservoir sedimentology based method, reservoir seismic prediction, kriging and stochastic reservoir modeling. The application of Markov chain models in the analogue of sedimentary strata is introduced in the third of the paper. The concept of Markov chain model, N-step transition probability matrix, stationary distribution, the estimation of transition probability matrix, the testing of Markov property, 2 means for organizing sections-method based on equal intervals and based on rock facies, embedded Markov matrix, semi-Markov chain model, hidden Markov chain model, etc, are presented in this part. Based on 1-D Markov chain model, conditional 1-D Markov chain model is discussed in the fourth part. By extending 1-D Markov chain model to 2-D, 3-D situations, conditional 2-D, 3-D Markov chain models are presented. This part also discusses the estimation of vertical transition probability, lateral transition probability and the initialization of the top boundary. Corresponding digital models are used to specify, or testify related discussions. The fifth part, based on the fourth part and the application of MRF in image analysis, discusses MRF based method to simulate the spatial distribution of catalogued reservoir variables. In the part, the probability of a special catalogued variable mass, the definition of energy function for catalogued variable mass as a Markov random field, Strauss model, estimation of components in energy function are presented. Corresponding digital models are used to specify, or testify, related discussions. As for the simulation of the spatial distribution of continuum reservoir variables, the sixth part mainly explores 2 methods. The first is pure GMRF based method. Related contents include GMRF model and its neighborhood, parameters estimation, and MCMC iteration method. A digital example illustrates the corresponding method. The second is two-stage models method. Based on the results of catalogued variables distribution simulation, this method, taking GMRF as the prior distribution for continuum variables, taking the relationship between catalogued variables such as rock facies, continuum variables such as porosity, permeability, fluid saturation, can bring a series of stochastic images for the spatial distribution of continuum variables. Integrating multiple data sources into the reservoir model is one of the merits of stochastic reservoir modeling. After discussing how to model spatial distributions of catalogued reservoir variables, continuum reservoir variables, the paper explores how to combine conceptual depositional models, well logs, cores, seismic attributes production history.

Study on methane hydration process in a semi-continuous stirred tank reactor

The methane hydration process is investigated in a semi-continuous stirred tank reactor. Liquid temperatures and reaction rates without stirrer are compared with those occurring with stirrer, while at the same time better stirring conditions of the methane hydration process are given by the experiments. Some basic data of fluid mechanics, for example, stirring Reynolds number, Froucle number and stirrer power, are calculated during the methane hydration process, which can be applied to evaluate stirrer capacity and provide some basic data for a scaled up reactor. Based on experiment and calculations in this work, some conclusions are drawn. First, the stirrer has great influence on the methane hydration process. Batch stirring is helpful to improve the mass transfer and heat transfer performances of the methane hydration process. Second, induction time can be shortened effectively by use of the stirrer. Third, in this paper, the appropriate stirring velocity and stirring time were 320 rpm and 30 min, respectively, at 5.0 MPa, for which the storage capacity and reaction time were 159.1 V/V and 370 min, respectively. Under the condition of the on-flow state, the initial stirring Reynolds number of the fluid and the stirring power were 12,150 and 0.54 W, respectively. Fourth, some suggestions, for example, the use of another type of stirrer or some baffles, are proposed to accelerate the methane hydration process. Comparing with literature data, higher storage capacity and hydration rate are achieved in this work. Moreover, some fluid mechanics parameters are calculated, which can provide some references to engineering application.

An application of the analytic hierarchy process and fuzzy logic inference in a decision support system for forage selection

Forage selection plays a prominent role in the process of returning cultivated lands back into grasslands. The conventional method of selecting forage species can only provide attempts for problem-solving without considering the relationships among the decision factors globally. Therefore, this study is dedicated to developing a decision support system to help farmers correctly select suitable forage species for the target sites. After collecting data through a field study, we developed this decision support system. It consists of three steps: (1) the analytic hierarchy process (AHP), (2) weights determination, and (3) decision making. In the first step, six factors influencing forage growth were selected by reviewing the related references and by interviewing experts. Then a fuzzy matrix was devised to determine the weight of each factor in the second step. Finally, a gradual alternative decision support system was created to help farmers choose suitable forage species for their lands in the third step. The results showed that the AHP and fuzzy logic are useful for forage selection decision making, and the proposed system can provide accurate results in a certain area (Gansu Province) of China.

A Semi-Empirical Equation of Penetration Depth on Concrete Target Impacted by Ogive-Nose Projectiles

In this paper, the penetration process of ogive-nose projectiles into the semi-infinite concrete target is investigated by the dimensional analysis method and FEM simulation. With the dimensional analysis, main non-dimensional parameters which control the penetration depth are obtained with some reasonable hypothesis. Then, a new semi-empirical equation is present based on the original work of Forrestal et al., has only two non-dimensional combined variables with definite physical meanings. To verify this equation, prediction results are compared with experiments in a wide variation region of velocity. Then, a commercial FEM code, LS-DYNA, is used to simulate the complex penetration process, that also show the novel semi-empirical equation is reasonable for determining the penetration depth in a concrete target.

A Semi-Empirical Equation of Penetration Depth on Concrete Target Impacted by Ogive-Nose Projectiles

In this paper, the penetration process of ogive-nose projectiles into the semi-infinite concrete target is investigated by the dimensional analysis method and FEM simulation. With the dimensional analysis, main non-dimensional parameters which control the penetration depth are obtained with some reasonable hypothesis. Then, a new semi-empirical equation is present based on the original work of Forrestal et al., has only two non-dimensional combined variables with definite physical meanings. To verify this equation, prediction results are compared with experiments in a wide variation region of velocity. Then, a commercial FEM code, LS-DYNA, is used to simulate the complex penetration process, that also show the novel semi-empirical equation is reasonable for determining the penetration depth in a concrete target.

Simulation of growth process of thin film on non-planar substrate

In (2 + 1) dimension, growth process of thin film on non-planar substrate in Kuramoto-Sivashinsky model is studied with numerical simulation approach. 15 x 15 semi-ellipsoids arranged orderly on the surface of substrate are used to represent initial rough surface. The results show that at the initial stage of growth process, the surface morphology of thin film appears to be grid-structure, and the interface width constantly decreases with the growth time, then reaches minimum. However, the grid-structure becomes ambiguous, and granules of different sizes distribute evenly on the surface of thin film with the increase of growth time. Thereafter, the average size of granules and the interface width gradually increase, and the surface morphology of thin film presents fractal properties. The numerical results of height-height correlation functions of thin film verify the surface morphology of thin film to be fractal for a longer growth time. By fitting of the height-height correlation functions of thin film with different growth times, the growth process is described quantitatively. (c) 2004 Elsevier B.V. All rights reserved.

Compact Reconfigurable Binary-Decision-Diagram Logic Circuit on a GaAs Nanowire Network

We describe a reconfigurable binary-decision-diagram logic circuit based on Shannon's expansion of Boolean logic function and its graphical representation on a semiconductor nanowire network. The circuit is reconfigured by using programmable switches that electrically connect and disconnect a small number of branches. This circuit has a compact structure with a small number of devices compared with the conventional look-up table architecture. A variable Boolean logic circuit was fabricated on an etched GaAs nanowire network having hexagonal topology with Schottky wrap gates and SiN-based programmable switches, and its correct logic operation together with dynamic reconfiguration was demonstrated.

Realization of extremely broadband quantum-dot superluminescent light-emitting diodes by rapid thermal-annealing process

The first demonstration, to our knowledge, of the creation of ultrabroadband superluminescent light-emitting diodes using multiple quantum-dot layer structure by rapid thermal-annealing process is reported. The device exhibits a 3 dB emission bandwidth of 146 nm centered at 984 mm with cw output power as high as 15 mW at room temperature corresponding to an extremely small coherence length of 6.6 mu m. (C) 2008 Optical Society of America.

Optimizing the GaAs capping layer growth of 1.3 mu m InAs/GaAs quantum dots by a combined two-temperature and annealing process at low temperatures

We report on optimizing the GaAs capping layer growth of 1.3 mu m InAs quantum dots (QDs) by a combined two-temperature and annealing process at low temperatures using metalorganic chemical vapor deposition. The initial part (tnm) of the capping layer is deposited at a low temperature of 500 degrees C, which is the same for the growth of both the QDs and a 5-nm-thick In0.15Ga0.85As strain-reducing capping layer on the QDs, while the remaining part is grown at a higher temperature of 560 degrees C after a rapid temperature rise and subsequent annealing period at this temperature. The capping layer is deposited at the low temperatures (<= 560 degrees C) to avoid postgrowth annealing effect that can blueshift the emission wavelength of the QDs. We demonstrate the existence of an optimum t (=5 nm) and a critical annealing time (>= 450s) during the capping, resulting in significantly enhanced photoluminescence from the QDs. This significant enhancement in photoluminescence is attributed to a dramatic reduction of defects due to the optimized capping growth. The technique reported here has important implications for realizing stacked 1.3 mu m InAs/GaAs QD lasers. (C) 2008 Elsevier B.V. All rights reserved.

Enhancement of the spectral width of high-power 1.5 mu m integrated superluminescent light source by quantum well intermixing process

A novel type of integrated InGaAsP superluminescent light source was fabricated based on the tilted ridge-waveguide structure with selective-area quantum well (QW) intermixing. The bandgap structure along the length of the device was modified by impurity free vacancy diffusion QW intermixing, The spectral width was broadened from the 16 nm of the normal devices to 37 nm of the QW intermixing enhanced devices at the same output power level. High superluminescent power (210 mW) was obtained under pulsed conditions with a spectral width of 37 nm.

1.55 mu m spot-size converter with monolithically integrated laser diode by conventional process

A novel 1.55 mum laser diode (LD) with monolithically integrated spot-size converter (SSC) is designed and fabricated using conventional photolithography and the chemical wet etching process. For the laser diode, a ridge double-core structure is employed. For the spot-size converter, a buried double-waveguide structure is incorporated. The laterally tapered active core is designed and optically combined with the thin passive core to control the size of the mode. The threshold current was measured to be 40 mA together with high slope efficiency of 0.35 W A(-1). The beam divergence angles in the horizontal and vertical directions were as small as 14.9degrees and 18.2degrees, respectively.