构造性神经网络在测井岩性识别中的应用

测井岩性识别是石油勘探中十分重要的基础工作,准确的岩性识别结果可以为勘探和开发提供可靠的依据。人工神经网络方法可对复杂的高维数据进行非线性映射,在模式识别领域得到越来越广泛的应用。针对传统的BP神经网络算法存在收敛速度慢、隐层数以及隐层节点数难以确定等缺点,研究了一种构造性神经网络学习算法---级联算法（Cascade Correlation Algorithm,简称CC算法）及其在石油工程中的应用。采用该算法对甘肃镇原-泾川地区进行岩性识别研究,通过与BP神经网络的识别结果进行比较,体现出构造性神经网络的优越性。     

靖边气田分离器分离效果对比分析

从理论计算分离器处理能力和现场开展试验两方面入手，对靖边气田目前在用的不同规格分离器的分离效果进行了对比分析，通过对现场数据统计和整理，得到了靖边气田不同规格分离器合理的处理气量，给气田开发建设中分离器的选型和维护改造中分离器的调整提供了依据。     

排气结构对导叶式旋风管分离性能的影响

通过对比实验测定了不同排气结构下导叶式旋风管的分离效率及压降,同时用智能型五孔球探针测试仪测定了不同排气结构的旋风管内流场分布,结合性能实验与流场测量的结果,研究了排气芯管内径及其结构型式对分离性能的影响,为旋风管结构优化和开发及其分离机理的研究提供了一定的实验基础。     

金属/聚合物复合材料对中能电子的屏蔽作用研究

运用MULASSIS计算程序模拟计算了不同组成的屏蔽材料对空间辐射环境的屏蔽效果,在此基础上进行屏蔽材料的初步设计,并制备了一系列不同组成的金属/聚合物复合材料,结合中能电子照射的实验结果,讨论了组成对金属/聚合物复合材料的屏蔽效果的影响.研究发现,高原子序数和低原子序数金属混合添加的复合物样品具有较好的屏蔽效果.     

Absolute Measurements of the Magnetic Field Generated by Different Coils in the Center of EGYPTOR Tokamak

The present work is devoted to measure the absolute magnetic field produced by different coils in the EGYPTOR tokamak using a calibrated pickup coil. Scaling these measurements in different equations connected with the discharge currents from each supply system are performed. The pickup coil used in the present study is well calibrated with Helmholz coils at the IPP in Prague, Czech Republic. A 0.2% deviation has been found between an evaluation done in the present study and the calibration using Helmholz coils. Experimental measurements of the toroidal magnetic field are in good agreement with calculations to within 2%. Very low values of stray magnetic field components arising from TF and OH coils are recorded which proves that the compensation coils for these components are sufficient.     

Performance prediction of a hybrid‐excitation synchronous machine with axially arranged excitation poles and permanent‐magnet poles

This paper presents a method of predicting the steady‐state performance of a new hybrid‐excitation synchronous machine (HESM) theoretically. The field pole of this HESM is axially divided into two parts; one is an excitation part and the other a permanent‐magnet (PM) part. A nonlinear equivalent circuit, which can include the saliency of the rotor and the magnetic saturation due to the iron core, is derived. Based on this equivalent circuit, the steady‐state performance of the HESM is calculated, and the results are confirmed through experiments. © 2004 Wiley Periodicals, Inc. Electr Eng Jpn, 150(2): 43–49, 2005; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.20033     

Inference and learning in hybrid probabilistic network

This paper proposed a novel hybrid probabilistic network, which is a good tradeoff between the model complexity and learnability in practice. It relaxes the conditional independence assumptions of Naive Bayes while still permitting efficient inference and learning. Experimental studies on a set of natural domains prove its clear advantages with respect to the generalization ability.     

A genetic‐based neuro‐fuzzy controller for blind equalization of time‐varying channels

This paper presents a neuro‐fuzzy network (NFN) where all its parameters can be tuned simultaneously using genetic algorithms (GAs). The approach combines the merits of fuzzy logic theory, neural networks and GAs. The proposed NFN does not require a priori knowledge about the system and eliminates the need for complicated design steps such as manual tuning of input–output membership functions, and selection of fuzzy rule base. Although, only conventional GAs have been used, convergence results are very encouraging. A well‐known numerical example derived from literature is used to evaluate and compare the performance of the network with other equalizing approaches. Simulation results show that the proposed neuro‐fuzzy controller, all parameters of which have been tuned simultaneously using GAs, offers advantages over existing equalizers and has improved performance. From the perspective of application and implementation, this paper is very interesting as it provides a new method for performing blind equalization. The main contribution of this paper is the use of learning algorithms to train a feed‐forward neural network for M‐ary QAM and PSK signals. This paper also provides a platform for researchers of the area for further development. Copyright © 2008 John Wiley & Sons, Ltd.     

NUMERICAL ANALYSIS AND EXPERI- MENT OF UNSTEADY THERMAL FIELD OF ROTOR PLATE FOR EDDY CURRENT RETARDER

The physical model based on heat transfer theory and virtual boundary method for analyzing unsteady thermal field of rotor plate for eddy current retarder used in automobile is established and boundary conditions are also defined. The finite element governing equation is derived by Galerkin method. The time differential item is discrete based on Galerkin format that is stable at any condition. And a new style of varying time step method is used in iteration process. The thermal field on the rotor plate at the radial and axle directions is analyzed and varying temperature at appointed points on two side-surfaces is measured. The testing and analytical data are uniform approximately. Finite element method can be used for estimating thermal field of the rotor plate at initial design stage of eddy current retarder.