LPNN-based approach for LASSO problem via a sequence of regularized minimizations

In compressive sampling theory, the least absolute shrinkage and selection operator (LASSO) is a representative problem. Nevertheless, the non-differentiable constraint impedes the use of Lagrange programming neural networks (LPNNs). We present in this article the -LPNN model, a novel algorithm that tackles the LASSO minimization together with the underlying theory support. First, we design a sequence of smooth constrained optimization problems, by introducing a convenient differentiable approximation to the non-differentiable -norm constraint. Next, we prove that the optimal solutions of the regularized intermediate problems converge to the optimal sparse signal for the LASSO. Then, for every regularized problem from the sequence, the -LPNN dynamic model is derived, and the asymptotic stability of its equilibrium state is established as well. Finally, numerical simulations are carried out to compare the performance of the proposed -LPNN algorithm with both the LASSO-LPNN model and a standard digital method.     

埋地管道泄漏内封堵装置设计与研究

针对流体输送埋地管道泄漏问题,设计了一种利用管道机器人携带封堵气囊进行快速应急封堵修复的埋地管道泄漏内封堵装置。采用矩阵变换方法建立了牵引系统驱动轮过弯方程,利用MATLAB软件对过弯方程进行了验证,同时利用ADAMS软件仿真分析过弯路径与驱动轮转角对牵引系统行走速度的影响。研究结果表明:由两个串联封堵器组成的应急封堵系统可满足复杂工况下的管道泄漏封堵要求;牵引系统驱动轮在弯管内部行走时,单轮速度呈周期性变化,但三个驱动轮整体周期运动特性一致;驱动轮转角在25°~40°时,牵引系统行走速度与驱动轮转角成正比,且转角为30°时驱动效果最好。该内封堵装置的结构设计可为管道泄漏应急封堵领域装备的研发提供重要参考。     

基于注意力机制与文本信息的用户关系抽取

随着社交媒体的发展,用户之间的关系网络对于社交媒体的分析有很大的帮助。因此,该文主要研究用户好友关系检测。以往的关于用户好友关系抽取的研究主要基于社交媒体上的结构化信息,比如其他好友关系,用户的不同属性等。但是,很多时候用户本身并没有大量的好友信息存在,同时也不一定有很多确定的属性。因此,我们希望基于用户发表的文本信息来对用户关系进行预测。不同于以往的潜在好友推荐算法,该文提出了一种基于注意力机制以及长短时记忆网络(long short-term memory,LSTM)的好友关系预测模型,将好友之间的评论分开处理,通过分析用户之间的评论来判断是否具备一定的好友关系。该模型将好友双方信息拼接后的结果作为输入,并将注意力机制应用于LSTM的输出。实验表明,用户之间的评论对于好友关系预测确实有较大的实际意义,该文提出的模型较之于多个基准系统的效果,取得了明显的提升。在不加入任何其它非文本特征的情况下,实验结果的准确率达到了77%。     

Food recognition using neural network classifier and multiple hypotheses image segmentation

ABSTRACT

This paper proposes the multiple-hypotheses image segmentation and feed-forward neural network classifier for food recognition to improve the performance. Initially, the food or meal image is given as input. Then, the segmentation is applied to identify the regions, where a particular food item is located using salient region detection, multi-scale segmentation, and fast rejection. Then, the features of every food item are extracted by the global feature and local feature extraction. After the features are obtained, the classification is performed for each segmented region using a feed-forward neural network model. Finally, the calorie value is computed with the aid of (i) food volume and (ii) calorie and nutrition measure based on mass value. The experimental results and performance evaluation are validated. The outcome of the proposed method attains 0.947 for Macro Average Accuracy (MAA) and 0.959 for Standard Accuracy (SA), which provides better classification performance.     

动态评价技术在塔河碳酸盐岩缝洞型油气藏中的应用

塔河油田奥陶系油气藏是大型碳酸盐岩溶洞型油气藏，其储渗空间主要为大小不同的溶洞、裂缝带、溶蚀孔隙等组成，该油气藏具有极强的非均质性，单纯用静态资料来认识这类油气藏是非常困难。文章提出利用生产动态资料和信息进行该类型油气藏研究的新思路，利用人工神经网络技术在处理非线性相关参数预测方面的优势，并以渗流理论为基础，结合试井成果，选用已知油井的产量、油嘴、油压、含水率、气油比、原油密度等6个开发动态参数作为样品输入数据，推导出影响油气藏开发的重要参数（地层系数）与生产信息的关系，建立了人工神经网络预测储层参数的结构模型。通过塔里木盆地塔河油气田实例研究，说明了利用动态信息评价油气藏技术在碳酸盐岩缝洞型油气藏储层预测和非均质性分析等方面具有较高的实用价值。     

埋地管道在冻土带的热力变形分析

埋地油气管道通过永冻土地带时,由于管内介质与周围的冻土发生了热力作用,使冰土解冻,在管道周围形成融化圈。文章分析了永冻土的特点,建立了永冻土地带土壤传热数学模型:融化圈内外传热方程、管道向土壤和土壤表面向大气所放热量的守恒关系、融化圈界面处土壤温度、相变界面处固相变为液相的传热条件。还建立了埋地管道模型,并以穿过永冻土地带土壤的输油管道为例,分析并计算了各种应力的最大值所在的位置和大小,提出了在永冻土地带减小管道热力变形的措施。     

利用HFC网络开展VoIP业务

本文首先介绍了VoIP的原理、相关协议．然后介绍了VoIP业务在HFC网络应用的业务环境和体系结构。并对其中的功能实体进行描述和说明。最后介绍一个具体的业务实例。     

Inhibition of microbial growth in ready‐to‐eat food stored at ambient temperature by modified atmosphere packaging

This study was undertaken to develop a modified atmosphere package to control microbial growth in ready‐to‐eat (RTE) products stored at ambient temperature. Ethanol and/or limonene associated with modified atmosphere (CO₂ : O₂ : N₂ = 30% : 5% : 65%) was used to inhibit the growth of total air‐borne microorganisms and Escherichia coli in RTE products stored at 25°C. The results indicated that 0.05% ethanol vapour in the headspace was effective to inhibit the growth of air‐borne microorganisms and E. coli at 25°C for 72 h in a model study, and the effectiveness was related to ethanol content. Both 73 ppm limonene and 0.05% ethanol vapour enhanced the bacteriostatic effect of modified atmosphere in RTE sushi roll products, and no off‐flavour was detected using this formulated gas; however, no significant inhibitory effect was observed for RTE cold noodle products. This study concludes that combinations of carbon dioxide, ethanol or limonene vapours are effective to inhibit microbial growth in RTE food at ambient temperature, and the outcome may be due to the hurdle effect. Copyright © 2003 John Wiley & Sons, Ltd.     

Parallel algorithm for numerical self-consistent field theory simulations of block copolymer structure

An efficient algorithm is presented for numerically evaluating a self-consistent field theoretic (SCFT) model of block copolymer structure. This algorithm is implemented on a distributed memory parallel cluster in order to solve the SCFT equations on large computational grids. Simulation results are presented for a two-component molten mixture of a symmetric ABA triblock copolymer with an A homopolymer. These results illustrate a case in which simulating a large system is required to resolve features with a wide range of length scales.     

Impact on the photothermal emission from single wall nanotubes by some alkali halide salts

We demonstrate that alkali-halide salts, particularly potassium bromide, can reduce the photothermal emission (PTE) from single walled carbon nanotubes (SWNT). PTE is a prominent spectral feature in Raman spectroscopy when a near infrared laser is used to analyze a dark colored sample. We subsequently show that trapping salts inside SWNT and coating SWNT with the salt has a more pronounced impact on not only reducing PTE, but also enhancing the intensity of the Raman spectral features. The effect, which we have called nanotube enhanced Raman spectroscopy (NERS), has differences and similarities to the widely studied surface enhanced Raman spectroscopy (SERS).