A Novel Beam Search to Improve Neural Machine Translation for English-Chinese

Neural Machine Translation (NMT) is an end-to-end learning approach for automated translation, overcoming the weaknesses of conventional phrase-based translation systems. Although NMT based systems have gained their popularity in commercial translation applications, there is still plenty of room for improvement. Being the most popular search algorithm in NMT, beam search is vital to the translation result. However, traditional beam search can produce duplicate or missing translation due to its target sequence selection strategy. Aiming to alleviate this problem, this paper proposed neural machine translation improvements based on a novel beam search evaluation function. And we use reinforcement learning to train a translation evaluation system to select better candidate words for generating translations. In the experiments, we conducted extensive experiments to evaluate our methods. CASIA corpus and the 1,000,000 pairs of bilingual corpora of NiuTrans are used in our experiments. The experiment results prove that the proposed methods can effectively improve the English to Chinese translation quality.     

The Crosstalk of Adipose-Derived Stem Cells (ADSC), Oxidative Stress,and Inflammation in Protective and Adaptive Responses

The potential use of stem cell-based therapies for the repair and regeneration of various tissues and organs is a major goal in repair medicine. Stem cells are classified by their potential to differentiate into functional cells. Compared with other sources, adipose-derived stem cells (ADSCs) have the advantage of being abundant and easy to obtain. ADSCs are considered to be tools for replacing, repairing, and regenerating dead or damaged cells. The capacity of ADSCs to maintain their properties depends on the balance of complex signals in their microenvironment. Their properties and the associated outcomes are in part regulated by reactive oxygen species, which mediate the oxidation-reduction state of cells as a secondary messenger. ADSC therapy has demonstrated beneficial effects, suggesting that secreted factors may provide protection. There is evidence that ADSCs secrete a number of cytokines, growth factors, and antioxidant factors into their microenvironment, thus regulating intracellular signaling pathways in neighboring cells. In this review, we introduce the roles of ADSCs in the protection of cells by modulating inflammation and immunity, and we develop their potential therapeutic properties.     

XJM-KS型浮选机矿化器负压产生机理研究

浮选机矿化器是用于浮选矿浆预处理的重要设备，其性能的优劣直接影响到浮选机的工作效果。文章通过流体力学仿真软件FLUENT对XJM-KS型浮选机矿化器的内部流场进行数值仿真，重点研究了聚焦式喷嘴布置所带来的流场特征。模拟结果表明，高速的矿浆射流能够在混合室内产生-2.43×10Pa的负压区，从而吸入大量的空气，实现空气与矿浆的有效混合与预矿化，与传统的机械搅拌式矿浆预处理装置相比，该矿化器能够大幅度提高预矿化效果。     

Stimuli-Responsive Nanocomposite Hydrogels for Biomedical Applications

The complex tissue-specific physiology that is orchestrated from the nano- to the macroscale, in conjugation with the dynamic biophysical/biochemical stimuli underlying biological processes, has inspired the design of sophisticated hydrogels and nanoparticle systems exhibiting stimuli-responsive features. Recently, hydrogels and nanoparticles have been combined in advanced nanocomposite hybrid platforms expanding their range of biomedical applications. The ease and flexibility of attaining modular nanocomposite hydrogel constructs by selecting different classes of nanomaterials/hydrogels, or tuning nanoparticle-hydrogel physicochemical interactions widely expands the range of attainable properties to levels beyond those of traditional platforms. This review showcases the intrinsic ability of hybrid constructs to react to external or internal/physiological stimuli in the scope of developing sophisticated and intelligent systems with application-oriented features. Moreover, nanoparticle-hydrogel platforms are overviewed in the context of encoding stimuli-responsive cascades that recapitulate signaling interplays present in native biosystems. Collectively, recent breakthroughs in the design of stimuli-responsive nanocomposite hydrogels improve their potential for operating as advanced systems in different biomedical applications that benefit from tailored single or multi-responsiveness.     

Survey on Passivity Based Control of Induction Machine

Induction machines (IM) constitute a theoretically interesting and practically important class of nonlinear systems. They are frequently used as wind generators for their power/cost ratio. They are described by a fifth‐order nonlinear differential equation with two inputs and only three state variables available for measurement. The control task is further complicated by the fact that IM are subject to unknown (load) disturbances and the parameters can be of great uncertainty. One is then faced with the challenging problem of controlling a highly nonlinear system, with unknown time‐varying parameters, where the regulated output, besides being unmeasurable, is perturbed by an unknown additive signal. Passivity‐based control (PBC) is a well‐established structure‐preserving design methodology which has shown to be very powerful to design robust controllers for physical systems described by Euler‐Lagrange equations of motion. PBCs provide a natural procedure to "shape" the potential energy yielding controllers with a clear physical interpretation in terms of interconnection of the system with its environment and are robust vis á vis to unmodeled dissipative effects. One recent approach of PBC is the Interconnection and Damping Assignment Passivity‐Based Control (IDA‐PBC) which is a very useful technique to control nonlinear systems assigning a desired (Port‐Controlled Hamiltonian) structure to the closed‐loop. The aim of this paper is to give a survey on different PBC of IM. The originality of this work is that the author proves that the well known field oriented control of IM is a particular case of the IDA‐PBC with disturbance.     

煤矿地面轨道运输矿车自动避障技术研究

介绍了煤矿地面轨道运输矿车自动避障技术。首先介绍了基于雷达的矿车前方障碍物检测技术及其相关算法,然后详细介绍了基于机器视觉的矿车前方障碍物检测系统和常用的图像处理方法以及相关的算法,最后介绍了雷达和机器视觉融合的障碍物检测技术在煤矿地面轨道运输矿车自动避障过程中的应用。     

Healing Pattern Analysis for Dental Implants Using the Mechano-Regulatory Tissue Differentiation Model

(1) Background: Our aim is to reveal the influence of the geometry designs on biophysical stimuli and healing patterns. The design guidelines for dental implants can then be provided. (2) Methods: A two-dimensional axisymmetric finite element model was developed based on mechano-regulatory algorithm. The history of tissue differentiation around eight selected implants can be predicted. The performance of the implants was evaluated by bone area (BA), bone-implant contact (BIC); (3) Results: The predicted healing patterns have very good agreement with the experimental observation. Many features observed in literature, such as soft tissues covering on the bone-implant interface; crestal bone loss; the location of bone resorption bumps, were reproduced by the model and explained by analyzing the solid and fluid biophysical stimuli and (4) Conclusions: The results suggested the suitable depth, the steeper slope of the upper flanks, and flat roots of healing chambers can improve the bone ingrowth and osseointegration. The mechanism related to solid and fluid biophysical stimuli were revealed. In addition, the model developed here is efficient, accurate and ready to extend to any geometry of dental implants. It has potential to be used as a clinical application for instant prediction/evaluation of the performance of dental implants.     

低觉醒脑电识别与唤醒的可穿戴系统研究

为智能化地识别警戒作业人员出现的低觉醒、注意力下降的生理状态，本文介绍了一种基于FPGA和脑电信号处理的低觉醒状态检测与唤醒系统，系统通过传感器从大脑头皮采集脑电信号，转换为数字信号，经傅里叶变换获取了脑电信号的θ相对能量、α相对能量、重心频率、谱熵等4个特征量，由4个特征量表征低觉醒状态并运用支持向量机对低警戒状态进行识别，当识别出低觉醒状态时采用声音报警模块发出声音，唤醒警戒作业人员。设计系统能够较好地识别出低觉醒状态，识别率达90.8%，可为提高警戒作业工作绩效提供一种可穿戴的智能装备。     

智能变电站设备管控大数据分析系统研究

结合数据挖掘、大数据技术在电力系统中的应用现状，分析和设计了一种智能变电站设备管控大数据分析系统。对变电站设备运行状态的大数据特征及基本框架进行了分析，并重点阐述了变电站设备运行状态大数据分析系统的数据集成与预处理、数据存储与处理、设备状态评估模型以及数据可视化展示。实验结果表明，所设计的大数据分析系统能够大幅度提升变电站设备数据分析智能化水平，为变电站管控稳定运行提供更加有效和实用性的技术支撑。     

基于聚类的数据加权优化在犯罪预测中的应用

近年来，我国传统暴力犯罪与成年人犯罪呈下降态势，但是，犯罪案由层出不穷。为有效提升公安实践工作中犯罪预测能力，打击各类违法犯罪事件,本文针对犯罪数据，提出一种新型犯罪预测模型。利用密度聚类分析方法将犯罪数据分类，然后进行数据降维提取关键属性生成特征数据，继而对特征数据进行加权优化并采用机器学习的方式对特征数据进行学习，从而预测犯罪案由。实验结果表明，与传统方法相比，本文方法具有更好的预测效果，为公安实践工作中类似案件的侦破和预防，提供新的路径支撑。