蓄冷空调系统的负荷计算

分析了常规空调系统和蓄冷空调系统负荷计算的不同及逐时负荷分布对蓄冷系统负荷计算的重要性，指出了在蓄冷系统负荷计算时应该注意的问题。     

Distributed adaptive output consensus control of second-order systems containing unknown non-linear control gains

In this paper, we address the output consensus problem of tracking a desired trajectory for a group of second-order agents on a directed graph with a fixed topology. Each agent is modelled by a second-order non-linear system with unknown non-linear dynamics and unknown non-linear control gains. Only a subset of the agents is given access to the desired trajectory information directly. A distributed adaptive consensus protocol driving all agents to track the desired trajectory is presented using the backstepping technique and approximation technique of Fourier series (FSs). The FS structure is taken not only for tracking the non-linear dynamics but also the unknown portion in the controller design procedure, which can avoid virtual controllers containing the uncertain terms. Stability analysis and parameter convergence of the proposed algorithm are conducted based on the Lyapunov theory and the algebraic graph theory. It is also demonstrated that arbitrary small tracking errors can be achieved by appropriately choosing design parameters. Though the proposed work is applicable for second-order non-linear systems containing unknown non-linear control gains, the proposed controller design can be easily extended to higher-order non-linear systems containing unknown non-linear control gains. Simulation results show the effectiveness of the proposed schemes.     

一种单位代价收益决策树剪枝算法

目前关于决策树剪枝优化方面的研究主要集中于预剪枝和后剪枝算法。然而,这些剪枝算法通常作用于传统的决策树分类算法,在代价敏感学习与剪枝优化算法相结合方面还没有较好的研究成果。基于经济学中的效益成本分析理论,提出代价收益矩阵及单位代价收益等相关概念,采用单位代价收益最大化原则对决策树叶节点的类标号进行分配,并通过与预剪枝策略相结合,设计一种新型的决策树剪枝算法。通过对生成的决策树进行单位代价收益剪枝,使其具有代价敏感性,能够很好地解决实际问题。实验结果表明,该算法能生成较小规模的决策树,且与REP、EBP算法相比具有较好的分类效果。     

可变形战斗部破片增益研究

通过试验得到可变形战斗部破片飞散分布规律,运用LS-DYNA对可变形战斗部变形和爆轰驱动破片飞散过程进行数值模拟,得到与试验吻合较好的结果;对相同结构口径的偏心起爆战斗部、周向均匀战斗部的爆轰驱动破片飞散过程进行了数值模拟。结果表明,在给定破片飞散角度内可变形战斗部整体性能优于偏心起爆战斗部。     

Sudden gains during therapy of social phobia.

The present study investigated the phenomenon of sudden gains in 107 participants with social phobia (social anxiety disorder) who received either cognitive-behavioral group therapy or exposure group therapy without explicit cognitive interventions, which primarily used public speaking situations as exposure tasks. Twenty-two out of 967 session-to-session intervals met criteria for sudden gains, which most frequently occurred in Session 5. Individuals with sudden gains showed similar improvements in the 2 treatment groups. Although cognitive-behavioral therapy was associated with more cognitive changes than exposure therapy, cognitive changes did not precede sudden gains. In general, the results of this study question the clinical significance of sudden gains in social phobia treatment. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

一种基于语义距离的关键词获取方法

关键词作为文本的概要信息,是信息检索,特别是针对海量Web信息处理的基础。由于手工分配关键词是一项耗时性的工作,因此不需要训练集的非监督关键词获取方法研究受到了关注。本文以语义聚类为模型,以词在给定语境和词义本身两种语义距离的计算为基础,给出一种关键词获取算法。实验结果表明,该算法得到较好的查准率。     

An automated multi-step calibration procedure for a river system model

Predicted climate change impact on future water availability in the Murray–Darling Basin (MDB) has highlighted the need for a whole of basin model that incorporates various physical and management characteristics for planning and operational purposes. Modelling platforms such as eWater Source Integrated Modelling System (Source) offer a useful framework in this regard, but at present lack automated calibration techniques to parameterise river system models.This paper presents an automated river system calibration procedure which is robust, repeatable, transparent and systematic. The procedure allows for river network calibration (as opposed to isolated reach by reach calibration), since this has more utility for basin planning and prediction. The calibration procedure routs upstream flow, estimates ungauged inputs via rainfall–runoff (RR) models, and estimates flow based split (distributary) functions and loss functions in complex river systems.This procedure was tested in the Northern Murray–Darling Basin (MDB) and results from the Border Rivers catchment are presented. The results from the Border Rivers case study demonstrate the applicability of the procedure with median calibration and evaluation NSE values of 0.88 and 0.79, respectively. The use of this procedure in the Border Rivers region has highlighted the likelihood of changing stream channel connections at higher flows in the lower reaches of the river network.     

基于非线性增益递归滑模的船舶轨迹跟踪动态面自适应控制

针对三自由度全驱动船舶存在模型不确定和未知外部环境扰动的情况,设计出一种基于非线性增益递归滑模的船舶轨迹跟踪动态面自适应神经网络控制方法.该方法综合考虑船舶位置和速度误差之间关系设计递归滑模面,引入神经网络对船舶模型不确定部分进行逼近,设计带σ-修正泄露项的自适应律对神经网络逼近误差与外界环境扰动总和的界进行估计,并应用一种非线性增益函数构造动态面控制律,选取李雅普诺夫函数证明了该控制律能够保证轨迹跟踪闭环系统内所有信号的一致最终有界性.最后,基于一艘供给船进行仿真验证,结果表明,船舶轨迹跟踪响应速度快、精度高,所设计控制器对系统模型参数摄动及外界扰动具有较强的鲁棒性.     

Funnel‐like prescribed tracking control for uncertain nonlinear stochastic switched systems

This paper deals with the funnel‐like prescribed tracking control problem for a class of uncertain nonlinear stochastic switched systems. An improved performance technique is developed to restrain the fluctuation at the moment of switches and a new algorithm is proposed to address the funnel‐like prescribed tracking problem. First, a dynamic gain‐based switched K‐filter is constructed to estimate the unmeasured state information of the switched system. Subsequently, the performance technique is applied to prescribe output tracking error and restrain fluctuations of the system. Thereafter, the dynamic output feedback switched controller is designed by the use of the backstepping method. Moreover, based on the Lyapunov stability theory, it is proved strictly that all signals of the resulting closed‐loop system are bounded in probability if the switching signal satisfies the average dwell time. Finally, a numerical simulation is presented to illustrate the effectiveness of the proposed theoretical results.     

一种改进的基于蓝噪声的误差分散算法

传统的误差分散算法大大促进了数字半调技术的发展。但由于该算法采用类似扫描仪的处理路线,简单的误差分散核,固定的阈值,并且忽略了关键灰度、点增益、图像色调,很容易出现比较明显的人工纹理。论文提出了一种改进的基于蓝噪声的误差分散算法。该算法采用可变的处理路线,多灰度误差分散核,并且根据图像色调、点增益自动调整量化阈值和关键灰度的误差分散系数,达到了较好的视觉效果。