Thermodynamic analysis of representative power generation cycles for low‐to‐medium temperature applications

This study is focused on the analysis of representative thermodynamic cycles for power generation at low‐to‐medium temperatures (with the highest cycle temperature from 450 to 700 K). The natural working fluid of carbon dioxide is selected for the current tests and comparisons with suitable operation ranges. Energy balance and exergy loss models are established and applied to 10 selected representative thermodynamic cycles. One modified efficiency parameter is also defined for better comparison of performances, which has considered the effects of both specific thermodynamic process and cycle complexity. Based on the modified efficiency parameter, it is found that Rankine cycle yields the highest performance at 450–500 K among the 10 representative cycles, while regenerative Brayton cycle shows better behavior for 550–700 K. Detailed behaviors and optimal principals of regenerative Brayton cycles are also identified and compared in this study. Also, a new cycle is also proposed in this study, which combines the advantages of Rankine cycle and Brayton cycle. The new cycle is proved to have better work output potential but higher system complexity factor. In addition, based on the thermodynamic analysis, possible future directions of low‐to‐medium temperature power cycles are summarized. It is hoped that the results can be of help for related power generation system designs. Copyright © 2014 John Wiley & Sons, Ltd.     

Wind speed variability across the UK between 1957 and 2011

By using surface data from 57 UK meteorological stations, a national [British Atmospheric Data Centre (BADC)‐57] and regional wind index for the UK has been calculated for the period 1983–2011. For a subset of seven stations, an additional national index (BADC‐7) has been calculated for the period 1957–2011. The indices show an annual variability of 4% over their respective periods, corresponding to a variation in typical wind turbine capacity factor of 7%. These indices are compared with indices calculated from other sources, namely an index generated using a gridded dataset of observed values interpolated across the UK, an index calculated from an area bounding the UK using the ERA‐40 re‐analysis dataset, indices calculated from bilinear interpolation of the ERA‐40 re‐analysis dataset to the 57 and 7 stations and another independent UK wind index. The indices show variation in trends, with all showing some level of decline with the exception of that generated using the ERA‐40 re‐analysis dataset averaged over the UK, which shows a significant increase. The various indices show varying degrees of agreement with correlation coefficients, after trends are removed, ranging between 0.611 and 0.979. The effect of changes in site exposure, instrument bias and measuring height was considered for the BADC‐7 and BADC‐57 indices. The change in instrument measurement height appears to have a significant biasing effect, and it is likely that this along with changes in exposure at urban sites has caused the decline in annual wind speeds observed for some of the indices. There does not appear to be evidence for significant changes in large‐area (mesoscale) surface roughness. The correlation between annual mean wind speeds at the seven surface station sites used to calculate the BADC‐7 index is seen to be quite weak, indicating very localized variations in inter‐annual variability. When regional differences in the index are investigated, it is seen that wind speeds show a very slight decline across the UK in all regions except the south‐east, which shows a slight increase. The greatest decrease is seen in the north‐west. These changes are in the same direction as the tentative predictions given by climate models for future changes in wind speed across the UK, although the uncertainty is large given the large degree of inter‐annual variation. Copyright © 2013 John Wiley & Sons, Ltd.     

用于排架结构的铅剪切阻尼器试验及数值分析

为进一步促进铅阻尼器在工程中的应用,针对常见的排架结构,研制了一种铅剪切阻尼器,通过低周往复加载试验,逐个分析阻尼材料铅的有效剪切面积、受剪区厚度、剪切截面形式、加载位移幅值、加载频率、加载次数等因素对阻尼器耗能性能的影响。试验结果表明,设计的阻尼器的阻尼力—位移滞回曲线饱满,耗能性能良好,符合双线性力学模型,并对试验中两个典型型号的阻尼器进行了数值模拟分析,通过对比数模值和试验值,证明了数值计算模型的实用性,为阻尼器在结构体系中的减震应用奠定了基础。     

某水电站进水口右岸楔形体边坡稳定性分析

针对西藏地区水电工程前期设计阶段受地质资料限制,对楔形体边坡稳定性分析往往需要大量试算验算,设计工作效率低的问题,以某水电站坝址楔形体XT2为分析对象,采用数值计算软件ABAQUS对初始地应力场进行反演,并将反演出的地应力场应用于大型有限差分分析软件FLAC3d,进行楔形体边坡稳定性计算与分析。结果表明,该方法既保证了计算结果的可靠性,又大大缩短了计算时间。     

转子动力学在纺织机械动态分析中的应用

简要介绍了纺织机械中高速转子的特点以及近年来国内外学者应用转子动力学的基本理论、分析计算方法、测试手段和参数识别技术等对纺织高速转子进行动态分析的概况，涉及临界转速和回转形态、动力响应、运动稳定性、转子动力特性系数研究等领域．     

Drip loss,peroxidase and sensory changes in kiwi fruit slices during frozen storage

The biochemical, sensory and drip loss changes that occur during processing and prolonged frozen storage of kiwi fruit slices (cvs Abbott and Hayward) were studied. Fruit slices were frozen at ?40°C, packed in polyethylene bags and stored at ?18°C for 11 months. Maturity characteristics (pH, acidity, dry matter, soluble solids) were determined on raw fruit. Objective (proteins, peroxidase (EC 1.11.1.7) activity, drip loss) and subjective (sensory tests) analyses were carried out during processing and storage, and indicated a good quality of the frozen kiwi fruit slices after 11 months of storage. There were significant differences (P < 0.05) between the studied varieties with respect to drip loss during frozen storage and colour after 11 months. Best results were obtained with cv Hayward. This variety showed less drip loss after thawing and after 11 months storage presented the same green colour as after freezing, while cv Abbott became yellowish-green.     

设备诊断系统的建立及应用

介绍了一套设备诊断系统的基本构成，通过实例证明了系统的可行性；展示了实施设备状态监测和故障诊断这种先进设备管理维修方式良好的发展前景．     

基于TSCM模型的网络短文本情感挖掘

针对网络短文本情感挖掘问题,提出一种新的基于LDA和互联网短评行为理论的主题情感混合模型TSCM,TSCM模型中的整篇评论中每个句子的主题分布是不同的,TSCM产生词的流程是先确定词的情感极性,再确定词的主题,TSCM考虑了词与词之间的联系.真实数据集Movie与Amazon上的大量实验表明,与代表性算法JST、S-LDA、D-PLDA和SAS相比较,TSCM模型能对用户真实情感与讨论主题进行更加有效的分析建模.     

基于稀疏补分析模型的近似最优子空间追踪

为了从含噪声的测量矢量中重构原始信号,研究了稀疏补分析模型下近似最优子空间追踪信号重构算法.针对直接采用稀疏综合模型下子空间追踪过程非最速梯度下降和信号重构概率不高的缺点,根据稀疏补分析模型下不同类型分析字典的结构特点来设计近似目标优化函数;改进了迭代追踪过程;优化了稀疏补取值方法;提出并实现了基于稀疏补分析模型的近似最优分析子空间追踪算法.仿真实验证明,当稀疏补运算符分别采用随机紧支框架和二维全变分矩阵时,算法的完全重构信号概率均明显高于ASP、AHTP、AIHT、AL1、GAP算法的完全重构信号概率;对于含高斯噪声的输入信号,算法的重构信号综合平均PSNR比相应的ASP、AHTP、AIHT算法分别提高了0.8dB、1.38dB、3.13 dB,但比GAP和AL1算法降低了0.32 dB和0.6dB.算法的完全重构概率与综合重构性能有了明显提高,收敛充分条件得到进一步简化.     

基于迁移学习的唐诗宋词情感分析

随着计算社会学的兴起,利用数据挖掘分析社会情感是近期的研究重点.当前的研究主要针对现代文本,对于古代诗歌这类短文本的情感分析相对较少.本文提出了一个基于短文本特征扩展的迁移学习模型CATL-PCO,通过分析诗歌情感对当时社会及文化进行进一步了解.该模型首先基于频繁词对对古文特征向量进行扩展,再通过迁移学习方式,建立三个分类器并投票得出最后的情感分析结果.CATL-PCO模型首先能够解决古文短文本特征稀疏的问题,在此基础上进一步解决由于现代译文信息匮乏所导致的古代诗歌情感分析困难问题,从而准确的分析古诗词情感倾向,从计算社会学的角度,增进对中国历史的认识.实验表明,当训练集为中国唐诗时,本文提出方法能够准确的对唐代诗歌进行情感分类,并能应用于唐代和宋代各个时期情感分析及代表流派分析.