天然气分析特殊情况时合峰和乙烷峰的分离及含量测试

进行天然气常量组分分析 ,当乙烷含量很低时 ,合峰 (甲烷和空气 )与乙烷峰分不开 ,无论怎样改变操作参数都无效 ,出现这种情况时 ,向试样中加入定量标样进行分析 ,再通过一定的计算就可测出乙烷含量。方法简便易行。     

一种基于GA-NSVM的自适应洛伦兹分峰拟合识别红外光谱吸收重叠峰的方法

提出一种基于遗传算法-非线性支持向量机的自适应洛伦兹分峰拟合识别红外光谱吸收重叠峰的方法。利用单质特征吸收线型的根本性差异,将混合物光谱分解为满足特征吸收线型的多个洛伦兹单峰,采用非线性支持向量机对多个拟合单峰进行多分类筛选确定特定目标组分的谱峰。采集了400个混合烷烃气体样本的红外光谱数据,论证了该方法在高相似分子结构的光谱识别分类的可行性。实验结果表明该方法能有效分离烷烃中甲烷、乙烷、丙烷的红外吸收单峰,具有良好的准确性和鲁棒性,模型参数的解释能力更强。该方法能够加速光谱检测技术在生物制药、食品化工、油气勘探等领域的应用,尤其是在含同系有机物混合物的分析及应用场合。     

气测录井各种单根峰的识别

气测录井作为一种实时测量地层气体的手段,其原始资料中记录下了许多由外界因素引起的气测异常数据,如后效、单根峰、停泵峰、双单根峰等,它们均是已钻过地层在特殊情况下的表现,不是实时钻进时地层的破碎气。通过它们可以间接地了解储集层的含油气情况、地层压力及钻井液使用情况,因此它们只是一种辅助资料,而对气测真正有意义的是实时钻进时破碎气的测量值,所以在整理回放资料时要把假值剔除掉。但剔除工作要格外慎重,一定要判断准确,否则会把地层的真实气测值显示给删除掉。后效、单根峰、停泵峰容易判断,一般不会出什么问题,但双单根峰的识别难度较大,需要掌握一定的气测理论基础知识。下面本文结合作者工作经验,通过实际例证来进行具体分析。     

正常儿童汉语拼音元音的共振峰测算方法对比

本文介绍 ３０例正常学龄儿童汉语拼音元音ａ、ｏ、ｅ、ｉ、ｕ、ü６个音的功率谱语频分析。用直接计算语频包络中振幅最大的谐波作为共振峰频率值和计算包络中最大谐波的偏峰值作为共振峰频率值。比较两种不同方法的差异。经ｔ检验证实，此两法测得的共振峰频率值在统计学上有显著的差别。作者提出，在进行语频分析时，应采用共振峰频率偏移值的计算方法。它能对语频作精确的评价。     

基于运行策略的某电子工厂冰蓄冷空调系统选型研究

通过分析空调系统的冷负荷特点，确定制冷主机的的选型及数量，并确定主机的运行模式。根据峰谷电价的价差，提出高电价部分削峰、高电价全削峰以及双削峰三种冰蓄冷的方案；对三种冰蓄冷方案进行经济分析，并给出了年运行费用的计算方法，得出年运行费用，通过比较投资回收周期，从而得出最佳运行策略。     

X射线衍射法测定Fe-Mn-Si形状记忆合金层错几率的研究

Ｆｅ－Ｍｎ－Ｓｉ合金的形状记忆效应来源于马氏体相变,而马氏体相变则通过奥氏体内形成每隔一层｛１１１｝面上的堆垛层错来完成。与层错能相关的层错几率可能控制马氏体的相变机制。本文根据Ｗａｒｅｎ的衍射理论,用Ｘ射线衍射峰位移和峰宽化两种方法测定了Ｆｅ－Ｍｎ－Ｓｉ合金的层错几率,其结果表明随锰含量增加,层错几率降低。本文着重对提高衍射峰位移和峰宽化两种测定方法的精度及其影响因素进行详细的研究,并认为峰位移法更简便     

重叠峰解卷积在能谱面分布中的新应用

为研究重叠峰解卷积技术对能谱面分布分析结果准确性的影响,以具有重叠峰的Ti-Ba材料为研究对象,对采用重叠峰解卷积技术前后的元素能谱面分布结果进行比较。实验结果表明:未采用谱峰解卷积时,Ti元素和Ba元素的能谱面分布基本相同,但元素定量分析结果显示Ti和Ba元素并非分布在相同的区域,即未采用谱峰解卷积的Ti-Ba元素能谱面分布分析有误;而采用谱峰解卷积技术后,可以清晰地观察Ti元素和Ba元素分布在不同的区域。由此可知,对于具有重叠峰的元素能谱面分布(重叠峰元素未形成化合物),可采用重叠峰解卷积来解决元素的谱峰重叠问题。     

某气田原料气中CO2含量对RSV乙烷回收工艺的影响分析

新疆油田某气田处理装置进行深冷提效改造,采用RSV凝液回收工艺,回收天然气中的乙烷、LPG及稳定轻烃。装置制冷深度-105℃,乙烷收率95%。初始原料气中不含CO₂,有利于提高乙烷收率。随着气田滚动开发,新的区块天然气中发现含有少量的CO₂,由于装置制冷深度较深,CO₂含量的上升有可能造成设备的冻堵。因此该技术对原料气CO₂含量的变化分析,研究其对脱甲烷塔及产品乙烷收率的影响。经技术研究及HYSYS软件模拟计算,得出结论 ：当原料气中CO₂含量超过1%时,及时调整操作参数降低乙烷收率维持装置正常运行,或考虑增加脱碳装置以维持乙烷收率不变。     

面积归一化法分析混合气体中各组分含量

色谱分析中常用的定量方法有很多,标准曲线法、内标法及峰面积归一法等。峰面积归一化法是比较常用的一种定量分析方法。本文中所介绍的实验就采用了面积归一法来进行混合气体中各组分含量的定量分析。     

放电等离子烧结TiB_2过程中Pa第二峰研究

放电等离子烧结ＴｉＢ２过程中真空室气压Ｐａ有二个峰值，本文对Ｐａ第二峰进行了研究．研究结果表明：Ｐａ第二峰是烧结过程中化学反应导致的气相沉积所致．原始粉末和烧结工艺对Ｐａ第二峰有影响，而Ｐａ第二峰也影响烧结材料的微观结构．     

Peak management

Peaks occur when a firm accustomed to operating under market constraint conditions occasionally encounters peaks in market demand that temporarily exceed the firm's supply capacity. This paper defines the demand peak, classifies peak management (PM) conditions and prescribes a framework for PM. The Theoryof-Constraints (TOC) is expanded to handle the PM concept, introducing the two-policy concept: one policy for peak periods and another for non-peak periods. The paper outlines techniques to elevate the firm's peak performance by loading during the peak period. Distinct pricing policies for peak and offpeak periods are presented and a comprehensive methodology is suggested to deal with the PM problem, including labour policy, outsourcing policy and pricing policy. The paper demonstrates the methodology and techniques in a broad range of real-life management situations. Under TOC principles, the present analysis shows the different decision rules for the two periods and their substantial impact on the firm's overall strategy. Finally, the paper discusses the ramifications of operating under two distinct sets of rules, as well as the rules for making the transitions between an off-peak and a peak period.     

离子交换波导折射率分布峰值的移动

本文提出利用表面等离子波共振(SPR)技术测量离子交换波导的表面折射率的方法.这种方法基于分析转移矩阵方法和表面等离子波技术,避免了传统方法中确定表面折射率的自由度问题,所以得到的波导表面折射率要比逆WKB法更为精确.用此方法研究了不同交换时间的样品,我们发现当交换达到一定的时间后,表面折射率会随交换时间的增长而下降,折射率分布峰值移向波导内部,说明这时存在一个银离子反扩散的过程,这与理论分析结论相吻合.     

Microstructural Cause of the Peak Effect in Superconducting Y-Ba-Cu-O Melt-Grown Bulks

Superconducting Y-Ba-Cu-O bulks with J _c as high as 30 kA /cm ² (at 77 K and 1 T) containing both Y-123 and Y-211 phases were utilized in the present work . With different post-annealings, we obtained samples with different degrees of peak effect in the J _c -vs- H characteristics. Using specimens with TEM observable areas as wide as 20 × 20 (m) ² , large-area high-resolution transmission-electron-microscopic (LA-HRTEM) observations were made for the samples with different degrees of peak effect. Statistical analyses were made for the size of Y-211 particles and the distance between adjacent twin boundaries. It is revealed that (1) Y-211 particles including those of very small diameters have nothing to do with-the peak effect, while (2) the inter-twin-boundary distance is undoubtully correlated with the degree of peak effect. Further investigation on the twin boundary structure indicates that lower T _c regions may be formed in the vicinity of twin boundaries to provide fluxons with mesoscopically distributed weak pinning centers.     

Peak effect versus skating in high-temperature nanofriction

The physics of sliding nanofriction at high temperature near the substrate melting point, TM, is so far unexplored. We conducted simulations of hard tips sliding on a prototype non-melting surface, NaCl(100), revealing two distinct and opposite phenomena for ploughing and for grazing friction in this regime. We found a frictional drop close to TM for deep ploughing and wear, but on the contrary a frictional rise for grazing, wearless sliding. For both phenomena, we obtain a fresh microscopic understanding, relating the former to 'skating' through a local liquid cloud, and the latter to linear response properties of the free substrate surface. We argue that both phenomena occur more generally on surfaces other than NaCl and should be pursued experimentally. Most metals, in particular those possessing one or more close-packed non-melting surfaces, such as Pb, Al or Au(111), are likely to behave similarly.     

Peak channel temperature of graphene-based transistors

In this study, graphene was mechanically deposited on SiO2/Si substrate, followed by ohmic metallization using electron-beam lithography. Finite element analysis was employed to characterize the operating temperature of graphene-based devices using the experimentally determined current-voltage data. The temperature of the hottest spot where the underlying SiO2 layer was 300 nm thick was elevated up to about 70 degrees C at a 10 mW dissipated power. However, the operating temperature dropped to about 50 degrees C when the 300 nm thick SiO2 layer was replaced with a 20 nm thick SiO2 layer. Thermal management is very critical in the reliability of graphene-based high speed electronic devices because the high operating temperature can degrade the device performance.     

连续电泳分离过程中峰值偏差的分析

针对目前微流控芯片上连续电泳实验过程中所产生的峰值重复性差的现象,建立了芯片的沟道模型,并对分离过程中沟道内的流场进行CFD仿真.仿真发现分离时缓冲液被推入样品池从而稀释了下次实验的样品,进而造成了连续电泳分离过程的峰值重复性差.最后通过实验验证了该仿真结果并在此基础上优化了分离时的电参数,将连续实验峰值的RSD由20%-30%降到10%以内,极大地提高了连续实验的重复性和稳定性.     

冰蓄冷空调与电网调峰

文章分析了冰蓄冷空调系统的工作原理、优越的调峰性能、广阔的应用前景和良好的社会经济和环境效益 ,提出我国应拉大峰谷电价比 ,大力发展和推广使用冰蓄冷空调。     

模式识别相关滤波器设计的峰值锐化技术

提出一种锐化相关峰的滤波器设计方法，把初始滤波函数乘上一个因子，实现相关输出的拉普拉斯运算，从而得到更税利的相关峰和更好的识别效果。