The rapid detection for methane of ZnO porous nanoflakes with the decoration of Ag nanoparticles

Realizing the real-time detection of CH₄ is important for the safety of human life. A facile hydrothermal method was used to synthesize Ag nanoparticles-decorated ZnO porous nanoflakes (PNFs) in this study. The characterization results confirmed that Ag nanoparticles had been decorated in ZnO nanoflakes with the thickness of ~10 nm. The gas-sensing properties of Ag-decorated ZnO nanoflakes were also investigated. While the gas-sensing performances of ZnO were remarkably improved by decorating Ag nanoparticles on the surface of ZnO nanoflakes, the response of the Ag-decorated ZnO sensor to 3000 ppm CH₄ is almost 1.3 times as high as that of pristine ZnO sensor. The obtained Ag/ZnO sensor exhibits better long-term stability and shorter response recovery time (5/38 s) in the comparison with pristine ZnO, demonstrating the possibility for the actual detection of CH₄. The enhanced CH₄ sensing performance can be attributed to the synergism between the unique hierarchical porous structure and the sensitizing actions utilized by the Ag nanoparticles.     

Hierarchically porous carbon derived from the activation of waste chestnut shells by potassium bicarbonate (KHCO3) for high-performance supercapacitor electrode

3D hierarchical porous carbon consisting of micropores, mesopores and macropores was successfully prepared through the activation of chestnut shell via KHCO₃. The influence of KHCO₃/chestnut shell ratio on the textural properties was carefully investigated and the structure, micrographs, and surface chemical status of the materials were expressed via X-Ray diffraction (XRD), transmission electron micrographs (TEM), scanning electron microscope (SEM), X-ray photoelectron spectroscopy (XPS), respectively. With the optimized amount of KHCO₃, high specific pore surface area (2298 m² g^-1) and high total pore volume (1.51 cm³ g^-1) were achieved, endowing an advantageous electrochemical characteristics for the electrode as ultracapacitors in three-electrode system. At 2 A g^-1, a high specific electric capacity of 387 F g^-1 was reached. The remarkable electrochemical performances are mainly due to hierarchical porous structure with the high specific surface area (SSA) and the eminent total pore volume. It means that this hierarchical porous carbon prepared by activated by using KHCO₃ would have more promising foreground in the field of energy storage.     

基于分等级搜索的可持续进化算法研究

通过分析标准遗传算法在可持续性进化方面的缺陷，提出了HFC模型及其运算流程，井给出了详细的数学描述。HFC算法的分等级搜索方式改进了标准遗传算法的缺陷，是一种能实现可持续性进化的算法模型。HFC模型不仅能确保种群的多样性，而且保证了个体间的公平竞争。实例验证了HFC模型和标准遗传算法在相同问题上的不同效果。     

面向冲裁模设计的CBR系统的实例表示

根据冲裁模的设计要求和结构特点，将冲裁模设计实例分为冲裁件和冲裁模两个层次。提出了层次化实例描述策略，建立了面向对象的基于框架的实例描述语言，并以此对实例进行外部描述。而设计实例的内部表示则用编码实现，并给出了相应的编码规则。     

重型数控机床热误差建模及预测方法的研究

重型数控机床的热误差已经成为影响其加工精度的一个关键问题。针对一台典型的重型落地铣镗床,以机床热误差测量试验为依据,分析该类机床温度场的特点;据此提出一种旨在完成高效温度测点优化的改进系统聚类方法,该方法使用一种兼顾欧氏距离和相关系数的系统聚类准则,可以有效地降低优化后温度测点之间的共线性。基于优化后的温度测点,利用多元线性回归分析,构建了机床的热误差预测模型。现场试验数据表明,该方法可以将热误差预测的均方根误差降低到10μm以下,相较于其他方法有着更高的热误差预测精度,有望在其他重型数控机床的热误差建模和预测研究中得到更大的推广应用。     

多元统计分析在产品设计要素分析中的应用

研究多元统计分析的理论,利用感性工学的方法,建立适用于大多数产品的数学模型,为产品设计要素分析提供依据.使用系统聚类分析法对相近感性意象的产品图片进行筛选,选出具有典型感性意象的图片;利用主成份分析法和聚类法将描述产品图片的感性语义进行分类、合并,获得正式调研的产品语义.以实例分析,证实是一种比较适用可行的方法.     

基于数据分布的故障分类器核函数参数优化

针对基于支持向量机的故障分类器的参数优化问题，研究了基于数据分布的支持向量机核函数参数优化的基本原理．提出了一种简化算法。依据该简化算法，实现了故障分类器的参数优化，并应用于汽轮发电机组的故障分类。测试结果表明，该参数优化方法可以提高故障分类器的分类能力，具有算法简单、优化效率高等优点，但其通用性有待进一步提高。     

基于网络光盘库的优先排队模型分析

本文深入研究了网络光盘库的排队网络模型,根据用户访问请求的不同建立了优先排队模型,并对其进行了仿真模拟,为光盘库在海量存储系统中的应用、设计和性能优化提供理论方法和实验依据.     

集装箱码头出口箱集港堆存模型研究

介绍了与集装箱堆场相关的业务,针对非自动化集装箱码头最常见的岸桥-集卡-轮胎吊装卸系统,以最复杂的出口箱场箱位分配为切入点,提出了从计划分配到动态分配的三阶段箱位分配法,给出了各阶段的优化目标.重点研究了三阶段中的具体箱位分配问题,详细分析了该问题的影响因素,提出了基于不同优先级的决策模型框架.各级子模型适用于不同的堆场密度,可分别构建并求解,并可嵌入到集装箱码头的生产营运信息系统中,为最终实现箱位分配智能决策支持系统奠定了理论基础.