Agent-based simulation of competitive and collaborative mechanisms for mobile service chains

A new paradigm for a mobile service chain’s competitive and collaborative mechanism is proposed in this study. The main idea of the proposed approach is based on a multi-agent system with optimal profit of the pull, push, and collaborative models among the portal access service provider (PASP), the product service provider (PSP), and the mobile service provider (MSP). To address the running mechanism for the multi-agent system, an integrated system framework is proposed based on the agent evolution algorithm (AEA), which could resolve all these modes. To examine the feasibility of the framework, a prototype system based on Java-Repast is implemented. The simulation experiments show that this system can help decision makers take the appropriate strategies with higher profits. By analyzing the expectations and variances (or risks) of each player’s profit, the interaction between and among entities in the chain is well understood. It is found that in the situation where a collaborative mechanism is applied, the performance of players is better as compared to the other two situations where a competitive mechanism is implemented. If some constraints are applied, the risk will be kept at a low level.     

欧盟EuP指令对高强度气体放电灯的能效要求分析

2009年3月，欧盟委员会公布了EuP指令有关高强度气体放电灯的实施措施。本文对该实施措施有关高强度气体放电灯的能效要求进行了介绍，就主要技术指标与我国国标进行比对，同时探讨其对我国高强度气体放电灯企业产生的影响。     

蒽醌降解物再生新型催化剂的制备与表征

以氢氧化铝直接快速脱水为原料制备蒽醌降解物再生新型催化剂。运用表面原位生长技术,采用固体碱与催化剂掺杂复配技术,制备出了孔容≥0.55 mL/g、比表面积≥180 m2/g、再生活性≥5 g/L、使用寿命≥90 d的条状蒽醌降解物再生新型催化剂。表面原位生长技术有利于氧化铝一次粒子长大,增大间隙孔,从而制备出大孔容、高比表面积的活性氧化铝;固体碱与催化剂掺杂可以强化碱性活性组分与载体的相互作用,避免碱液活性组分的快速流失,从而大大提高新型催化剂的活性和使用寿命。     

结合区域生长与图割算法的冠状动脉CT血管造影图像三维分割

针对图割算法适用于小幅图像,且在分割结构较复杂、感兴趣区域较小的三维CT血管造影(CTA)冠状动脉图像时效率较低的问题,实现了将区域生长和图割结合分割冠状动脉的算法.首先,利用基于阈值的区域生长算法将图像划分为若干区域,去除无关像素,得到结构简化、感兴趣区域较突出的图像;其次,对简化后的图像,结合灰度和空间信息构造网络图;最后,利用图割理论实现网络图分割,得到冠状动脉分割图像.实验结果表明,与传统的图割方法相比: 在分割效率上,区域生长和图割结合的分割算法降低了计算复杂度,效率提高了51.7%; 在绘制质量上,得到的冠状动脉分割图像目标区域完整,有助于医师对病变的正确分析.     

Influence of Mg Addition on Graphite Particle Distribution in the Al Alloy Matrix Composites

Al alloy matrix composites reinforced with copper-coated graphite particle have been prepared by melt stirring process in this work. The effect of the addition of Mg on distribution of the graphite particles has been investigated. Scanning electron microscopy (SEM) was used to observe the micro-morphology of Al alloy matrix composites reinforced with graphite particles. Meanwhile, the content of graphite was analyzed in the different position of casting by dissolution method and the mechanical properties of the composites were detected. The results show that the content of graphite increase with increasing Mg content; the graphite particles distribute uniformly in the particle reinforced metal matrix composites (PMMC) with 0.6 wt pct Mg; however, the agglomeration of the graphite particles is observed obviously in the matrix when Mg content is more than 1.0 wt pct. In addition, the proper Mg addition amount is beneficial to enhance the mechanical properties of the graphite particles reinforced Al alloy matrix composites and the abrasion resistance of the materials due to a reduce friction coefficient.     

基于密度峰值的网格聚类算法

2014年提出的密度峰值聚类算法,思想简洁新颖,所需参数少,不需要进行迭代求解,而且具有可扩展性。基于密度峰值聚类算法提出了一种网格聚类算法,能够高效地对大规模数据进行处理。首先,将N维空间粒化为不相交的长方形网格单元;然后,统计单元空间的信息,利用密度峰值聚类寻找中心点的思想确定中心单元,即中心网格单元被一些低局部密度的数据单元包围,而且与比自身局部密度高的网格单元的距离相对较大;最后,合并与中心网格单元相近网格单元,从而得出聚类结果。在UCI人工数据集上的仿真实验结果表明,所提算法能够较快得出聚类中心,有效处理大规模数据的聚类问题,具有较高的效率,与原始的密度峰值聚类算法相比,在不同数据集上时间损耗降低至原来的1/100~1/10,而精度损失维持在5%~8%。     

智能水下机器人多传感器信息融合的一种方法

首先叙述了自行研制的水下机器人的数据融合目的，通过分析水下智能机器人采用的传感器的性能，对各种传感器的信息进行可信度分配，然后采用Ｄｅｍｐｓｔｅｒ－Ｓｈａｆｅｒ证据理论方法，把各类传感信息作为证据，判断机器人前方是否存在障碍物。     

Walnut‐Like Multicore–Shell MnO Encapsulated Nitrogen‐Rich Carbon Nanocapsules as Anode Material for Long‐Cycling and Soft‐Packed Lithium‐Ion Batteries

Metal oxide‐based nanomaterials are widely studied because of their high‐energy densities as anode materials in lithium‐ion batteries. However, the fast capacity degradation resulting from the large volume expansion upon lithiation hinders their practical application. In this work, the preparation of walnut‐like multicore–shell MnO encapsulated nitrogen‐rich carbon nanocapsules (MnO@NC) is reported via a facile and eco‐friendly process for long‐cycling Li‐ion batteries. In this hybrid structure, MnO nanoparticles are uniformly dispersed inside carbon nanoshells, which can simultaneously act as a conductive framework and also a protective buffer layer to restrain the volume variation. The MnO@NC nanocapsules show remarkable electrochemical performances for lithium‐ion batteries, exhibiting high reversible capability (762 mAh g^?1 at 100 mA g^?1) and stable cycling life (624 mAh g^?1 after 1000 cycles at 1000 mA g^?1). In addition, the soft‐packed full batteries based on MnO@NC nanocapsules anodes and commercial LiFePO₄ cathodes present good flexibility and cycling stability.     

Gold nanoparticles for high-throughput genotyping of long-range haplotypes

Completion of the Human Genome Project and the HapMap Project has led to increasing demands for mapping complex traits in humans to understand the aetiology of diseases. Identifying variations in the DNA sequence, which affect how we develop disease and respond to pathogens and drugs, is important for this purpose, but it is difficult to identify these variations in large sample sets. Here we show that through a combination of capillary sequencing and polymerase chain reaction assisted by gold nanoparticles, it is possible to identify several DNA variations that are associated with age-related macular degeneration and psoriasis on significant regions of human genomic DNA. Our method is accurate and promising for large-scale and high-throughput genetic analysis of susceptibility towards disease and drug resistance.