一种基于自适应进化策略采样的粒子滤波算法

提出了一种自适应进化策略算法(AES),该算法利用适应度值控制变异步长的自适应调整,从而提高了进化策略的搜索效率和精度。将AES算法和粒子滤波(PF)相结合,提出了基于自适应进化策略采样的粒子滤波算法(AESPF)。该算法将AES应用于粒子重采样,以保证粒子的有效性和多样性。通过仿真计算表明,提出的算法可以有效提高滤波性能。     

被动多传感器改进交互多模型粒子滤波算法

将改进的粒子滤波算法即基于均匀重采样的粒子滤波(AUPF)与交互式多模型算法(IMM)相结合,提出交互式多模型均匀重采样粒子滤波算法(IMM-AUPF),并将其应用于被动多传感器的机动目标跟踪中。均匀重采样粒子滤波在标准粒子滤波的基础上通过改进重采样过程,在解决粒子退化问题的同时,增加了粒子的多样性,提高了滤波性能。在多模型中应用均匀重采样粒子滤波,提高被动多传感器系统的机动目标跟踪精度。将该方法与交互式多模型粒子滤波算法(IMM-PF)进行仿真对比,结果表明该方法具有更好的跟踪性能。     

基于UKF和线性优化的改进粒子滤波算法

针对标准粒子滤波算法存在的粒子退化问题,提出了一种改进的粒子滤波算法,该算法将不敏卡尔曼滤波算法（UKF）、线性优化的思想和基本粒子滤波算法相结合,运用不敏卡尔曼滤波算法获得重要性概率密度函数,提高了粒子的使用效率;运用线性优化的思想,保证了所有粒子都以一定的概率对状态估计作出贡献,提高了粒子的多样性。仿真结果表明,改进的算法很好的解决了基本粒子滤波存在的粒子退化问题,具有更高的状态估计精度。     

自适应网格交互多模型不敏粒子滤波算法

将一种变结构多模型算法——自适应网格交互多模型(AGIMM)算法和不敏粒子滤波(UPF)算法相结合,提出了自适应网格交互多模型不敏粒子滤波算法(AGIMMUPF).该算法通过自适应网格实现了模型自适应,从而以较小的模型集合覆盖了目标大范围的机动,并以此来克服固定结构交互多模型粒子滤波(IMMPF)算法存在的缺陷,同时各模型滤波算法采用不敏粒子滤波(UPF)算法,使重要性密度函数融合了最新量测信息,更好地逼近真实状态的后验概率分布.通过计算机仿真证明,提出的算法可以有效提高IMMPF的费效比.     

一种改进的粒子滤波算法

为提高非线性滤波的性能,提出一种改进的粒子滤波算法(MUPF),该算法将新构造的辅助模型和无迹卡尔曼滤波(UKF)相结合产生新的建议分布函数。与传统的无迹粒子滤波(UPF)相比,该算法充分利用了新的测量信息。在非高斯条件下与其他滤波器进行了仿真对比,分析了跟踪性能和均方根误差。仿真结果表明,MUPF不仅具有很高的跟踪精度,而且用很少的粒子就可以达到更好的滤波性能。     

基于粒子群算法的微带天线特征方程求解方法

粒子群算法是根据鸟群觅食过程中的迁徙和群集模型而提出的用于解决优化问题的一类新兴的随机优化算法.文中充分利用粒子群算法的全局收敛特性,并与参数跟踪策略相结合形成一种新的求解复超越方程算法.基于该算法对微带天线的特征方程进行了求解,求解结果与该问题的已有结论相吻合,说明该算法的有效性.     

改进自适应惯性权重粒子群算法及其在核动力管道布置中的应用北大核心CSCD

[目的]旨在研究非线性自适应惯性权重粒子群优化算法,实现船用核动力一回路系统管道路径的布置优化设计。[方法]根据船用核动力一回路系统的管道布局设计特点,建立一回路系统的管道布局空间模型、约束条件和评价函数;基于管道节点数量,提出一种粒子群优化(PSO)算法的新型定长编码方法,然后结合该编码方法建立方向引导机制;在此基础上,针对粒子群优化算法易陷入局部最优解、收敛速度慢的缺点,结合辅助线性变化的学习因子,提出一种基于非线性自适应惯性权重的改进粒子群优化算法;将改进粒子群优化算法与协同进化算法相结合,提出一种用于求解分支管道布局问题的协同进化粒子群优化算法,以用于核动力一回路系统的管道布局优化。[结果]仿真结果显示,所提的改进算法与标准算法相比收敛速度提高了40%~50%,不仅能够得到更好的管道布局效果,还解决了标准粒子群优化算法容易陷入局部最优解的问题。[结论]研究成果可为船用核动力一回路系统管道布置的优化设计提供有益的参考。     

基于PF-DC的多传感器目标跟踪算法

针对传统的粒子滤波(PF)的多传感器目标跟踪算法存在计算量过大的问题,研究了一种基于PF和数据压缩(PF-DC)的多传感器目标跟踪算法.仿真结果表明,与传统的串行多传感器PF算法相比较,该算法可以极大的降低算法的计算量,从而有效地提高算法的实时性.     

基于一种改进粒子滤波算法的目标跟踪研究

为了解决非线性、非高斯系统目标跟踪问题,研究了一种新的滤波方法——高斯粒子滤波算法。通过基于重要性采样和蒙特卡罗模拟方法得到一高斯分布来近似未知状态变量的后验分布。并讨论了此算法在机动目标非线性转弯运动中的跟踪应用,与粒子滤波算法相比,其优点是不需要重采样步骤。在闪烁噪声下比较了高斯粒子滤波器、粒子滤波器和扩展卡尔曼滤波器在滤波精度、运算时间等方面的差异,仿真结果表明该算法性能优于其他算法。     

用Vivado HLS实现粒子滤波算法的硬件加速

针对传统现场可编程门阵列(field-programmable gate array,FPGA)开发流程中,以硬件实现粒子滤波算法(particle filter,PF)的开发周期长、过程繁琐和灵活性差等缺陷,提出了一种基于Vivado高层次综合(high level synthesis,HLS)工具的PF算法硬件实现新方法,直接高效地将C语言描述的PF算法综合为RTL硬件模块.文中以二维纯方位跟踪(2 dimensional bearing-only tracking,2-D BOT)为应用背景,以高斯粒子滤波(Gaussian particle filter,GPF)为目标算法,根据其数据结构,给出相应的并行流水线策略,综合出具有高并行度的FPGA运算模块.C/RTL协同仿真表明:该运算模块能较好实现对动目标的跟踪,并且能达到相当于简化粒子滤波器1.5倍的运算速度;同时该实现方法对其他复杂软件算法的硬件化加速具有指导意义.     

Errata

正In the paper"Influence of Fouling Assemblage on the Corrosion Behaviour of Mild Steel in the Coastal Waters of The Gulf of Mannar,India"in Vol.12,No.4,Page:509,References were lost,and the two authors’biographies were identical.The correct text is shown below.We apologize to the authors and our readers for any inconvenience caused by the errors.     

Journal of Marine Science and Application 2014 Sum. Contents

正~~     

第34届海洋、离岸及极地工程国际会议(英文)

正St.John's,Newfoundland,Canada,May 31-June 5,2015 OMAE2015 is the ideal forum for researchers,engineers,managers,technicians and students from the scientific and industrial communities from around the world to: meet and present advances in technology and its scientific support;     

联合作战计划和执行系统

全球指挥控制系统(GCCS)实施当前美国海军网络中心战信息基础设施的联合计划网络。联合作战计划和执行系统(JOPES)支持GCCS实现联合计划。JOPES有两类计划：时间不限的精密预案计划生成作战计划、方案计划或职能计划；时间敏感的危机行动计划生成作战命令或战役方案。前者在和平时期创建的作战计划是后者的计划基础，加速应付危机的能力。     

Eastbound or westbound ?

In terms of equal sailing distances, where is the inflexion when ships depart from ports in the Asian Continent to New York via Suez and/or Panama？
The answer is China South Port Group （Hong Kong, Shenzhen and Guangzhou）.     

Recent progress on wave induced loading measurements in warships

Recent measurements of wave induced hull strain and flexure in RN warships are presented together with the derivation of the current design criteria for extreme hull girder bending loads. The history of the development of the shipboard instrumentation used is given and recent developments to improve the quality and ease of analysis of the data are described. An unexpectedly high transverse asymmetry in the longitudinal strains measured in destroyers is shown to be the result of a combination of vertical and lateral bending in oblique seas. Finally recent theoretical comparisons between the loading of Deep-Vee hulls and conventional UK rounded bilge hulls are presented which demonstrate the higher loading experienced by this type of hull form.     

桥梁防撞设施物理模型试验

物理模型试验是预报防撞设施所受撞击力及优化设计方案的主要方法之一。结合杭州湾大桥柔性防撞设施及东海大桥独立式防撞体设计方案,阐述了船—防撞体撞击试验的基本原理、方法、试验方案及相关试验结果。提出了柔性防撞系统的优化方案。试验结果表明,该方案经济有效,可供预报类似桥梁防撞设施撞击力时参考。     

印尼某电厂码头水工建筑物设计

印尼某电厂项目码头水工建筑物包括码头、防波堤和护岸等主要组成部分。在该项目水工建筑物的设计过程中,综合考虑当地的自然条件、施工能力和材料来源等因素,不断优化结构选型和结构断面。根据不同使用要求,护岸分别采用直立式和斜坡式2种结构型式,推荐的设计方案节省投资、施工方便,可供同类工程设计参考。     

Impact of regulation on the performances of long-distance transport services: A comparison of the different approaches in Sweden and Norway

Despite the many socio-economic similarities between Sweden and Norway, differences in jurisdiction, organisation, cooperation, and financing of long-distance passenger train and coach services have led to the development of four distinctively different ways of serving the markets. This paper describes how the train and coach markets have developed in the two countries, with emphasis on regulatory and industrial structure and a couple of performance variables.Looking at passenger rail, both countries separated infrastructure from operation over a decade ago. However, while Norwegian rail is characterised by an almost monopoly supplier, rail services in Sweden are partly decentralised to the responsibility of county authorities and are widely subjected to competitive tendering. The rest of the network is about to be opened up for on-the-track competition. Swedish Rail (SJ) has spent the last decades consolidating its core business (passenger rail) and sold out its other businesses. In contrast, the Norwegian state rail (NSB) has expanded its business to become a major bus operator and property owner, with extensions also into the Swedish market.The coach industry was more recently deregulated in both countries. The Swedish coach market is dominated by privately owned companies operating services to and from Stockholm. In Norway, state-owned NSB is a major coach operator on medium distance routes, and is also the largest partner of Nor-Way Bussekspress which totally dominates long-distance coach services. Further, the Norwegian coach market is characterised by cross-ownership and cooperation which has enabled an extensive route network which covers most of Norway.We find distinct differences in achievements in the two modes and in the two countries. Swedish rail services have succeeded in winning market shares and in renewing and developing both infrastructure and service levels to a greater extent than the Norwegian model. On the other hand, the Norwegian coach market seems to be more developed and efficient compared to its Swedish counterpart.The paper concludes with a discussion on the possible links between the different approaches and the performance observed, with the aim to stimulate further and more detailed research on some important issues.     

开放式数据挖掘系统模型

目前国内已经出现了少量的数据挖掘相关产品。国内数据挖掘软件无论从数量还是质量上比较，都与国外软件存在着较大的差距。随着数据仓库技术的普及，数据挖掘应用的需求越来越强烈，如何缩短这种差距，研发数据挖掘软件产品成为国内业界的一个重要问题。通过引用目前国际通用数据挖掘标准技术来构建开放式的数据挖掘系统模型，该系统模型从可扩展性、可重用性、易用性等方面得到了加强。