非稀疏欠定盲分离及其在语音分离中的应用

本文提出一种基于马尔科夫链蒙特卡洛方法(MCMC)的贝叶斯非稀疏盲源分离算法。用广义高斯分布(GGD)来拟合源信号的分布,通过MCMC抽样得到GGD参数和隐变量的估计,并由此得到源信号的最小均方误差估计(MMSE),解决了GGD参数估计容易陷入局部极值点、鲁棒性差的问题。根据语音信号的局部平稳性,提出基于非稀疏度评判准则的盲分离算法,用MCMC方法分离非稀疏区的语音信号,进一步提高了语音信号分离精度。仿真实验证明,本文方法改善了非稀疏信号和语音信号的分离效果,而且具有更好的鲁棒性。     

铁路正交异性钢桥面板典型疲劳裂纹寿命估算

针对铁路正交异性钢桥面板中典型疲劳裂纹形式,建立计算模型。采用有限元数值方法模拟钢桥面板应力分布,确定各典型疲劳裂纹最不利位置;利用有限元子模型技术模拟各疲劳裂纹位置焊接细节,分析焊缝引起应力集中程度及对疲劳裂纹产生所造成影响;依据断裂力学揭示疲劳裂纹扩展速率与裂纹周围应力场关系,对几种典型疲劳裂纹进行疲劳寿命估算。结果表明:在桥面板与纵肋连接处,桥面板疲劳裂纹寿命较短,而纵肋疲劳裂纹寿命较长,这与国内外现场实测及试验结果相吻合;在横梁与纵肋连接处,其主应力较大且应力集中效应明显,极易产生疲劳裂纹。     

大跨张弦梁结构特点及施工方法

张弦梁结构是近年来大跨空间结构领域兴起的一种新的结构类型,以其结构受力合理,跨越能力大,施工方便且建筑造型优美受到越来越多的关注。针对张弦梁结构和基本的施工方法,介绍福州火车南站雨棚张弦梁施工技术。     

OSCAR statement of methods

OSCAR (Outil de Simulation du CAptage pour la Reconnaissance des défauts) is the pantograph–catenary dynamic software developed by Société Nationale des Chemins de fer Français (SNCF) since 2004. A three-dimensional finite element (FE) mesh allows the modelling of any catenary type: alternating current (AC) or direct current (DC) designs, and conventional or high-speed lines. It is a representative of the real overhead line geometry, with contact wire (CW) irregularities, staggered alignment of the CW, dropper spacing, wire tension, etc. Nonlinearities, such as slackening of droppers and unilateral contact between the pantograph and the CW, are taken into account. Several pantograph models can be used, with a complexity level growing from the three-lumped-mass model to the multibody model. In the second case, a cosimulation between the FE method catenary and the multibody pantograph models has been developed. Industrial features for pre- and post-treatments were developed to increase robustness of results and optimise computation time. Recent developments include volume meshing of the CW for stress computation or statistical analysis and lead to new fields of studies such as fatigue failure or design optimisation. OSCAR was fully validated against in-line measurements for its different AC and DC catenary models as well as its different pantograph models (with independent strips for instance) and has continuously been certified against EN50318 since 2008.     

基于ALE的矿粉货物液化晃荡问题并行数值模拟

含水量较高的矿粉货物在海上运输过程中易出现液化,形成自由表面并使晃荡现象加剧,严重威胁船舶运输安全。针对船载液化矿粉晃动和舱壁冲击问题,采用ALE有限元方法对其进行了细致建模和计算模拟,从三维角度考察了在船舱一定装载率和运动状态下液化矿粉的晃荡现象和特性;同时,借助不同仿真软件,对计算结果的合理性和准确性进行了相互比对和分析。模型求解借助了高性能计算资源,以解决问题求解时间长和多组计算工况带来的大规模计算需求;结合所建计算模型特点和流固耦合特性,研究了多核环境下两种不同区域分解策略和实现方式,通过并行计算性能数据比较分析,以探求更为合理的并行加速策略。     

基于云计算的海上信息系统

海洋信息系统能够分别向船舶公司、船员、港口工作工人等组织和人员提供实时的海洋数据,为他们的工作提供巨大的便利,在近年来得到了广泛应用。然而,在传统海洋信息系统构建方法中,不同组织和领域的系统往往孤立,互相之间不具备数据交换和访问能力,一方面浪费了大量存储的计算资源,另一方面为各部门之间的协作造成障碍。为解决以上问题,本文提出一种新型的基于云计算的海上信息系统,以弥补不同信息系统之间信息共享的障碍,将云计算技术应用于海洋信息领域,取得了较为明显的效果。     

简议膨胀土路堤处治

分析总结了国内外膨胀土路堤的主要处治方法,针对南友路膨胀土的性质,进行了室内掺灰改性试验、模拟封闭条件下路基土的土性试验,由此提出了一套以封闭为主的膨胀土路堤处治方法。     

基于振型动能法的大跨度斜拉桥振型分析和模态识别

针对拉索的自身振型以及与塔梁相互间的模态耦合作用被人为忽略和单凭振型图难以识别出主梁的主振型两大问题,采用多段索单元模拟拉索以便能准确地计算结构模型。通过动力计算分析振型参与的动能比例,实现了振型识别功能,得出斜拉桥的自振频率和振型。分析结果表明:斜拉桥的自振特性表现出明显的三维性和相互耦合的特点,主梁、桥塔、斜拉索之间相互影响;在一个较宽的频率范围内,许多振型都可能被动力荷载激起强烈的振动;为识别出主梁振动为主的振型和主振型方向分析振型参与的动能比例,应采用10阶以上的振型情况分析,为动力测试提供理论数据。     

基于粗糙集交通信息提取计算的城市道路行程时间预测

针对城市道路的行程时间预测问题进行研究。由于城市道路交通问题具有不确定性和不精确性,故采用基于粗糙集的交通信息提取计算理论建立城市道路行程时间预测模型。模型建立后,利用在荷兰代尔夫特市采集到的实际数据,对该预测模型进行检验。检验结果表明:如果不进行原始数据的前期处理,那么得到的预测误差在35%左右;而在剔除了质量较差的数据后,预测精度明显提高;同时,条件属性和决策属性的分类,显著影响到预测的精度。通过计算得到分类范围值,该模型能够较好的对交通状态进行物理解释同时预测精度能够达到可以接受的范围。     

Revisiting the application of simultaneous perturbation stochastic approximation towards signal timing optimization

Simultaneous Perturbation Stochastic Approximation (SPSA) has gained favor as an efficient optimization method for calibrating computationally intensive, “black box” traffic flow simulations. Few recent studies have investigated the efficiency of SPSA for traffic signal timing optimization. It is important for this to be investigated, because significant room for improvement exists in the area of signal optimization. Some signal timing methods and products perform optimization very quickly, but deliver mediocre solutions. Other methods and products deliver high-quality solutions, but at a very slow rate. When using commercialized desktop signal timing products, engineers are often forced to choose between speed and solution quality. Real-time adaptive control products, which must optimize timings within seconds on a cycle-by-cycle basis, have limited time to reach a high-quality solution. The existing literature indicates that SPSA provides the potential for upgrading both off-line and on-line solutions alike, by delivering high-quality solutions within seconds. This article describes an extensive set of optimization tests involving SPSA and genetic algorithms (GAs). The final results suggest that GA was slightly more efficient than SPSA. Moreover, the results suggest today's signal timing solutions could be improved significantly by incorporating GA, SPSA, and “playbooks” of preoptimized starting points. However, it may take another 5–10 years before our computers become fast enough to simultaneously optimize coordination settings (i.e., cycle length, phasing sequence, and offsets) at numerous intersections, using the most powerful heuristic methods, at speeds that are compatible with real-time adaptive solutions.