基于MSWF的初始投影向量估计方法研究

分析了初始期望信号存在误码差时,信号经两级维纳滤波得到的估计值,给出了估计初始投影向量的方法。该方法可以不利用接收信号的先验信息,经过两级维纳滤波后构造初始投影向量,从而进行信号子空间的估计,降低了初始期望信号存在误差时对多级维纳滤波的影响,提高了子空间估计的精度。     

基于地磁熵及灰关联分析的地磁匹配新方法

利用地磁定位和惯性导航来确定载体的准确位置的方法,是近年来发展起来的新技术。其关键问题之一是如何进行精确地磁匹配,为解决此问题,文章提出综合利用地磁熵及灰关联分析的新方法,即在计算出待匹配区域的地磁熵值的基础上,选择相应的灰关联匹配的阈值,从而找出最佳匹配点,进而确定载体位置。文章分析了该算法的理论基础并提出了具体的实现算法,经过仿真实验表明,此算法的运算速度快、匹配精度高、具有良好的抗测量误差干扰的能力。     

基于金字塔分解的分层光学相关匹配策略

在光电混合相关识别系统中,由于光电和电光转换的限制,相关运算不能以光速来进行,降低了系统实际的处理速度。该文将金字塔分解用于光学相关匹配,以实现分层光学相关,来提高海量图像的处理速度。首先分析了构建图像金字塔过程中的滤波方法和目标尺寸对相关结果的影响,从而选定了合适的图像金字塔类型,确定了系统所容许的目标尺寸,然后阐述了分层相关匹配的实现方法。利用该方法处理速度至少可提高约4倍。     

桥梁汽车活载特征分析与设计值研究

该文对北京市近年来车辆荷载的类型及轴载数据进行了详细的分析与处理,选取了具有代表性的实测荷载数据,采用一定的计算方法对其进行分析后,得出实际车辆荷载的计算代表值,并与规范设计值进行了对比,总结了两者之间车辆荷载的特征与差异,并提出了北京地区的汽车设计荷载建议值,可为北京地区的桥梁维修加固及未来北京地区制定汽车荷载设计规范的地方标准提供参考。     

武汉地铁通用管片楔形量研究

盾构隧道通用管片简化了模板设计,施工中能够较好地控制隧道掘进轴线和管片成环质量,通用性强.武汉地铁4号线一期工程为了便于统一供应管片,盾构区间设计采用通用管片,根据线路平面和纵断面设计条件,研究了采用不同楔形量通用管片对应的线路拟合误差分布情况,并根据分析结果优选出该线通用管片楔形量的合理取值.     

智能网联环境下基于混合深度学习的交通流预测模型

为适应未来智能网联环境下精细化交通流预测需求，提出一种基于混合深度学习 (Hybrid Deep Learning, HDL)的车道级交通流速度预测模型. 模型以智能网联系统强大的数据采集和计算能力为基础，采用集成经验模态分解算法将原始速度序列分解为多个固有模态函数分量和残差分量，并将所得分量重构为模型输入；利用双向长短期记忆神经网络和注意力机制，构建深度学习模型框架；为检验模型预测精度和可靠性，选择北京市二环路多个连续车道断面速度数据进行算法验证. 结果表明，HDL模型在不同车道均有理想的预测结果，单步和多步预测精度均显著优于对比模型.     

A multi-layered risk exposure assessment approach for the shipping industry

Maritime administration and coastal states have become more aware of the need to enhance risk mitigation strategies primarily due to increased worldwide shipping activities, changing safety qualities of the world fleet and limited resources to deploy mitigation strategies. This paper introduces an innovative multi-layered framework to assess, predict and mitigate potential harm. The proposed approach addresses known restrictions of risk assessments in shipping. These restrictions are the lack of scalability to apply risk assessments over large areas using an automated routine, the absence of recognizing that the world fleet is heterogeneous, the lack of integrating location specific environmental conditions such as wind, currents or waves and most importantly, the lack of recognizing the uncertainties associated with each factor especially for predictions. The proposed framework is based on the idea of integrating various layers representing the most important factors that can influence risk in order to estimate and predict risk exposure for a given area. As proof of concept of the underlying ideas, the outcome of a pilot project with the Australian Maritime Safety Authority is presented which demonstrates the integration of the first two layers and is based on a unique and comprehensive combination of data. The results of selected endpoints of risk exposure compare well with observations. The article also discusses the integration of the remaining layers including the recognition and addition of uncertainties in the future.     

关于新时代我国综合交通运输发展的若干思考

为进一步明确交通强国背景下我国综合交通运输发展战略，从理论和实践两个层面论证交通强国与综合交通运输的辩证关系，在此基础上分析了供给侧结构性改革、国家战略支撑、区域协调发展、科技创新、扩大对外开放、治理体系和治理能力现代化等对综合交通运输发展的新要求。以习近平关于综合交通运输的重要论述为根本遵循，研究提出新时代我国综合交通运输发展安全、便捷、高效、绿色、经济的价值导向、总体思路，探讨基础设施、运输服务、科技创新、治理体系四个方面的战略重点，为交通强国背景下我国综合交通运输发展构建了总体战略框架。     

The capitalization of subway access in home value: A repeat-rentals model with supply constraints in Beijing

Urban rail transit enhances accessibility of the communities it serves and often contributes to the value of local real properties. A small number of studies have gone beyond the traditional cross-sectional hedonic analysis to use repeat-sales data in the US for more robust estimation of rail transit’s impact on property value. While the empirical literature generally supports the positive capitalization of rail transit in home value, it has produced widely varied estimates and failed to incorporate the theoretical insight that more elastic housing supply would reduce the extent of local amenities’ capitalization in home value. Observing Beijing’s rapid expansion of subway lines during the late 2000s, this study uses repeat rental transactions from 2005 to 2011 to investigate how home value reacts to the change in the home’s distance to nearest subway station. Repeat-rentals estimates suggest a rent-distance elasticity of −0.02, 70% below the cross-sectional hedonic estimate. Moreover, using the unique history of state-owned enterprise relocations as instrument for intra-city variation in land supply, we find that the capitalization of subway proximity in home value is weaker where land supply is more elastic. Our findings support the significant bias caused by omitted variables in the hedonic estimates and confirm housing supply’s role in explaining the intra-city spatial variation in estimates.     

基于自适应奇异值标准谱和EMD的柴油机故障诊断

针对柴油机多发故障,提出了自适应奇异值标准谱和经验模态分解(Empirical Mode Decomposition,EMD)相结合的故障诊断模型。通过计算平均最近邻域发散度和奇异值标准谱的方法自适应地选择奇异值分解的嵌入维数和重构阶数,提高了奇异值分解降噪的精度。对降噪后的信号进行EMD分解,并利用调整余弦相似度标准提取反映信号真实特征的主固有模态函数(Intrinsic Mode Function,IMF),进而提取故障特征参数。将此模型应用于F3L912柴油机进气门漏气、单缸失火和多缸失火等故障的诊断,通过提取峭度和过零率作为故障特征,获得了较高的故障分类准确率。