基于主成分和BP神经网络的汽车保有量预测

针对汽车保有量预测建模中输入因子过多而导致神经网络规模过大、泛化能力差的问题,通过主成分分析法和贝叶斯正则化方法对BP神经网络进行改进,可简化网络结构,增强泛化能力。对此,以我国汽车保有量预测为例进行的仿真计算表明,结果令人满意,这也同时证明了该方法用于汽车保有量预测的可行性与有效性。     

汉字知多少

汉字是记录汉语的文字,也是世界上最古老的文字之一,已有六千年左右的历史。现存最古可识的是三千多年前殷商的甲骨文和稍后的金文。现用汉字是从甲骨文、金文（钟鼎文）演变而来的。     

可持续发展的绿色交通

本文在分析了现有交通运输系统的不可持续性的基础上,提出了构建绿色交通体系,促进城市交通的可持续发展,并诠释了绿色交通与可持续发展之间的关系,最后列举了发展绿色交通的几条措施。     

汽车门框密封条设计解析与研究

本文通过从密封条的产品性能、结构设计、产品材料三个方面对车门门框密封条的设计过程进行了全方位的解析与研究。     

动力总成悬置系统性能评价

主要介绍汽车动力总成悬置系统性能的评价方法,包含设计阶段的解耦目标分析、动力总成刚体模态分析,重点分析整车阶段基于试验的悬置系统隔振率与整车振动的关系,得出针对悬置系统隔振性能的较为有效、合理的评价方法。     

重型柴油机瞬态工况排放和EGR影响的试验研究

本文中对一重型柴油机瞬态工况下的NOx与碳烟排放和引入EGR技术的影响进行试验研究。结果表明,引入EGR后,NOx的瞬态排放有所改善;但碳烟的瞬态排放恶化;关闭EGR阀可缓解碳烟瞬态排放的恶化。通过直流电机EGR阀关闭阶跃试验,揭示了EGR阀的关闭对发动机各项性能的影响规律,为重型柴油机瞬态排放的EGR控制提供了依据。     

基于用户相关数据的乘用车试验场变速强化试验方法

为避免以往乘用车传动系可靠性试验的盲目性,提出一种与用户实际使用工况相关联的变速强化试验方法,该方法采集强化路和用户典型路面的试验数据,计算传动轴各级转矩对应的累积疲劳损伤。利用Weibull分布和Monte-Carlo仿真模拟90%用户的总损伤,最终获得试验场变速试验强化系数,为科学评价乘用车传动系可靠性和可靠性试验规范的制订提供技术依据。     

基于分布式智能的船舶电力推进系统运行控制与管理策略研究

运行控制和管理策略是船舶电力推进控制系统设计的核心技术,是实现船舶电力推进控制系统自动化、网络化、信息化的基础.本文对基于分布式智能船舶电力推进系统的运行特征进行了分析,梳理推进系统设备以及与其它系统之间的控制逻辑,研究了满足任务需求的电力推进系统运行控制和管理策略及体系结构.为水面船舶电力推进监控系统标准化设计奠定了良好的基础.     

Discussion on the assessment of initial stress field of railway rock massEI北大核心

Research purposes: In actual work, according to the ground stress data for assessment, most railway tunnels are under very high ground stress and high ground stress state, which causes adjustment of surrounding rock and engineering measures to strengthen, thus these will cause large increased investment, even cause large dispute and chaos in analysis causes of tunnel lining cracking and deformation, but the engineering case that rock burst and large deformation occurred in actual excavation of tunnel is very rare. In this paper, according to rock mass mechanical theory, the ground stress information and engineering case, combined the relevant provisions of similar industry, the assessment of initial stress fields of rock mass in railway specification is discussed. Research conclusions: (1) The assessment of railway specification to initial ground stress field is not realistic and should be revised, preliminary judgment when the measured in-situ stress >30 MPa, the possibility of rockburst is increased. (2) Assessment conclusion of initial ground stress field should be discreet, the increased investment should be incorporated into the dynamic design. (3) The stress tests of survey design stage should be reasonable and appropriate. (4) The research results can be applied to the modification of railway norms and tunnel exploration.     

Risk‐based stochastic equilibrium assignment model in augmented urban railway network

This study developed a methodology to model the passenger flow stochastic assignment in urban railway network (URN) with the considerations of risk attitude. Through the network augmentation technique, the urban railway system is represented by an augmented network in which the common traffic assignment method can be used directly similar to a generalized network form. Using the analysis of different cases including deterministic travel state, emergent event, peak travel, and completely stochastic state, we developed a stochastic equilibrium formulation to capture these stochastic considerations and give effects of risk aversion level on the URN performance, the passenger flow at transfer stations through numerical studies. Copyright © 2012 John Wiley & Sons, Ltd.