一种顾及多因素影响的公路客运量预测模型

提出采用双线性模型预测公路客运量。预测模型包含时序自回归项、线性回归项和双线性项三部分。为了研究模型参数的时变特性,将公路客运量预测分为模型参数的预测和在此基础上的客运量预测,采用多层递阶方法计算模型的时变参数,然后进一步分析拟合参数的变化曲线,计算后续时段的参数预测值,并以此进行公路客运量预测。实例分析表明,双线性动态预测模型能很好地反映公路客运量的发展特性,具有较高的预测精度和实用价值。     

基于非完备贝叶斯网络的车型识别方法

针对BP神经网络等车型识别算法不能很好地适应我国车型复杂状况，分析了贝叶斯网络（IDS-BN）算法中完备数据集的问题，提出了基于非完备数据集的贝叶斯网络车型识别方法和车型识别系统结构模块，基于该模块拟定了车型特征变量．构建了车型识别网络模型，并给出了模型参数学习、车型分类器等算法，包括MDL评分和贪婪搜索的结构学习、EM参数学习、随机模拟采样推理等。实验表明，该方法识别率较高，鲁棒性好，满足我国车型识别的实际要求。     

边坡稳定分析及其设计中的有关问题探讨

从工程角度出发对边坡稳定分析方法的实质及选用、强度参数的取值问题、雨水对边坡稳定的影响等进行了探讨,提出了公路边坡稳定与防护的措施。     

A method for estimation of vehicle inertial parameters

In this paper, a new method is presented for estimating the current sprung mass inertial parameters of a vehicle, such as the mass, pitch and roll mass moments of inertia, and lateral and longitudinal centre of gravity locations. The method measures the sprung mass response when the vehicle is driven over an unknown and unmeasured random road profile. From these measurements, the equivalent free-decay responses are extracted and modal analysis techniques used to estimate the sprung mass natural frequencies, damping ratios and mode shapes. This information is combined with a simplified vehicle estimation model, least squares analysis and known equivalent stiffness parameters to estimate the vehicles’ inertial parameters. The results obtained from several simulation examples show that estimates of the inertial parameters generally have small relative errors.     

The dynamic study of locomotives under saturated adhesion

In order to study the dynamic behaviours of locomotives under saturated adhesion, the stability and characteristics of stick–slip vibration are analysed using the concepts of mean and dynamic slip rates. The longitudinal vibration phenomenon of the wheelset when stick–slip occurs is put forward and its formation mechanism is made clear innovatively. The stick–slip vibration is a dynamic process between the stick and the slip states. The decreasing of mean and dynamic slip rates is conducive to its stability, which depends on the W/R adhesion damping. The torsion vibration of the driving system and the longitudinal vibration of the wheelset are coupled through the longitudinal tangential force when the wheelset alternates between the stick and the slip states. The longitudinal oscillation frequencies of the wheelset are integral multiples of the natural frequency of torsion vibration of the driving system. A train dynamic model integrated with an electromechanical and a control system is established to simulate the stick–slip vibration phenomenon under saturated adhesion to verify the theoretical analysis. The results show that increases of the longitudinal axle guidance stiffness and the motor suspension stiffness are beneficial to the stick–slip vibration stability and the locomotive's traction ability. The optimised matching of the longitudinal axle guidance stiffness and the motor suspension stiffness are helpful to avoid longitudinal resonance when the stick–slip vibration occurs.     

浅析FTA在摩托车重要部品中的应用

常用的可靠性分析有故障模式、影响及危害性分析(FMEA)和故障树分析(FTA)两种.用FTA来解决设计开发时重要部品的重要性能问题和生产过程中重要工艺参数的质量控制,对提高我国摩托车重要部品和整车的可靠性能具有重要的指导和现实意义.     

湘江隧道大跨、浅埋段超前预注浆加固技术

针对长沙营盘路湘江隧道大跨段存在岩土交界面、板岩地层遇水崩解、稳定性差、埋深浅的施工难题,采用超前预注浆技术分部对隧道进行加固,通过注浆过程控制及效果检查确保注浆加固效果,并分析和总结了在全-强风化板岩、砂卵石地层中通过注浆技术堵水防渗,从而保证隧道开挖安全。     

基于模糊正交试验的纤维混凝土纤维参数优选

该文以钢纤维体积率、长径比、外形和聚丙烯纤维体积率为因子,安排正交试验研究了混杂纤维各因素对高性能混凝土抗压强度和劈裂抗拉强度的影响,并利用模糊数学理论对正交试验结果进行了分析,得到了混杂纤维混凝土的较优配比。结果表明:纤维各物理参数对纤维高性能混凝土综合性能的影响程度从大到小依次为:钢纤维外形、聚丙烯体积率、钢纤维体积率、钢纤维长径比;基准混凝土中加入长径比为70、体积率为1.5%的端钩形钢纤维和体积率为0.055%的聚丙烯纤维混杂时,对高性能混凝土增强和增韧效果最优。     

验证荷载条件下既有梁桥可靠度分析

分析了抗力服从不同先验分布时其后验分布的数字特征变化规律,建立了验证荷载条件下的抗力概率模型;结合条件概率的方法,建立了验证荷载条件下的结构可靠指标计算公式.并从验证荷载前后结构功能函数相关系数、验证荷载取值等方面对一实例桥梁进行了可靠度分析,同时研究了恒载参数、活载参数对考虑验证荷载的既有桥梁结构可靠度的影响.结果表...     

基于Advisor的纯电动汽车整车性能参数化分析

提高纯电动汽车整车动力性能和续驶里程,电池和电机的选用非常关键。文章介绍了纯电动汽车结构;对比分析了在纯电动汽车上常用的电机和电池类型及性能参数;在Advisor软件中分别针对电机和电池的参数对续驶里程和动力性的影响进行仿真,得出了其对整车性能的影响;进一步推导出了电池电量与续驶里程的函数关系,为实际应用中选用电池类型．核算电量成本等提供了经验公式。