干道信号协调控制优化方法

针对干道信号协调控制中控制策略及其控制优化参数设置上存在的问题,考虑到智能控制中Sugeno模糊推理所具有的复杂系统动态性能表达能力和神经网络控制方法所具有的学习能力,结合两者的优势建立了1种基于模糊神经网络的交通信号协调控制模型,实现对绿信比与相位差的优化,使用微观交通仿真软件Vissim对干线交通进行了仿真研究,仿真结果表明,该方法能更为有效地减小平均车辆延误。     

平台工作船舱室噪声预报分析

基于统计能量法(SEA),采用VA One软件,建立某85 m平台工作船的SEA模型,进行不同工况下的舱室噪声预测,并与实测值进行比较。对比分析平台工作船在航行工况下不同主机负荷时的舱室噪声规律以及动力定位(DP)工况下侧推桨45%负荷时的舱室噪声规律;对比不同激励对舱室噪声的影响,并分析舱室噪声预测时应考虑的一些因素。通过对比不同工况下的噪声预测值与实测值发现,艏侧推桨是平台工作船最重要的噪声源,尤其是在侧推桨出现空泡情况下,噪声超标严重。分析结果可为该类型船舶的降噪措施提供参考。     

Distribution Laws of Earth Pressure Acting on a Shield Tunnel Based on Measured Data北大核心CSCD

The earth pressure acting on a shield tunnel is intrinsically related to synchronous grouting, the position of the shield machines, the rigidity of the segment and stratum, etc. Existing theoretical formulas and empirical methods couldn′t exactly reflect the actual state of the segment of a shield tunnel during construction due to the com⁃plexities mentioned above. Based on earth pressure data measured at 52 monitoring sections in 35 tunnels, the distri⁃bution laws and influential factors are analyzed. The results show that: 1) the earth pressure scope is 0 to 400 kPa for over 90% of the total samples, but is recommended to be 0.3 to 0.4 MPa for the empirical grouting pressure; 2)the earth pressure on the segment and the buried depth approximately have an exponential relationship, with the dif⁃ference between the max. earth pressure and the stable earth pressure decreasing with an increase of the buried depth; 3) the lateral pressure coefficient range is 0.5 to 2.3, which is partially out of the scope of recommended values(0.5 to 1.0) of surrounding rock of grade Ⅵ, so it is not appropriate to follow the recommended values in the specifi⁃cation; 4) the relationship of the earth pressure and the segment/soil stiffness ratio presents a quadratic function, and the most reasonable value for the stiffness ratio of the segment/soil is 1.0; and 5) there are four typical stages of the spatial distribution of earth pressure on a segment in clay stratum—the erection stage, the grouting influence stage,the consolidation contraction stage and the recovery stage, with the distribution of the circumferential earth pressure being asymmetrical. The distribution curves of the earth and water pressure in sand and soil layers are classified as a“weak decay pulse fluctuation”and“double hump,”respectively. After stabilizing, the circumferential earth pressure is notably in a symmetrical distribution, and the water pressure exhibits a light bulb shaped distribution (large at the lower part and small at the upper part). These conclusions provide a reference for studying the mechanism of earth pressure functions and perfecting the design of segments for shield tunnelling. © 2018, Editorial Office of "Modern Tunnelling Technology". All right reserved.     

基于Bayesian的舰船战场损伤评估模型研究

针对舰船战场损伤评估的多元信息特点,建立了在观测操作条件下的舰船战场损伤评估的贝叶斯网络推理算法.分析了舰船战场损伤过程的不确定性,在所建立的贝叶斯网络模型的基础上,提出运用删减法来降低网络的复杂性.实例表明了该算法的适用性、有效性和可操作性.     

基于模糊推理的区域路网交通状态分析方法

为准确判别区域路网的交通状态,对区域路网交通流的宏观特性进行了详细分析,建立了路网交通状态评估指标体系。结合区域路网的拓扑结构与交通流特征,基于模糊推理技术提出了路网交通状态分析方法。以城市主干道路网为例,验证了路网交通状态分析方法的有效性。该方法可应用于在线交通状态分析和历史数据库交通运行特征的提取,为交通管理决策提供基础信息。     

基于白适应神经模糊推理系统的跟驰模型研究

由于驾驶行为的不确定性,难以建立精确的车辆跟驰模型。针对这一问题,应用自适应模糊神经推理系统（ANFIS）建立跟驰模型,以跟随车与前车速度差及行车间距为输入量、跟随车的加速度为输出量,建立25条模糊推理规则,将模糊推理规则产生的数据作为车辆跟驰ANFIS模型的训练数据,并利用MATLAB编程对其进行训练。最后,设计了基于车载高精度GPS的跟驰试验,并结合试验数据分别对自适应模糊神经推理系统跟驰模型和传统跟驰模型进行仿真。结果表明,前者输出的跟驰车辆加速度值更接近于真实值。     

舰炮交流伺服系统模糊滑模控制器设计

该文针对舰炮永磁同步电机交流伺服系统 ,设计了一种新型滑模控制器 ,以前馈信号作为平均控制 ,采用模糊推理来调节开关控制的幅度。仿真研究表明 ,该控制器能较好地实现对指令信号的跟踪 ,具有较强的抗参数摄动和抗干扰能力 ,表现出良好的控制性能。     

Bayesian committee of neural networks to predict travel times with confidence intervals

Short-term prediction of travel time is one of the central topics in current transportation research and practice. Among the more successful travel time prediction approaches are neural networks and combined prediction models (a ‘committee’). However, both approaches have disadvantages. Usually many candidate neural networks are trained and the best performing one is selected. However, it is difficult and arbitrary to select the optimal network. In committee approaches a principled and mathematically sound framework to combine travel time predictions is lacking. This paper overcomes the drawbacks of both approaches by combining neural networks in a committee using Bayesian inference theory. An ‘evidence’ factor can be calculated for each model, which can be used as a stopping criterion during training, and as a tool to select and combine different neural networks. Along with higher prediction accuracy, this approach allows for accurate estimation of confidence intervals for the predictions. When comparing the committee predictions to single neural network predictions on the A12 motorway in the Netherlands it is concluded that the approach indeed leads to improved travel time prediction accuracy.     

基于自适应神经模糊推理系统的平行泊车路径规划

针对常见的平行泊车场景,提出一种基于自适应神经模糊推理系统的平行泊车路径规划方法.以基于优化算法的泊车路径规划方法得出的泊车路径作为训练样本,利用Python脚本语言建立以自适应遗传算法和拟牛顿法为内核的自动化训练框架,使自动训练后的自适应神经模糊推理系统既可继承基于优化算法的泊车路径规划方法适用范围更广的优势,又有效...     

���ڶ�������Ķ�Ӧ�������ź����ȿ����о�

本文对不同进口方向的多辆应急车辆在同一时段内通过同一信号交叉口的协调控制问题进行研究,提出一种基于多智能体的多应急车辆信号优先控制系统.在该系统中引入了相位智能体和管理智能体,同时采用模糊推理理论实现各类智能体的内在逻辑,以及模块之间的协调机制.基于Starlogo多智能体仿真软件,针对特定十字交叉口的多应急车辆优先信号协调控制案例进行仿真实验.结果表明,多应急车辆信号优先控制策略能实现应急车辆在交叉口时间通行权上的优先,同时能够减少应急车辆对其他车辆的干扰,对于提高城市范围内的紧急救援效率,具有一定的实用价值.