基于人工神经网络的混合梁斜拉桥智能诊断方法研究

研究目的:本文以天津市河北大街混合梁斜拉桥为工程背景,基于人工神经网络模型,提出适用于混合梁斜拉桥的分步识别方法,分别采用概率和径向基函数神经网络对子结构和钢主梁子结构局部构件进行损伤识别.此外还提出适用于钢主梁局部构件识别的动-静组合损伤指标,并建立相应的径向基函数网络模型,分别针对单损伤、双损伤和三损伤的不同损伤情况进行数值模拟.研究结论:识别结果表明:(1)本文所提出的分步识别方法具有较高的识别精度,网络识别速度快,适用于大型混合梁斜拉桥的智能诊断过程;(2)所提出的动-静组合损伤指标对混合梁斜拉桥的局部损伤识别也较为敏感;(3)单处损伤测试工况中,识别精度几乎高达100％;(4)在两处和三处损伤测试工况中,位置识别正确率分别达到82.61％和78.3％.     

动态交通流量预测方法研究

随着智能运输系统的广泛应用,实时交通流量预测的重要性也日益显著。本文介绍了预测模型发展过程中比较重要的几个模型,并由此引出人工神经网络。介绍误差逆传播(BP)模型的相关理论。指出传统BP神经网络的缺陷,并提出提高预测精度的措施引进高阶神经网络。建立普通BP神经网络的预测模型,利用误差反传播算法实现这些影响因素到输出变量的复杂映射,再用高阶神经网络构建另一预测模型。利用交叉口实测数据进行预测,并用实际数据进行比较验证。     

基于BP神经网络的旅游交通安全评价分析研究

阐述了BP神经网络的基本理论、算法、缺陷及改进方法，通过实例对旅癖交通安全进行研究。选择合适的学习样本及参数进行训练，并获得较好的评价结果。为游客的旅游安全及相关部门的管理者提供了理论依据。     

公路隧道交通量的预测及分析

公路隧道交通量的预测对隧道通风系统的节能以及降低隧道运营成本有很重要的意义。分别利用多元统计分析法和BP神经网络两种方法对公路隧道交通量进行了预测,并对两类预测数据进行分析比较,得出了多元统计分析法适用于车流量少而且稳定的公路隧道的预测,而BP神经网络法则适用于车流量大而且不稳定的公路隧道的预测的结论。     

Predicting asphalt pavement crack initiation following rehabilitation treatments

Prolongation of the service life of pavements requires efficient prediction of the performance of their structural condition and particularly the occurrence and propagation of cracking of the asphalt layer. Although pavement performance prediction has been extensively investigated in the past, models for predicting the cracking probability and for quantifying impacts of associated explanatory factors following pavement treatment, have not been adequately investigated in the past. In this paper the probability of alligator crack initiation following pavement treatments is modeled with the use of genetically optimized Neural Networks, The proposed methodological approach represents the actual (observed) relationships between of probability of crack initiation and the various design, traffic and weather factors as well as the different rehabilitation strategies. Data from the Long Term Pavement Performance (LTPP) Data Base and the Specific Pavement Study 5 (SPS-5) are used for model development. Results indicate that the proposed approach results in accurately predicting the probability of crack initiation following treatment; furthermore it provided information on the relationship between external factors and cracking probability that can help pavement managers in developing appropriate rehabilitation strategies.     

Travel mode estimation for multi-modal journey planner

For route planning and tracking, it is sometimes necessary to know if the user is walking or using some other mode of transport. In most cases, the GPS data can be acquired from the user device. It is possible to estimate user’s transportation mode based on a GPS trace at a sampling rate of once per minute. There has been little prior work on the selection of a set of features from a large number of proposed features, especially for sparse GPS data. This article considers characteristics of distribution, auto- and cross-correlations, and spectral features of speed and acceleration as possible features, and presents an approach to selecting the most significant, non-correlating features from among those. Both speed and acceleration are inferred from changes in location and time between data points. Using GPS traces of buses in the city of Tampere, and of walking, biking and driving from the OpenStreetMap and Microsoft GeoLife projects, spectral bins were found to be among the most significant non-correlating features for differentiating between walking, bicycle, bus and driving, and were used to train classifiers with a fair accuracy. Auto- and cross-correlations, kurtoses and skewnesses were found to be of no use in the classification task. Useful features were found to have a fairly large (>0.4) correlation with each other.     

轨道交通站点购票客流预测及优化技术研究

将地铁中高峰时段购票客流看成排队论系统,结合排队论知识进行数据优化,利用MATLAB编制BP神经网络模型的程序,建立仿真并预测站点购票客流及自动售票机的需求量,从而达到应急售票服务的优化,提升乘客服务满意度.并以某市大客流站点的客流数据进行论证.     

隔挡式背斜构造区隧道涌突水量的BP网络预测

川东隔挡式背斜区具有其特殊的岩溶地质构造,已修建的数十条隧道每每揭露背斜区,均发生较大地下水涌突水灾害,对这种特殊构造下涌突水灾害的研究具有重要现实意义.针对岩溶地下水系统具有强烈的非线性特征,建立合适的BP神经网络,评价某在建公路华蓥山隧道的涌突水灾害危险等级.结果显示,背斜两翼非可溶岩层等级为Ⅰ～Ⅲ级;核部可溶岩地层为Ⅲ～Ⅴ级,且越靠近核部危险性等级越高;西翼涌突水危险性等级高于东翼.评价结果与勘察阶段的研究相互印证.     

基于Prescan 的信号灯路口跟停场景虚拟重构

为满足我国智能驾驶汽车测试场景库的搭建和ADAS(高级驾驶辅助系统)功能研发和验证的需求,设计了一种基于Prescan的交通信号灯路口车辆跟停场景虚拟重构方法,该方法由道路环境建设模块、初始条件设定模块和车辆控制模块组成。道路环境建设模块通过输入道路参数信息构建虚拟道路,初始条件设定模块通过输入本车和目标车的初始位置、初始速度信息确定零时刻车辆和道路的空间位置及状态信息,车辆控制模块依据车辆速度位置等信息,利用训练的神经网络控制本车加速度,实现跟停场景的虚拟重构。仿真结果表明,该方法可以实现交通信号灯路口车辆跟停场景的虚拟重构。     

基于神经网络的智能驾驶模式识别研究

为满足智能驾驶汽车高级驾驶辅助系统(ADAS)功能研发和验证的需求,提高ADAS功能的准确性,设计了一款基于神经网络的智能驾驶模式识别程序,该程序由数据采集、目标检测、场景识别预测3个模块组成。数据采集模块利用ESR毫米波雷达、前置摄像头对交通环境及周围车辆的数据信息进行采集;目标监测模块通过控制算法选择判断触发各类ADAS功能场景的最可疑目标;场景识别处理模块以汽车制造商提供的大量自然驾驶数据的场景挖掘结果为依据,利用神经网络学习各类ADAS场景的特征行为,并通过约束条件对各类ADAS功能场景的识别结果进行实时判定。通过开放道路试验进行验证,结果表明,该程序的场景识别结果准确率可达到99.86%。