Multiple mental tasks classification based on nonlinear parameter of mean period using support vector machines

Mental tusk classification is one of the most important problems in Brain-computer interface. This paper studies the classification of five-class mental tusks. The nonlinear parameter of mean period obtained from frequency domain information was used as features for classification implemented by using the method of SVM （support vector machines）. The averaged classification accuracy of 85. 6% over 7 subjects was achieved for 2-second EEG segments. And the results for EEG segments of 0.5s and 5.0s compared favorably to those of Garrett＇s. The results indicate that the parameter of mean period represents mental tusks well for classification, Furthermore, the method of mean period is less computationally demanding, which indicates its potential use for online BCI systems.     

基于轨迹数据的共享电动自行车逆行风险行为影响因素研究

为改善电动自行车带来的交通安全问题,研究逆行风险行为与其影响因素间的相关关系。基于长沙市芙蓉区共享电动自行车GPS轨迹数据,实现逆行行为的精准识别,采用机器学习CatBoost模型与SHAP可解释机器学习框架,从道路条件、交通状态、土地利用性质等方面开展逆行行为影响要素挖掘及作用解析。研究结果表明：CatBoost模型能够有效预测路段逆行频次并提取逆行行为的重要影响因素,主要包括出行时段、公共交通设施、土地利用性质、道路条件及交通状态等;从出行时段来看,工作日、早晚高峰时段更容易发生逆行;从公共交通设施与土地利用性质来看,道路周围公交站地铁站出口数量及餐饮、公司、购物等设施数量与逆行频次呈现非线性影响关系,在一定范围内设施数量与逆行行为存在正影响作用;从道路条件来看,过街通道间距在50~400 m时不易发生逆行,在非机动车道无物理隔离设施或过街通道间距在400~600 m时容易发生逆行,间距大于600 m时作用不稳定;从路段机非分隔形式来看,护栏分隔的逆行概率较低,绿化带分隔的逆行概率较高;从交通状态来看,当骑行速度、加速度较低或较高时与逆行行为负相关,当骑行速度在6~16 km·h^-1及加速度在0.3~1.0 m·s^-2时与逆行行为正相关。研究成果可为共享电动自行车风险骑行行为辨识、非机动车交通安全管理提供有效的技术支持。     

Buderus组合磨床磨削质量问题的解决

针对ｕｄｅｒｕｓ组合磨床投产一段时期后，齿轮加工质量不稳定的状况，对机床主轴精度，夹具，砂轮的安装、平衡、修整及冷却进行了检查与调整；并对内孔磨削参数和面 改进；从而使锥面圆度、内孔圆柱度得以稳定，且在公差范围内。     

富水砂卵石地层矩形顶管机的研究及应用——结合成都川大下穿人民南路人行通道工程

以成都川大下穿人民南路人行通道矩形顶管项目为依托,对富水砂卵石地层矩形顶管机关键技术进行研究。1)为探索同平面多刀盘联合开挖矩形断面技术,采用ANSYS Workbench有限元仿真分析方法对刀盘结构进行优化,得出:刀盘采用箱体式4辐条结构外加大圆环支撑形式,可满足强度需求的同时有效提高刀盘开口率。2)研究针对大粒径卵石的双螺旋输送机排渣技术,利用ANSYS仿真分析对螺旋叶片厚度进行优化,通过工业性试验表明:大直径、大节距螺旋叶片设计可有效提高卵石地层的排渣效果,实现450 mm粒径卵石的顺利排出。3)采用正交试验对掘进过程中渣土改良工艺进行研究,获得适应该地层的渣土改良方案。     

A real-time posture monitoring method for rail vehicle bodies based on machine vision

Monitoring vehicle operation conditions has become significantly important in modern high-speed railway systems. However, the operational impact of monitoring the roll angle of vehicle bodies has principally been limited to tilting trains, while few studies have focused on monitoring the running posture of vehicle bodies during operation. We propose a real-time posture monitoring method to fulfil real-time monitoring requirements, by taking rail surfaces and centrelines as detection references. In realising the proposed method, we built a mathematical computational model based on space coordinate transformations to calculate attitude angles of vehicles in operation and vertical and lateral vibration displacements of single measuring points. Moreover, comparison and verification of reliability between system and field results were conducted. Results show that monitoring of the roll angles of car bodies obtained through the system exhibit variation trends similar to those converted from the dynamic deflection of bogie secondary air springs. The monitoring results of two identical conditions were basically the same, highlighting repeatability and good monitoring accuracy. Therefore, our monitoring results were reliable in reflecting posture changes in running railway vehicles.     

Data-driven train set crash dynamics simulation

Traditional finite element (FE) methods are arguably expensive in computation/simulation of the train crash. High computational cost limits their direct applications in investigating dynamic behaviours of an entire train set for crashworthiness design and structural optimisation. On the contrary, multi-body modelling is widely used because of its low computational cost with the trade-off in accuracy. In this study, a data-driven train crash modelling method is proposed to improve the performance of a multi-body dynamics simulation of train set crash without increasing the computational burden. This is achieved by the parallel random forest algorithm, which is a machine learning approach that extracts useful patterns of force–displacement curves and predicts a force–displacement relation in a given collision condition from a collection of offline FE simulation data on various collision conditions, namely different crash velocities in our analysis. Using the FE simulation results as a benchmark, we compared our method with traditional multi-body modelling methods and the result shows that our data-driven method improves the accuracy over traditional multi-body models in train crash simulation and runs at the same level of efficiency.     

基于多尺度核独立元分析与核极限学习机的柴油机故障诊断

为提高柴油机故障诊断速度和精度,提出了基于改进多尺度核独立元分析与量子粒子群优化核极限学习机的故障诊断方法。首先利用固有时间尺度分解对缸盖振动信号进行多尺度时频分解,并根据故障敏感度参数筛选有效分量以实现振动冲击特征增强;然后利用核独立元分析消除有效分量间的频带混叠,分离故障敏感频带,并提取各频带的AR模型参数、多尺度模糊熵和标准化能量矩构造联合故障特征向量;最后建立基于量子粒子群优化的核极限学习分类器实现柴油机故障诊断。试验结果表明,该方法有效增强了缸盖振动信号中的故障敏感特征,提高了柴油机故障诊断速度和精度,故障分类准确率达到98.45%。     

基于机器学习的地铁轨道几何劣化规律个性化建模研究

较高的轨道平顺性是保障地铁列车安全舒适运行的基础,准确掌握地铁轨道的劣化规律对保障轨道质量具有重要意义。根据地铁线路特点,选择影响地铁轨道质量劣化的7类异质性因素,给出赋值模型,并基于机器学习方法建立轨道质量指数(track quality index,TQI)短时预测前馈神经网络模型。为了验证模型,采集了北京地铁1号线的线路设备数据及2016年8月15日至2019年2月18日间的17次TQI检测数据,形成训练数据集和测试数据集,并采取深度学习技术,利用训练数据集对该模型进行训练。基于测试数据集的模型预测值的可决系数为0.938,平均绝对百分比误差为4.80%,结果表明该模型是有效的且具有较高的预测精度。     

直流电动转辙机动作电流的智能分析

微机监测系统可以采集并记录道岔动作过程电流的变化情况。人工分析时可以根据波形特点定性判断道岔状态。为实现定量分析,提出了分3个阶段分析电流波形的基本方法。介绍了计算机进行智能分析的基本分析方法和流程,强调了智能分析的重要性。这是信号专家系统关于道岔状态分析的方案。     

京沪高速铁路废弃泥浆处理现场试验

结合京沪高速铁路工程现场实际情况,提出了在泥浆池中进行混凝沉淀固液分离＋排出上清液＋回填的废弃泥浆现场处理工艺流程;利用研制的废弃泥浆处理装置在京沪高速工程现场进行了试验,表明在泥浆池中处理废弃泥浆的技术路线是可行的,研制开发的废弃泥浆处理装置满足设计要求,可用来处理废弃泥浆。