Improving rail network velocity: A machine learning approach to predictive maintenance

Rail network velocity is defined as system-wide average speed of line-haul movement between terminals. To accommodate increased service demand and load on rail networks, increase in network velocity, without compromising safety, is required. Among many determinants of overall network velocity, a key driver is service interruption, including lowered operating speed due to track/train condition and delays caused by derailments. Railroads have put significant infrastructure and inspection programs in place to avoid service interruptions. One of the key measures is an extensive network of wayside mechanical condition detectors (temperature, strain, vision, infrared, weight, impact, etc.) that monitor the rolling-stock as it passes by. The detectors are designed to alert for conditions that either violate regulations set by governmental rail safety agencies or deteriorating rolling-stock conditions as determined by the railroad.Using huge volumes of historical detector data, in combination with failure data, maintenance action data, inspection schedule data, train type data and weather data, we are exploring several analytical approaches including, correlation analysis, causal analysis, time series analysis and machine learning techniques to automatically learn rules and build failure prediction models. These models will be applied against both historical and real-time data to predict conditions leading to failure in the future, thus avoiding service interruptions and increasing network velocity. Additionally, the analytics and models can also be used for detecting root cause of several failure modes and wear rate of components, which, while do not directly address network velocity, can be proactively used by maintenance organizations to optimize trade-offs related to maintenance schedule, costs and shop capacity. As part of our effort, we explore several avenues to machine learning techniques including distributed learning and hierarchical analytical approaches.     

Relationship between condition and recruitment success of red shrimp (Aristeus antennatus) in the Balearic Sea (Northwestern Mediterranean)

This study evaluates the reproductive potential and condition of the red shrimp (Aristeus antennatus) population inhabiting the waters around the Balearic Islands (Northwestern Mediterranean Sea), between 1991 and 2004. Red shrimp is one of the most valuable exploited demersal fishery resources in the western Mediterranean. To assess the condition of this species, we use two indices based on weight and length data, one including the gonad weight and another excluding it. Different biological parameters of the population dynamics and reproductive biology, such as sex-ratio, maturity, gonadosomatic index and presence of the spermatophore in females have also been analysed. A negative relationship was found between the gonadosomatic index and the condition of red shrimp, indicating that energy reserves are transferred from the body to the gonad during the reproductive period. The condition of adults reached minimum values during the maturation and spawning period after mating, when the gonadosomatic index, the spermatophore presence in females and the proportion of females in the population were highest. The relationship between the condition of adults during the months prior to spawning and the number of recruits in the following year was significant and positive. This relationship was stronger when only male condition was considered, suggesting that males have an important role on the reproductive potential of this species. Overall, our results suggest that condition of red shrimp, particularly males, is an important aspect for the reproductive and recruitment success of this species. The observed decreasing trend in male condition over years may raise concern on the future reproductive potential of that population.     

Seasonal and short spatial patterns in European hake (Merluccius merluccius L.) recruitment process at the Balearic Islands (western Mediterranean): The role of environment on distribution and condition

This study evaluates the link between the recruitment process of European hake (Merluccius merluccius L.) of the Balearic Islands (western Mediterranean) and the environmental and physiological conditions. Spatio-temporal variation of abundance and condition of fish were evaluated at two locations each with different oceanographic conditions, one in the north (Sóller, SO) and another in the south (Cabrera, CA) of Mallorca Island. Environmental variables explored were hydrography, sediment characteristics, phytoplankton pigment concentration (ppc) and the trophic resources of hake. Individuals were divided in three life stages: recruits, post-recruits and young adults. Hepatosomatic index (HSI), relative condition index (K_n), gonadosomatic index (GSI) and fullness index (FI) were analysed for the three life stages. Recruitment starts in February with the incorporation of smaller hakes, and it can be followed through spring and early summer with a peak in April. However, some spatial heterogeneity in the recruitment process has been found between north and south of the Island. The main pulse of recruitment occurred at a different time in the two areas. Spatial heterogeneity was also consistent with the condition of hake recruits, with higher values of K_n and HSI at SO than at CA. Maximum values of K_n were found in February at SO and in April at CA, coinciding with the start of the different recruitment pulses to the fishing grounds. Post-recruits and young adults also showed higher condition at SO than at CA. The arrival in spring of the Western Winter Intermediate Waters (WIW) drives the spatial-temporal variation in abundance and condition of hake. Ppc was highly correlated with recruit abundance with a time lag of two months, while for post-recruits the time lag was three months. The observed differences in the condition of hake between areas could be a consequence of the fact that the waters to the north of Mallorca are comparatively more under the seasonal influence of WIW which is formed in more productive areas. Thus, this study characterises the short temporal and spatial variability in the hake recruitment process off the Balearic Islands, both in terms of abundance and fish condition. This pattern is explained on the basis of the mesoscale environmental variability observed between north and south of Mallorca and the ecological adaptive strategy of recruiting in the optimal environmental season.     

新能源汽车电机驱动器母线电容综述

直流母线(DC Link)电容是新能源汽车电机驱动器的关键零部件之一,其可靠性会影响整个电驱动系统的可靠性和安全性。状态监测作为评估母线电容健康状态的常用方法之一,可及时发现母线电容的退化征兆或故障特征以避免故障发生。基于此,文章首先介绍了铝电解电容器的基本特性,其次针对目前电容器的状态监测方法进行分类,将其分为电容器纹波电流传感器法、电路模型法及智能监测法,介绍了每种方法的原理,对比分析了各类方法的优缺点,为后续电容器状态监测提供了研究方向。     

暂扣车辆有关问题的研究

文章针对当前道路运政执法现状,分析了运政执法人员暂扣车辆的条件、处理依据、期限及保管问题,对如何正确合法地暂扣车辆,提出了相应的意见和对策。     

基于SVM的桥梁状态监测方法

随着交通领域的迅速发展,保障桥梁安全、降低维护费用成为普遍关注的问题。以振动分析为基础、广泛借助信息技术的理论和方法,进行桥梁状态监测成为当前的研究热点。考虑到的桥梁损伤样本很难获得、个体差异使得损伤数据难以共享的实际问题,本文将桥梁状态监测归结为异常监测问题,引入基于SVM的一类学习算法从长期监测数据中获取正常状态的模式,实现异常状态的精确报警。文中采用香港汀九桥400小时实测数据,验证了这种方法的实际效果。     

基于BP神经网络的船用柴油机振动状态监测

通过监测柴油机表面振动信号,用时间序列分析方法提取柴油机故障的振动特征参数,以此建立相应的神经网络,用于船用柴油机的状态监测,提高诊断的准确性。试验研究在中速四冲程增压柴油机上进行。文中以柴油机气阀间隙异常的诊断和柴油机负荷状态的识别为例阐述了该方法的实现过程,并给出了振动信号的特征参数与柴油机工作状态之间的关系。研究表明,利用神经网络监测柴油机运行状态的变化是可行的和有效的。     

校内实训基地使用情况及存在问题分析

实训基地是学校对学生实施职业技能训练和职业素质培养的场所,它分为校内和校外两种。文章以问卷调查为基础,以《综合货代实训》课程为例,探讨了校内实训基地的使用情况,并对目前校内实训基地建设和使用中存在的主要问题进行分析。     

严肃执行《建筑法》,规范管理工程造价

不规范的承发包方式、竞价方式会给国家和社会带来多大危害,规范工程造价(报价)已刻不容缓,工程价值和造价相背离的现象不能再延续下去了。     

基于再次均匀采样策略改进轨道车辆二系悬挂系统参数估计方法

基于粒子滤波和卡尔曼滤波的轨道车辆二系悬挂系统参数估计方法存在因粒子枯竭而无法对系统参数变化进行监测的缺陷,所以引入再次均匀采样策略对其进行改进.采用轨道车辆系统横向动力学模型,建立轨道车辆二系悬挂系统的横向动态空间模型,并运用改进后的参数估计算法和Matlab软件对二系横向阻尼和抗蛇行阻尼2个系统参数进行仿真计算.结果表明:在无法获得轨道横向不平顺统计特性的情况下,参数估计方法仍具有良好的鲁棒性;在轨道车辆二系悬挂系统发生突发或老化故障的情况下,改进后的参数估计方法可以准确估计系统参数的变化,且误差值始终保持在10％以下,能够实现对系统状态的实时监测.