基于随机森林结合前向序列选择算法的铁路隧道防水板检测数据关键影响因素识别北大核心

针对参数特征复杂度高的铁路产品,提出了一种基于随机森林(RF)结合前向序列选择(SFS)算法的铁路产品检验检测数据关键影响因素识别方法,以辅助基于经验的识别方法。创新使用RF-SFS算法,将其应用于铁路隧道防水板检验检测数据关键影响因素的识别研究。根据多年铁路隧道防水板检测数据,建立RF模型,获得了影响铁路隧道防水板检测结果的特征关键性评分序列。随后,结合SFS方法得出关键性评分序列的阈值,将排名前6位的影响因素识别为关键特征,模型的预测能力达到99.98%。为验证关键特征识别方法的有效性,对比分析3种模型在使用不同特征子集时的预测能力。当仅选用关键特征时,3种模型的预测能力均达到最佳,加入冗余特征后模型的预测能力逐渐降低。     

公共自行车使用时空特性挖掘及租还需求预测

基于宁波市公共自行车刷卡数据、POI(Point of Interest)数据、气象和空气质量等数据，从数据驱动视角，深入挖掘公共自行车使用的时空特征及站点租还车需求预测。在时间上，采用KMeans算法，将站点聚为5类，探讨各类站点的时变需求规律及影响因素；在空间上，提出基于POI 数据的站点用地类型识别方法，将站点分为居住类、交通设施类、办公类和商业休闲类。构建以 15，30，60 min 为间隔，以租还车需求为目标变量的随机森林预测模型，并与常用的 BP (Back Propagation)神经网络、K最近邻方法进行比较。结果表明，随机森林模型的精度更高，适用性更强。以30 min为间隔的站点租还车需求预测精度最高，考虑站点土地利用类型后能有效提高模型的预测精度。本文结果可作为未来站点平衡调度的依据并推广应用于共享单车系统，为改善服务水平提供技术和理论支撑。     

Real-time trip purpose prediction using online location-based search and discovery services

The use of smartphone technology is increasingly considered a state-of-the-art practice in travel data collection. Researchers have investigated various methods to automatically predict trip characteristics based upon locational and other smartphone sensing data. Of the trip characteristics being studied, trip purpose prediction has received relatively less attention. This research develops trip purpose prediction models based upon online location-based search and discovery services (specifically, Google Places API) and a limited set of trip data that are usually available upon the completion of the trip. The models have the potential to be integrated with smartphone technology to produce real-time trip purpose prediction. We use a recent, large-scale travel behavior survey that is augmented by downloaded Google Places information on each trip destination to develop and validate the models. Two statistical and machine learning prediction approaches are used, including nested logit and random forest methods. Both sets of models show that Google Places information is a useful predictor of trip purpose in situations where activity- and person-related information is uncollectable, missing, or unreliable. Even when activity- and person-related information is available, incorporating Google Places information provides incremental improvements in trip purpose prediction.     

海堤内坡种植防护林的安全影响分析及尺度论证

海堤从外海堤脚至内侧护堤之间的滩涂围垦空间,既是陆地边缘,也是水的边缘。外侧可构建滨海景观,内侧结合随塘河形成亲水景观,具有得天独厚的优势。内侧平台的防护林及景观有机结合,可构建人们休闲旅游的"黄金绿带"。结合台州某海堤生态提升工程,提出海堤生态提升总体技术方案,重点对海堤内坡种植景观防护林对海堤安全的影响进行分析,明确种植平台高程、尺度布置等主要技术参数,为类似海堤生态改造工程提供参考。     

Physical–biological coupling in spore dispersal of kelp forest macroalgae

The physical–biological linkages controlling the dispersal of spores produced by macroalgae that reside in kelp forests are complicated and laced with feedbacks. Here we discuss the fundamental elements of these interactions. Biological considerations include spore swimming and sinking speeds, their periods of viability in the plankton, and the height of spore release above the seafloor, which together determine the durations over which spores can be swept by horizontal currents before they contact the seafloor. Morphologies and material properties of canopy forming kelps may also influence the drag exerted on passing waters by the kelps, the plants' ability to persist in the face of rapid flows, and thereby the degree to which impinging currents are redirected around, or slowed within, kelp forests. Macroalgal life histories, and the size of spore sources as controlled by the dimensions of kelp forests and the density and fecundity of individuals within them, influence effective dispersal distances as well. Physical considerations encompass the mean speed, direction, and timescales of variability of currents relative to spore suspension times, the interaction of surface gravity waves with currents in producing turbulence in the benthic boundary layer, wind-driven surface mixing, water stratification, and shoreline bathymetry and substratum roughness, all of which can affect the interplay of vertical and horizontal transport of macroalgal spores. Intricate within-forest processes may induce attenuation of current speeds and consequent reductions in seabed shear, along with simultaneous production of small-scale turbulence in kelp wakes. Slower mean currents and smaller eddy scales in turn may attenuate vertical mixing within forests, thus extending spore suspension times. Further complexities likely arise due to changes in the relative rates of horizontal and vertical dispersion, modifications to the overall profiles of vertical mixing, and the creation of fine-scale secondary flows around kelp individuals and substratum features. Under conditions of more rapid currents, submergence of the surface canopy and the establishment of skimming flows at the canopy–fluid interface may introduce additional coherent flow structures that alter rates of fluid exchange to and from the forest. Many of these coupled physical–biological processes are just beginning to be examined in a rigorous fashion in kelp forests, but their potential importance is clear.     

The shoaling of nutrient-enriched subsurface waters as a mechanism to sustain primary productivity off Central Baja California during El Niño winters

Using CalCOFI data for coastal shallow stations (above 100 m depth), higher than expected nitrate concentrations were detected in near-surface high-temperature waters off of Central Baja California during some El Niño winters. Though recent data are not available for Central Baja California, past El Niño data, though limited, showed nitrate concentrations above 16 μM at temperatures above 16 °C, and nitrate concentrations between 1 and 2 μM at 19 °C, while the previously established relationship of temperature and nitrate for California Current waters predicts nitrate depletion above 14 or 15 °C. The anomalous, high temperature–high nitrate enrichment events documented in Central Baja California were detected as shallow as 9 m and as deep as 73 m, were associated with low-oxygen (between 2 and 4 ml/l) and high-salinity (between 33.8 and 34.3 psu) waters, and occurred during the winter months of an El Niño year. Using recent data for San Diego, CA, similar but weaker enrichment events were detected for the El Niño winter of 1997–1998. The periodic shoaling of a subsurface subtropical water mass of high temperature, high salinity, low oxygen and high nutrients during some El Niño winters is proposed to cause periodic enrichment and to maintain productivity during warming events in this area. Enrichment events were not detected off Ensenada, in Northern Baja California, possibly due to the amplification of the onshore flow during El Niño there, or due to the Ensenada front. The proposed mechanism of periodic enrichment of nutrient-depleted surface waters during some El Niño winters by subsurface waters from the California Undercurrent may explain the following: (1) survival of giant kelp forests at their southern limit in Central Baja California documented during past El Niño events in warm waters, (2) the rapid recovery and high carrying capacity of giant kelp documented after the mass disappearance during El Niño 1997–1998, and (3) the increase in the extent of mesotrophic chlorophyll detected in the area during the 1997–1998 and 1982–1983 El Niño events.     

基于车辆运行数据的疲劳驾驶状态检测

疲劳驾驶是导致交通事故的主要原因之一，及时检测疲劳驾驶，并提醒驾驶员集中注意力，对保证安全行车具有重要意义.本文基于CAN(Controller Area Network)总线采集的车辆运行状态数据，提取了18项与驾驶行为相关的特征，并采用随机森林算法对疲劳驾驶进行识别，结果表明整体的识别准确率为0.785，其中召回率为0.61，即61%的疲劳驾驶状态可被识别出来.实验表明，基于车辆运行状态的疲劳驾驶检测具有一定的效果，且与其他客观的疲劳驾驶检测方法(基于驾驶员生理指标和图像面部特征)相比，具有简单方便，不影响驾驶，且成本低的优势.     

退耕还林政策及可持续性的经济学分析

依据产权理论，本文提出了资产价值特征概念，以分析树木的资产价值特征为基础，分析了政府和个体面对退耕还林的理性选择，阐述了退耕还林制度出台的内在逻辑及实质。在此基础之上本文还通过分析双方的行为，对退耕还林的可持续性进行了思考，并提出了相应的政策建议。     

阿里山森林铁路屏遮那隧道施工特色与技术难题

阿里山森林铁路为世界闻名的高山铁路之一,惟受2009年莫拉克台风重创,森林铁路60k+550~61k+450二万坪路段发生大规模崩塌,路基严重流失,致使铁路停驶至今。为使森林铁路早日复驶,并能永续经营,采用费用较高、工期较长的迂回隧道方式进行修复。本案例无任何道路可供通达,施工前须辟建高难度的施工便道,以供后续人员、材料、机具进出; 台风、豪雨期间,崩塌区落石及土石流不断,施工便道维护及人员避难不易; 洞口腹地狭窄、崩积层进洞困难及林地生态保护等施工特色与技术难题,更非原规划设计足以因应。经业主、监造及施工团队数次应变处理后,勉以通过,其经验与心得可供后续类似案例参考。     

基于多源数据的公交专用道效能评价方法与影响模型

施画公交专用道是落实公交优先策略、改善公交服务质量,提升交通系统效率的重要策略.近年来,如何提高公交专用道设置的科学性、提升服务效能受到广泛关注.为科学定量评价公交专用道设置的正、负效能,建立基于多维运行影响的公交专用道效能评价指标体系,提出基于主观层次分析法和客观离差最大化法相结合的组合评价方法,实现对公交专用道效能...