基于Kinect的地铁乘客不安全行为识别方法与实验*

为解决地铁视频监控技术对乘客不安全行为只记录不识别且较少考虑识别精确度的问题,提出1种基于Kinect传感器的高效识别方法。以Pelvis为向量起点和动作活动高频关节为终点构建识别特征向量;运用余弦定理获得标准动作与测试动作关节的最大角度差序列;以最大角度差为动作特征量建立相似度计算模型,运用动态时间规整算法（DTW）将初始结果转换为动作相似度。以相连关节法为对照组开展对比实验,结果表明:前者在抽烟、挥拳、挥手呼救等行为识别的准确度分别为91.7%,86.9%,89.2%,平均比对照组高4%以上,显著提高了地铁乘客不安全行为的识别率,可为地铁智能安全管控提供理论与技术依据。     

外界热辐射作用下木材热解和着火的预测模型

在突发的火灾过程中，可燃性固体材料的着火时间和热解特性是非常关键的参数。建立了预测木材在外界热辐射作用下热解和着火时间的数值模型，可以给出固体的质量损失速率、内部温度场随时间的变化，并根据模型计算得到的表面温度随时间的变化对着火时间进行预测。计算结果和实验测量值相符较好。     

Analysis of Daily Peaking and Run‐of‐River Operations with Flow Variability Metrics,Considering Subdaily to Seasonal Time Scales

Environmental flows are an important consideration in licensing hydropower projects as operational flow releases can result in adverse conditions for downstream ecological communities. Flow variability assessments have typically focused on pre‐ and post‐dam conditions using metrics based on daily averaged flow values. This study used subdaily and daily flow data to assess environmental flow response to changes in hydropower operations from daily peaking to run‐of‐river. An analysis tool was developed to quantify flow variability metrics and was applied to four hydropower projects. Significant differences were observed between operations at the 99% confidence level in the median flow values using hourly averaged flow datasets. Median daily rise and fall rates decreased on average 34.5 and 27.9%, respectively, whereas median hourly rise and fall rates decreased on average 50.1 and 50.6%, respectively. Differences in operational flow regimes were more pronounced in the hourly averaged flow datasets and less pronounced or nonexistent in the daily averaged flow datasets. These outcomes have implications for the development of ecology‐flow relationships that quantify effects of flow on processes such as fish stranding and displacement, along with habitat stability. Results indicate that flow variability statistics should be quantified using subdaily datasets to accurately represent the nature of hydropower operations, especially for daily peaking facilities.     

A Hydrologic Data Screening Procedure for Exploring Monotonic Trends and Shifts in Rainfall and Runoff Patterns

A thorough understanding of past and present hydrologic responses to changes in precipitation patterns is crucial for predicting future conditions. The main objectives of this study were to determine temporal changes in rainfall‐runoff relationship and to identify significant trends and abrupt shifts in rainfall and runoff time series. Ninety‐year rainfall and runoff time series datasets from the Gasconade and Meramec watersheds in east‐central Missouri were used to develop data screening procedure to assess changes in the rainfall and runoff temporal patterns. A statistically significant change in mean and variance was detected in 1980 in the rainfall and runoff time series within both watersheds. In addition, both the rainfall and runoff time series indicated the presence of nonstationary attributes such as statistically significant monotonic trends and/or change in mean and variance, which should be taken into consideration when using the time series to predict future scenarios. The annual peak runoff and the annual low flow in the Meramec watershed showed significant temporal changes compared to that in the Gasconade watershed. Water loss in both watersheds was found to be significantly increasing which is potentially due to the increase in groundwater pumping for water supply purposes.     

基于空间距离和时间成本的中小文化旅游城市可达性研究

在区域旅游发展中,中小文化旅游城市是其重要内容之一。除旅游景点本身的吸引力之外,旅游目的地的可达性及旅游景点基础设施的完善是其旅游发展的重要内容,因此中小文化旅游城市的可达性研究具有重要意义。基于GIS软件,论文从空间距离和时间成本角度对全国34个中小文化旅游城市的可达性进行研究。结果表明:我国中小文化旅游城市的可达性表现出非均衡的空间分布特征,其3 h可达服务范围(即游客市场)并未覆盖全国。其中,东北地区、华南地区、内蒙古高原、青藏高原等地区因中小文化旅游城市分布较少而出现可达服务范围的空白区域,而长江三角洲、成渝、鲁西南和海峡西岸等地区中小文化旅游城市的可达服务范围已绵延成片,具有较大的区域旅游合作潜力。此外,通过对中小文化旅游城市的旅游收入、旅游人数与可达范围内社会经济属性进行相关性分析,发现旅游收入与1 h可达范围内的GDP相关性最高,而旅游人数与50 km范围内的GDP相关性最高。     

一体式MBR污泥特性及对其过滤性能的影响

研究了一体式MBR长期不排泥运行的污泥特性变化,考察了水力停留时间、曝气量等因素对污泥增长特性、污泥颗粒粒径分布及膜过滤性能的影响,并探讨了这些因素之间的关系。结果表明:水力停留时间越短,MLSS与MLVSS增加越快,膜污染加速,膜过滤阻力增加越快,但MBR中污泥浓度的增长、变化对出水COD的影响不明显;污泥平均粒径随SRT的延长而减小,实验结束污泥粒径最小值为0.6μm大于实验用膜的孔径0.2μm,表明混合液中的微细颗粒污泥未对膜通量的下降起决定性作用,膜表面污泥的沉积对膜通量影响更大一些,增加曝气强度可以明显改善MBR的过滤性能。     

基于安全疏散的高速铁路桥梁救援定点站台设计参数研究

应用安全疏散性能化设计的理论方法,研究了桥梁救援定点疏散站台的尺寸参数 ,以保证设计更为安全、经济、合理。首先,采用FDS软件模拟火灾警戒线随疏散时间 的发展趋势,分析ASET对站台长度的影响;其次,采用EVACNET4软件模拟了列车中部和 端部火灾情境下,乘客在站台的疏散运动,研究站台宽度对疏散时间的影响。结果表明 :从火灾警戒范围考虑,站台长度取值为450 m是经济合理的;有效宽度1.75 m为站台 疏散时间特征变化的分界点,有效宽度为1.75 m时的疏散时间均小于有效宽度为1.5 m 时的疏散时间,而当有效宽度大于1.75 m后,疏散时间的变化不大。研究成果可为救援 定点疏散站台的尺寸设计提供参考和依据。     

考虑人群拥堵的疏散出口选择行为研究及建模

出口选择是疏散过程中最重要和复杂的决策行为之一,受到空间结构、人群分布和行人认知等多方面因素的影响,为了使疏散仿真能够合理地模拟出行人的寻路和逃生过程,提出了考虑人群拥堵的出口选择模型。该模型基于多元Logit离散选择模型,考虑到影响个体疏散时间的因素以及对不同出口类型的偏好。研究发现个体的疏散时间随着拥堵人群位置的不同而有所差异,从而以决策者而不是出口的角度定义了一个计数区域来估计影响决策者排队时间的人数。并且研究出口重新选择的触发条件和程序。将出口选择模型和程序引入BuildingEXODUS软件进行疏散仿真,通过对比显示,仿真结果更符合真实情况,特别是在具有障碍物和人群分布不均匀的场景中更显优势。     

The time process of post‐earthquake recovery: the Yao'an earthquake in China

Post‐disaster recovery is a constantly changing and developing process. The authors conducted three real‐name follow‐up surveys at 1, 12 and 18 months after the Yao'an earthquake, which had a surface wave magnitude of 6.0. They also calculated recovery ratios at different times and drew post‐earthquake domestic life recovery curves. Based on the recovery curves, the time trajectory of domestic life recovery takes on an approximate S‐type development and change process. The recovery time process of domestic life can be divided into four periods: emergency period (weeks 0–2(5)), early recovery period (weeks 2(5)–24), rapid recovery period (weeks 24–34) and late recovery period (weeks 34–60(80)).     

Organic matter balances in anaerobic digestors

Organic carbon and COD balances in a system of 12 anaerobic units operating at organic loading levels 0.4–0.8 kg COD/m³.d and hydraulic detention times 8–20 d resulted in a mean CH₄ production of 341 ml/g COD converted and a mean CH₄ + CO₂ production 1815 ml/g OC converted. The gas retained in the liquid anaerobic effluent was mainly carbon dioxide (94–98%).