In this study,the moment tensor of transversely isotropic shale was analyzed using a discrete element method-acoustic emission model(DEM-AE model).Firstly,the failure modes of the shale obtained from the acoustic emis...In this study,the moment tensor of transversely isotropic shale was analyzed using a discrete element method-acoustic emission model(DEM-AE model).Firstly,the failure modes of the shale obtained from the acoustic emission(AE) events and physical experiments were compared.Secondly,the relationships between AE events and seismic magnitudes,and AE events and the resulting cracks were analyzed.Finally,a moment tensor T-k chart describing the seismic source was introduced to demonstrate the differences in the transversely isotropic shale.The results showed that,for different anisotropy angles,a linear logarithmic relationship existed between the cumulative AE events and the seismic magnitude in the concentration area of the AE events.A normal distribution was observed for the number of AE events as the seismic magnitude changed from small to large.The moment tensor T-k chart indicated that the number and proportion of linear tension cracks in the shale were highest.When θ = 30°,the peak seismic magnitude was at a minimum.The average seismic magnitude in the concentration area of the AE events was also relatively small.Points close to the U=-1/3V line and the number of cracks included in a single AE event were at a minimum,and the corresponding peak stress also reached its lowest level.In contrast,when θ=90°,all related parameters were contrary to the above θ = 30° case.The DEM-AE model and the moment tensor T-k chart are suitable for analyzing the distribution of shale cracks appearing during the loading process.This study can provide constructive references for future research on the fracturing treatment of shale.展开更多
科学的轨道交通出行模式分析是运营决策优化的重要依据。为挖掘城市轨道交通时空流动特征及其影响机理,提出一种基于非负张量分解的OD客流强度时空分布计算方法,采用融合SHAP归因分析的极端梯度提升树(eXtreme Gradient Boosting,XGBoo...科学的轨道交通出行模式分析是运营决策优化的重要依据。为挖掘城市轨道交通时空流动特征及其影响机理,提出一种基于非负张量分解的OD客流强度时空分布计算方法,采用融合SHAP归因分析的极端梯度提升树(eXtreme Gradient Boosting,XGBoost)对各模式OD客流强度进行拟合预测。使用城市轨道交通AFC(automatic fare collection system,AFC)系统数据,从空间、时段以及出行日3个维度构建3阶客流OD张量,采用交替非负最小二乘法(alternating non negative least squares,ANLS)实现非负CP张量分解。基于张量分解结果,从北京轨道交通344个站点连续1周16266966条出行数据中,提取出晨高峰长距离通勤、早高峰中短通勤、平峰休闲中转出行、晚归出行4种出行模式的时、空分布特征。基于可解释性机器学习模型,对各模式OD客流进行预测。结果表明XGBoost与CatBoost、LightGBM、OLS相比更具优势。根据OD起终点站域环境特征,考虑起终点缓冲区内各类兴趣点(point of interest,POI)数量、小区住户数、房价、人口数量、站点偏离距离以及出行距离等指标,构建OD强度关联指标体系,解释各指标对OD客流强度的正负反馈效应。SHAP归因分析说明,居民更倾向于14站以内的中短途出行,并分别得到了就业类POI数目对晨、早通勤客流正向影响,以及餐饮类POI数目对休闲中转出行客流正向影响的临界阈值。该方法可为轨道交通精细化出行引导和客流组织提供数据支撑,优化城市轨道交通供需平衡及服务水平。展开更多
慢性下腰痛(chronic low back pain,CLBP)作为全球致残率最高的肌肉骨骼疾病之一,其临床表现除持续性疼痛和运动功能障碍外,常伴随注意力缺陷、执行功能下降和工作记忆障碍等认知损害。多模态磁共振成像研究为阐明CLBP的神经机制提供了...慢性下腰痛(chronic low back pain,CLBP)作为全球致残率最高的肌肉骨骼疾病之一,其临床表现除持续性疼痛和运动功能障碍外,常伴随注意力缺陷、执行功能下降和工作记忆障碍等认知损害。多模态磁共振成像研究为阐明CLBP的神经机制提供了重要证据:功能磁共振成像(functional magnetic resonance imaging,fMRI)显示前额叶、扣带回和岛叶等认知控制脑区激活异常;静息态fMRI分析揭示默认模式网络、前额叶-顶叶控制网络及显著性网络功能连接失衡;弥散张量成像发现额叶-顶叶通路和胼胝体白质纤维完整性下降,且与认知测评表现相关;磁共振波谱提示N-乙酰天冬氨酸降低及谷氨酸/γ-氨基丁酸平衡失调,反映神经元功能和兴奋-抑制调节受损。现有证据支持CLBP通过“疼痛-情绪-认知”环路和三大脑网络失衡机制导致认知障碍。现有研究主要存在以下局限:大多数为横断面设计,无法确立因果关系;缺乏长期随访数据;样本代表性有限。基于这些局限,未来研究应当:(1)开展纵向追踪和干预性研究以验证神经机制与认知损害的因果关系;(2)整合多模态MRI技术与精细认知行为评估,建立CLBP认知损害的预测模型;(3)探索影像学生物标志物的临床转化价值,为早期识别和干预提供依据。通过解决这些问题,有望为改善CLBP患者的认知结局提供新思路。本综述系统梳理了CLBP相关认知损害的神经影像学研究进展,旨在为从事慢性疼痛与认知神经机制研究的科研人员及临床医生提供理论参考与研究思路,推动该领域从现象描述向机制探索与临床干预的转化。展开更多
随着数据量的爆炸式增长,边缘计算在大数据处理中的作用愈加重要.现实应用中产生的数据通常建模表示成高阶增量式张量的形式,增量式张量Tucker分解是一种高效挖掘高阶海量数据中隐藏信息的方法.针对传统增量式张量分解忽视张量模特征对...随着数据量的爆炸式增长,边缘计算在大数据处理中的作用愈加重要.现实应用中产生的数据通常建模表示成高阶增量式张量的形式,增量式张量Tucker分解是一种高效挖掘高阶海量数据中隐藏信息的方法.针对传统增量式张量分解忽视张量模特征对分解过程的影响、分解结果不能较好保留原始数据特征的问题,提出一种基于模特征的增量式张量Tucker分解方法 ITTDMC (incremental tensor tucker decomposition based on mode characteristics).首先,用模长增量决定增量因子矩阵更新顺序,以此降低更新顺序带来的重构误差;其次,根据模熵变化比决定增量因子矩阵更新权重,使分解结果更准确保留各模特征;然后,将过往时刻的模特征和更新参数记录在指导张量中,遇到模特征相似的增量数据时直接使用来指导张量中参数的更新,避免重复计算,降低时间开销;最后,在合成和真实数据集上进行大量的实验,实验结果表明ITTDMC在模特征明显的数据集上能显著降低(最高可达29%)增量式张量的重构误差.展开更多
基金Financial support for this work is provided by the National Natural Science Foundation of China (no.51474208)the National Key Research and Development Program of China (2016YFC0600904)+1 种基金a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD)The fnancial support provided by China Scholarship Council (CSC,Grant no.201606420013)
文摘In this study,the moment tensor of transversely isotropic shale was analyzed using a discrete element method-acoustic emission model(DEM-AE model).Firstly,the failure modes of the shale obtained from the acoustic emission(AE) events and physical experiments were compared.Secondly,the relationships between AE events and seismic magnitudes,and AE events and the resulting cracks were analyzed.Finally,a moment tensor T-k chart describing the seismic source was introduced to demonstrate the differences in the transversely isotropic shale.The results showed that,for different anisotropy angles,a linear logarithmic relationship existed between the cumulative AE events and the seismic magnitude in the concentration area of the AE events.A normal distribution was observed for the number of AE events as the seismic magnitude changed from small to large.The moment tensor T-k chart indicated that the number and proportion of linear tension cracks in the shale were highest.When θ = 30°,the peak seismic magnitude was at a minimum.The average seismic magnitude in the concentration area of the AE events was also relatively small.Points close to the U=-1/3V line and the number of cracks included in a single AE event were at a minimum,and the corresponding peak stress also reached its lowest level.In contrast,when θ=90°,all related parameters were contrary to the above θ = 30° case.The DEM-AE model and the moment tensor T-k chart are suitable for analyzing the distribution of shale cracks appearing during the loading process.This study can provide constructive references for future research on the fracturing treatment of shale.
文摘科学的轨道交通出行模式分析是运营决策优化的重要依据。为挖掘城市轨道交通时空流动特征及其影响机理,提出一种基于非负张量分解的OD客流强度时空分布计算方法,采用融合SHAP归因分析的极端梯度提升树(eXtreme Gradient Boosting,XGBoost)对各模式OD客流强度进行拟合预测。使用城市轨道交通AFC(automatic fare collection system,AFC)系统数据,从空间、时段以及出行日3个维度构建3阶客流OD张量,采用交替非负最小二乘法(alternating non negative least squares,ANLS)实现非负CP张量分解。基于张量分解结果,从北京轨道交通344个站点连续1周16266966条出行数据中,提取出晨高峰长距离通勤、早高峰中短通勤、平峰休闲中转出行、晚归出行4种出行模式的时、空分布特征。基于可解释性机器学习模型,对各模式OD客流进行预测。结果表明XGBoost与CatBoost、LightGBM、OLS相比更具优势。根据OD起终点站域环境特征,考虑起终点缓冲区内各类兴趣点(point of interest,POI)数量、小区住户数、房价、人口数量、站点偏离距离以及出行距离等指标,构建OD强度关联指标体系,解释各指标对OD客流强度的正负反馈效应。SHAP归因分析说明,居民更倾向于14站以内的中短途出行,并分别得到了就业类POI数目对晨、早通勤客流正向影响,以及餐饮类POI数目对休闲中转出行客流正向影响的临界阈值。该方法可为轨道交通精细化出行引导和客流组织提供数据支撑,优化城市轨道交通供需平衡及服务水平。
文摘慢性下腰痛(chronic low back pain,CLBP)作为全球致残率最高的肌肉骨骼疾病之一,其临床表现除持续性疼痛和运动功能障碍外,常伴随注意力缺陷、执行功能下降和工作记忆障碍等认知损害。多模态磁共振成像研究为阐明CLBP的神经机制提供了重要证据:功能磁共振成像(functional magnetic resonance imaging,fMRI)显示前额叶、扣带回和岛叶等认知控制脑区激活异常;静息态fMRI分析揭示默认模式网络、前额叶-顶叶控制网络及显著性网络功能连接失衡;弥散张量成像发现额叶-顶叶通路和胼胝体白质纤维完整性下降,且与认知测评表现相关;磁共振波谱提示N-乙酰天冬氨酸降低及谷氨酸/γ-氨基丁酸平衡失调,反映神经元功能和兴奋-抑制调节受损。现有证据支持CLBP通过“疼痛-情绪-认知”环路和三大脑网络失衡机制导致认知障碍。现有研究主要存在以下局限:大多数为横断面设计,无法确立因果关系;缺乏长期随访数据;样本代表性有限。基于这些局限,未来研究应当:(1)开展纵向追踪和干预性研究以验证神经机制与认知损害的因果关系;(2)整合多模态MRI技术与精细认知行为评估,建立CLBP认知损害的预测模型;(3)探索影像学生物标志物的临床转化价值,为早期识别和干预提供依据。通过解决这些问题,有望为改善CLBP患者的认知结局提供新思路。本综述系统梳理了CLBP相关认知损害的神经影像学研究进展,旨在为从事慢性疼痛与认知神经机制研究的科研人员及临床医生提供理论参考与研究思路,推动该领域从现象描述向机制探索与临床干预的转化。
文摘随着数据量的爆炸式增长,边缘计算在大数据处理中的作用愈加重要.现实应用中产生的数据通常建模表示成高阶增量式张量的形式,增量式张量Tucker分解是一种高效挖掘高阶海量数据中隐藏信息的方法.针对传统增量式张量分解忽视张量模特征对分解过程的影响、分解结果不能较好保留原始数据特征的问题,提出一种基于模特征的增量式张量Tucker分解方法 ITTDMC (incremental tensor tucker decomposition based on mode characteristics).首先,用模长增量决定增量因子矩阵更新顺序,以此降低更新顺序带来的重构误差;其次,根据模熵变化比决定增量因子矩阵更新权重,使分解结果更准确保留各模特征;然后,将过往时刻的模特征和更新参数记录在指导张量中,遇到模特征相似的增量数据时直接使用来指导张量中参数的更新,避免重复计算,降低时间开销;最后,在合成和真实数据集上进行大量的实验,实验结果表明ITTDMC在模特征明显的数据集上能显著降低(最高可达29%)增量式张量的重构误差.
