基于分形理论的黄河内蒙古段河势演变特征及其与凌汛灾害关联研究

寒区河道凌汛灾害河势“弯道效应”的量化评估十分重要。基于分形理论提出河道横断面-纵剖面-平面多维度河势分形维数计算方法及其物理机制，并探讨黄河内蒙古段不同维度河势演变分形特征及其与凌汛灾害的关联关系。结果表明，黄河内蒙古段不同维度河势均具有多尺度自相似分形特征，且具有多年记忆周期的长程相关性；冰坝（严重性冰塞）发生频次与河道主槽弯曲分形维数呈正相关指数型函数关系，与河相系数、深泓点高程和河段平均底坡分形维数负相关，与水深-面积分形维数正相关，总体表明冰坝灾害更易发生于主槽偏移摆动大、蜿蜒曲折、河湾发育程度高的宽浅型弯曲河道，研究成果可为凌汛期冰塞冰坝灾害易发河段诊断及预测提供重要理论依据。     

岩移沉降观测中数据缺失的一种估算方法

在矿区的岩移沉降观测过程中,由于环境、人为等原因导致观测数据缺失问题,损害了整体观测数据的完整性,影响后续沉降趋势预测.文中提出了一种最大期望算法,通过计算极大似然估计,交替执行E步和M步,最终得到观测缺失数据填补值.利用某矿区实测数据进行实验,对比分析发现最大期望算法对缺失数据填补更接近实际测量值,在实际应用中有一定参考价值.     

基于卷积神经网络的5G蜂窝网络无线定位方法

5G蜂窝网络发展迅猛，其覆盖面积将逐渐增大，因此使用5G蜂窝网络进行定位是有研究潜力的研究方向。本文提出一种新的深度学习技术来实现高效、高精度和低占用的定位，以代替传统指纹定位过程中繁重的指纹库生成以及距离计算。该方法建立了一个特殊的卷积神经网络，并根据5G天线信号的接收信号强度指示、相位和到达角等特征量，选择合适的输入数据格式构造样本组建训练集，对该卷积神经网络进行训练。训练得到的卷积神经网络可以替代指纹定位中的庞大指纹库，非常有利于直接在5G移动设备端实现定位。虽然卷积神经网络在训练过程中需要大量时间，但在训练完毕后直接进行分类定位的速度非常快，可以保障定位实现的实时性。本文所实现的卷积神经网络权重与偏置所占内存不到0.5 MB，且能够在实际应用环境中以95%的定位准确率以及0.1 m的平均定位精度实现高精度定位。     

基于形态特征提取的急性下壁心肌梗死BiLSTM网络辅助诊断算法

急性下壁心肌梗死是一种病发急、进展快、致死率高的心脏疾病,该文提出一种新颖的基于形态特征提取的BiLSTM神经网络分类的急性下壁心肌梗死辅助诊断算法,可大幅度提高医生对急性下壁心肌梗死疾病的诊断效率并有助于及时确诊.算法包括:对胸痛中心数据库心拍信号进行降噪及心拍分割;根据临床心内科医学诊断指南提取了12导联波形距离特征和分导联波形幅值特征;依据提取的特征搭建LSTM与BiLSTM神经网络进行心拍的分类识别;使用PTB公开数据库和胸痛中心数据库多临床中心进行交叉验证.实验结果表明,加入胸痛中心真实临床数据后,基于形态特征提取BiLSTM神经网络的急性下壁心肌梗死辅助诊断算法准确率达到99.72％,精度达到99.53％,灵敏度达到100.00％,同时F1-Score达到99.76.该算法比其他现有算法准确率提高至少1％,该项研究具有非常重要的临床应用价值.     

磨料有序排布电镀砂轮的制备及其把持力研究

为解决电镀砂轮磨削加工中容屑空间不足的问题，采用点胶微粘接的方法制备了磨料有序排布的电镀砂轮，分析了磨料粘接效果和镀层力学性能。通过SEM分析了磨料/镀层/导电胶的结合界面，并进行了干磨削试验。研究结果表明，直径约为磨料粒径40%的胶点可粘接住磨料，单个胶点上粘接多颗磨料的占比小于6%；双脉冲电镀工艺制备的镀层显微硬度大于500HV，表层残余应力小于100MPa，磨料/镀层/导电胶之间的界面贴合紧密，无明显缺陷；砂轮在磨削时没有出现磨料脱落现象。     

Estimating long-term time-resolved indoor PM2.5 of outdoor and indoor origin using easily obtainable inputs

To evaluate the separate impacts on human health and establish effective control strategies, it is crucial to estimate the contribution of outdoor infiltration and indoor emission to indoor PM_2.5 in buildings. This study used an algorithm to automatically estimate the long-term time-resolved indoor PM_2.5 of outdoor and indoor origin in real apartments with natural ventilation. The inputs for the algorithm were only the time-resolved indoor/outdoor PM_2.5 concentrations and occupants’ window actions, which were easily obtained from the low-cost sensors. This study first applied the algorithm in an apartment in Tianjin, China. The indoor/outdoor contribution to the gross indoor exposure and time-resolved infiltration factor were automatically estimated using the algorithm. The influence of outdoor PM_2.5 data source and algorithm parameters on the estimated results was analyzed. The algorithm was then applied in four other apartments located in Chongqing, Shenyang, Xi'an, and Urumqi to further demonstrate its feasibility. The results provided indirect evidence, such as the plausible explanations for seasonal and spatial variation, to partially support the success of the algorithm used in real apartments. Through the analysis, this study also identified several further development directions to facilitate the practical applications of the algorithm, such as robust long-term outdoor PM_2.5 monitoring using low-cost light-scattering sensors.     

基于ERNIE-BiGRU模型的摘要语步自动识别研究

学术文献的摘要是对文献主要内容的浓缩,摘要不同部分的语步具有不同的信息,语步的自动识别和抽取对于学术摘要的后续研究有着重要的应用价值,而目前语步识别的研究相对较少,并且相关算法的效果还需要提高。针对上述问题,该文提出了一种基于ERNIE-BiGRU模型的语步识别算法。该算法首先结合中文句法分析理论提出基于句法依存关系的多语步结构拆分法,对学术文献摘要多语步结构进行自动拆分,获得多个单语步结构;然后构建用于训练的单语步结构语料库,并利用知识增强语义表示预训练模型,训练出句子级词向量;最后将训练出的单语步结构词向量信息输入双向门限循环单元(BiGRU)进行摘要语步自动化识别,取得了良好的效果。实验结果表明,该算法具有较好的鲁棒性和较高的识别精度,在结构化和非结构化摘要上的识别准确率分别达到了96.57%和93.75%。     

Authenticity green coffee bean species and geographical origin using near-infrared spectroscopy combined with chemometrics

To prevent the adulteration of agricultural resources and provide a solution to enhance the green coffee bean supply chain, authentication using the near-infrared spectroscopy (NIRS) technique was investigated. Partial least square with discrimination analysis (PLS-DA) models combined with various preprocessing methods were built from NIR spectra of 153 Vietnamese green coffee samples. The model combined with the standard normal variate and the first order of derivative yielded excellent performance in predicting coffee species with the error cross-validation of 0.0261. PLS-DA model of mean centre and first-order derivative spectra also yielded good performance in verifying geographical indication of green coffee with the error of 0.0656. By contrast, the predicting abilities of post-harvest methods were poor. The overall results showed a high potential of the NIRS in online authentication practices.