基于改进CLDNN的辐射源信号识别

传统辐射源信号识别方法往往需要人工提取特征,不仅对专业知识要求较高,而且人为选择的特征不能够保证适用于大多数类型信号的识别,识别精度和识别速度也不能兼顾。针对上述问题,将语音处理领域常用的深度学习模型——卷积长短时深度神经网络(convolutional long short-term deep neural network, CLDNN)引入到辐射源信号的识别中,并将该模型中的长短时记忆层改为双向门控循环单元层。模型的输入为原始时间序列数据,特征提取和分类识别过程均在网络中进行,避免了人工选择特征的不完备性。实验结果表明,所提模型在低信噪比情况下也能够有效识别信号类型,同时与其他模型相比,实现了识别精度和识别速度之间的平衡。     

基于特征选取与树状Parzen估计的入侵检测

针对目前网络空间安全形势快速变化带来的新风险和新挑战, 提出一种基于相关性分析的特征选取和树状Parzen估计优化的入侵检测方法。首先, 通过基于相关性分析的数据特征选取方法对数据维度进行压缩。其次, 对原始数据集进行特征筛选, 生成新的特征子集。最终, 使用序列模型优化算法中的树状Parzen估计算法对随机森林算法进行模型优化。实验结果表明, 相比其他应用机器学习算法的入侵检测方法, 所提方法在提升综合性能的同时拥有更高的检测效率, 有效地提升了入侵检测技术的实用性。     

基于大津算法和深度学习的开集声纹识别自适应阈值计算方法

针对开集声纹识别的自适应阈值计算问题, 提出一种基于大津算法和深度学习的阈值计算方法. 首先, 采用三层受限Boltzmann机堆叠而成的深度置信网络从Mel倒谱系数中提取语音深层特征; 其次, 通过Gauss混合模型计算特征的相似度值, 对特征的相似度值使用大津算法计算阈值. 实验结果表明, 该方法识别效果较理想, 与等错误率计算阈值方法相比, 具有更高的识别准确率.     

基于域适应与分类器差异的滚动轴承跨域故障诊断

基于数据驱动方法诊断滚动轴承故障时，不同工况下的数据特征分布差异会导致模型诊断性能严重下降.针对这一问题，提出了基于域适应与分类器差异的滚动轴承跨域故障诊断方法.首先利用卷积神经网络对带标记的源域样本和无标记的目标域样本进行特征提取;然后通过2个全连接分类器进行故障分类;最后通过分步优化分类损失、域最大平均差异损失和分类器差异损失，实现源域和目标域之间的域分布对齐，从而实现无标记目标域样本的故障诊断.实验结果表明，所提方法与主流的域适应方法相比具有更高故障诊断准确率，验证了该方法的合理性和可行性.     

某岩质陡崖边坡崩塌地质灾害成因及治理设计分析

以某岩质陡崖边坡崩塌地质灾害治理工程项目为依托,对崩塌地质灾害形成的主要诱发因素,如地形条件、地层岩性、降雨以及人类活动等进行了比较详细的介绍,为事故原因的定性提供相关依据。同时,针对现场实际情况及边坡稳定性分析结果提出一系列治理措施,为今后的具体工程实践提供相关理论依据及技术参考。     

Haversian bone-mimicking bioceramic scaffolds enhancing MSC-macrophage osteo-imunomodulation

The interaction between immune cells and bone forming cells plays a vital role in maintaining the homeostasis of the skeletal system, and is regulated by the three-dimensional structure of tissues. Whether the construction of biomaterials can activate or reproduce this spatial “cross-talk” between immune cells and bone forming cells in bone natural formation process is a prerequisite for successful fracture healing and bone regeneration. Herein, a bone marrow mesenchymal stem cells (MSCs)/macrophages-laden Haversian bone-mimicking bioceramic scaffold was successfully prepared through the biomimetic design of biomaterials and 3D printing technology. MSCs and macrophages were respectively distributed in the cancellous bone and Haversian canals of the scaffold to simulate the three-dimensional structure regulation of the cell spatial distribution and multiple intercellular interaction in natural bone tissue, and worked in concert to modulate the scaffold material-mediated osteo-immune microenvironment. The in vitro study revealed that the pro-inflammatory response of macrophages was more significantly inhibited when distributed with MSCs in the scaffolds at a cell ratio of 0.5:1 for co-culture, in comparison with multicellular culture at other ratios and unicellular culture. Meanwhile, MSCs exhibited the relatively high osteogenic potential, most likely via the activation of certain key signaling pathways mediated by macrophages-derived paracrine signaling mediators (OSM, BMP-2, and WNT10b). This work not only establishes a bionic platform for the regulation of multicellular osteo-immune response and regeneration but also offers a promising tissue-engineered biomimetic scaffold with improved immunomodulatory function for promoting bone tissue regeneration.     

海洋放射性现场探测谱仪的机械优化设计

为提高海洋放射性现场探测NaI(Tl)谱仪的探测性能，采用蒙特卡罗仿真和统计计算方法，研究和分析谱仪封装、材料和尺寸等机械设计对探测性能的影响规律。综合考虑伽马射线衰减和密封、耐压及耐腐蚀等要求，开展谱仪的机械优化设计和海上现场检测实验。优化谱仪达到了对40K最小探测活度约0.30 Bq/L，现场测量相对误差为0.45%。该研究对海洋放射性现场探测谱仪的性能优化设计和现场应用具有重要的指导意义。     

Neural network structure identification in inflation forecasting

Neural networks (NNs) are appropriate to use in time series analysis under conditions of unfulfilled assumptions, i.e., non‐normality and nonlinearity. The aim of this paper is to propose means of addressing identified shortcomings with the objective of identifying the NN structure for inflation forecasting. The research is based on a theoretical model that includes the characteristics of demand‐pull and cost‐push inflation; i.e., it uses the labor market, financial and external factors, and lagged inflation variables. It is conducted at the aggregate level of euro area countries from January 1999 to January 2017. Based on the estimated 90 feedforward NNs (FNNs) and 450 Jordan NNs (JNNs), which differ in variable parameters (number of iterations, learning rate, initial weight value intervals, number of hidden neurons, and weight value of the context unit), the mean square error (MSE), and the Akaike Information Criterion (AIC) are calculated for two periods: in‐the‐sample and out‐of‐sample. Ranking NNs simultaneously on both periods according to either MSE or AIC does not lead to the selection of the ‘best’ NN because the optimal NN in‐the‐sample, based on MSE and/or AIC criteria, often has high out‐of‐sample values of both indicators. To achieve the best compromise solution, i.e., to select an optimal NN, the preference ranking organization method for enrichment of evaluations (PROMETHEE) is used. Comparing the optimal FNN and JNN, i.e., FNN(4,5,1) and JNN(4,3,1), it is concluded that under approximately equal conditions, fewer hidden layer neurons are required in JNN than in FNN, confirming that JNN is parsimonious compared to FNN. Moreover, JNN has a better forecasting performance than FNN.     

一种面向IPv6的定制化路由备份机制

针对网络定制化能力和可靠路由问题，提出一种面向IPv6的定制化路由备份机制.该机制设计了路由定制化流程，使网络在满足多种应用差异化需求的基础上，考虑用户和网络服务提供商对网络的满意度最大化问题，并结合IPv6崭新的包头格式设计路由备份机制来增强网络的可靠性.仿真实验和性能分析表明，本机制通过二人博弈在选路由时均衡了用户和网络服务提供商的效用，使路由具备了定制化能力.在可靠性上，相比不存在定制的原IPv6路由，本机制在路由时间开销增加16.8%的基础上，预留了备份路径资源，使路由恢复响应时间缩短15%以上.     

矿山复杂巷道网络组件式三维建模方法

煤矿复杂巷道(曲线巷道、交汇巷道、变断面巷道)建模是巷道三维建模的难点.现有方法在基于导线测量数据重建复杂巷道网络、尤其是矿山常见的曲线交汇巷道仍存在不足.提出了一种巷道组件式三维建模方法.该方法针对煤矿巷道典型导线测量数据，通过巷道中心线建模、巷道双边线建模、巷道组件三维建模三个步骤，自动构建复杂巷道网络三维模型.实验结果表明，提出的方法鲁棒性好，建模时间随数据量增长呈线性稳定增长.建立的巷道网络三维模型可为智慧矿山建设中的空间、几何、力学等多种分析提供支持.