多深度学习模型决策融合的齿轮箱故障诊断分类方法

针对齿轮故障诊断中单一传感器采集信息不完全、容错性不佳及一种神经网络模型具有局限性,传统信号处理技术提取特征困难等问题,提出了多深度学习模型决策融合的齿轮箱故障诊断分类方法,构建了基于卷积神经网络(convolutional neural networks, CNN)和改进堆叠降噪自动编码器(stacked denoising autoencoders, SDAE)的混合网络模型,根据改进的Dempster-Shafer(D-S)证据理论实现决策级融合诊断。以时频信号作为CNN的输入,以频域信号作为SDAE的输入,采用Adam优化算法和dropout、批量归一化技术训练该混合模型。实验结果表明：利用该融合方法对齿轮进行故障诊断相比单个的网络模型CNN和SDAE诊断正确率有所提高,为齿轮故障智能诊断分类提供了新路径。     

大型活动散场期间地铁车站短时进站客流预测

大型活动散场期间的地铁车站客流属于可预知的非常规客流,采用常规客流的统计预测方法难以准确预测其客流需求.基于深度学习,将历史客流规律、大型活动数据与实时自动售检票系统数据相结合,提出了一种适用于大型活动散场期间地铁车站的短时客流预测模型.首先对历史客流数据进行了拆分及降噪处理,并分析了活动客流特征.之后,基于深度学习框...     

基于多级分类器的自由手写数字在线识别

设计了一种用于手写字符在线识别的多级分类器模型。在该模型中,通过对四个独立的分类算法进行集成,将脱机方法与联机方法,神经网络方法与传统方法有机地结合起来构成一个完整的手写字符在线识别系统。理论分析和实验结果表明,该集成系统比单一系统具有明显的优越性。     

基于神经网络的道路季节分类

为了合理确定路面结构设计时的输入参数,引入自组织特征映射神经网络,结合Matlab软件对神经网络进行权值训练,将网络训练是否收敛来作为分类的依据,根据温度、交通量和降雨量等几个重要参数对道路进行季节分类,最后按照分类结果进行路面结构分析与材料设计.实践证明,该方法分类效果良好,能很好地解决路面设计参数的合理确定问题,从而大大延长路面的使用寿命,提高道路投资的经济效益.     

Modelling Operational Risk Losses with Graphical Models and Copula Functions

The management of Operational Risk has been a difficult task due to the lack of data and the high number of variables. In this project, we treat operational risks as multivariate variables. In order to model them, copula functions are employed, which are a widely used tool in finance and engineering for building flexible joint distributions. The purpose of this research is to propose a new methodology for modelling Operational Risks and estimating the required capital. It combines the use of graphical models and the use of copula functions along with hyper-Markov law. Historical loss data of an Italian bank is used, in order to explore the methodology’s behaviour and its potential benefits.      

Chinese Speech Recognition Model Based on Activation of the State Feedback Neural Network

IntroductionIn recentyears,speech recognition has made greatprogress.Commercial systems such as Via Voice( IBM Company) and Naturally Speaking 1 .0( Dragon System Company ) lead the field.Although the recognition ratio has greatlyimproved,many issues still need further research,such as,real- time in processing,systemcomplexity[4 6] ,very large scale integrated circuit( VLSI) implementation,etc.Among these issues,the implementation in VLSI is the most criticalchallenge for wide use of s…     

企业网上运行MIS系统的安全措施

基于有关计算机网络安全的重要性，结合某企业MIS系统的设计，介绍了企业网络实施的安全策略，特别介绍了企业网上安全运行MIS系统所实施的具体方案。     

Concurrent transmission with minimal SR-pair interference in WiMAX mesh networks

Performance of wireless mesh networks can be greatly improved by spatial reuse (SR), therefore, lots of algorithms have been proposed to maximize the number of concurrent transmissions. However, most methods only considered the interference brought by the sender nodes. In this paper, we study the interference model in WiMAX mesh networks and propose a minimal SR-pair interference model to perform concurrent transmissions in both uplink and downlink directions. Simulation results show that our model is more efficient and stable, i.e., it increases the number of concurrent transmissions and its performance is constant and independent of transmission directions. Biography: XIONG Qing(1978–), male, Ph. D., research direction: wireless mesh networks.     

Universal Distributed Estimation Scheme for a Bandwidth Constrained Ad Hoc Sensor Network

This paper describes a distributed estimation scheme (DES) for a bandwidth constrained ad hoc sensor network.The DES is universal in the sense that operations on all sensors are identical and independent of noise distribution.The scheme requires each sensor to transmit just a 1-bit message per observation.Simulation results show that the scheme achieves much better mean-squares error (MSE) performance than the simplified isotropic universal DES and even outperforms the isotropic universal DES which requires more than twice the bandwidth of this scheme.     

不具备M-矩阵条件的时滞Hopfield神经网络模型的收敛性

考虑了一类具有时滞的Hopfield神经网络模型解的收敛性．在不需要M-矩阵条件的前提下，获得了该网络的所有解当t→∞时都趋向于平衡点．其结果补充和完善了已有文献的相应结果．