二维编织C/SiC复合材料的热膨胀系数预测

根据二维编织 C/ SiC复合材料的细观结构及其制备工艺特点 , 提出了一种预测该材料面内热膨胀系数的单胞模型。模型充分考虑了编织结构复合材料中的纤维束弯曲和 CVI工艺制备陶瓷基复合材料产生的孔洞对热膨胀系数的影响。利用单胞模型预测了二维编织 C/ SiC的结构参数、 纤维体积含量、 孔洞含量对复合材料热膨胀系数的影响规律 , 结果表明 : 随着纤维束扭结处产生间隙与纱线宽度比值的增大 , 热膨胀系数增大 ; 当其它参数不变时 , 随着纤维体积含量的增大 , 热膨胀系数反而下降; 随着孔洞含量的增加 , 热膨胀系数也出现了下降的趋势。利用 DIL402C热膨胀仪测试了二维编织 C/ SiC复合材料纵向热膨胀系数 , 试验结果与模型预测结果吻合较好。     

基于模型比较的液压伺服作动器故障监控仿真研究

针对某型液压伺服作动器,提出一种基于模型比较的方法实现故障监控。阐述了该型作动器的组成及工作原理,之后对其故障模式进行分析,采用基于模型比较的监控方法建立液压伺服作动器的数学模型和仿真模型,运用Matlab软件进行仿真验证,证明本文建立的仿真模型可靠,与实际状态接近程度高,为液压伺服作动器的故障监控提供了可靠依据。     

施工现场安全危险源实时监控与安全风险预测方法研究

对国内外建筑业施工现场安全风险的研究进行了文献综述,针对目前研究的不足之处,提出了施工现场安全危险源实时监控和安全风险实时预测的示意性模型,并详细解释了该模型的含义和方法。研究将提供一种基于前馈信号的施工现场安全危险源实时监控和安全风险实时预测的方法;并通过将现有研究的视角引入到施工现场关键安全危险源的前馈信号上,为进一步的研究打下良好的理论基础。     

公安高校网安全问题对策研究

对公安高校网的技术策略进行研究,确地选择多级分布式检测模型作为公安高校网的一种内网外联监控模型。根据该模型的设计,可以很好的实现对公安高校网中的内网外联情况进行监控。     

二维编织C/SiC陶瓷基复合材料的热传导系数预测

根据二维编织C/SiC复合材料的细观结构及其制备工艺特点,提出了一种预测该材料热传导系数的单胞模型。模型简化了编织结构纱线的实际构型,充分考虑了编织结构复合材料由于化学气相渗透(CVI)工艺制备陶瓷基复合材料产生的孔洞对热传导系数的影响。利用单胞模型预测了二维编织C/SiC的结构参数、纤维体积含量、孔洞体积含量对复合材料热传导系数的影响规律。结果表明: 随着纤维束扭结处产生间隙与纱线宽度比值的增大,热传导系数减小;当其它参数不变时,热传导系数随着纤维体积含量和孔洞体积含量的增加而下降。利用Hot Disk热测量仪采用瞬变平面热源法测试了二维编织C/SiC复合材料面内的热传导系数,试验结果与模型预测结果吻合较好。     

最小二乘支持向量机和小波电铲供电机组状态预测

针对电铲供电机组振动时间序列是个非线性、非平稳的复杂时间序列,难以用单一预测方法进行有效预测的问题,建立了一种基于小波分解和最小二乘支持向量机混合模型进行状态预测的方法.首先通过小波分解,将原始振动时间序列分解到不同层次,然后根据分解后各层次分量的特点选择不同的嵌入维数和LS-SVM参数分别进行预测,最后重构得到原始序列的预测值.对某电铲供电机组振动趋势的预测结果表明,该模型的预测性能好于单一的支持向量机预测方法.     

轻型客车复合材料板簧模态的预测和分析

为了预测某轻型客车复合材料板簧的模态并预判复合材料板簧的模态是否会与相关激励耦合发生共振,在ABAQUS软件中建立了针对复合材料板簧模态计算问题的有限元模型。对复合材料板簧的有限元模型进行了计算模态分析,根据计算模态分析结果预测了复合材料板簧的模态。对复合材料板簧的样件进行了试验模态分析,通过对比模态预测结果和试验模态分析结果的方法验证了模态预测结果的准确性和有限元模型的正确性。根据复合材料板簧的模态分析结果,设计的复合材料板簧能够避免共振现象的发生。利用经过验证的复合材料板簧有限元模型分析了各设计变量与复合材料板簧一阶模态频率之间的关系。分析结果表明,选用0°铺层角度、较低密度的复合材料和较高的纤维体积含量能够降低复合材料板簧发生共振的可能性。得到的研究结果可显著降低复合材料板簧的研发风险和成本。     

高速轿车气流噪声模拟研究

采用大涡模拟（LES）方法计算了某轿车模型的瞬态外流场，并研究了车辆表面脉动压力和流态，然后采用FW-H声学模型，预测了车外场点的噪声特性。根据流场和声学模拟结果对某轿车模型进行修改，并进行了噪声测量，模型修改后气流噪声有显著降低。     

履带车辆不平路面越野平均速度预测方法研究

为了研究路面不平度对履带车辆越野平均速度的影响规律,本文建立了履带车辆多体动力学模型,提出了振动响应指标并通过试验验证了模型的可信性;根据试验设计进行仿真计算,建立振动响应与车速、路面不平度间的近似模型;以振动响应指标的门限值为约束条件,采用目标寻优方法拟合了路面不平度与车速间的数学关系,提出了随机不平路面条件下的越野速度预测方法。应用该方法计算了车辆通过某试验场综合路面,结果表明：所建模型可以反映路面不平度对车速的影响,为车辆机动性预测提供了有效的量化手段。     

履带车辆不平路面越野平均速度预测方法研究

为了研究路面不平度对履带车辆越野平均速度的影响规律,本文建立了履带车辆多体动力学模型,提出了振动响应指标并通过试验验证了模型的可信性;根据试验设计进行仿真计算,建立振动响应与车速、路面不平度间的近似模型;以振动响应指标的门限值为约束条件,采用目标寻优方法拟合了路面不平度与车速间的数学关系,提出了随机不平路面条件下的越野速度预测方法。应用该方法计算了车辆通过某试验场综合路面,结果表明：所建模型可以反映路面不平度对车速的影响,为车辆机动性预测提供了有效的量化手段。     

Characterization of the traditional Cypriot spirit Zivania by means of Counterpropagation Artificial Neural Networks

Multivariate chemometric techniques, such as Principal Component Analysis and Discriminant Analysis, were previously used to determine the authenticity of the Cypriot traditional spirit Zivania, but these techniques revealed difficulties in making this characterization. In the present paper, a non-linear classification model has been built by means of Counterpropagation Artificial Neural Networks. The aim of this model is the characterization of Zivania and the differentiation of this alcoholic beverage from other, similar, beverages from all over the world, especially Europe. This procedure may be an ideal tool for describing Zivania's uniqueness, since the mapping based on the Neural Networks has shown acceptable predictive capabilities. Moreover, the role of each variable in the classification model has been considered: Counterpropagation Artificial Neural Network results have been analysed by means of Principal Component Analysis, in order to study which variables have a real discriminant role in the classification model. This procedure appeared as a promising tool to study the relationship between variables and classes in a global way and not variable by variable, and to obtain a multivariate overview of variable behaviour in the classification model.     

Prognose von 15 Zeitreihen der DGOR mit Neuronalen Netzen

In 1982, the working group “Forecasting Methods” of the Deutsche Gesellschaft für Operations Research (DGOR) carried out a forecasting comparison between 12 various models which were applied to 15 time series. The results of this study can be considered as a good benchmark for further prediction techniques. This paper reports upon the prediction of these 15 time series by using a Neural Network which was developed by the Backpropagation algorithm. The four highest autocorrelated lag-variables were used as the input variables of the Neural Network. The results show that the Neural Network delivered worse predictions than the other methods including the naive prediction by forecasting non-stationary time series. Stationary time series could be predicted better than the naive prediction, but in comparison to the other techniques the results were only average. After regarding the problem of non-stationarity by using the Dickey-Fuller-Test, first differences were chosen as the input-variables of the Neural Network. In this case, there was a considerable improvement, but the best method (Box-Jenkins' ARIMA technique) could not be surpassed.     

基于粗神经网络的企业组织创新风险预警

提出了基于粗神经网络的企业组织创新风险预警模型,通过粗集减少属性的数量,提取主要的特征属性,降低神经网络构成系统的复杂性及计算时间;结合神经网络系统的容错能力、并行处理能力、抗干扰能力及处理非线性问题能力,将粗集与神经网络进行串行结合.实例研究表明,将Rough Set与BP神经网络结合起来应用于企业组织创新风险预警,大大简化了BPNN的结构,减少了网络的计算量,加快了收敛速度.     

Parameter estimation using compensatory neural networks

Proposed here is a new neuron model, a basis for Compensatory Neural Network Architecture (CNNA), which not only reduces the total number of interconnections among neurons but also reduces the total computing time for training. The suggested model has properties of the basic neuron model as well as the higher neuron model (multiplicative aggregation function). It can adapt to standard neuron and higher order neuron, as well as a combination of the two. This approach is found to estimate the orbit with accuracy significantly better than Kalman Filter (KF) and Feedforward Multilayer Neural Network (FMNN) (also simply referred to as Artificial Neural Network, ANN) with lambda-gamma learning. The typical simulation runs also bring out the superiority of the proposed scheme over Kalman filter from the standpoint of computation time and the amount of data needed for the desired degree of estimated accuracy for the specific problem of orbit determination.     

基于神经网络的柴油机声音品质评价预测

采用多元线性回归与人工神经网络系统（Artificial Neural Network，简称ANN）分别建立了柴油发动机噪声声音品质预测模型，并将两种模型的预测值与实测值进行比较。试验结果表明,该系统可以反映客观参数和主观满意度间的非线性关系，可以用来预测和描述柴油发动机噪声的声音品质。     

用神经网络BP算法预测钨合金材料抗拉强度

为减少实验量,降低实验成本,采用人工神经网络BP算法处理了钨合金材料的抗拉强度的实验数据,包括钨含量、变形量对材料抗拉强度的影响,给出了在不同钨含量条件下变形量对材料抗拉强度的关系曲线,和不同变形量条件下钨含量对材料抗拉强度的关系曲线.通过本文的分析可知,采用BP算法来处理钨合金的实验数据是可行的.     

基于Kohonen网络的轴系故障诊断研究

本文试从Kohonen神经网络在船舶铀系故障诊断中应用研究出发，提出了船舶轴系模拟试验台的结构和监测原理，通过数据采集系统和开发的软件系统，获得轴系振动倍号并求得其快速傅里叶变换(FFT)频谱，从频谱图中提取信号能量分布的特征矢量，用Kohonen神经网络对特征矢量进行识别，最终得到正确的故障诊断结果。介绍了Kohonen神经网络的学习和工作算法，并研究了在轴系故障诊断中的具体实施方法。实例验证表明，Kohonen网络是一种很有价值的轴系故障诊断手段。     

A transiently chaotic neural network approach to the design of cellular manufacturing

The design of Cellular Manufacturing Systems (CMS) has attained the significant attention of academicians and practitioners over the last three decades. Minimizing intercellular movements while maximizing utilization of machines are the main objectives of interest in designing CMS and are considered in present research. In this paper, the drawbacks of former neural networks-based approaches to cell formation are discussed. The standard version of cell formation problem is formulated and a 'Transiently Chaotic Neural Network' (TCNN) with supplementary procedures is introduced as a powerful rival. A simplified network is constructed. After developing the related equations the approach is tested using the proposed algorithm with 18 problems selected from literature. The results are compared with various other approaches including ART1, Extended-ART1, Ortho-Synapse Hopfield Neural Network (OSHN), etc. The main advantages of our proposed method are: (1) the ability to avoid the local optima trap, (2) the ability to solve problems of different sizes with the same set of values for parameters, and (3) the less computation time. The results also indicate considerable improvement in grouping efficiency through the proposed approach.     

智能神经网络开发系统的实现技术

对比了智能神经元模型和传统的神经元模型，论述了智能神经网络系统的组成原理，给出了智能神经网络开发系统的基本模型，并具体地阐述了智能神经网络开发系统基本模型中的各个组成部分。利用智能神经网络开发系统，研究人员可以较为容易地开发神经网络应用程序。     

Spatial-Resolution Independent Object Detection Framework for Aerial Imagery

Earth surveillance through aerial images allows more accurate identification and characterization of objects present on the surface from space and airborne platforms. The progression of deep learning and computer vision methods and the availability of heterogeneous multispectral remote sensing data make the field more fertile for research. With the evolution of optical sensors, aerial images are becoming more precise and larger, which leads to a new kind of problem for object detection algorithms. This paper proposes the “Sliding Region-based Convolutional Neural Network (SRCNN),” which is an extension of the Faster Region-based Convolutional Neural Network (RCNN) object detection framework to make it independent of the image’s spatial resolution and size. The sliding box strategy is used in the proposed model to segment the image while detecting. The proposed framework outperforms the state-of-the-art Faster RCNN model while processing images with significantly different spatial resolution values. The SRCNN is also capable of detecting objects in images of any size.