Logistic人口增长模型的混沌数据系统及反馈控制研究

针对Logistic人口增长模型由稳定状态,通过控制参量的变化进入倍周期分岔及混沌,提出了一种基于Lyapunov指数的状态反馈控制策略,使混沌状态的Logistic模型变成为可测。该控制策略可将混沌状态的Logistic人口增长模型快速的收敛于给定的任意值,并进行了数字仿真。所有计算及作图由MathCad7.0实现。     

隐式曲面的整数型生成算法

隐式曲面易实现几何造型，但较难绘制，章提出一个隐式曲面的象素级生成算法，并针对隐式曲面的特点实现了隐藏线消隐，由于是象素级算法，所以该算法绘制的曲面非常细致和平滑，该算法只使用整数运算，所以具有很快的速度。     

Shear-Resistant Behavior Analysis of Light Composite Shear Walls

IntroductionAt present,there are two kinds of thermalinsulation methods for external walls,i. e.,external or internal insulation of the external wall.These two methods have no relation between thethermal- insulating layer and the structural layer ofthe wall. The thermal- insulating layer isafterwards glued to the structural layer of thewall,which makes construction complicated andthe thermal- insulating layer is easily damaged orfalls off.Italso brings difficulty to the interior andexterior tr…     

利用说话人自适应实现基于DNN的情感语音合成

为了提高情感语音合成的质量,提出一种采用多个说话人的情感训练语料,利用说话人自适应实现基于深度神经网络的情感语音合成方法。该方法应用文本分析获得语音对应的文本上下文相关标注,并采用WORLD声码器提取情感语音的声学特征;采用文本的上下文相关标注和语音的声学特征训练获得与说话人无关的深度神经网络平均音模型,用目标说话人的目标情感的训练语音和说话人自适应变换获得与目标情感的说话人相关的深度神经网络模型,利用该模型合成目标情感语音。主观评测表明,与传统的基于隐马尔科夫模型的方法比较,该方法合成的情感语音的主观评分更高。客观实验表明,合成的情感语音频谱更接近原始语音。所以,该方法能够提高合成情感语音的自然度和情感度。     

基于MHMM的脱机手写体字符识别

对隐马尔可夫模型(HMM)的训练方法及模型参数的选取进行了探讨,并将HMM理论用于脱机手写体识别中,建立了一种基于字符投影变换图像的边界链码特征的多重隐马尔可夫模型(MHMM).实验结果表明,该方法是可行的,且具有良好的兼容性和灵活性,可应用于手写体字符的自动识别中.     

DHMM的动态模式识别理论在旋转机械故障诊断中的应用

对于旋转机械启动过程的动态模式,提出了一种基于离散隐马尔可夫模型(DHMM)的旋转机械故障诊断新方法.该方法对旋转机械启动过程的局部振动信号进行FFT特征提取,然后利用自组织特征映射对提取的特征矢量进行预分类编码,把矢量编码作为观测序列引入到DHMM中进行机器学习和故障诊断实验.实验表明,提出的方法对旋转机械启动过程进行诊断是十分有效的.     

新型方钢管混凝土柱与混凝土梁的节点破坏机理

结合我国现行钢管混凝土柱在实际工程中的应用状况,提出了用于方钢管混凝土柱-钢筋混凝土梁框架结构的新型连接节点形式非穿心暗钢牛腿方钢管混凝土柱与混凝土梁节点,采用了X型加载试验方案,制作了1∶3的试验模型,进行了低周反复荷载作用下的试验研究,得到了相关的钢筋应力应变发展规律和相应的滞回曲线,通过对试验结果节点破坏的机理分析,提出了进一步的改进措施.试验研究结果表明,这种节点具有可靠的受力性能,与常规的连接方式相比相对简捷而可靠.     

一种改进的多层前馈神经网络结构研究

通过将传统多层前馈神经网络结构改造为具有跨层连接的网络,提出了相应的多层前馈神经网络改进算法.通过分析网络误差函数,从理论上证明了:相对于无跨层连接网络,有跨层连接网络能以更加简洁的结构逼近理想状态.最后,用一个隐层神经元解决了传统网络必须用两个隐元才能解决的异或问题.     

Empirical analysis and forecasting of volatility dynamics in high‐frequency returns with time‐varying components

We study intraday return volatility dynamics using a time‐varying components approach, and the method is applied to analyze IBM intraday returns. Empirical evidence indicates that with three additive components—a time‐varying mean of absolute returns and two cosine components with time‐varying amplitudes—together they capture very well the pronounced periodicity and persistence behaviors exhibited in the empirical autocorrelation pattern of IBM returns. We find that the long‐run volatility persistence is driven predominantly by daily level shifts in mean absolute returns. After adjusting for these intradaily components, the filtered returns behave much like a Gaussian noise, suggesting that the three‐components structure is adequately specified. Furthermore, a new volatility measure (TCV) can be constructed from these components. Results from extensive out‐of‐sample rolling forecast experiments suggest that TCV fares well in predicting future volatility against alternative methods, including GARCH model, realized volatility and realized absolute value. Copyright © 2009 John Wiley & Sons, Ltd.     

Puzzling over orphan enzymes

Despite the current availability of several hundreds of thousands of amino acid sequences, more than 39% of the well-defined enzyme activities (EC numbers) are not associated with any sequence in major public databases. This wide gap separating knowledge of biochemical function and sequence information is found in nearly all classes of enzymes. Thus, there is an urgent need to explore the 1525 orphan enzymes (EC numbers without associated sequences), in order to progressively bridge this unwanted gap. Improving genome annotation could unveil a significant proportion of sequenceless enzymes. Peptide mass mapping and further genome mining would be useful to identify proper sequence for enzymes found in species for which genetic tools are missing. Finally, the whole community must help major public databases to begin addressing the problem of missing or incomplete information. Received 31 October 2005; received after revision 8 December 2005; accepted 20 December 2005