一种新的变步长LMS自适应滤波算法

本文通过建立步长因子μ与误差信号之间的非线性函数关系,提出了一种新的变步长LMS(LeastMean Square)算法.该算法具有初始阶段和未知系统时变阶段步长自动增大而稳态时步长很小的特点,且克服了S函数变步长LMS算法(简称SVSLMS算法)在自适应稳态阶段μ(n)取值偏大的缺陷.理论分析和计算机仿真结果表明该算法的性能优于SVSLMS算法.     

分数阶Fourier变换域中网络流量的自相似特性分析

通过分析网络流量数据在FrFT域的统计特性发现,实际网络流量在FrFT域满足自相似性,进一步地,针对网络流量在FrFT域的“时域”和“频域”展开,分别给出了基于改进的整体经验模态分解—去趋势波动分析（MEEMD-DFA）的Hurst指数估计法以及基于加权最小二乘回归（WLSR）的Hurst指数自适应估计法。实验结果表明,相比于现有估值算法,MEEMD-DFA法具有较高的估计精度,但计算复杂度高;而FrFT自适应估计法则具有更优的估计顽健性,且计算复杂度较低,可作为一种实时在线估计真实网络数据Hurst指数的方法。     

长方格子光子晶体带隙特性

基于平面波展开法,以长方格子光子晶体带隙特性为研究对象,改变长方格子空气圆柱半轴a和b的大小,得到当a=b时形成较宽的光子带隙,随着b的增大,光子带隙逐渐减小.改变构成长方格子光子晶体基底的材料,得到Ge构成基底时形成较宽的光子带隙.研究结果为光子晶体器件的制作提供参考.     

On-line parameter identification of SPM motors based on MRAS technique

This article proposes a new parameter estimation method for surface-mounted permanent magnet (SPM) motors by employing the model reference adaptive system (MRAS) technique. The proposed algorithm is applied to an SPM motor used in washing machines. To determine washing and dewatering patterns, a washer tries to detect the size of load in the beginning of the operation. Hence it requires an accurate estimation of system parameters. The proposed estimation method identifies stator inductance and flux-linkage by using d- or q-axis current error, and its value is used for estimation independently with respect to each parameter. Therefore, the proposed scheme can get rid of the instability in inductance and flux estimation caused by the cross-coupling between d- and q- quantities. In this article, the stator resistance is detected by using the recursive least square (RLS) method to reduce measurement noises and to eliminate dead-time effect. Simulation results show that the proposed scheme tracks the parameters well even when there is transient states in rotor speed and the load conditions, and experimental results verify the validity and efficacy of the proposed algorithm.     

Potential of fluorescence spectroscopy in detection of low-levels of gluten in flour: A preliminary study

Celiac disease is caused due to the ingestion of gluten containing food products. Eating gluten free diet is only way to avoid its occurrence. Gluten free products may get contaminated during the handling and processing operation which leads to the adverse impact on the gluten intolerant patients. Three different gluten free mixes were adulterated with two different types of wheat flours to get the gluten concentration in the range of 0–5 % by making 94 sample combinations. The obtained fluorescence spectra from these samples were transformed using multiplicative scatter correction (MSC) pre-processing to develop partial least square regression (PLSR) model for the prediction of low-levels of gluten. The obtained results showed a high value of 0.90 correlation coefficient (R²) with 0.46 % root mean square error of cross-validation (RMSECV). The calibration and prediction models also showed the same R² with root mean square error of calibration (RMSEC) and prediction (RMSEP) of 0.41 % and 0.46 % respectively. Hence fluorescence spectroscopy can be used to estimate the level of adulteration which is helpful for developing a sensor for determination of low-levels of gluten as a rapid and non-invasive method in food applications.     

X射线能谱分析中对铁的百分含量修正的理论推导和实验测定

在KYKY-1000B扫描电镜上进行X射线能谱分析时,常出现附加的铁峰。本文从理论上对产生附加铁峰的原因作出了解释,并推导出附加铁峰强度和样品原子序数的函数关系的解析表达式,同时对附加的铁的百分含量进行了实验测量。由实验数据拟合得出的实验曲线和理论分析得出的表达很好地符合。     

基于广义最小均方的自适应递归滤波算法

本文提出了一种广义最小均方(GLMS)算法,分析了GLMS算法的性能,并根据GLMS算法,导出一种自适应递归滤波(ARF)算法。在推导中,ARF算法避开了超稳定定理,从而考虑自适应系统的严格正实条件和Popov不等式是不必要的。计算机模拟表明ARF算法具有很高的收敛速度。     

基于回波序列最小二乘拟合的高分辨率SAR运动目标速度估计

运动目标速度估计是机载单天线高分辨率合成孔径雷达(SAR)实现运动目标成像和定位的关键环节。针对现有方法运算量大、易受距离徙动干扰等缺点,该文提出一种基于回波序列最小二乘拟合的速度估计方法。利用该方法,首先通过包络相关提取相邻回波序列的距离变化量,然后对其做最小二乘线性拟合,目标的距离向速度和方位向速度可由拟合系数计算得到。与传统方法相比,该方法不仅计算量小,而且无须先做距离徙动校正(RCMC)。该文给出了新方法的数学模型和参数选取原则,分析了该方法的估计精度、计算量和适用条件,并通过仿真和实际数据处理验证了该方法的有效性。

     

基于matlab的环境数据曲线拟合在某子母弹引信机构仿真中的应用

以某子母弹为对象 ,分别论述不同的离散曲线的解析式近似表达方法 ,最后以不同方法的拟合结果作为仿真模型的输入 ,得到仿真结果进行分析     

Shape optimization of inclined ribs as heat transfer augmentation device

This work presents numerical optimization techniques for the design of a rectangular channel with inclined ribs toenhance turbulent heat transfer.The response surface method with Reynolds-averaged Navier-Stokes analysis isused for optimization.Shear stress transport turbulence model is used as a turbulence closure.Computational re-sults for local heat transfer rate show a reasonable agreement with the experimental data.Width-to-rib height ratioand attack angle of the rib are chosen as design variables.The objective function is defined as a linear combina-tion of heat-transfer and friction-loss related terms with the weighting factor.Full-factorial experimental designmethod is used to determine the data points.Optimum shapes of the channel have been obtained in a range of theweighting factor.