基于DEPSO-RBF神经网络的锌银电池SOC估计

电池容量是判断电池性能状态的重要指标。针对锌银电池的荷电状态估计问题,利用电池放电过程中放电时间、放电电流和电池电压3个参数作为径向基神经网络的输入,电池荷电状态为输出,建立电池放电的径向基神经网络模型;为克服径向基神经网络收敛精度不高、易陷入局部极小值的缺点,采用差分进化算法和粒子群算法结合的混合算法优化RBF神经网络;MATLAB仿真结果表明,经过混合优化算法优化的径向基神经网络与仅使用粒子群优化的径向基神经网络相比,估计精度得到大大提高。     

基于改进教学优化算法的Hermite正交基神经网络混沌时间序列预测

针对传统预测模型对混沌时间序列预测精度低、收敛速度慢及模型结构复杂的问题, 提出了基于改进教学优化算法的Hermite正交基神经网络预测模型. 首先, 将自相关法和Cao方法相结合对混沌时间序列进行相空间重构, 以获得重构延迟时间向量; 其次, 以Hermite正交基函数为激励函数构成Hermite正交基神经网络, 作为预测模型; 最后, 将模型参数优化问题转化为多维空间上的函数优化问题, 利用改进教学优化算法对预测模型进行参数优化, 以建立预测模型并进行预测分析. 分别以Lorenz 系统和Liu系统为模型, 通过四阶Runge-Kutta法产生混沌时间序列作为仿真对象, 并进行单步及多步预测对比实验. 仿真结果表明, 与径向基函数神经网络、回声状态网络、最小二乘支持向量机及基于教学优化算法的Hermite正交基神经网络预测模型相比, 所提预测模型具有更高的预测精度、更快的收敛速度和更简单的模型结构, 验证了该模型的高效性, 便于推广和应用.     

一种新型广义RBF神经网络在混沌时间序列预测中的研究

提出了一种新颖的广义径向基函数神经网络模型,其径向基函数(RBF)的形式由生成函数确定.然后,给出了易实现的梯度学习算法,同时为了进一步提高网络的收敛速度和网络性能,又给出了基于卡尔曼滤波的动态学习算法.为了验证网络的学习性能，采用基于卡尔曼滤波算法的新型广义RBF网络预测模型对Mackey-Glass混沌时间序列和Henon映射进行了仿真.结果表明,所提出的新型广义RBF神经网络模型能快速、精确地预测混沌时间序列,是研究复杂非线性动力系统辨识和控制的一种有效方法. 关键词： 广义径向基函数神经网络 卡尔曼滤波 梯度下降学习算法 混沌时间序列 预测     

基于果蝇算法和灰色神经网络的电子装备故障预测方法研究

针对电子装备故障数据小样本、非线性的特点,在相空间重构处理原始时间序列数据的基础上,基于k折交叉验证和果蝇算法优化灰色神经网络模型参数,从而提出一种基于果蝇算法和灰色神经网络的故障预测方法,并以某型雷达高压电源监测数据仿真结果为例验证其模型性能;实验结果表明,相比已有方法,该方法在全局优化、收敛速度、预测精度方面都具有一定优势。     

基于一种新型聚类算法的RBF神经网络混沌时间序列预测

运用两阶段学习方法构建径向基函数(RBF)神经网络模型预测混沌时间序列.在利用非监督学习算法确定网络隐层中心时,提出了一种基于高斯基的距离度量,并联合输入输出聚类的策略.基于Fisher可分离率设计高斯基距离度量中的惩罚因子,可以提高聚类的性能.而输入输出聚类策略的引入,建立了聚类性能与网络预测性能之间的联系.因此,根据本文方法构建的网络模型,一方面可以加快网络训练的速度,另一方面可以提高预测性能.将该方法对Mackey-Glass, Lorenz和Logistic混沌时间序列进行了预测仿真研究,仿真结果表明了该方法的有效性. 关键词： 混沌时间序列 预测 径向基神经网络 聚类     

基于高斯过程的混沌时间序列单步与多步预测

针对混沌时间序列单步和多步预测,提出基于复合协方差函数的高斯过程 (GP)模型方法.GP模型的确立由协方差函数决定,通过对训练数据集的学习,在证据最大化框架内,利用矩阵运算和优化算法自适应地确定协方差函数和均值函数中的超参数.GP模型与神经网络、模糊模型相比,其可调整参数很少.将不同复合协方差函数的GP模型应用在混沌时间序列单步及多步提前预测中,并与单一协方差函数的GP、支持向量机、最小二乘支持向量机、径向基函数神经网络等方法进行了比较.仿真结果表明,基于不同复合协方差函数的GP方法能精确地预测混沌时间序 关键词： 高斯过程 混沌时间序列 预测 模型比较     

基于遗传神经网络的伺服机构健康状态预测

为了及时把握伺服机构的健康状态,为装备的管理维护与任务执行提供必要的决策支持,从装备的自然退化趋势出发,提出了一种基于遗传算法优化BP神经网络的预测模型。利用BP神经网络优秀的非线性映射能力构造预测模型,将神经网络初始权值阈值编码,利用改进的自适应遗传算法确定最优解。将该模型应用到伺服机构的健康状态预测上,并与标准BP神经网络及径向基神经网络做比较。结果表明该模型有更好的预测精度及收敛速度,从而验证了模型的有效性。     

基于SOM和RBF网络的软测量方法研究

针对氢粉碎过程中钕铁硼粉碎状态不可知,为有效预测合金的反应状态,提出了一种基于自组织特征映射(SOM)神经网络和径向基函数(RBF)神经网络结合构建的网络模型。在该模型中,SOM神经网络作为聚类网络,采用无教师学习算法对输入样本进行自组织分类,并将分类中心及其对应的权值向量传递给RBF神经网络,作为径向基函数的中心;RBF神经网络作为基础网络,采用高斯函数作为径向基函数实现从输入到隐含层的非线性映射,输出层则采用有教师学习算法训练网络的权值,从而实现输入层到输出层的线性映射。并以钕铁硼氢粉碎过程合金中氢含量为检测对象,运用上述方法在MATLAB平台上建立了合金中氢含量预测模型,并完成了仿真验证。     

基于自适应变异果蝇优化算法和广义回归神经网络的布里渊散射谱特征提取

针对布里渊光时域反射光纤传感系统散射谱的高精度特征提取的要求,提出了一种基于自适应变异果蝇优化算法和广义回归神经网络的布里渊散射谱特征提取算法。不仅利用了广义回归神经网络在逼近能力、学习速度、模型的泛化等方面具有的优势,而且采用搜索能力较强的自适应变异果蝇优化算法进一步增强了神经网络的学习能力,从而提高了布里渊散射谱的拟合度和频移提取的准确度。在布里渊散射谱中心频率为11.213 GHz,线宽为40～50,30～60和20～70 MHz的散射谱白噪声实验模型中,将新算法分别与基于有限元分析的Levenberg-Marquardt拟合法、粒子群优化和拉凡格式混合拟合法、最小二乘法进行预测比较,新算法获得的最大拟合频移误差为0.4 MHz,平均拟合度为0.991 2,均方根误差为0.024 1。仿真结果表明所提出的算法拟合度较好,绝对误差小。因此,将此算法用于基于布里渊光时域反射的分布式光纤传感系统,可有效提高布里渊散射谱的拟合度和频移提取的准确度。     

Liley模型的模拟EEG信号的非线性预测和分析

分析Liley模型的模拟脑电(Electroencephalogram,EEG)信号的非线性预测和径向基函数(Radial Basis Functions,RBF)神经网络预测,利用相图分析和非线性正交预测(Nonlinear Cross-Prediction,NLCP)方法研究模拟EEG信号.结果发现:①RBF神经网络预测的效果要好于非线性预测;②NLCP方法对含有强周期分量的高维系统具有较好的适用性;③支持了EEG中存在混沌运动的观点.     

Guided-Wave Two-Dimensional Acousto-Optic Scanner Using Proton-Exchanged Lithium Niobate Waveguide

Two-dimensional scanning of a 0.6328 mum guided-light beam has been realized using noncolinear acousto-optic (AO) coplanar Bragg diffraction together with colinear AO guided-mode to substrate radiation-mode conversion in a Z-cut Xpropagation LiNbO3 proton-exchanged (PE) waveguide. The two surface acoustic (SAW) waves utilized are at the center frequencies of 500 and 200 MHz, propagating in the Y and X axes, respectively. Two-dimensional scanning of approximately 720 resolvable light beam spots, namely, 18 40 (horizontal vertical) scanning, has been demonstrated using a light beam of 1.0 mm aperture. The total number of resolvable beam spots can be greatly increased from 720 by simply utilizing SAW transducers of larger bandwidth and a light beam of greater aperture. It should also be possible to significantly increase the diffraction efficiency from 3 % by optimizing the parameters of the PE waveguide and the SAWs.     

Hybrid Wavelength and Code Division Multiple Access in Optical Networks

A review of schemes for multiple access in fiber optic networks shows that a hybrid of wavelength and code division multiple access (WCDMA) combines the best features of both. In particular, the hybrid scheme retains the large information carrying capacity of wavelength division multiple access (WDMA) and flexibility of code division multiple access (CDMA). In this paper WDMA, optical CDMA (OCDMA), and WCDMA networks are discussed. In OCDMA networks, concept of incoherent and coherent coding including inverse decoding and matched filter is introduced. The delay performance of both WDMA and WCDMA networks, under the simple suboptimum access protocols based on cyclic search, is computed. It has been shown quantitatively that tuning delay significantly affects the delay performance of both WDMA and WCDMA networks. Futhermore, delay performance of WCDMA networks is always better than the WDMA networks for the same tuning delay, load, and number of users.     

Effect ofγ-radiation on the spectral luminescence characteristics of impure cadmium tungstate crystals

Results of a study of the effect of γ-radiation on the spectral luminescence properties of cadmium tungstate crystals doped with silver, bismuth, and molybdenum cations are presented. Spectral characteristics of the nondnnnoped crystals are briefly described. Absorption and photo and X-ray luminescence spectra of the crystals taken before and after exposure to γ-radiation (5.5·10⁴ Gy) are compared. It is found that the spectral characteristics of the crystals doped with silver, bismuth, and molybdenum cations do not change markedly after the exposure. The relation between the type of impurity-induced defects, individual characteristics of the impurity cations, and the character of the effect of γ-radiation on the spectral luminescence properties of impure crystals is analyzed (preliminarily). Translated from Zhurnal Prikladnoi Spektroskopii, Vol. 64, No. 1, pp. 55–60, January–February, 1997.     

Research on beamforming method for dual-line array with space mismatch between the two line arrays

The performance of port/starboard discrimination of dual-line array will be seriously degraded when the space between the two line arrays of the dual-line array is mismatched. The directivity function of the delay-sum and delay-subtract beamforming is deduced and the port/starboard discrimination of the delay-sum and delay-subtract beamforming is also studied both under the ideal status and the status under which the space between the two line arrays of the dual-line array is mismatched. Analysis results show the delay-subtract beamforming has better performance of port/starboard discrimination. At last, the sea trial results demonstrate the theoretical results.     

Retrieval of ultraviolet skylight radiances and O3 slant column densities from balloon-borne limb spectrometer

With a novel light-weight and absolutely calibrated ultraviolet （UV） spectrometer, UV skylight radiances and O3 slant column densities are measured by balloon-borne limb measurements in Xinjiang area, China. UV skylight radiances measured at the height of 31 km are compared with the results from Modtran in the wavelength range from 290 to 420 nm. O3 slant column densities are derived from radiance spectra in the Huggins bands （320 - 335 nm） using differential optical absorption spectroscopy method. And the parameter exhibits a good correlation with the same value simulated by radiative transfer model （Tracy）. The O3 profile simultaneously measured by an O3 sonde is used as input in Tracy calculations. The O3 sonde is launched on the same balloon.     

Design optimization of highly sensitive LSPR enhanced surface plasmon resonance biosensors with nanoholes

For breaking through the sensitivity limitation of conventional surface plasmon resonance (SPR) biosensors, novel highly sensitive SPR biosensors with Au nanoparticles and nanogratings enhancement have been proposed recently.But in practice, these structures have obvious disadvantages.In this study, a nanohole based sensitivity enhancement SPR biosensor is proposed and the influence of different structural parameters on the performance is investigated by using rigorous coupled wave analysis (RCWA).Electromagnetic field distributions around the nanohole are also given out to directly explain the performance difference for various structural parameters.The results indicate that significant sensitivity increase is associated with localized surface plasmons (LSPs) excitation mediated by nanoholes.Except to outcome the weakness of other LSP based biosensors, larger resonance angle shift, reflectance amplitude, and sharper SPR curves' width are obtained simultaneously under optimized structural parameters.     

Polymer gratings based on photopolymerization for low-order distributed feedback polymer lasers

Novel polymer distributed feedback(DFB)gratings are fabricated based on photopolymerization to reduce lasing threshold of polymer lasers.A photopolymer formulation sensitive to 355-nm ultraviolet(UV)light is proposed for the fabrication of polymer gratings and it can be used to form polymer films by spin-coating process.A very low surface-relief depth ranging from 12.5 to about 1.0 nm has been demonstrated with a refractive-index modulation of about 0.012.The experimental results indicate that such polymer gratings have promising potentials for the fabrication of low-order DFB organic semiconductor lasers.     

A novel methodology for photometric compensation of projection display on patterned screen

We propose a novel methodology based on the projector-camera (ProCam) system to address the photometric compensation issue for the projection display on the patterned screen.The patterned screen is treated as the combination of a perfect white screen and a color modulator.The perfect white screen is used to automatically and accurately characterize the ProCam system offline using the polynomial model,and the parameters of the color modulator can be efficiently recovered by employing only two gray images based on the linear reflectance model.The experimental results show that the color artifacts of the display image can be greatly improved with this methodology,which demonstrates its feasibility and validity in the photometric compensation.     

Direction-of-arrival estimation using matrix spatial prefiltering approach

An approach to direction-of-arrival （DOA） estimation for multiple narrowband farfield signals is proposed. The technique uses a novel matrix spatial prefiltering approach. Specifically, a matrix filter is designed to spatially filter the incoming data snapshots. The un-wanted components arriving from the stopband angular sectors are attenuated and the desired components from the angular sector of interest pass with minimal distortion. The matrix filter spatially filters the element-space data and the output reserves the element-space data property, which makes it very useful by passing sensor data through a spatial prefilter prior to applying many other array processors to attenuate interferences and improve system performance. Several examples of DOA estimation problem are presented to illustrate the performance of the proposed spatial prefiltering approach. Results of simulation and real data show that the pre-filter can efficiently attenuate the spatial interferences and significantly improve the estimation and resolution capability of DOA estimators at low signal-to-noise ratios for the sources located inside the passband sector. In addition, the use of spatial prefilter makes it possible to estimate DOAs for multiple sources more than the number of the elements of an array.     

Quantum efficiency calibration of opto-electronic detector by means of correlated photons method

A new calibration method of detectors can be realized by using correlated photons generated in spontaneous parametric down-conversion (SPDC) effect of nonlinear crystal.An absolute calibration system of detector quantum efficiency is performed.And its principle and experimental setup are introduced.A continuouswave (CW) ultraviolet (351 nm),diode-pumped,frequency-doubled,and solid-state laser is used to pump BBO crystal.The quantum efficiencies of the photomultiplier at 633,702,and 789 nm are measured respectively.The coincidence peaks are observed using coincidence circuit.Some measurement factors including the filter bandwidth of trigger channel,the detector position alignment and polarization of the pump light are analyzed.The uncertainties of this calibration method are also analyzed,and the relative uncertainties of total calibration are less than 5.8%.The accuracy of this method could be improved in the future.