Study on the optical path difference of aero-optical window

To study the optical path difference (OPD) of optical widow in which light beams transmit under the aerodynamic thermal environment, a grid model is introduced. Based on it, the aero-optical effect of the side-mounted window is analyzed in detail combining the optical transmission theory and the numerical simulation method. Firstly, the temperature and stress of the window are simulated, and then the OPD is obtained through the refractive index. To demonstrate it, temperature field, deformation field and stress field of the window are calculated and analyzed. The influences of thermal shock and atmosphere pressure on the OPD are discussed.     

湍流边界层激励下高速列车车窗参数研究*

高速列车运行过程中,由于结构表面的不平整性产生的湍流边界层激励（TBL）对车厢内部的声场环境产生了一定的影响。采用非相关壁面平面波技术（UWPWs）将随机激励（TBL）转化为一组壁面平面波,利用该激励模型模拟获取了TBL激励下高速列车车窗表面的壁面压力,并将其加载到车窗有限元结构上构建了TBL激励下高速列车车窗模型。分别调查了空腔厚度、双侧玻璃厚度比以及空腔阻尼损耗因子三个参数对车窗声振特性的影响。结果表明：当空腔厚度为20mm,双侧玻璃厚度比为9:5,空腔阻尼损耗因子为0.05时,分别为各自参数下调查工况中的最优解。基于参数调查的结果对车窗结构进行了优化,优化后车窗结构的速度级响应在分析频段内较之前有了显著的削弱,辐射声功率级整体下降了2.8dB。     

Roughness sub-layer investigation of a turbulent boundary layer

Particle image velocimetry (PIV) measurements at relatively low Reynolds numbers in a turbulent boundary layer over a three-dimensional roughened surface, consisting of pyramidal rows, have been presented. Measurements have been taken in a streamwise wall-normal plane intercepting the apex of a row of pyramids and the diagonal of the square base. The results shown in this paper point out the non-homogeneity of the flow in the roughness sub-layer. The different flow behaviour along the ascendant and the descendent part of the pyramids and in the region between two consecutive pyramids has been visualised. Low values of the streamwise component of the mean velocity and high values of the streamwise and wall-normal component of the Reynolds normal stresses, turbulent kinetic energy and Reynolds shear stresses are present in the downstream part of the pyramids, near their base. The mean representation of the flow shows swirling patterns in correspondence of the top of the pyramids. It is suggested that this representation is produced mainly by vortical structures travelling along the wall, retarded and intensified when interacting with the flow around the roughness element and by swirling patterns originating from the interaction of the incoming flow with the pyramids. In order to understand the origin of the high turbulence activity observed downstream the pyramids, near their base, a conditional analysis based on the quadrant method has been performed. It has been shown that sweep events are the major contributor to the very high values of the Reynolds shear stresses observed in this region and are due to the combination of the vortical flow at the top of the pyramid and a large scale (with respect to the roughness element) inflow motion.     

Adaptive control of chaotic systems based on a single layer neural network

This Letter presents an adaptive neural network control method for the chaos control problem. Based on a single layer neural network, the dynamic about the unstable fixed period point of the chaotic system can be adaptively identified without detailed information about the chaotic system. And the controlled chaotic system can be stabilized on the unstable fixed period orbit. Simulation results of Henon map and Lorenz system verify the effectiveness of the proposed control method.     

正补态光学神经网络模型性能评估

本文对一种多值光学神经网络模型——正补态模型的性能,存贮容量、容错性及收敛性进行了评估,并在N=16的网络上进行了计算机模拟.结果表明,正补态模型的性能比其它光学神经网络模型有所改善.     

Universal form of the power spectrum of the aero-optical aberration caused by the supersonic turbulent boundary layer

Based on the measurement of one-dimensional （1D） optical path difference （OPD） of the supersonic turbulent bound- ary layer, an analytical form for the power spectrum of the two-dimensional （2D） OPD is obtained with its structure function and under the locally homogeneous isotropic assumption. The universality of this spectrum is argued, and its validity is checked by the comparison with experimental result. The potential applications of this model in theoretical and numerical studies are emphasized. Another contribution of this work is around the application of correlation function to analyzing the statistics of OPD. Based on our results and other results published elsewhere, we show that the OPD is often not stationary, and one should be cautious about using this tool.     

Synchronization regimes in a map-based model neural network

The dynamical activity of a neural network model composed of electrically connected map-based neurons is investigated. After detailing the behavior of the isolated neuron for a wide parameter range, collective network states are depicted using the activity, spatial correlation and time phase distribution as measures. A detailed discussion on the stability of global and partial synchronization states is presented.     

Temporal evolution of optical path difference of a supersonic turbulent boundary layer

The density distribution of a supersonic turbulent boundary layer is measured with the nanoparticle-based planar laser scattering technique, and the temporal evolution of its optical path difference (OPD) in a short time interval is characterized by proper orthogonal decomposition (POD). Based on the advantage of POD in capturing the energy of a signal, a temporal evolution model is suggested for the POD coefficients of OPD. In this model, the first few coefficients vary linearly with time, and the others are modeled by Gaussian statistics. As an application, this method is used to compute the short-exposure optical transfer function.     

The influence of boundary layer on sound propagation in optical fibers

Physical models of two-layered and three-layered fiber were built. These models were used to calculate acoustic properties of optical fibers. Acoustic properties of fibers with boundary layer and without boundary layer calculated from these model were compared. The propagation of acoustic cylindrical symmetric waves in optical fibers was considered. This problem was treated analytically and numerically in continual approximation. The possibility of simultaneous propagation of two cylindrical symmetric waves in optical fibers with a boundary layer was shown (in a three-layered model). The physical phenomenon of the presence of two waves in the fiber with boundary layer is proposed for study of the boundary layer. Formulas for the calculation of the properties of the boundary layer from the acoustic experiment, verified by numerical calculations, are represented.     

多值神经网络改进模型及其光学实现

本文提出一种改进的光学神经网络模型,并利用空间光调制器PROM构成的光学系统实现了这种模型的联想记忆运算.计算机模拟和实验结果表明,改进模型提高了光学神经网络的识别能力,并在—定程度上提高了存贮容量.     

Temporal evolution of the optical path difference of a supersonic turbulent boundary layer

The density distribution of a supersonic turbulent boundary layer is measured with the nanoparticle-based planar laser scattering technique, and the temporal evolution of its optical path difference (OPD) in a short time interval is characterized by proper orthogonal decomposition (POD). Based on the advantage of POD in capturing the energy of a signal, a temporal evolution model is suggested for the POD coefficients of the OPD. In this model, the first few coefficients vary linearly with time, and the others are modeled by Gaussian statistics. As an application, this method is used to compute the shortexposure optical transfer function.     

A multi-target approach for isolated word recognition using artificial neural network

I.TntroductionStatisticalandneuralnetworkmcthodsforpatternclassiflcationusesignificantlydifTerentapproachesintrainingaclassificr.Inthestatistica1approach,thcformationofaclassifier1arge1ydependsonthestatisticsofthetrainingpatternsand,insomecases,theassumptionsaboutthedistributionofthepopulation.Theneuralnetworkmethodisnon-parametricandcanbeadaptivcinthetrainingprocessl'l.Becauseofitssimplicityandflexibility,theneuralnetworkhasbecnincreasing1yusedforpatternclassiflcation.Itisnowwe11knownthatan…     

Global Hopf bifurcation analysis on a BAM neural network with delays

A delayed differential equation that models a bidirectional associative memory (BAM) neural network with four neurons is considered. By using a global Hopf bifurcation theorem for FDE and a Bendixon's criterion for high-dimensional ODE, a group of sufficient conditions for the system to have multiple periodic solutions are obtained when the sum of delays is sufficiently large.     

An enhanced non-uniformity correction algorithm for IRFPA based on neural network

Influenced by detector materials’ non-uniformity, growth and etching techniques, etc., every detector’s responsivity of infrared focal plane arrays (IRFPA) is different, which results in non-uniformity of IRFPA. And non-uniformity of IRFPA generates fixed pattern noises (FPN) that are superposed on infrared image. And it may degrade the infrared image quality, which greatly limits the application of IRFPA. Non-uniformity correction (NUC) is an important technique for IRFPA. The traditional non-uniformity correction algorithm based on neural network and its modified algorithms are analyzed in this paper. And a new improved non-uniformity correction algorithm based on neural network is proposed in this paper. In this algorithm, the desired image is estimated by using three successive images in an infrared sequence. And blurring effect caused by motion is avoided by applying implicit motion detection and edge detection. So the estimation image is closer to real image than the estimation image estimated by other algorithms, which results in fast convergence speed of correction parameters. A comparison is made to these algorithms in this paper. And experimental results show that the algorithm proposed in this paper can correct the non-uniformity of IRFPA effectively and it prevails over other algorithms based on neural network.     

A novel multilevel coding technique for high speed optical fiber communication systems

This paper delineates a novel coding technique, a combination of coding and multiplexing technique, known as mapping multiplexing technique (MMT). A mapping algorithm is used to multiplex four channels, each of which running at 10 Gbit/s. Up to four bits per symbol at a symbol rate of 10 Gboud was successfully transmitted over a single wavelength. Compare to the conventional return to zero (RZ) and none-return to zero (NRZ)-time division multiplexing (TDM) more than 75% and 50% reduction in the spectral width are achieved, respectively. Calculated chromatic dispersion tolerance of ±130 ps/nm is achieved for the 40 Gbit/s MMT transmission system at the expense of ∼2 dB degradation in the receiver sensitivity (compared to NRZ) due to an increase in the number of levels and in the signal dependency of signal-spontaneous beat noise. In comparison with conventional 4-ary system, the proposed system depicts ∼6.5 dB improvement in terms of receiver sensitivity.     

Nonlinear experimental dye-doped nematic liquid crystal optical transmission spectra estimated by neural network empirical physical formulas

In this paper, two complementary objectives related to optical transmission spectra of nematic liquid crystals (NLCs) were achieved. First, at room temperature, for both pure and dye (DR9) doped E7 NLCs, the 10-250 W halogen lamp transmission spectra (wavelength 400-1200 nm) were measured at various bias voltages. Second, because the measured spectra were inherently highly nonlinear, it was difficult to construct explicit empirical physical formulas (EPFs) to employ as transmittance functions. To avoid this difficulty, layered feedforward neural networks (LFNNs) were used to construct explicit EPFs for these theoretically unknown nonlinear NLC transmittance functions. As we theoretically showed in a previous work, a LFNN, as an excellent nonlinear function approximator, is highly relevant to EPF construction. The LFNN-EPFs efficiently and consistently estimated both the measured and yet-to-be-measured nonlinear transmittance response values. The experimentally obtained doping ratio dependencies and applied bias voltage responses of transmittance were also confirmed by LFFN-EPFs. This clearly indicates that physical laws embedded in the physical data can be faithfully extracted by the suitable LFNNs. The extraordinary success achieved with LFNN here suggests two potential applications. First, although not attempted here, these LFNN-EPFs, by such mathematical operations as derivation, integration, minimization etc., can be used to obtain further transmittance related functions of NLCs. Second, for a given NLC response function, whose theoretical nonlinear functional form is yet unknown, a suitable experimental data based LFNN-EPF can be constructed to predict the yet-to-be-measured values.     

Response of a panel to a supersonic turbulent boundary layer: studies on a theoretical model

The response of a clamped panel to a supersonic turbulent boundary layer is studied on the basis of a recently developed theoretical model. This model, in the form of an integro-differential equation, incorporates the effect of coupling between the panel motion and the flow of the surrounding fluid. A Ritz-Galerkin method is used to obtain approximate solutions for the statistics of the panel response to the turbulence. Comparisons of the numerical results with previous experimental data are presented and assessments of the theoretical model in the light of such comparisons are made.     

Complex object 3D measurement based on phase-shifting and a neural network

An accurate phase-height mapping algorithm based on phase-shifting and a neural network is proposed to improve the performance of the structured light system with digital fringe projection. As phase-height mapping is nonlinear, it is difficult to find the best camera model for the system. In order to achieve high accuracy, a trained three-layer back propagation neural network is employed to obtain the complicated transformation. The phase error caused by the non-sinusoidal attribute of the fringe image is analyzed. During the phase calculation process, a pre-calibrated phase error look-up-table is used to reduce the phase error. The detailed procedures of the sample data collection are described. By training the network, the relationship between the image coordinates and the 3D coordinates of the object can be obtained. Experimental results demonstrate that the proposed method is not sensitive to the non-sinusoidal attribute of the fringe image and it can recover complex free-form objects with high accuracy.