基于电子倍增电荷耦合器件的哈特曼-夏克波前传感器质心探测误差分析

哈特曼-夏克(Hartmann-Shack,H-S)波前传感器的探测误差是自适应光学系统中的一个主要误差源.本文分析了电子倍增电荷耦合器件(electron multiplying charge-coupled devices,EMCCD)的各种噪声源对H-S波前传感器质心探测误差的影响,推导了基于EMCCD的H-S波前传感器的质心探测误差的数学表达式,并进行了数值仿真和实验研究.结果表明,基于EMCCD的H-S波前传感器的质心探测误差与增益直接相关,在增益小于P_max/(Q·N_mean)时质心探测误差随着增益的增大而减小,在增益大于P_max/(Q·N_mean)时质心探测误差随着增益的增大而增大,其中P_max是EMCCD的峰值电荷,Q是量子效率,N_mean是平均入射光强. 关键词： 自适应光学 电子倍增电荷耦合器件 波前传感器 质心探测误差     

Ab initio study on N,N′,N″‐triaminoguanidine

Electronic structure calculations and second‐order delocalizations in N,N′,N′′‐triaminoguanidine ( TAG ) have been studied by employing ab initio MO and density functional methods. There are total 10 rotational isomers on the potential energy (PE) surface of TAG . The effect of three amino groups substitution on guanidine ( Gu ) has been studied in terms of the primary and the secondary electron delocalizations in TAG by employing Natural Population Analysis (NPA). An increased electron delocalization is observed in protonated triaminoguanidine ( TAGP ) due to the three strong intramolecular hydrogen bonds and hence accounts for its extra stability. The increase in the electron delocalization upon protonation in TAG can be compared to that in guanidine. The absolute proton affinity (APA) of TAG is less than that of Gu . HOMA and NICS studies have been carried out to understand electron delocalization in TAGP . Copyright © 2007 John Wiley & Sons, Ltd.     

Deterministic Motion of the Controversial Piston in the Thermodynamic Limit

We consider the evolution of a system composed of N non-interacting point particles of mass m in a cylindrical container divided into two regions by a movable adiabatic wall (the adiabatic piston). We study the thermodynamic limit for the piston where the area A of the cross-section, the mass M of the piston, and the number N of particles go to infinity keeping A/M and N/M fixed. The length of the container is a fixed parameter which can be either finite or infinite. In this thermodynamic limit we show that the motion of the piston is deterministic and the evolution is adiabatic. Moreover if the length of the container is infinite, we show that the piston evolves toward a stationary state with velocity approximately proportional to the pressure difference. If the length of the container is finite, introducing a simplifying assumption we show that the system evolves with either weak or strong damping toward a well-defined state of mechanical equilibrium where the pressures are the same, but the temperatures different. Numerical simulations are presented to illustrate possible evolutions and to check the validity of the assumption.     

The performance of in situ adaptive tabulation in computations of turbulent flames

This paper presents a detailed characterization of the local and global errors associated with the in situ adaptive tabulation (ISAT) algorithm, which is used in conjunction with a transported PDF method. Calculations of a non-premixed turbulent methane/air piloted jet flame (Sandia flame D) using a skeletal chemical mechanism were performed using ISAT coupled with the computational fluid dynamics (CFD) code FLUENT. The three strategies implemented in ISAT for the growing of the ellipsoids of accuracy (EOAs) are discussed, and the cumulative distribution function (CDF) of the local error is presented for each of the three growing strategies. Computations are also performed to characterize the global error in the ISAT/PDF calculation. The computations used to characterize the global error were performed in parallel to achieve substantial savings in computational time.

In general the local error is well controlled, but there is a small probability of relatively large errors. Results from the investigation suggest that large retrieve errors are due to the region of accuracy (ROA) being non-convex, where the ROA is the connected region for which the error does not exceed the error tolerance, ?_tol. The global error in ISAT is found to be small compared to statistical error for ?_tol ≤ 10^?4, and is found to vary linearly with ?_tol.     

Dependence of the profile of a hologram recorded on color centers in a CaF2 crystal on the type of hologram-forming centers

A hologram recorded in a CaF₂ crystal by modulating the concentration of simple centers (F, M, R, and N) and highly aggregated (mainly colloidal) color centers has been exposed to incoherent 365-nm light to a dose of about 4000 J/cm² at a temperature of 80°C. As a result of this exposure, the color centers were almost completely transformed into quasi-colloidal centers. The hologram was retained, and its diffraction efficiency increased; the intensity and number of observed diffraction orders increased as well. Analysis of the angular dependences of the diffraction response made it possible to construct a spatial profile of the hologram, which turned out to be much narrower in comparison with the initial-hologram profile; both profiles differ from a sinusoidal one. The hologram??s stability to such a large exposure to incoherent light at an elevated temperature is related to the peculiar diffusion-drift mechanism of hologram recording on color centers in ionic crystals.     

Ground State and Spin-Glass Phase of the Large-N Infinite-Range Spin Glass via Supersymmetry

The large-N infinite-range spin glass is considered, in particular, the number of spin components k needed to form the ground state and the sample-to-sample fluctuations in the Lagrange multiplier field on each site. The physical significance of k for the correlation functions is discussed. The difference between the large-N and spherical spin glass is emphasized; a slight difference between the average Lagrange multiplier of the large-N and spherical spin glasses is derived, leading to a slight increase in the energy of the ground state compared to the naive expectation. Further, there is a change in the low-energy density of excitations in the large-N system. A form of level repulsion, similar to that found in random matrix theory, is found to exist in this system, surviving interactions. Even though the system is an interacting one, a supersymmetric formalism is developed to deal with the problem of averaging over disorder.     

Chemical reactions as dynamical systems on the interval

We consider the most general chemical reaction of the typen ₁ A ₁+...+n _N A _N m ₁ B ₁+...+m _M B _M whereN,M1,n ₁,...,n _N andm ₁,...,m _M are positive integers defining the stoichiometry, andA ₁,...,A _N andB ₁,...,B _M are the names of chemicals or ions. We assume that _i=1 ^N n_i= _j=1 ^M m_j. The time evolution of the concentrations is given by the law of mass action and leads to a dynamical system (with discrete or continuous time) which is governed by a polynomial map of the interval [B, C], where B0 and C1. We define the physically meaningful range for the parameters of the map, and we show that, within such a range, the map has a unique fixed point, which is stable and a global attractor, with the exception of one particular case, where bifurcation is observed.     

基于微种群遗传算法和自适应BP算法的遥感图像分类

介绍了采用微种群遗传算法和自适应BP算法相结合的混合遗传算法来训练前向人工神经网络(BPNN)的方法。即先用微种群遗传学习算法进行全局训练,再用自适应BP算法进行精确训练,以达到加快网络收敛速度和避免陷入局部极小值的目的。将此算法用于遥感图像分类,网络的训练速度及分类结果表明,该算法收敛速度较快,预测精度较高。     

Modal prediction of atmospheric turbulence wavefront for open-loop liquid-crystal adaptive optics system with recursive least-squares algorithm

For an open-loop liquid-crystal adaptive optics system, its performance is mainly limited by the time delay. We propose a new modal recursive least-squares (RLS) predictive algorithm to overcome this problem. The RLS algorithm has a simple architecture, low computational complexity and a high converging speed. The impact of the number of the foreword prediction frame N and the number of the predictor order M of the RLS predictor is analyzed in detail. The results show that the best foreword prediction frame N must be equal to the system loop delay frame SLDF. The appropriate predictor order M of the RLS predictor is equal to 2 or 3 when there is no measurement noise and it depends on noise ratio N_R when the measurement error cannot be neglected. We present numerical simulations to show the significant improvements brought by the RLS predictor.     

Hartree theory for matter in a photon or phonon field

For a Dicke-type Hamiltonian describing M two-level systems coupled to N modes of a Boson field, the expectation value is minimized with respect to unrestricted products of 2^M-level states and field states. In the resulting Hartree ground state(s) ? ? Π, ? is ground state of a cubic Schrödinger equation in $\text{C}$^2M and Π is the product of N coherent one-mode states depending on ?. It is proved that ? factorizes as well, into M two-level states determined by a nonlinear equation in $\text{R}$^M; that for weak coupling the Hartree ground state is unique and independent of the parameters in H; and that for strong coupling there are consecutively 2,…,2^L (and possibly even more) Hartree ground states where 1≤L≤min{M, 2N} counts certain reflection symmetries. Details of this symmetry-breaking bifurcation (such as structural stability of relaxed two-level systems) and connections to the true ground state(s) of H are worked out.     

Stability of Ultraviolet-Cutoff¶Quantum Electrodynamics with Non-Relativistic Matter

We prove that the quantum-mechanical ground state energy of a system consisting of an arbitrary number, M, of static nuclei of atomic number ≤Z and of an arbitrary number, N, of Pauli electrons interacting with the quantized, ultraviolet-cutoff radiation field is bounded below by $-K.M, where K is a finite constant depending on Z, on the finestructure constant α and on the ultraviolet cutoff Λ, with , as , and K' independent of Λ. Received: 4 September 1996/ Accepted: 9 April 1997     

Study of bifurcations in a model Hartree-Fock calculation

The Hartree-Fock procedure is used to study the behaviour of the ground state of a system ofM spinless electrons distributed overN equivalent and equidistant sites (M ⩽N) as a function of the strength of the mutual repulsion between the electrons. Below a critical strength, all initial configurations are seen, after repeated iterations, to converge to a unique solution. Above this critical strength, in addition to the initial configurations which lead to a unique solution, there exist configurations which on repeated iterations give rise to stable two-period solutions. Although the number of independent stable two-period solutions depends on the coupling strength, for no value of the coupling are stable solutions of periodicity higher than two seen.     

Rotation selectivity of random phase encoding in volume holograms

We present the study of angular selectivity for holographic multiplexing based on random phase encoding by a ground glass. The rotational selectivity of the volume hologram is calculated theoretically and coincides with the experimental measurement. By controlling the parameters including rotational center, effective numerical aperture of both volume hologram and the ground glass, we can obtain different rotational selectivity applied to random phase encoding in volume holographic storage.     

Local and interfacial magnetic properties of inhomogeneous finite linear chains

The Hubbard Model has been used to study the local and interfacial magnetic properties of finite inhomogeneous cluster systems. These are generally of the type, NMMMMMN, where N and M, respectively, refer to non-magnetic and magnetic atoms, of a 7-site finite chain. The applicability of the Hartree–Fock (HF) approximation is gauged via direct comparison of the ground state magnetization results derived from exact diagonalization methods. The underlying HF and exact mechanisms are compared as a function of the model parameters, with particular attention being paid to the local and interfacial (N/M interface) magnetic properties. Regimes, which exhibit favourable comparison between HF and exact results are found. Detailed inspection of the HF prediction is made and general trends established as a function of system size and model parameters.     

Hologram multiplexing in a highly photosensitive photopolymerizable material in a sol-gel matrix

In this paper, we present a photopolymerizable material in a sol-gel matrix suitable for recording a large number of multiplexed diffraction gratings with low total exposure energy. The matrix of this material is an inorganic porous silica glass, a material that does not shrink when radical photopolymerization is initiated. The photosensitive component is based on acrylamide, N,N′-methylenebisacrylamide, triethanolamine and yellow eosine as a photoinitiator. In the reported study, 50 holograms were angularly multiplexed with diffraction efficiencies between 0.1 and 1.9% resulting in a dynamic range of M/# = 3.9. The total exposure was 4.5 mJ/cm² and the energy used in recording each hologram was 95 μJ/cm². This indicates a very high sensitivity for this material in the range of 3.3 to 15.5 cm/mJ. Due to this good holographic performance, the material is suitable for holographic data storage applications.     

Feasibility of gem identification using reflectance spectra coupled with artificial intelligence

Abstract

Standard traditional gem identification requires expert supervision, while sophisticated modern methods are time-consuming and expensive. In contrast, reflectance spectroscopy coupled with artificial intelligence is economical and convenient and does not require specialist supervision. This study established an artificial neural network model that consists of standard multilayered, feed-forward, and back-propagation neural networks, and obtained reflectance spectra of a transparent gem (almandine), an opaque gem (turquoise), several almandine imitations (agate, plastic, and glass), and several treated turquoise samples (dyed, impregnated, and Zachery treated) using an Analytical Spectral Devices spectrometer. The acquired spectra were used to train and test the artificial neural network model. The results show that the model can effectively discriminate between genuine and imitation gems of different classes. However, discrimination between natural and treated gems of same class is not as effective as discrimination of gems of different classes. The results suggest that an artificial neural network based on reflectance spectroscopy could serve as a useful tool for preliminary gem identification, and the advanced identification needs further training and investigation.     

The statistical error of green's function Monte Carlo

The statistical error in the ground state energy as calculated by Green's Function Monte Carlo (GFMC) is analyzed and a simple approximate formula is derived which relates the error to the number of steps of the random walk, the variational energy of the trial function, and the time step of the random walk. Using this formula it is argued that as the thermodynamic limit is approached withN identical molecules, the computer time needed to reach a given error per molecule increases asN ^h where 0.5 <b < 1.5 and as the nuclear chargeZ of a system is increased the computer time necessary to reach a given error grows asZ ^5.5. Thus GFMC simulations will be most useful for calculating the properties of lowZ elements. The implications for choosing the optimal trial function from a series of trial functions is also discussed.     

Retrieval Properties of Bidirectional Associative Memories

We study a two-layer neural network made of N and M(N) neurons, producing a two-way association search for a family of p(N) patterns, where each pattern is a pair of two independent sub-categories of information having respectively N and M(N) components. In terms of the ratio =lim _N M(N)/N, we study the retrieval capability of this network and show that there exists, at least, three regimes of association for which we determine the evolution of the threshold _c () of the storage capacity =lim _N p(N)/N.     

Complex Anti-Self-Dual Connections on a Product of Calabi–Yau Surfaces and Triholomorphic Curves

We study the adiabatic limit of a sequence of Ω-anti-self-dual connections on unitary bundles over a product of two compact Calabi–Yau surfaces M×N by scaling metrics to shrink N to a point. We show that after fixing gauge transformations, a subsequence of the N-components of these connections converges to a triholomorphic curve from M away from a Cayley cycle in M×N to the moduli space of instantons on M×N modulo gauge equivalence in the Hausdorff topology, and converges on the blow-up locus to a family, which is parameterized by the Cayley cycle, of triholomorphic curves from C ² to . Received: 22 May 1998 / Accepted: 26 August 1998