遗传算法BP神经网络的预报研究和应用

针对目前 BP神经网络在实际气象预报应用中 ,网络结构难以确定以及网络极易陷入局部解问题 ,用遗传算法优化神经网络的连接权和网络结构 ,并在遗传进化过程中采取保留最佳个体的方法 ,建立基于遗传算法的 BP网络模型 ,并以广西的月降水量进行实例分析 ,计算结果表明 ,该方法预报精度高、而且稳定 .     

一个数论函数的均值

研究了三进制中数字和函数的均值性质，给出了一般情况下均值A1(N)，A2(N)，A3(N)的求和公式．     

On the relations among best approximation and Fourier coeffcients

We discuss the relations among the best approximation En(f) and the Fourier coeffcients of a function, f(n) ∈ C,n = 0,±1,±2,..., under the conditions that {f (n)}∞n=0 ∈ MVBVS* and {f (n) f(n)}∞n=0 ∈ MVBVS*, where MVBVS* is the class of the so-called Strong Mean Value Bounded Variation Sequences.     

Parabolic Equations with Measurable Coefficients

We investigate the unique solvability of second order parabolic equations in non-divergence form in , p ≥ 2. The leading coefficients are only measurable in either one spatial variable or time and one spatial variable. In addition, they are VMO (vanishing mean oscillation) with respect to the remaining variables. The second author was partially supported by NSF Grant DMS-0140405.     

Chernoff's Theorem and Discrete Time Approximations of Brownian Motion on Manifolds

Let be a one-parameter family of positive integral operators on a locally compact space . For a possibly non-uniform partition of define a finite measure on the path space by using a) for the transition between any two consecutive partition times of distance and b) a suitable continuous interpolation scheme (e.g. Brownian bridges or geodesics). If necessary normalize the result to get a probability measure. We prove a version of Chernoff's theorem of semigroup theory and tightness results which yield convergence in law of such measures as the partition gets finer. In particular let be a closed smooth submanifold of a manifold . We prove convergence of Brownian motion on , conditioned to visit at all partition times, to a process on whose law has a density with respect to Brownian motion on which contains scalar, mean and sectional curvatures terms. Various approximation schemes for Brownian motion on are also given.      

Mean value formulas for classical solutions to uniformly parabolic equations in the divergence form with non-smooth coefficients

We prove surface and volume mean value formulas for classical solutions to uniformly parabolic equations in the divergence form with low regularity of the coefficients. We then use them to prove the parabolic strong maximum principle and the parabolic Harnack inequality. We emphasize that our results only rely on the classical theory, and our arguments follow the lines used in the original theory of harmonic functions. We provide two proofs relying on two different formulations of the divergence theorem, one stated for sets with almost C¹-boundary, the other stated for sets with finite perimeter.     

Weighted isoperimetric inequalities in warped product manifolds

We prove some sharp isoperimetric type inequalities in warped product manifolds, or more generally, multiply warped product manifolds. We also relate them to inequalities involving the higher order mean-curvature integrals. Applications include some sharp eigenvalue estimates, Pólya-Szegö inequality, Faber-Krahn inequality, Sobolev inequality and some sharp geometric inequalities in some warped product spaces.     

A Complex Mode Method for Wind-Induced Responses of 6-Parameter Practical Viscoelastic Damping Energy Dissipation Structures Based on the Davenport Wind Speed Spectrum北大核心CSCD

针对六参数实用黏弹性阻尼耗能结构,基于Davenport风速谱系列响应问题进行了系统的研究.首先,利用六参数黏弹性阻尼器的微分型本构关系,建立了耗能结构基于Davenport风速谱激励下的运动方程;然后,运用复模态法将耗能结构的运动方程由二阶微分方程转化为一阶方程,获得了耗能结构系统对风振激励响应的频域解和功率谱密度函数表达式;最后,利用数学恒等式,基于随机振动理论获得了耗能结构系统在Davenport风速谱激励下的响应和阻尼器受力的解析解.该文方法不仅考虑了结构系统在风振激励作用下全振型展开的结果,表达式较现有结果更为简便,效率及精度更高,且适用于非经典阻尼结构.     

氧化铟锡薄膜的磁感应热声分析

基于优良导体在磁场下的涡流效应理论和固体的热声效应理论,建立了氧化铟锡(ITO)导电薄膜磁-热-声效应的理论模型,推导了导电薄膜热致发声的温度振荡和输出声压表达式。对有基底的ITO导电膜进行了磁场下的热声理论计算和实验测试,结果表明：薄膜的温度振荡值随频率呈上升趋势,与电-热-声模型相比趋势相反;薄膜声压的理论值与实验值在频域内的变化趋势基本吻合,验证了理论模型的正确性。进而,根据磁-热-声的理论模型,分析了线圈相关参数对薄膜声压级的影响,结果表明：薄膜声压级随着线圈匝数的增加而增大,随着薄膜与线圈中心距离的增加而减小,随着线圈半径的增加而减小。文中的研究结果拓展了导电薄膜在扬声器等领域的应用。