激光雷达反演大气边界层高度的优化方法

由于大气中气溶胶垂直分布复杂,常用的梯度法和小波协方差变换法在边界层高度自动连续识别方面仍具有较大的不确定性.选取在浙江金华两个时间段内观测得到的双波长激光雷达数据,将二维矩阵方法应用于复杂大气垂直分布情况下边界层高度的反演,从距离和时间两个维度优化边界层高度的反演结果.研究结果表明,在气溶胶垂直分布出现多层结构的情况...     

融合K-means和熵权法的高鲁棒性大气边界层高度估计方法

针对常用激光雷达边界层高度估计方法在云层或悬浮气溶胶层等复杂大气结构下会产生误判的问题,提出一种融合K-means和熵权法的高鲁棒性大气边界层高度估计方法。选取美国大气辐射测量项目南部大平原站点的微脉冲激光雷达数据,将K-means算法和熵权法应用于多种条件下的边界层高度估计,从初始参数选取和距离计算两个方面提升基于聚类分析的边界层高度的估计性能。实验结果表明：与常用激光雷达边界层高度估计方法相比,所提方法具有较强的抗干扰能力,能更好地追踪复杂大气结构下的边界层高度日变化过程;在晴朗无云天气和复杂大气结构下,其边界层高度的估计值与无线电探空仪边界层高度的测量值基本一致,相关系数分别为0.9718和0.9175。所提方法具有较高的鲁棒性,可以可靠地估计多种条件下的大气边界层高度。     

黑体辐射数值反演的快速稳定算法

讨论采用Tikhonov正则化方法进行黑体辐射数值反演的相关技术问题,特别是将双参数模型提供的相当好的正则参数初值与求解Morozov偏差方程的一个三阶收敛格式相结合,给出了一个新的、具有高效稳定等特点的混合数值反演算法,并进行了数值试验.这一算法在原则上也适合其它许多物理参数识别问题(如Fermi体系的反演、各种类型的Abel方程的数值反演等).     

不确定混沌系统的多项式函数模型补偿控制

针对不确定混沌系统控制问题, 研究了一种基于共轭梯度法(conjugate gradient algorithm, CGA)的多项式函数模型 (polynomial-basis-functions model, PBFM)的补偿控制方法. 该方法首先用PBFM对混沌系统的动力学特性进行拟合, 然后用拟合好的PBFM模型对不确定混沌系统进行前馈补偿控制. 该方法的特点是不需要被控混沌系统的数学模型, 可以快速跟踪任意给定的参考信号. 数值仿真试验表明了该方法不仅具有响应速度快、控制精度高, 而且具有较强的抑制混沌系统参数摄动能力和抗干扰能力. 关键词： 混沌控制 多项式函数模型 共轭梯度法     

一种全球定位系统接收机单点定位的新算法及数值试验

借助于Bancroft方法快速决定全球定位系统中接收机的大概位置，可作为观测方程线性化的初值.采用最小二乘法来求解方程时，观测数据存在较大误差(即出现粗差)的情况下，最小二乘解会出现不稳定性，即问题呈不适定性.为了克服解的不适定性，引入了正则化方法，采用正则化方法及正则化参数最优选择来提高抗粗差能力.实验结果表明，利用这种定位方法的抗粗差能力有一定的改进，这对实时、快速定位有重要意义. 关键词： 全球定位系统 Bancroft方法 最小二乘法 正则化     

基于联合正则化及压缩传感的MRI图像重构

基于压缩传感的MRI图像重构利用图像稀疏的先验知识能从很少的投影值重构原图像。目前MRI重构算法只利用MRI图像稀疏性表示或只利用基于其局部光滑性的先验知识,重构效果不理想。针对此问题,结合两种先验知识,提出一种基于联合正则化及压缩传感的MRI图像重构方法。利用块坐标下降法将求解联合正则化问题转化为交替求解二次凸优化、稀疏正则化和全变差正则化三个简单的优化问题。并提出分别采用共轭梯度法、二元自适应收缩法以及梯度下降法对以上优化问题求解。实验结果表明,该算法重构效果比现有算法有明显地提高。     

基于变分方法的有限区域风场分解与重构I: 理论框架和仿真实验

众所周知, 风场分解与重构最有效的方法就是引入速度势和流函数, 其一般通过求解两个Poisson 方程得到. 由于速度势和流函数在边界上的耦合性质,有限区域风场分解是不唯一的, 这对风场分解带来了很大困难. 本文采用变分伴随结合正则化方法来克服风场分解的不唯一性, 其核心是把速度势和流函数的边值作为控制变量来反演. 目标泛函由两部分组成, 一是衡量重构风场误差大小的观测项; 二是保证风场分解问题适定的正则化项, 其目的在于寻求具有气象意义的稳定正则化解. 数值试验结果表明, 在正确选取正则化参数后, 利用变分伴随结合正则化方法进行有限区域风场分解与重构是有效可行的.     

New expressions for the surface roughness length and displacement height in the atmospheric boundary layer

An alternative model for the prediction of surface roughness length is developed. In the model a new factor is introduced to compensate for the effects of wake diffusion and interactions between the wake and roughness obstacles. The experiments are carried out by the use of the hot wire anemometry in the simulated atmospheric boundary layer in a wind tunnel. Based on the experimental data, a new expression for the zero-plane displacement height is proposed for the square arrays of roughness elements, which highlights the influence of free-stream speed on the roughness length. It appears that the displacement height increases with the wind speed while the surface roughness length decreases with Reynolds number increasing. It is shown that the calculation results based on the new expressions are in reasonable agreement with the experimental data.     

Near-surface turbulence in the atmospheric boundary layer

Measurements of the near-surface turbulence in the atmospheric boundary layer have been made using hot-wire probes above the salt flats of northwestern Utah, where the momentum thickness Reynolds number, R_θ, is O(10⁶), and the surface is smooth and nearly devoid of flow obstructions. The measurements were made with arrays of up to 24 parallel straight sensors and with a modular 12-sensor probe capable of measuring all of the components of the instantaneous velocity vector and velocity gradient tensor. Measurements were also made in a laboratory wind tunnel at R_θ=1730 using 22 straight sensors. The data analysis focuses on the effects of the Reynolds number on turbulence properties and on the physics of the dissipation rate of turbulent kinetic energy.Some properties are found to be dependent on the Reynolds number when normalized with inner variables, while others are not. Among those that show the significant Reynolds number dependence are the rms and the skewness factor of the streamwise velocity fluctuations.Significant differences in flow structure, particularly those related to high rates of dissipation, are implied by the data. The joint PDF and covariance integrand of streamwise and wall normal vorticity fluctuations show less preferred orientation of the vorticity vector in the buffer layer at R_θ of O(10⁶) than at R_θ=1070. The largest contribution to the dissipation rate, at O(10⁶) is by the ∂w/∂z velocity gradient, while this term makes a quite small contribution to the dissipation rate at low R_θ. Here w and z are the spanwise velocity fluctuations and direction, respectively. Conditional analysis in the streamwise-wall normal (x−y) plane based on high instantaneous dissipation rate shows that the typical high dissipation rate events are generally similar at high and low Reynolds numbers, but display some significant differences.     

激光雷达探测北京城区夏季大气边界层

为了研究北京城区夏季大气边界层结构变化特征及大气边界层内气溶胶消光特性,2004年8月利用便携式米散射激光雷达对北京城区夏季大气边界层进行了系统观测。反演了观测站上空大气气溶胶的消光特性垂直分布以及大气边界层的高度。分析了气象条件和人类活动对大气边界层结构的影响。观测数据表明： 北京城区夏季大气边界层有明显的日变化特征,早晚比较低,日间有一个从低变高再变低的过程,中午前后达到最高。结合气象参数对测量数据进行的统计分析表明：北京城区夏季大气边界层高度相对稳定,多分布在1.8km以下,平均值为0.68km;大气边界层内存在浓度较高的气溶胶粒子,平均光学厚度（3km以内）在0.30左右。     

大气边界层各向异性的室内模拟研究

 边界层的各向异性研究对于了解实际大气的湍流特性和经典理论的适用范围均具有较大的意义。利用室内水槽实验模拟了实际大气边界层,使用小波变换方法分析了对数光强起伏谱的特征,将获得的标度指数与理论值进行比较,结果表明混合层在水平方向较为接近各向同性,而垂直方向则呈现一定的各向异性。     

大气边界层各向异性的室内模拟研究

边界层的各向异性研究对于了解实际大气的湍流特性和经典理论的适用范围均具有较大的意义。利用室内水槽实验模拟了实际大气边界层,使用小波变换方法分析了对数光强起伏谱的特征,将获得的标度指数与理论值进行比较,结果表明混合层在水平方向较为接近各向同性,而垂直方向则呈现一定的各向异性。     

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.     

Accurate temperature profiling of the atmospheric boundary layer using an ultraviolet rotational Raman lidar

An ultraviolet rotational Raman lidar (RRL) system at an eye-safe wavelength of 354.7 nm is demonstrated for profiling the vertical temperature of the atmospheric layer. Two sets of narrow-band interference filters, which central wavelengths are located at 353.9 and 352.5 nm, respectively, are employed to block the elastic Mie- and Rayleigh-scattering signals and to separate two required rotational Raman signals for temperature retrieval. Experiments were carried out for verifying the feasibility of the prototype of the ultraviolet RRL from May to November in 2008, and the system calibrations were performed with the radiosonde data which were obtain from the local meteorological bureau, and good agreements were got in the experiment results. The observation results show that a statistical temperature error of less then 1 K was obtained up to the heights of 2.5 km for nighttime and 2.0 km for daytime with 300 mJ laser energy and ∼8 min observation time.     

Observations of the atmospheric boundary layer across the land-sea transition zone using a scanning Mie lidar

Optical properties of the atmospheric boundary layer (ABL) above the land-sea transition interface were measured using a scanning Mie lidar located 30 km away from the Adriatic coast. Based on the two-dimensional range-height-indicator scans, detailed information on the ABL was obtained, including parameters such as atmospheric optical depth, aerosol extinction coefficient and the height of the ABL. The presented case study indicates that the height of the ABL in the land-sea transition zone and the adjacent mountainous region was changing rapidly due to highly variable atmospheric conditions.     

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.