STUDY ON CANONICAL AUTOREGRESSION PREDICTION OF METEOROLOGICAL ELEMENT FIELDS*

Through extension of canonical correlation to the analysis of meteorological element fields (MEF), a concept from combination of canonical autocorrelation with canonical autoregression (CAR) is developed for short-term climatic prediction of MEFs with a formulated scheme. Experimental results suggest that the scheme is of encouraging usefulness to a weak persistence MEF,i.e., rainfall field and, in particular, to a strong persistance one like a SST field.     

系统辨识(3)：辨识精度与辨识基本问题

系统辨识是研究建立系统数学模型的理论与方法.讨论系统辨识涉及的一些基本问题,包括辨识精度、辨识方法的提出,辨识输入信号的设计,参数可辨识性与系统可辨识性,开环可辨识性与闭环可辨识性,可辨识性与能控性和能观测性的关系,可辨识性与输入信号的关系,以及与辨识方法收敛性相关的激励信号与激励条件,辨识算法收敛分析的基本工具,典型辨识算法的收敛结论等.     

丙溴磷对二种海洋微藻的GPx活性及GSH、CAR含量的影响

实验以三角褐指藻和青岛大扁藻为材料 ,进行了丙溴磷农药对海洋微藻的抗氧化防御系统成份— GPx(glutatione Peroxidase)活性和 GSH(glutathione)、CAR(carotinoid)含量的影响研究 ,结果表明 ,在丙溴磷胁迫下 ,微藻的 GPx活性呈现下降趋势 ,GSH和 CAR含量也表现为下降趋势 ,并且胁迫的时间越长、胁迫的强度越大它们下降的幅度也越大。因此微藻细胞对抗氧化防御系统的能力被削弱 ,从而可能导致活性氧在体内的过量产生与积累 ,进而对藻细胞造成更大的伤害。     

Assessing the relationships between elevation and extreme precipitation with various durations in southern Taiwan using spatial regression models

A spatially autocorrelated effect exists in precipitation of a mountainous basin. This study examines the relationship between maximum annual rainfall and elevation in the Kaoping River Basin of southern Taiwan using spatial regression models (i.e. geographically weighted regression (GWR), simultaneous autoregression (SAR), and conditional autoregression (CAR)). Results show that the GWR, SAR, and CAR models can improve spatial data fitting and provide an enhanced estimation for the rainfall–elevation relationship than the ordinary least squares approach. In particular, GWR achieves the most accurate estimation, and SAR and CAR achieve similar performance in terms of the Akaike information criterion. The relationship between extreme rainfall and elevation for longer duration is more concise than that for short durations. Results show that the spatial distribution of precipitation depends on elevation and that rainfall patterns in study area are heterogeneous between the southwestern plain and the eastern mountain area. Copyright © 2011 John Wiley & Sons, Ltd.     

STUDY ON CANONICAL AUTOREGRESSION PREDICTION OF METEOROLOGICAL ELEMENT FIELDS

Through extension of canonical correlation to the analysis of meteorological element fields(MEF), a concept from combination of canonical autocorrelation with canonical autoregression(CAR) is developed for short-term climatic prediction of MEFs with a formulated scheme. Experi-mental results suggest that the scheme is of encouraging usefulness to a weak persistence MEF,i.e., rainfall field and, in particular, to a strong persistance one like a SST field.     

Single epoch ambiguity resolution for Galileo with the CAR and LAMBDA methods

This paper investigates single epoch ambiguity resolution performance using Galileo four frequency data. Two commonly used ambiguity resolution methods are used in the tests, including the Cascade Ambiguity Resolution (CAR) and the Least-Squares Ambiguity Decorrelation Adjustment (LAMBDA) methods. For CAR method, four optimal combinations are carefully selected according to their wavelength to noise ratios and success rate for ambiguity fixing. The test results show, in general, the LAMBDA method performs better than the CAR method. The speed of ambiguity resolution is closely related to the carrier phase measurement precision. With carrier phase measurement precision of 3 mm, single epoch ambiguity resolution can be achieved at every epoch with simulated 1-s interval 24-h Galileo data (total epochs 86,400). With the increase of carrier phase noise, ambiguity resolution performances become worse. When the noise level is increased to 12 mm, single epoch ambiguity resolution can only be achieved about 50% of epochs.     

系统辨识(4)：辅助模型辨识思想与方法

辅助模型辨识思想、多新息辨识理论、递阶辨识原理、耦合辨识概念是该文作者提出的研究辨识问题的原创性新方法,已经被用在很多辨识问题的研究中,形成了不同的辨识方法族,可以用于解决许多线性或非线性模型的自适应信号处理、自适应参数估计、自适应滤波和预测、自适应控制等问题.由于客观事物具有双重属性:一些特征变量是可观测的;一些是不可测的.如果表征系统特征的观测变量都是可测的,就容易建立描述其运动规律的数学模型.客观事物的不可测属性给建立系统数学模型带来特别的困难.在这种情况下,如何利用系统的可测信息,实现对系统未知变量的估算,来建立系统的数学模型,是辨识领域极具挑战性的研究课题.辅助模型辨识思想就是在这样的背景下发展起来的.该文介绍辅助模型辨识思想和一些基于辅助模型的辨识方法.     

CAR模型的阶与时滞的强一致估计

研究随机线性反馈控制系统的结构辨识问题。在已知时滞的下界和模型阶的上界的假定下，通过使修改的Ｂａｙｅｓｉａｎ信息准则最小化，推导出由多输入多输出ＣＡＮ模型描述的系统的未知阶与时滞的估计算法，证明了算法是强一致收敛的，且能在有限步内达到其模型结构参数的真值。讨论了当模型的参数矩阵不满秩时减弱条件Ｈ’ｓ的强一致估计算法。     

Cloud-Aerosol-Radiation (CAR) ensemble modeling system: Overall accuracy and efficiency

The Cloud-Aerosol-Radiation (CAR) ensemble modeling system has recently been built to better understand cloud/aerosol/radiation processes and determine the uncertainties caused by different treatments of cloud/aerosol/radiation in climate models. The CAR system comprises a large scheme collection of cloud, aerosol, and radiation processes available in the literature, including those commonly used by the world's leading GCMs. In this study, detailed analyses of the overall accuracy and efficiency of the CAR system were performed. Despite the different observations used, the overall accuracies of the CAR ensemble means were found to be very good for both shortwave (SW) and longwave (LW) radiation calculations. Taking the percentage errors for July 2004 compared to ISCCP (International Satellite Cloud Climatology Project) data over (60°N, 60°S) as an example, even among the 448 CAR members selected here, those errors of the CAR ensemble means were only about-0.67% (-0.6 W m-2 ) and-0.82% (-2.0 W m-2 ) for SW and LW upward fluxes at the top of atmosphere, and 0.06% (0.1 W m-2 ) and -2.12% (-7.8 W m-2 ) for SW and LW downward fluxes at the surface, respectively. Furthermore, model SW frequency distributions in July 2004 covered the observational ranges entirely, with ensemble means located in the middle of the ranges. Moreover, it was found that the accuracy of radiative transfer calculations can be significantly enhanced by using certain combinations of cloud schemes for the cloud cover fraction, particle effective size, water path, and optical properties, along with better explicit treatments for unresolved cloud structures.     

辨识方法的计算效率(1):递推算法

算法的计算量可用其乘法运算次数和加法运算次数表示(除法作为乘法对待,减法作为加法对待).一次乘法运算或一次加法运算称为一个flop,即一次浮点运算.作为"辨识方法的计算效率"系列3篇连载论文的第1篇,主要了讨论递推辨识算法的计算量,包括向量和矩阵基本运算的flop数,以及线性回归系统、多元线性回归系统、多变量系统的随机梯度辨识算法、最小二乘辨识算法、递推最小二乘辨识算法的最经济计算量,即实现算法的最少flop数.