广义岭估计的谱分解式和Liu型广义岭估计

首先从多参数岭估计,即广义岭参数人手,给出其谱分解形式。然后就岭估计会使估值偏差增大的问题,在Liu估计的基础上,推导出一种新的有偏估计方法——Liu型广义岭估计,并给出该方法的模型、解式。Liu型广义岭估计可以在尽量不增大偏差的同时改进病态性,获得更好的估值。最后通过算例进行了比较分析。     

抗差Helmert方差分量估计在水下声学定位中的应用

水下声学定位观测数据中不可避免地存在粗差,随机模型解算广泛地采用等权模型,模型实现简单,但与实际不符,且不能抑制粗差影响。针对该问题,提出一种基于IGG3方案的抗差Helmert方差分量估计方法。该方法通过水深和观测距离将观测值分为两类,利用Helmert方差分量估计确定不同类观测值的权比,抗差解决了粗差导致Helmert方差分量估计模型失效的问题。实验结果表明,相比于传统解算方法,抗差Helmert方差分量估计方法可以合理确定各观测值权比,削弱粗差影响,提高水下定位精度和可靠性。     

一种两步抗差估计的设计及其解算

抗差估计的关键是选择适当的权函数和合理的调制因子,针对这两个重要问题,在研究当前调制因子取值方法的基础之上,提出了可动性调制因子这一新概念,并基于可动性调制因子、IGG方案权函数,设计了两步抗差估计。两步抗差估计解决了抗差估计中传统的调制因子、方差因子、初值三者取值时存在的一系列问题,尤其当三者取值不能协调配制时出现的秩亏问题,从而使抗差估计方法更具有实际意义和广泛意义。     

附加约束条件的序贯平差抗差估计

推导了附加约束条件的序贯平差抗差估计解法,对其两种解法进行了分析比较,这两种方法上是一致的。计算算例表明：附加约束条件的抗差序贯平差能有效地抵制粗差的影响。     

空间直角坐标系统转换的抗差算法研究

为了解决空间直角坐标系间相互转换时,由于公共点的坐标精度差直接导致坐标转换精度差的问题.探讨利用抗差估计理论解决公共点坐标粗差的影响,结合具体的坐标转换算例,分别采用Tukey、IGG1及IGG3三种权函数进行迭代计算,完成坐标转换.结果表明,抗差估计原理用于空间直角坐标系间的相互转换,可以降低公共点粗差的影响,提高坐...     

基于验后方差估计原理探测与剔除粗差

将抗差估计的思想融入到粗差探测的算法中,设计出对模型误差特别是粗差具有抵抗能力的粗差探测算法。应用最小二乘法来计算观测值的参数、残差、协因数阵及单位权的初值,再根据残差和有关的参数,按所选择的权函数,计算每个观测值的权,经过迭代计算求得观测值的残差,按照统计检验的方法剔除粗差。通过实验证明,基于验后方差估计原理导出的选权迭代具有很强的粗差探测能力。     

经典选权迭代法研究与两步抗差估计的提出

随着权函数的不同出现了不同的抗差估计方法,相应不同的抗差估计方法在一定程度上存在着不足或者缺陷。详细地论述了几种经典的选权迭代抗差估计方法的不足,并从理论出发,提出了两步抗差方案。     

带乘性噪声广义系统的观测噪声最优估计算法

研究带乘性噪声广义系统的观测噪声最优估计问题。在假设系统正则的情况下,针对乘性噪声为一般随机矩阵即各观测通道乘性噪声同时刻相关的情况,通过受限等价变换和状态扩维的方法,给出了线性最小方差意义下的观测噪声最优滤波算法和最优平滑估计算法。数字仿真结果表明了算法的有效性。     

一种融合抗差的共线秩亏平差算法

传统卫星测高数据处理中共线平差的精度极大影响最终解算结果,针对经初步处理后卫星测高数据仍可能存在粗差的问题,提出一种融合抗差的共线秩亏平差算法。该算法结合传统共线平差及抗差估计的优点,保留了秩亏自由网平差不需要基准控制点的优良性质,并有效地剔除秩亏自由网中存在的粗差,实现更高精度的解算结果。实验表明,相比较传统共线秩亏平差算法而言,提出的算法解算结果精度更高。     

选权迭代抗差估计的初值选取方法分析

粗差探测是测量数据处理领域的重要研究课题,分别基于最小二乘(LS)、一次范数最小(L1)、最小平方中位数(LMS)和最小截平方和(LTS)方法选取选权迭代法的初值,联合IGGⅢ方案进行抗差估计。通过算例从筛选粗差个数、参数差值的平方和、验后单位权中误差等3个指标,对几种初值选取方法的的抗差能力进行比较分析。结果表明:LMS、LTS方法选取初值的方法效果最好,L1方法抗粗差能力次之,LS方法只能在粗差较少的情况下有效。     

The Role of GNSS Vertical Velocities to Correct Estimates of Sea Level Rise from Tide Gauge Measurements in Greece

In this study, we show how the Global Navigation Satellite System (GNSS)-derived vertical velocities contribute to the correction of tide gauge (TG) measurements used for the sea level rise estimation in Greece. Twelve sites with records of local sea level heights are processed in order to estimate their trend. Certain error sources related to TGs, e.g. equipment changes, data noise, may lead to biased or erroneous estimations of the sea level height. Therefore, it would be preferred to follow a robust estimation technique in order to detect and reduce outlier effects. The geocentric sea level rise is estimated by taking into account the land vertical motion of co-located GNSS permanent stations at the Hellenic area. TGs measure the height of the water relative to a monitored geodetic benchmark on land. On the other hand, using GNSS-based methods the vertical land motion can be derived. By means of extended models fitted to the GNSS time-series position, obtained from seven years of continuous data analysis, periodic signals are well described. The synergy of the two co-located techniques results in the correction of TG relative sea level heights taking into account the GNSS vertical velocities and consequently obtaining the conversion to absolute (geocentric) sea level trend.     

Response-based estimation of sea state parameters — influence of filtering

Reliable estimation of the on-site sea state parameters is essential to decision support systems for safe navigation of ships. The wave spectrum can be estimated from procedures based on measured ship responses. The paper deals with two procedures—Bayesian Modelling and Parametric Modelling—which both use complex-valued frequency response functions (FRF) to estimate the wave spectrum. It is therefore interesting to investigate how the filtering aspect, introduced by FRF, affects the final outcome of the estimation procedures. In order to do this, extensive numerical simulations—with known wave parameters—are carried out for a large container vessel. The study shows that filtering has an influence on the estimations, since high-frequency components of the wave excitations are not estimated as accurately as lower frequency components.     

Uncertainty analysis on extreme value analysis of significant wave height at eastern coast of Korea

In this study, we considered the problem of estimating long-term predictions of design wave height based on the observation data collected over 10–15 years along the eastern-coast of the Korean peninsula. We adopted a method that combines Bayesian method and extreme value theory. The conventional frequency analysis methods must be reconsidered in two ways. First, the conventional probability distributions used in the frequency analysis should be evaluated to determine whether they can accurately model the variation in extreme values. Second, the uncertainty in the frequency analysis should also be quantified. Therefore, we performed a comparative study of the Gumbel distribution and GEV distribution to show the higher efficiency of the latter. Further, we compared the Bayesian MCMC (Markov Chain Monte Carlo) scheme and the MLE (Maximum Likelihood Estimation) with asymptotic normal approximation for parameter estimation to confirm the advantage of the Bayesian MCMC with respect to uncertainty analysis.     

Performance Analysis for Matched-Field Source Localization: Simulations and Experimental Results

Matched-field methods concern estimation of source locations and/or ocean environmental parameters by exploiting full wave modeling of acoustic waveguide propagation. Typical estimation performance demonstrates two fundamental limitations. First, sidelobe ambiguities dominate the estimation at low signal-to-noise ratio (SNR), leading to a threshold performance behavior. Second, most matched-field algorithms show a strong sensitivity to environmental/system mismatch, introducing biased estimates at high SNR. In this paper, some theoretical developments on matched-field performance analysis are summarized, including Bayesian performance bounds and probabilistic ambiguity analysis, both incorporating environmental/system uncertainty/mismatch. Performance analysis is then implemented for source localization in a typical shallow water environment chosen from the Shallow Water Evaluation Cell Experiments (SWellEX). The performance predictions describe the simulations of the maximum-likelihood estimator (MLE) well, including the mean-square error (MSE) in all SNR regions as well as the bias at high SNR. The threshold SNR and bias predictions are also validated through SWellEX experimental data processing. The results suggest the current environmental, acoustic, and statistical modeling has developed to such a level that the optimum theoretical matched-field performance can be achieved in a well-controlled experiment.     

贝叶斯统计方法在海浪方向谱研究中的应用

较详细介绍了贝叶斯统计方法在海浪方向谱估计中的应用,指出先验分布的不同对估计结果没有影响,由于均匀分布与最大熵原则相对应,使其具有某种特殊性.数值模拟表明贝叶斯方法对双峰方向分布估计的有效性,将其应用于黑海实测资料,证明双峰方向分布的出现依赖于估计方法分辨力的大小,而用最大似然法得到的双峰方向分布值得怀疑.最后给出出现双峰方向分布的一种可能物理解释.     

基于广义极值分布的设计波高推算

简介了广义极值分布函数及其3种参数估计方法,包括极大似然(ML)、线性矩(LM)和间隔最大积(MPS)估计的计算方法。使用广义极值分布函数推算了北部湾涠洲岛海域3个波向的年波高极值序列设计波高,并与Weibull分布、Gumbel分布和皮尔逊Ⅲ型分布的推算结果加以对比。分析表明,涠洲岛海域极值波高服从于广义极值Ⅲ型分布,拟合优度检验结果表明广义极值分布能更好地拟合极值波高;MPS方法是一种优良的参数估计法,推算的设计波高可作为海岸环境工程设计的首要参考值。     

极值波高建模中不确定性的模糊量化方法

A non-traditional fuzzy quantification method is presented in the modeling of an extreme significant wave height. First, a set of parametric models are selected to fit time series data for the significant wave height and the extrapolation for extremes are obtained based on high quantile estimations. The quality of these results is compared and discussed. Then, the proposed fuzzy model, which combines Poisson process and generalized Pareto distribution(GPD) model, is applied to characterizing the wave extremes in the time series data. The estimations for a long-term return value are considered as time-varying as a threshold is regarded as non-stationary. The estimated intervals coupled with the fuzzy theory are then introduced to construct the probability bounds for the return values. This nontraditional model is analyzed in comparison with the traditional model in the degree of conservatism for the long-term estimate. The impact on the fuzzy bounds of extreme estimations from the non stationary effect in the proposed model is also investigated.     

应用贝叶斯方法对有效信息较少的渔业资源进行资源评估-以印度洋长鳍金枪鱼的资源评估为例

众所周知,对有效信息较少的渔业资源进行资源评估面临很大的挑战,而贝叶斯方法在数据数量较少、质量较差的情况下能利用其它种群高质量的数据或已知的先验信息提高资源评估结果的可靠性。由于印度洋长鳍金枪鱼的数据质量较差而数据量有限,长鳍金枪鱼的资源评估结果存在很大的不确定性,为此,本文以印度洋长鳍金枪鱼的资源评估为例,以调查贝叶斯方法在有效信息较少的资源评估中的优势。本文根据不同的先验假设与捕捞数据系列,共构建了8个贝叶斯动态产量模型,以评估长鳍金枪鱼资源。结果表明:(1)分析参数的后验分布能提高捕捞数据系列选择与参数假设的合理性; (2) 利用种群统计学方法为内禀增长率(r)构建有信息先验,能提高资源评估结果的可靠性。与传统方法相比,当基于贝叶斯框架时,能将已知的知识表示为先验信息并能分析参数的后验分布,从而在数据较少或数据质量较差的情况下,能利用各种信息提高参数估计的合理性与资源评估的可靠性。因此,对数据量较少或数据质量较差情况下的渔业资源评估而言,贝叶斯方法非常有效,如本文所示的印度洋长鳍金枪鱼的资源评估。     

Stochastic simulation of episodic soft coastal cliff recession

Probabilistic methods provide a means of demonstrating the potential variability in predictions of coastal cliff recession. They form the basis for risk-based land use planning, cliff management and engineering decision-making. A range of probabilistic methods for predicting soft coastal cliff recession has now been developed, including statistical techniques, process-based simulation and structured use of expert judgement. A new episodic stochastic simulation model is introduced, which models the duration between cliff falls as a gamma process and fall size as a log-normal distribution. The method is applied to cliff recession data from a coastal site in the UK using maximum likelihood and Bayesian parameter estimation techniques. The Bayesian parameter estimation method enables expert geomorphological assessment of the local landslide characteristics and measurements of individual cliff falls to be combined with sparse historic records of cliff recession. An episodic simulation model is often preferable to conventional regression models, which are based on assumptions that are seldom consistent with the physical process of cliff recession.