隧道支护结构荷载作用的随机反演

提出荷载－结构模式进行随机反分析方法，从隧道支护结构变形反算对应的荷载作用，可反推围岩抗力系数和覆土荷载的分布情况，以及作用于支护结构上各部位的外荷载。提出了以实测位移为基础的样条函数插值．以插值位移补充大量的实测值，满足了反分析所需众多的已知位移数。通过一施工实例对该方法的可行性进行了检验。     

地球物理反问题中的随机反演理论

本文从最大后验概率密度观点出发，在数据噪音向量和待求模型向量为具有零均值的独立高斯随机过程的假设前提下，建立起了随机反演的非线性系统方程；给出了模型方差估计的函数表达式，并在文章最后，证明了反演解的稀疏性，即解释了随机反演的输出解的高分辨率特征。文章在最小二乘反演方法的基础上，发展并完善了随机反演方法的理论基础；揭示了随机反演方法与最小二乘反演方法之间的本质区别；阐述了随机反演方法的优越性，并指出了其广阔的应用前景。     

Simulation of normal distributed smooth fields by Karhunen-Loéve expansion in combination with kriging

Simulation of multigaussian stochastic fields can be made after a Karhunen-Loéve expansion of a given covariance function. This method is also called simulation by Empirical Orthogonal Functions. The simulations are made by drawing stochastic coefficients from a random generator. These numbers are multiplied with eigenfunctions and eigenvalues derived from the predefined covariance model. The number of eigenfunctions necessary to reproduce the stochastic process within a predefined variance error, turns out to be a cardinal question. Some ordinary analytical covariance functions are used to evaluate how quickly the series of eigenfunctions can be truncated. This analysis demonstrates extremely quick convergence to 99.5% of total variance for the 2nd order exponential (‘gaussian’) covariance function, while the opposite is true for the 1st order exponential covariance function. Due to these convergence characteristics, the Karhunen-Loéve method is most suitable for simulating smooth fields with ‘gaussian’ shaped covariance functions. Practical applications of Karhunen-Loéve simulations can be improved by spatial interpolation of the eigenfunctions. In this paper, we suggest interpolation by kriging and limits for reproduction of the predefined covariance functions are evaluated.     

GPS Ambiguity Resolution: Impact of Time Correlation,Cross-Correlation and Satellite Elevation Dependence

In this contribution we discuss the geometry-free GPS single baseline model and show how the least-squares ambiguities are affected by changes in the stochastic model. We particularly pay attention to the effect of time correlation, cross-correlation and satellite elevation dependence. We also differentiate between the impact on the location of the ambiguity search space and the impact on the size and shape of the search space. The analysis is carried out for both the model in which the ionospheric delays are assumed absent, and for the model in which they are assumed present. The former model is applicable to short baselines only.     

Simulation of normal distributed smooth fields by Karhunen-Loéve expansion in combination with kriging

Simulation of multigaussian stochastic fields can be made after a Karhunen-Loéve expansion of a given covariance function. This method is also called simulation by Empirical Orthogonal Functions. The simulations are made by drawing stochastic coefficients from a random generator. These numbers are multiplied with eigenfunctions and eigenvalues derived from the predefined covariance model. The number of eigenfunctions necessary to reproduce the stochastic process within a predefined variance error, turns out to be a cardinal question. Some ordinary analytical covariance functions are used to evaluate how quickly the series of eigenfunctions can be truncated. This analysis demonstrates extremely quick convergence to 99.5% of total variance for the 2nd order exponential (‘gaussian’) covariance function, while the opposite is true for the 1st order exponential covariance function. Due to these convergence characteristics, the Karhunen-Loéve method is most suitable for simulating smooth fields with ‘gaussian’ shaped covariance functions. Practical applications of Karhunen-Loéve simulations can be improved by spatial interpolation of the eigenfunctions. In this paper, we suggest interpolation by kriging and limits for reproduction of the predefined covariance functions are evaluated.     

A Simple Parameterisation for Flux Footprint Predictions

Flux footprint functions estimate the location and relative importance of passive scalar sources influencing flux measurements at a given receptor height. These footprint estimates strongly vary in size, depending on receptor height, atmospheric stability, and surface roughness. Reliable footprint calculations from, e.g., Lagrangian stochastic models or large-eddy simulations are computationally expensive and cannot readily be computed for long-term observational programs. To facilitate more accessible footprint estimates, a scaling procedure is introduced for flux footprint functions over a range of stratifications from convective to stable, and receptor heights ranging from near the surface to the middle of the boundary layer. It is shown that, when applying this scaling procedure, footprint estimates collapse to an ensemble of similar curves. A simple parameterisation for the scaled footprint estimates is presented. This parameterisation accounts for the influence of the roughness length on the footprint and allows for a quick but precise algebraic footprint estimation.     

A coherency function model of ground motion at base rock corresponding to strike-slip fault

IntroductionEarthquakedamagesurveyandresearchresultshavedemonstratedthatspatialdistributiondifferenceofgroundmotionisoneoftheimportantreasonswhichcausedlongstructure(eglongspanbridge,undergroundpipe)destroy.Thathowtoprovideareasonableinputofgroundmotionfieldforaseismicdesignoflongstructureisaurgentprobleminearthquakeengineeringfield.Atpresent,themethodtostudyspatialvariationofgroundmotionsisadoptingstatisticanalysisbasedondensearrayrecordssuchasSMART-1array,etc,togetcoherencyfunctionofground…     

Postglacial topographic evolution of glaciated valleys: a stochastic landscape evolution model

The retreat of valley glaciers has a dramatic effect on the stability of glaciated valleys and exerts a prolonged influence on the subsequent fluvial sediment transport regime. We have studied the evolution of an idealized glaciated valley during the period following retreat of ice using a numerical model. The model incorporates a stochastic process to represent deep‐seated landsliding, non‐linear diffusion to represent shallow landsliding and an approximation of the Bagnold relation to represent fluvial sediment transport. It was calibrated using field data from several recent surveys within British Columbia, Canada. We present ensemble model results and compare them with results from a deterministic linear‐diffusion model to show that explicit representation of large landslides is necessary to reproduce the morphology and channel network structure of a typical postglacial valley. Our model predicts a rapid rate of fluvial sediment transport following deglaciation with a subsequent gradual decline, similar to that inferred for Holocene time. We also describe how changes in the model parameters affect the estimated magnitude and duration of the paraglacial sediment pulse. Copyright © 2005 John Wiley & Sons, Ltd.     

Study on simulating strong ground motion by fractal stochastic method

The time history of strong ground motion can be synthesized by empirical Green's function (EGF) method.Firstly a large seismic event is discretized into a series of subevents; secondly recordings of earthquakes with proper size and spatial distribution are chosen as time history (EGF) of those subevents; finally the EGFs are summated to get the time history of ground motion caused by the large event.