Variability of seismic response of soils using stochastic finite element method

The evaluation of seismic response of soil sites constitutes an important problem with respect to groundmotion amplification and soil instability because of liquefaction. The base motion generated during earthquake is a random process. In addition, the soil sites are usually homogenous with randomly varying characteristics. The uncertainties associated with the input motion and site characteristics may lead to a wide range of variability of the site response. In this paper, a Monte-Carlo based stochastic finite element method is used to study the variability of seismic response.     

基于Monte-Carlo方法的施工导流系统综合风险分析

施工期导流动态风险不仅与水文不确定性、水力不确定性有关,而且与坝体施工进度及其完工工期的不确定性有关。以风险分析为核心,将各种不确定性因素耦合起来研究施工导流系统的综合风险,提出施工进度计划风险分析模型和假定坝体施工进度确定条件下的施工导流风险分析模型,在此基础上,建立基于Monte-Carlo方法全面考虑水文水力不确定性以及施工进度不确定性的施工导流系统综合风险分析模型。实例分析说明,该施工导流系统综合风险研究方法和分析模型是可靠的、适用的。     

江淮梅雨特征及其与北太平洋海温的SVD分析

利用江淮地区1954～2001年48年梅雨量资料,采用谐波分析、EOF、合成分析和SVD分解等方法讨论了江淮梅雨的时空变化特征以及与北太平洋海温的关系.结果表明:江淮梅雨量时空分布不均,存在显著的年际和年代际变化特征.影响江淮梅雨的海温关键区是北太平洋西风飘流区,关键影响时段是当年的1～3月;当年1～3月北太平洋西风飘流区海温异常偏高,同年江淮大部地区梅雨量异常偏多,反之亦然.SVD分解结果与合成分析的结果相吻合,通过0.05的Monte-Carlo显著性检验.     

洪水频率分析稳健性方法研究

结合使用Huber的M估计和最小距离(MD)估计,构造了一种P-Ⅲ型总体下的频率分析方法。统计试验结果表明,当样本中有一定比例干扰数据的影响时,该方法具有较好的稳健性;同时,当样本中无干扰数据的影响时,该方法亦具有相对较优的统计性能。     

Guidance on the Choice of Threshold for Binary Forecast Modeling

This paper proposes useful guidance on the choice of threshold for binary forecasts. In weather forecast systems, the probabilistic forecast cannot be used directly when estimated too smoothly. In this case, the binary forecast, whether a meteorological event will occur or not, is preferable to the probabilistic forecast. A threshold is needed to generate a binary forecast, and the guidance in this paper encompasses the use of skill scores for the choice of threshold according to the forecast pattern. The forecast pattern consists of distribution modes of estimated probabilities, occurrence rates of observations, and variation modes. This study is performed via Monte-Carlo simulation, with 48 forecast patterns considered. Estimated probabilities are generated by random variate sampling from five distributions separately. Varying the threshold from 0 to 1, binary forecasts are generated by threshold. For the assessment of binary forecast models, a 2×2 contingency table is used and four skill scores （Heidke skill score, hit rate, true skill statistic, and threat score） are compared for each forecast pattern. As a result, guidance on the choice of skill score to find the optimal threshold is proposed.     

A Monte-Carlo game theoretic approach for Multi-Criteria Decision Making under uncertainty

Game theory provides a useful framework for studying Multi-Criteria Decision Making problems. This paper suggests modeling Multi-Criteria Decision Making problems as strategic games and solving them using non-cooperative game theory concepts. The suggested method can be used to prescribe non-dominated solutions and also can be used as a method to predict the outcome of a decision making problem. Non-cooperative stability definitions for solving the games allow consideration of non-cooperative behaviors, often neglected by other methods which assume perfect cooperation among decision makers. To deal with the uncertainty in input variables a Monte-Carlo Game Theory (MCGT) approach is suggested which maps the stochastic problem into many deterministic strategic games. The games are solved using non-cooperative stability definitions and the results include possible effects of uncertainty in input variables on outcomes. The method can handle multi-criteria multi-decision-maker problems with uncertainty. The suggested method does not require criteria weighting, developing a compound decision objective, and accurate quantitative (cardinal) information as it simplifies the decision analysis by solving problems based on qualitative (ordinal) information, reducing the computational burden substantially. The MCGT method is applied to analyze California’s Sacramento-San Joaquin Delta problem. The suggested method provides insights, identifies non-dominated alternatives, and predicts likely decision outcomes.     

使用物理方法反演中国陆地雨强和水凝物垂直结构

使用MM5模式模拟了11个降水个例,获得中国陆地的降水廓线,由蒙特卡罗辐射传输模式计算其上行辐射亮温,构成初始云一辐射数据库.采用对照表方法从中获得Db_ml和Db_m2两个数据集,作为GPROF反演算法的先验数据集.对2007年7月的3次强降水过程的地面雨强进行反演,与微波成像仪(TMI)及NESDIS、GSCAT、...     

Stochastic analysis of beams on reinforced earth beds

The present study comprises Monte-Carlo simulation assisted analysis of foundations resting on reinforced earth beds using the concept of beams on an elastic foundation, treating the modulus of subgrade reaction (MSR) as a stationary stochastic field characterised by mean, variance, autocorrelation function (ACF) and the autocorrelation distance (ACD). Realisations of the MSR, generated by solving a stochastic differential equation, are fed to a deterministic distributed parameter model to generate realisations of two dependent stochastic fields, namely deflection and bending moment in the foundation beam, and two random variables, namely the location of occurrence of maximum deflection and the bending moment. Subsequently these realisations are analysed to evolve probability distribution functions, variance and ACF of the dependent stochastic fields and the random variables. It is revealed that the ACF of these fields is independent of the ACF of the MSR. Further, variance of deflection is found to increase as the ACD of the MSR increases, implying requirement of a larger factor of safety when random soils display low frequency (macro level) variations. On the other hand, variance of the bending moment is larger at smaller ACDs of the MSR, indicating that for bending moments a larger factor of safety is required when the random soils display high frequency (micro level) variations.     

Influence of spatially random soil on lateral thrust and failure surface in earth retaining walls

The present study investigates the influence of spatial variability of soil properties on the lateral thrust and failure surface of a 6?m high frictionless rigid earth retaining wall in active condition. The drained friction angle of the soil is modelled as a log-normal anisotropic random field in two dimensions, using the Cholesky decomposition technique. The effect of horizontal and vertical scale of fluctuation is observed on the lateral thrust. The failure surface in the backfill is found to be more or less similar for different combinations of spatial variability in vertical and horizontal directions. Monte-Carlo simulation technique is used to compute the probability of failure and to obtain the worst-case spatial variability configuration.