最大熵在岩体结构面间距分布估计中的应用研究

结构面间距是表征岩体完整性的重要参数。由于结构面分布的不均匀性和地质构造的复杂性,常用的概率分布不能精细描述结构面间距的分布特征。对于由测线法测得的结构面间距样本,根据岩体内部能量平衡,在结构面间距的统计矩的约束下,采用最大熵原理推导结构面间距的概率分布,最大熵原理估计得到的结构面间距的概率分布具有无偏性和一致性;给出结构面间距常用的分布函数的最大熵形式;采用Monte Carlo模拟方法进行结构面三维网络模拟,采用K-S检验验证最大熵估计结构面间距概率分布的适用性和准确性。对隧道采样区的结构面间距样本进行分布估计,得到的最大熵概率分布显著优于常用的负指数分布或对数正态分布,能够准确反映结构面间距的真实分布情况。     

最大熵在岩体结构面间距分布估计中的应用研究EI北大核心CSCD

结构面间距是表征岩体完整性的重要参数。由于结构面分布的不均匀性和地质构造的复杂性,常用的概率分布不能精细描述结构面间距的分布特征。对于由测线法测得的结构面间距样本,根据岩体内部能量平衡,在结构面间距的统计矩的约束下,采用最大熵原理推导结构面间距的概率分布,最大熵原理估计得到的结构面间距的概率分布具有无偏性和一致性;给出结构面间距常用的分布函数的最大熵形式;采用Monte Carlo模拟方法进行结构面三维网络模拟,采用K-S检验验证最大熵估计结构面间距概率分布的适用性和准确性。对隧道采样区的结构面间距样本进行分布估计,得到的最大熵概率分布显著优于常用的负指数分布或对数正态分布,能够准确反映结构面间距的真实分布情况。     

基于全寿命成本的价值分析技术在房地产方案选择中的应用研究

应用基于全寿命成本的价值分析技术，将寿命周期中各项成本的不确定性也作为衡量方案优劣程度的因素，根据各项成本的概率分布，利用蒙特卡罗仿真技术计算各方案全寿命成本的分布函数，利用价值分析结果选择最优方案。克服了传统分析方法考虑因素单一、实际值和期望值存在较大偏差的缺点，在实际投资项目选择中具有一定的应用价值。     

建筑围护结构传热系数的反演研究

建筑围护结构传热系数的反演研究是典型的热传导反问题研究。针对围护结构传热特征,将其简化为一维热传导过程来处理。通过采用有限控制体积法(FCV)求解一维热传导模型为基础,采用Bayesian推理已知先验分布与条件概率分布求后验概率分布,而后验概率分布便是反问题的解。采用马尔科夫链蒙特卡罗(Markov Chain Monte Carlo,MCMC)算法,通过构造Markov Chian对传热系数进行反演估计。算例计算结果表明MCMC-Bayes方法对围护结构传热系数反演估计是可行的     

基于蒙特卡罗的天然气长输管道泄漏范围预测

为定量研究天然气管道泄漏危险区域范围,采用蒙特卡罗方法分别从管道介质局部压力降、破裂处气体泄漏速率以及大气条件下气体扩散等3个阶段对泄漏过程进行模拟,得到了泄漏位置处管道气体压力、泄漏速率和顺(横)风向下危险区域范围的概率分布函数。结果表明,管道压力、破裂形式和风速等事故条件决定着泄漏危险范围分布的特征,通过模拟可以得到已知事故场景环境变量概率条件下泄漏范围的概率值。     

岩体不连续面互相关产状模拟的Copula函数方法

为准确描述岩体不连续面产状(倾角和倾向)的统计特性,并考虑倾角和倾向之间的互相关性,提出不连续面互相关产状模拟的Copula函数方法,通过确定拟合不连续面产状实测数据最优的边缘概率分布和Copula函数(相关结构),建立不连续面倾角和倾向的二维联合概率分布函数。同时与传统的Fisher分布和二维经验分布方法的计算结果进行可视化对比分析,采用极射赤平投影方法在赤平投影图上比较实测产状和不同方法模拟的产状之间的差别。最后,通过4个案例验证了提出方法的有效性。结果表明：传统的Fisher分布和二维经验分布方法不能准确描述不连续面产状之间的互相关性,而提出的Copula函数方法能够基于少量的实测数据更加灵活地构造出具有任意边缘分布类型和相关结构的产状联合概率分布函数,更加真实地表征不连续面产状之间的互相关性,克服了以往构建不连续面产状概率分布模型时将倾角和倾向视作2个独立变量的局限性。     

基于模拟技术的工程工期风险分析

通过对传统工期分析方法(关键路径法CPM、计划评审技术PERT)和工期模拟技术(MCS-CB)的比较研究,发现模拟技术不仅能够克服CPM中未涉及的工期不确定性问题,而且能够解决PERT中假设过多、过分乐观估计完工概率的问题;继而对MCS-CB的原理、软件运用以及案例应用进行了系统研究,很好地实现了关键路线的选择、分项工程和总工期的概率分布选择、完工概率分析以及基于敏感性判断的工期优化等工作。     

桩基不重复随机抽样不合格桩数概率分布研究及其工程应用

用条件概率方法研究桩基质量随机抽样不合格桩数ｒ的概率分布ｐ_（ｎ,Ｎ）（ｒ）和数字特征的计算公式。然后,引用计算随机变量函数概率分布的方法,导出桩基抽样不合格率γ的概率分布ｐ_（ｎ,Ｎ）（γ）和数字特征的计算公式。这些公式为桩基质量检测的统计推断提供了理论基础。     

Probability distribution characteristics and estimation method of prestress of concrete structures based on Gaussian mixture model

为解决预应力混凝土框架结构预应力筋数量众多、有效应力不确定性强且分布不均匀、抽样检测评估无法准确表征结构有效应力整体特征等结构评价难题，引入高斯混合模型研究结构有效应力概率分布特征，并据此提出了一种概率估计方法。该方法通过MonteCarlo模拟建立预应力筋集有效应力的高斯混合模型；通过中心化、去除离散点等方式修正模型以增强其高斯特性；基于模型高斯显著性检验准则确定正态分布近似估计范围，并以此为总体样本进行抽样检测，进而估计预应力筋集实时有效应力的概率分布；以某无黏结预应力混凝土框架结构为工程实例验证了该方法有效性。研究结果表明：既有结构预应力筋集有效应力概率分布由多个均值与方差相近的子分布组成，可采用近似正态的高斯混合模型统一表征；由设计差异导致的有效应力分布不均匀性可采用中心化的数学方法进行消除，进而提升模型近似估计的准确性；既有结构施工阶段以张拉工艺为主的建造误差上限可通过模型计算结合实测值统计分析综合推定。     

施工项目成本估测中的建模研究

本文运用灰色系统理论建立预测模型群进行施工项目成本估测，并依据概率论三值估计法，三角形分布法，推出成本概率分布的期望值和方差，确定成本随机变化的概率分布曲线。     

Cost simulation in an item-based project involving construction engineering and management

Despite the extensive use of simulation in management, the continuous simulation model for cost estimation remains unexploited, especially for construction engineering and management. This study introduces streamlining Monte Carlo simulation procedures with evaluation of stochastic processes and input probability distribution selection via hypothesis testing, and specification of correlations between simulated variates. By using self-developed algorithms and a spreadsheet-add-on program, this investigation uses historical construction projects as case study data to create an early-stage cost distribution for budget allocation. While establishing the applicability of the proposed simulation procedures, this study demonstrates that the simulated cost results present superior simulation accuracy in addition to separating the principal work items and unit price component model. Generally, the precision and absolute error rates fall into acceptable ranges when the proposed systematic simulation procedures are adopted. The cost simulation approach offers a simplified decision tool for fairly assessing construction cost and uncertainties based on the experienced judgment of project managers.     

考虑操作效率差异的双资源项目调度模型研究

针对多资源项目计划与调度问题中未考虑资源之间关联性而与实际情况脱离的问题,基于人力资源和设备资源两种可更新资源,考虑工人操作不同设备存在效率差异,建立一种以完工周期和项目成本为优化目标的考虑差异性操作效率的双资源多目标项目计划与调度模型。针对模型特点设计相应的染色体编码方案,运用NSGA-II算法进行求解。通过工程实例进行仿真实验,得到多目标优化的Pareto解集,同时与仅考虑相同操作效率的Pareto解对比,仿真实验在一定程度上缩短了项目工期,降低了项目成本。     

Monte Carlo simulation approach to life cycle cost management

Public private partnerships (PPP) and private financial initiative (PFI) projects face the challenge of meeting unforeseen future risks. Life cycle cost estimate is a crucial part of PFI/PPP procurement. The traditional deterministic cost model can not take into consideration the uncertainty of future events let alone determine the contingency allowance for the projects. The large number of cost items in the life cycle cost model of building projects makes cost control difficult. Monte Carlo simulation method is applied to the Quantitative Risk Assessment of life cycle costing risk management. A PFI school project was chosen as a case study to demonstrate a new simulation approach to life cycle cost management. The lives and replacement cost rates of building elements are the inputs for the simulation model, while the cumulative life cycle cost are the outputs of the analysis. The sensitivity analysis revealed the cost significant items, which provides the most efficient way for cost control. The results of the analysis identify the high risk life cycle assumptions and provide a variation reference for the decision-makers to define risk and contingency allowance in PFI/PPP projects. The approach can also be applied to other types of building PFI projects.     

建筑业劳务工人职业流动对工人技术水平的影响及对策研究

在我国建筑业,劳务工人的频繁流动使其缺乏连续、逐步提高的技术训练,直接导致这一群体的技术水平和能力难以有效提高。通过主体仿真技术建立建筑业劳务工人职业流动的仿真模型,并引入学习曲线对建筑业劳务工人的技术水平进行模拟,发现存在劳务工人职业流动频繁、用工成本飙升和劳务工人技术水平提升缓慢的现象,其根本原因在于劳务工人的自利倾向、低流动成本和用工单位的相互竞价。仿真实验还研究了技术准入对策和技术激励对策在引导劳务工人的自利倾向、增加劳务工人的流动成本两方面的作用。相应对策的仿真实验表明,技术准入和技术激励两种对策均能有效解决以上问题,其中技术激励的正面作用远大于其导致的成本增加,使激励变得更加重要。     

Optimization design of exhaust duct system in semiconductor factory using dynamic programming method

An exhaust duct system in semiconductor factory is designed using the dynamic programming method (DPM), which considers system pressure equilibrium and the least life-cycle cost to derive the duct size and fan capacity. An example of alkaline gas exhaust system is provided to understand the characteristics of DPM and to compare with the conventional duct design methods. Since DPM contains the concept of minimizing the life-cycle cost, the design results not only guarantee each path to share the same pressure, but also bear a smaller cost than other methods. The limit on duct diameter or flow velocity is added to the computation process. As a result, all the derived outcomes satisfy the requirements on the range of duct diameter or flow velocity. The differences between the design and simulation (actual operation) results under DPM are much lower than those of other methods. Thus, an exhaust duct system that best approximates the actual operation may be designed using DPM.     

Predicting Cost Escalation Pathways and Deviation Severities of Infrastructure Projects Using Risk‐Based Econometric Models and Monte Carlo Simulation

In the past decade, infrastructure‐related legislation in the United States has consistently emphasized the need to measure the variation associated with infrastructure project cost estimates. Such cost variability is best viewed from the perspective of the project development phases and how the project cost estimate changes as it evolves across these phases. The article first identifies a few gaps in the cost overrun literature. Then it introduces a methodology that uses risk‐based multinomial models and Monte Carlo simulation involving random draws to predict the probability that a project will follow a particular cost escalation pathway across its development phases and that it will incur a given level of cost deviation severity. The article then uses historical data to demonstrate how infrastructure agencies could apply the proposed methodology. Statistical models are developed to estimate the probability that a highway project will follow any specific cost escalation pathway and ultimately, a given direction and severity of cost deviation. The case study results provided some interesting insights. For a given highway functional class, larger project sizes are associated with lower probability of underestimating the final cost; however, such a trend is not exhibited by very large projects (total cost exceeding $30M). For a given project size, higher class roads were generally observed to have a lower probability of underestimating the final cost, compared to lower class roads and this gap in probability narrows as the project size increases. It was determined that a project's most likely pathway of cost escalation is not a guarantee that it will yield any particular direction of cost deviation. The case study results also confirmed the findings of a few past studies that the probabilities of cost escalation pathways and the cost overruns differ significantly across highway districts, and attributed this to differences in administrative culture and work practices across the districts. Infrastructure managers can use the developed methodology to identify which projects are likely to experience a particular pathway of cost escalation, the direction and severity of cost deviation, and to develop more realistic project contingency estimates.     

Simulation-Facilitated Model for Assessing Cost Correlations

Project cost becomes increasingly variable if many cost items for a construction project are correlated, and this can increase the uncertainty of completing a project within a target budget. This work presents a factor-based computer simulation model (COSTCOR) for evaluating project costs given correlations among cost items. Uncertainty in the total cost distribution of an item (grandparent) is transferred to several factor cost distributions (parents) according to qualitative estimates of the sensitivity of each cost item to each factor. Each cost distribution is then decomposed further into a family of distributions (children; costs given factor conditions), with each child corresponding to a factor condition. Correlations are retrieved by sampling from the child distributions with the same condition for a given iteration of the simulation. COSTCOR integrates the uncertainty effects caused by each factor at the project cost level, thus making it easier for management to determine what parts of the project need to be controlled.     

Locational competition under environmental regulation when input prices and productivity differ

The purpose of the paper is to outline an analytical framework which captures the ample scope of locational competition: cost differences, resulting from differences in factor prices including taxes, human capital, infrastructure services and total factor productivity. If cost differences are small, locational competition controls excessive government power. We have modeled locational competition by assuming that governments have a vital interest to keep mobile factors of production at home. We represent this aspect by restricting the usage of environmental instruments such that they will at most exhaust the cost difference to a competing foreign firm. If cost differences are large enough there is no binding restriction for the cost-benefit calculus of a national environmental policy. The tax will be below marginal damage due to strategic reasons of rent shifting. If small international cost differences do not allow taxation in accordance with marginal damage considerations, then locational competition restricts the size of the tax rate such that the firm is indifferent in relocating or staying at home.     

设计院牵头的 EPC 模式合作关系博弈研究

通过实地调研结合对国家出台的有关 EPC 项目政策研究,构建 EPC 项目中业主方、设计单位和施工单位的演化博弈模型,借助系统动力学仿真方法分析了三方演化路径和趋势。研究结果表明:业主激励成本和设计施工优化成本对合作关系的建立呈负向影响,公平合理的收益分享系数和额外优化收益能加速合作关系的建立。并根据仿真结果分析,针对 EPC项目三方实现合作共赢提出优化建议。     

Study on hybrid ground-coupled heat pump systems

Although ground-coupled heat pump (GCHP) systems are becoming attractive air-conditioning systems in some regions, the significant drawback for their wider application is the high initial cost. Besides, more energy is rejected into ground by the GCHP system installed in cooling-dominated buildings than the energy extracted from ground on an annual basis and this imbalance can result in the degradation of system performance. One of the available options that can resolve these problems is to apply the hybrid ground-coupled heat pump (HGCHP) systems, with supplemental heat rejecters for rejecting extra thermal energy when they are installed in cooling-dominated buildings. This paper presents a practical hourly simulation model of the HGCHP system by modeling the heat transfer of its main components. The computer program developed on this hourly simulation model can be used to calculate the operating data of the HGCHP system according to the building load. The design methods and running control strategies of the HGCHP system for a sample building are investigated. The simulation results show that proper HGCHP system can effectively reduce both the initial cost and the operating cost of an air-conditioning system compared with the traditional GCHP system used in cooling-dominated buildings.