Papers by Keyword: Sensitivity Analysis

Paper TitlePage

Authors: Lin Ke Li, Ai Jun Zhang, Jin Yu Liu, Chun Jiao Hou, Hao Dong Li
Abstract: A new method that 3D numerical simulation combined with anti-cut stability analysis is employed to analyze the stability of Majiabian gravity dam, and the results of the analysis conform to the measured value. And this method can be used in analysis of congener gravity dam. When the results of 2D sliding stability analysis satisfy Code’s requirements, elastic material can be used in 3D numerical simulation. The results of water level sensitivity analysis and deformation modulus of its foundation sensitivity analysis indicates that this gravity dam is insensitive to water level; grouting method can be adopted to enhance its foundation. The results of abutment under tension stress analysis indicate that tension stress of abutment must be considered in high gravity dam design, and arch can be used for reducing the tension stress.
Authors: Ri Cheng Liu, Bang Shu Xu, Bo Li, Yu Jing Jiang
Abstract: Mechanical behaviors of pile-soil effect and anchor-soil effect are significantly important in supporting engineering activities of foundation pit. In this paper, finite difference method (FDM) was utilized to perform the numerical simulation of pile-anchor system, composed of supporting piles and pre-stressed anchor cables. Numerical simulations were on the basis of the foundation pit of Jinan’s West Railway Station, and 3D simulation analysis of foundation pit has been prepared during the whole processes of excavation, supporting and construction. The paper also analyzed the changes of bending moments of piles and axial forces of cables, and discussed mechanical behaviors of pile-anchor system, through comparisons with field monitoring. The results show that the parameters concluding vertical gridding’s number, cohesion of pile and soil, and pile stiffness have robust influences on supporting elements’ behaviors. Mechanical behaviors of supporting pile and axial forces of anchor cable changed dramatically, indicating that the potential failure form was converted from toppling failure to sliding failure.
Authors: Li Zhe Jia, Zhong Dong Duan
Abstract: The spectral displacement and the damage state medians were assumed as deterministic parameters in HAZUS-Advanced Engineering Building Module (AEBM). As the historical material and experimental data were insufficient to establish an exactly probability density function model of uncertain variable, the damage assessment results by probabilistic model derived from many assumptions and simplification were unreliable. The uncertainties of the spectral displacement and damage state medians were modeled by the Envelope Bound Convex Model. Furthermore, a bound fragility analysis methodology is derived by integrating HAZUS-AEBM module with the convex set theory. The fragility of a hotel in southern China are calculated and compared to the damage derived by HAZUS-AEBM method, and the results of fragility analysis are more sensitive to the spectral displacement and the damage state medians.
Authors: Xi Yong Huang, M.H. Aliabadi
Abstract: In this paper a sensitivity formulation using the Boundary Element Method (BEM) is presentedfor analysis of structural reliability problems. The sensitivity formulation is based on implicit differentiation method where the first and second order derivatives of the random variables are obtained directly by differentiation of the discretised boundary integral equation. The structural reliability is assessed using the Monte Carlo Method and FORM with BEM sensitivity parameters. A benchmark example is presented to demonstrate the accuracy and efficiency of the BEM for both Monte Carlo and Sensitivity based FORM approaches.
Authors: Xue Li Zhang, Cun Wei Lu
Abstract: The aim of our research is to realize a high-sensitivity three-dimensional (3-D) shape measurement of target with an optical microscope. On the conventional method – shape from focus, it’s to regulate focal length of microscope to obtain 3-D shape information of the specimen. However, the method is used the vertical stratification method to obtain 3-D shape information of total surface of target, and the variable focal length of zoom lens is limited therefore the measurement accurate is confined on the vertical direction. In order to solve these problems, we propose a high-sensitivity 3-D shape measurement method based on microscope and laser projection. The method is based on the slit pattern projection technique and 3-D image processing technique. The proposal method can be used for medicine, pharmacy, life science, and material science.
Authors: Zhen Han Yao, Ying Tao Wei
Abstract: To enhance the efficiency for the shape optimization of complex nonlinear structure, based on the FE analysis combined with the sensitivity analysis, the gradient-based Kriging (GBK) method is applied. A new criterion to determine the likely best spatial correlation parameter in GBK based on Latin Hypercube Design (LHD) presented by authors is also applied. A more practical example for the shape optimization of automotive tire is presented to show the applicability of the GBK method.
Authors: Ping Liang, Qiang Zhang, Jun Wei
Abstract: Using the weighted inverse theory of matrices and the weighed inverse topological change method of finite element structures, a set of formulae used in the sensitivity analysis of the parameters of a structure are presented. Combined with the previous work of the author, it can analyze the static modification of the structures efficiently. The all formulae of this method are explicit ones and have advantages of simplicity and easily operation on the computer. An example is given to demonstrate the valid and the efficiency of the formulae.
Authors: Zi Yan Ren, Dian Hai Zhang, Minh Trien Pham, Chang Seop Koh
Abstract: A novel reliability calculation algorithm is developed and applied to a robust optimization of electromagnetic device with uncertain design variables taken into account. The proposed algorithm, to calculate the reliability of a given design, incorporates sensitivity analysis based on finite element method with Monte Carlo simulation (MCS) method so that it has higher numerical efficiency than conventional methods: MCS and first-order reliability methods. Through numerical examples, its applicability to a reliability-based robust optimal design of electromagnetic device is demonstrated.
Authors: Qing Chun Yang, Wei Lu, Ping Li
Abstract: Deep geological disposal (DGD) is selected for the long-term confinement of high-level radioactive waste (HLW) by many countries. Safety assessment of nuclear waste disposal in a deep geological repository requires understanding and quantifying radionuclide behavior through the hosting geological formation. This paper presents a numerical model to deal with the pore water composition in bentonite barrier in the evolving geochemical environment which includes bentonite, concrete and clay in a high level radioactive waste repository designed in clay formation, the model considers the following processes: advection, diffusion, aqueous complexation, mineral dissolution/precipitation and cation exchange. Bentonite porosity changes caused by mineral dissolution/precipitation reactions are taken into account in the model. The numerical sensitivity analysis to the accessible porosity of bentonite are performed, the results illustrate that the pH in bentonite is sensitive to changes in accessible porosity of bentonite, the concentrations of aqueous species are sensitive to bentonite porosity before 1000 years, and the effective diffusion coefficient of bentonite controls the extent of high pH plume in bentonite.
Authors: Fang Zhen Li, Cong Lu, Wen Rui Jiang
Abstract: This paper proposes a robust approach to solve the complex problem of concurrent tolerance allocation. The optimal tolerances are allocated based on the adjusted nominal values of the dimensions which are less sensitive to the product performance, and selection of manufacturing processes or machines from the alternatives have also been taken into account simultaneously. Immune genetic algorithm is used to solve the model and obtain the global optimal solution. The overrunning clutch assemble is illustrated as an example.
Showing 1 to 10 of 283 Paper Titles