The Experts below are selected from a list of 18999 Experts worldwide ranked by ideXlab platform
Francoisxavier Briol - One of the best experts on this subject based on the ideXlab platform.
-
bayesian probabilistic Numerical Integration with tree based models
Neural Information Processing Systems, 2020Co-Authors: Harrison Zhu, Zhichao Shen, Xing Liu, Ruya Kang, Seth Flaxman, Francoisxavier BriolAbstract:Bayesian quadrature (BQ) is a method for solving Numerical Integration problems in a Bayesian manner, which allows users to quantify their uncertainty about the solution. The standard approach to BQ is based on a Gaussian process (GP) approximation of the integrand. As a result, BQ is inherently limited to cases where GP approximations can be done in an efficient manner, thus often prohibiting very high-dimensional or non-smooth target functions. This paper proposes to tackle this issue with a new Bayesian Numerical Integration Algorithm based on Bayesian Additive Regression Trees (BART) priors, which we call BART-Int. BART priors are easy to tune and well-suited for discontinuous functions. We demonstrate that they also lend themselves naturally to a sequential design setting and that explicit convergence rates can be obtained in a variety of settings. The advantages and disadvantages of this new methodology are highlighted on a set of benchmark tests including the Genz functions, and on a Bayesian survey design problem.
A R Khoei - One of the best experts on this subject based on the ideXlab platform.
-
on the implementation of a multi surface kinematic hardening plasticity and its applications
International Journal of Plasticity, 2005Co-Authors: A R Khoei, N JamaliAbstract:Abstract This paper is concerned with an application of the multi-surface plasticity in solid mechanics and geotechnical problems. The model is of a von-Mises type with associated flow rule, originally proposed by Montans. The Mroz translation rule is implemented to the movements of the yield surfaces and the fully implicit scheme with radial mapping method is applied in Numerical computations. Algorithmic consistent tangent modulus with Numerical Integration Algorithm of constitutive equations is extracted. The model is developed in the class of kinematic hardening models, so the ‘Masing’ rule is preserved. The model is able to consider the plastic strain accumulation in constant axial stress state, such as ratcheting. The implementation is validated by means of a simple deformation path of combined extension and compression test, a pure shear test with pseudo-random loading, a test which demonstrates the capabilities of the model in simulation of cyclic loading and ratcheting, a cyclic shear test in saturated undrained sand and finally, the analysis of a plate with holes, which presents the shear band using the multi-surface plasticity model.
-
an implicit Algorithm for hypoelasto plastic and hypoelasto viscoplastic endochronic theory in finite strain isotropic kinematic hardening model
International Journal of Solids and Structures, 2003Co-Authors: A R Khoei, A Bakhshiani, Massood MofidAbstract:This paper is concerned with objective stress update Algorithm for elasto-plastic and elasto-viscoplastic endochronic theory within the framework of additive plasticity. The elastic response is stated in terms of hypoelastic model and endochronic constitutive equations are stated in unrotated frame of reference. A trivially incrementally objective Integration scheme for rate constitutive equations is established. Algorithmic modulus consistent with Numerical Integration Algorithm of constitutive equations is extracted. The implementation is validated by means of a set of simple deformation paths (simple shear, extension and rotation), two benchmark test in nonlinear mechanics (the necking of a circular bar and expansion of a thick-walled cylinder), a test which demonstrates the capabilities of the proposed model in simulation of cyclic loading and ratcheting in finite strain case (cyclically loaded notched bar) and finally, the analysis of a tensile test, which presents a shear band with a finite thickness independent of the finite element mesh using endochronic viscoplastic constitutive model.
Harrison Zhu - One of the best experts on this subject based on the ideXlab platform.
-
bayesian probabilistic Numerical Integration with tree based models
Neural Information Processing Systems, 2020Co-Authors: Harrison Zhu, Zhichao Shen, Xing Liu, Ruya Kang, Seth Flaxman, Francoisxavier BriolAbstract:Bayesian quadrature (BQ) is a method for solving Numerical Integration problems in a Bayesian manner, which allows users to quantify their uncertainty about the solution. The standard approach to BQ is based on a Gaussian process (GP) approximation of the integrand. As a result, BQ is inherently limited to cases where GP approximations can be done in an efficient manner, thus often prohibiting very high-dimensional or non-smooth target functions. This paper proposes to tackle this issue with a new Bayesian Numerical Integration Algorithm based on Bayesian Additive Regression Trees (BART) priors, which we call BART-Int. BART priors are easy to tune and well-suited for discontinuous functions. We demonstrate that they also lend themselves naturally to a sequential design setting and that explicit convergence rates can be obtained in a variety of settings. The advantages and disadvantages of this new methodology are highlighted on a set of benchmark tests including the Genz functions, and on a Bayesian survey design problem.
Yan Xin - One of the best experts on this subject based on the ideXlab platform.
-
ssct a simple sequential spectrum sensing scheme for cognitive radio
Global Communications Conference, 2009Co-Authors: Yan Xin, Honghai Zhang, Sampath RangarajanAbstract:Cognitive radio that supports a secondary and opportunistic access to licensed spectrum shows great potential to dramatically improve spectrum utilization. Spectrum sensing performed by secondary users to detect unoccupied spectrum bands, is a key enabling technique for cognitive radio. This paper proposes a truncated sequential spectrum sensing scheme, namely the sequential shifted chi-square test (SSCT). The SSCT has a simple test statistic and does not rely on any deterministic knowledge about primary signals. As figures of merit, the exact false-alarm probability is derived, and the miss-detection probability as well as the average sample number (ASN) are evaluated by using a Numerical Integration Algorithm. Corroborating Numerical examples show that, in comparison with fixed-sample size detection schemes such as energy detection, the SSCT delivers considerable reduction on the ASN while maintaining a comparable detection performance.
-
a simple sequential spectrum sensing scheme for cognitive radio
arXiv: Information Theory, 2009Co-Authors: Honghai Zhang, Yan XinAbstract:Cognitive radio that supports a secondary and opportunistic access to licensed spectrum shows great potential to dramatically improve spectrum utilization. Spectrum sensing performed by secondary users to detect unoccupied spectrum bands, is a key enabling technique for cognitive radio. This paper proposes a truncated sequential spectrum sensing scheme, namely the sequential shifted chi-square test (SSCT). The SSCT has a simple test statistic and does not rely on any deterministic knowledge about primary signals. As figures of merit, the exact false-alarm probability is derived, and the miss-detection probability as well as the average sample number (ASN) are evaluated by using a Numerical Integration Algorithm. Corroborating Numerical examples show that, in comparison with fixed-sample size detection schemes such as energy detection, the SSCT delivers considerable reduction on the ASN while maintaining a comparable detection performance.
L Malcher - One of the best experts on this subject based on the ideXlab platform.
-
an improved damage evolution law based on continuum damage mechanics and its dependence on both stress triaxiality and the third invariant
International Journal of Plasticity, 2014Co-Authors: L Malcher, E N MamiyaAbstract:Abstract In this contribution, it is suggested a damage evolution law which is based on Continuum Damage Mechanics (CDM) and dependent on the hydrostatic pressure, by the stress triaxiality, and the third invariant of deviatoric stress tensor, by the so-called normalized third invariant. The contribution has been motivated meanly by the reason that the accuracy in describe the mechanical behavior of materials and the predictive fracture onset ability of damage constitutive models are strongly dependent on the loading condition used to procedure the calibration of material parameters. Regarding classical damage models as Lemaitre and Gurson, the level of material degradation can be optimist or conservative for loading conditions far from the calibration point. In the first part of this paper, the suggested damage evolution law is presented and the new state and dissipation potential are determined. The plastic flow rules for associative and non-associative plasticity are derived and an implicit Numerical Integration Algorithm is suggested, based on the operator split methodology. The Numerical Algorithm is also implemented in an “in house” finite element framework and its robustness is tested for a set of Numerical simulations upon wide range of stress triaxiality. Numerical results are compared with experimental data presented in literature and parameters as reaction curve, evolution of the equivalent plastic strain and damage variable at fracture are analyzed. In a critical situation, the Numerical results have shown that the original damage models as Lemaitre’s model has a prediction of 68% in disagreement with experimental data and the proposed damage evolution law has around 1%, regarding the determinations of the displacement at fracture initiation.
-
an extended gtn model for ductile fracture under high and low stress triaxiality
International Journal of Plasticity, 2014Co-Authors: L Malcher, F Andrade M PiresAbstract:Abstract This contribution provides an improvement on GTN model upon the prediction of fracture location within low level of stress triaxiality. In the proposition, two distinct damage parameters are introduced as internal variables of the degradation process and an effective damage is calculated as a sum of both contributions in the post-processed step. In the beginning, the volume void fraction, based on conservation mass law, is assumed as the first damage parameter, similar to Gurson’s original model. This volumetric damage contribution is able to capture spherical void growth, which plays the main role in tensile loading condition. The second damage parameter is proposed as a new shear mechanism, based on geometrical and phenomenological aspects and is also a function of the equivalent plastic strain, Lode angle and stress triaxiality. The shear damage parameter is formulated independent of the volume void fraction and requires a new nucleation of micro-defects mechanism to trigger the shear growth contribution, and hence is able to capture elongated (and rotation) void growth which is present in simple shear and combined shear/tensile or shear/compression loading conditions. Furthermore, the first and the second damage parameters are coupled in the yield function in order to affect the hydrostatic stress and deviatoric stress contributions, separately. In the first part of this paper, a review of Gurson’s model and its most famous version as GTN’s model is done. After that, the new contribution is presented and an implicit Numerical Integration Algorithm is determined, based on the operator split methodology. The calibration strategy is discussed for determination of material parameters. Numerical tests are performed for a butterfly specimen using two types of materials (aluminum alloy 2024-T351 and steel 1045) under ranges of stress triaxiality between - 1 / 3 η 1 / 3 (shear/compression or shear/tensile). At the end, the behavior of internal variables is analyzed, such as: evolution of both damage parameters, evolution of the equivalent plastic strain, the reaction curve and the contour of the effective damage parameter. The results obtained are compared with experimental data and have shown that the present formulation performs well in the prediction of the fracture location and determination of the correct level of equivalent plastic strain at fracture under predominant shear loading condition.
