Type

Database

Creator

Date

Thumbnail

Search results

213 records were found.

In this paper we demonstrate that it is possible to gradually improve the performance of support vector machine (SVM) classifiers by using a genetic algorithm to select a sequence of training subsets from the available data. Performance improvement is possible because the SVM solution generally lies some distance away from the Bayes optimal in the space of learning parameters. We illustrate performance improvements on a number of benchmark data sets.
The Tree Augmented Naïve Bayes (TAN) classifier relaxes the sweeping independence assumptions of the Naïve Bayes approach by taking account of conditional probabilities. It does this in a limited sense, by incorporating the conditional probability of each attribute given the class and (at most) one other attribute. The method of boosting has previously proven very effective in improving the performance of Naïve Bayes classifiers and in this paper, we investigate its effectiveness on application to the TAN classifier.
We describe here the rheological response of dense, slowly deforming granular materials to shear in a cylindrical Couette cell. All components of the stress on the outer cylinder are measured pointwise as a function of the depth, for different methods of construction of the bed that presumably lead to distinct fabrics. The static stress profiles for the different construction protocols are different, but a stress profile that is independent of construction history emerges when the granular column is sheared for sufficient time, in accord with the predictions of plasticity theories. However the qualitative features of the the stress profile under shear differs radically from the predictions of plasticity theories and data reported in earlier studies. We discuss a hypothesis for the anomalous stress profiles that was proposed recently by...
We present a novel way to extract continuum fields from discrete particle systems that is applicable to flowing mixtures as well as boundaries and interfaces. The mass and momentum balance equations for mixed flows are expressed in terms of the partial densities, velocities, stresses and interaction terms for each constituent. Expressions for these variables in terms of the microscopic quantities are derived by coarse-graining the balance equations, and thus satisfy them exactly. A simple physical argument is used to apportion the interaction forces to the constituents. Discrete element simulations of granular chute flows are presented to illustrate the strengths of the new boundary/mixture treatment. We apply the mixture formulation to confirm two assumptions on the segregation dynamics in particle simulations of bidispersed chute flo...
Discrete particle simulations of granular materials under 2D, isochoric, cyclic pure shear have been performed and are compared to a recently developed constitutive model involving a deviatoric yield stress, dilatant stresses and structural anisotropy. The original model shows the cyclic response qualitatively, but suffers from an artificial drift in pressure. With a small modification in the definition of the stress anisotropy and an additional limit-pressure term in the evolution equation for the pressure, it is able to show the transient as well as the limit cycles. The overall goal – beyond the scope of the present study – is to develop a local constitutive model that is able to predict real life, large scale granular systems.
Our mesoscale simulation method [M. Robinson, S. Luding, and M. Ramaioli, submitted (2013)] for multiphase fluid-particle flows couples Smoothed Particle Hydrodynamics (SPH) and the Discrete Element Method (DEM) and enjoys the flexibility of meshless methods, such as being capable to handling free surface flows or flow around complex and/or moving geometries. We use this method to simulate three different sedimentation test cases and compare the results to existing analytical solutions. The grain velocity in Single Particle Sedimentation compares well (< 2% error) with the analytical solution as long as the fluid resolution is coarser than two times the particle diameter. The multiple particle sedimentation problem and Rayleigh Taylor Instability (RTI) also perform well against the theory, but it was found that the method is susceptibl...
General-purpose computation on Graphics Processing Units (GPU) on personal computers has recently become an attractive alternative to parallel computing on clusters and supercomputers. We present the GPU-implementation of an accurate molecular dynamics algorithm for a system of spheres. The new hybrid CPU-GPU implementation takes into account all the degrees of freedom, including the quaternion representation of 3D rotations. For additional versatility, the contact interaction between particles is defined using a force law of enhanced generality, which accounts for the elastic and dissipative interactions, and the hard-sphere interaction parameters are translated to the soft-sphere parameter set. We prove that the algorithm complies with the statistical mechanical laws by examining the homogeneous cooling of a granular gas with rotatio...
This thesis presents the top-down design of a control electronics that is used to control a thermal actuator system for autonomous alignment functionality. With special heating and cooling sequences, the actuator system is able to create displacements in small steps. So it is also called “thermal stepper system”. This system is ideal for aligning critical parts in complex, high-precision systems, like, for example, optical elements or the electrodes of capacitive sensors. The performances of the thermal stepper system relay on the control of temperatures. Therefore, a precise temperature-measurement and temperature-control system is designed and implemented with discrete components. It has three main functions, which are: auto-alignment, auto-calibration and stability compensation. The auto-alignment is realized by means of the therm...
Recently, tactile sensors mounted on robot fingers have been identified as essential sensory devices for the control of multifingered robotic hands. A basic tactile sensing task is to determine the force distribution on the contact area between the fingers and grasped object. To increase the grasp stability and to protect the fragile sensors, a kind of elastic material is required to cover the tactile sensors. This thesis derives the relationship between the surface force profile and the stress or strain profile measured by tactile sensors beneath the contract surface for simplified situations. This relationship can be described by integral equations of convolution type, or more generally, integral equations of the first kind with two unknown functions. The algorithms for numerical inversion in real time, an analog network for solving ...
This dissertation is concerned with dynamic modeling and kinematic control of constrained mechanical systems with symmetry from a geometric point of view. Constraints are defined via the characteristics of distributions or codistributions on the tangent bundle (velocity phase space) of configuration space. Lie symmetry groups acting on the systems are assumed to leave both Lagrangian and constraints invariant. As a special case of mechanical systems with holonomic constraints, we rigorously analyze the kinematics and dynamics of floating, planar four-bar linkages. The analyses include topological description of the configuration space, symplectic and Poisson reductions of the dynamics and bifurcation of relative equilibria. for kinematic control of nonholonomic systems, we mainly study the related optimal control problem for a system c...
Want to know more?If you want to know more about this cutting edge product, or schedule a demonstration on your own organisation, please feel free to contact us or read the available documentation at http://www.keep.pt/produtos/retrievo/?lang=en