Abstract: The paper presents a new hybridization methodology involving Neural, Fuzzy and Rough Computing. A Rough Sets based approximation technique has been proposed based on a certain Neuro – Fuzzy architecture. A New Rough Neuron composition consisting of a combination of a Lower Bound neuron and a Boundary neuron has also been described. The conventional convergence of error in back propagation has been given away for a new framework based on 'Output Excitation Factor' and an inverse input transfer function. The paper also presents a brief comparison of performances, of the existing Rough Neural Networks and ANFIS architecture against the proposed methodology. It can be observed that the rough approximation based neuro-fuzzy architecture is superior to its counterparts.
Abstract: In the numerical solution of the forward dynamics of a
multibody system, the positions and velocities of the bodies in the
system are obtained first. With the information of the system state
variables at each time step, the internal and external forces acting on
the system are obtained by appropriate contact force models if the
continuous contact method is used instead of a discrete contact
method. The local deformation of the bodies in contact, represented
by penetration, is used to compute the contact force. The ability and
suitability with current cylindrical contact force models to describe
the contact between bodies with cylindrical geometries with
particular focus on internal contacting geometries involving low
clearances and high loads simultaneously is discussed in this paper.
A comparative assessment of the performance of each model under
analysis for different contact conditions, in particular for very
different penetration and clearance values, is presented. It is
demonstrated that some models represent a rough approximation to
describe the conformal contact between cylindrical geometries
because contact forces are underestimated.