<p>Shiv Nadar University is a comprehensive, multidisciplinary, research-focused, and student-centric University offering a full range of academic programs at the undergraduate, postgraduate and doctoral level. With state-of-the-art infrastructure, the University comprises of academic wings, sophisticated labs, international standard sports facilities, amphitheaters, auditorium, conference rooms and smart classrooms. The University&rsquo;s goal is to become internationally recognized for the quality of its research and creative endeavors and their applicability to improving quality of life, generating new insights and expanding the boundaries of human knowledge creativity. Committed to excellence in teaching, research and service, the University aims to serve the higher education needs of India and the world beyond.</p>

Shiv Nadar University

Center for Informatics

To date, there has been considerable growth in the development and application of artificial neural networks. A new approach toward this application, so-called support vector machines (SVM), is proposed. The superior classification and regression abilities of this approach offers an attractive alternative to solve a number of process engineering problems.

World University root node

Chemical Engineering Progress

Support vector machines: A useful tool for process engineering applications

Na Adsorption on the Si(111)-(7 × 7) Surface: From Two-Dimensional Gas to Nanocluster Array

We have systematically investigated Na adsorption on the [Formula presented] surface at room temperature using scanning tunneling microscopy (STM). Below the critical coverage of 0.08 monolayer, we find intriguing contrast modulation instead of localized Na adsorbates, coupled with streaky noise in the STM images, which is accompanied by monotonic work function drop. Above the critical coverage, Na clusters emerge and form a self-assembled array. Combined with first-principles theoretical simulations, we conclude that the Na atoms on the ([Formula presented]) surface are, while strongly bound ([Formula Presented]) to the surface, highly mobile in “basins” around the Si rest atoms, forming a two-dimensional gas phase at the initial coverage and that the cluster at the higher coverage consists of six Na atoms together with three Si adatoms. {\textcopyright} 2003 The American Physical Society.

Physical Review Letters

Na adsorption on the [formula presented] surface: From two-dimensional gas to nanocluster array

Fragmentation of germanium and tin cluster ions (Gex+, Snx+: x = 4-20) in the low-energy collisions with a Si surface has been investigated by means of a tandem time-of-flight mass spectrometer. 1At low incident energies, smaller clusters fragmented by an atom loss process, whereas larger clusters decayed by fission. The favored fragmentation paths for both cluster ions were similar to those for Si cluster ions. The results support the structural similarities among Si, Ge and Sn clusters in the present size range. For tin cluster ions, low-energy fragmentation patterns were compared with those obtained from theoretical calculations using generalized gradient approximation (GGA) and the B3PW91 exchange-correlation functional. It has been found that the B3PW91 hybrid functional results are consistent with the experimental observations.

European Physical Journal D

Low-energy surface collision induced dissociation of Ge and Sn cluster ions

A Tabu search based approach for solving a class of bilevel programming problems in chemical engineering was presented. An algortihm was also tested for a number of benchmark problems. The results obtained showed superiority of the approach over the conventional methods in solving benchmark problems.

Journal of Heuristics

A Tabu Search Based Approach for Solving a Class of Bilevel Programming Problems in Chemical Engineering

Kernel PCA methodology, an elegant nonlinear generalization of the linear PCA, is illustrated by considering the examples of (i) denoising chaotic time series and, (ii) prediction of properties of polymer nanocomposites developed in our laboratory. Kernel PCA captures the dominant nonlinear features of the original data by transforming it to a high dimensional feature space. An appropriately defined kernel function allows the computations to be performed in the original input space and facilitates extraction of substantially higher number of principal components enabling excellent denoising and feature extraction capabilities. Use of simple matrix algebra in simulations makes the method an attractive alternative to some hard optimization based methodologies. © 2003 Elsevier Ltd. All rights reserved.

Chemical Engineering Science

Feature extraction and denoising using kernel PCA

Ab initio calculations have been performed using density-functional theory with the B3PW91 hybrid exchange-correlation functional and the Gaussian method to obtain the electronic and vibrational properties of the fullerene (f) and Frank-Kasper (FK) isomers of the metal-encapsulated silicon clusters M@Si16, M=Ti and Zr. The electron affinities of the two isomers are found to differ significantly and our result for FK-Ti@Si16 is in good agreement with recent experiments. The Raman and infrared vibrational spectra of the f and FK isomers show marked differences, due to their distinct bonding natures and structural features, that can be used unambiguously to identify the structures of these clusters experimentally. The polarizabilities, however, have similar values and lie above the bulk limit of silicon. The optical gaps and absorption spectra have been calculated using time-dependent density-functional theory. The lowest electronic excitation for the FK isomer lies in the deep blue region, while the one for the f isomer lies in the red region, making them attractive for optoelectronic applications. {\textcopyright} 2003 The American Physical Society.

Physical Review B - Condensed Matter and Materials Physics

Ab initio calculations of electronic structures, polarizabilities, Raman and infrared spectra, optical gaps and absorption spectra of M@Si16 (M=Ti and Zr) clusters

Ab initio prediction of nano-structured materials using supercomputer

This work provides a methodology for the control of complex non-linear systems by compensating for the non-linear part using support vector machines (SVM) and subsequently developing simple linear feedback law for control. The method tested for the benchmark Rössler and Lorenz chaotic oscillators shows excellent performance. © 2003 Elsevier B.V. All rights reserved.

Physics Letters, Section A: General, Atomic and Solid State Physics

Control of chaotic dynamical systems using support vector machines

The geometries of several Mn clusters in the size range Mn13-Mn23 are studied via the generalized gradient approximation to density functional theory. For the 13- and 19-atom clusters, the icosahedral structures are found to be most stable, while for the 15-atom cluster, the bcc structure is more favoured. The clusters show ferrimagnetic spin configurations.