Publication8

In this paper we address an aspect of speaker recognition task, viz. unsupervised speaker turn detection. A metric based approach with two-pass criteria is proposed for this task. A GMM-based modified Log Likelihood Ratio metric is used in the first pass; Bayesian Information Criterion (BIC) metric is used in the second pass to verify or discard the speaker turn points hypothesized in the first pass. We consider two cases: long speaker turn segments (> 2 sec.) and short speaker turn segments (< 2 sec.). We have evaluated our algorithm using TIMIT speech files. Our precision results range from 85% to 93%, recall ranges from 75% to 78%, and the F-ratio is in the range 80-85%. These results are better than what has been reported in the literature so far.

The success of a spacecraft mission depends to a great extent on the performance of the moving mechanical systems. The most common mode of failure in these systems is tribological. Tribological failures occur mainly due to nonavailability of lubricant at the working zone of the bearings as a result of degradation, evaporation, and creep. The life of these moving mechanical systems could be extended if lubricant is replenished by some means. Therefore, to ensure long-term uninterrupted performance of these systems, an efficient lubricant replenishment system is essential. This article describes the development of a novel lubricant supply system that can supply lubricant for more than 30 years at a controlled rate of a few micrograms/hour. Experimental evaluation of the lubricator was carried out under different operating conditions experienced in a spacecraft. The results of the experiments are compared with the theoretical results obtained from the mathematical model and computational fluid dynamics (CFD) simulation. The lowest flow rate obtained was 3.5 μg/h from a lubricator with a flow control orifice diameter of 2.3 μm. The results of the study show that the lubricator can provide a long-term uninterrupted supply of lubricant to the moving mechanical systems for a period of more than 30 years. An outstanding and most promising feature of this lubricator is that the flow rate at various stages of the life can be theoretically predicted using the developed mathematical model and the prediction can be experimentally verified before the system is put into service. © 2011 Society of Tribologists and Lubrication Engineers.

An experimental investigation of the non-normal nature of thermoacoustic interactions in an electrically heated horizontal Rijke tube is performed. Since non-normality and the associated transient growth are linear phenomena, the experiments have to be confined to the linear regime. The bifurcation diagram for the subcritical Hopf bifurcation into a limit cycle behavior has been determined, after which the amplitude levels, for which the system acts linearly, have been identified for different power inputs to the heater. There are two main objectives for this experimental investigation. The first one deals with the extraction of the linear eigenmodes associated with the acoustic pressure from experimental data. This is accomplished by the Dynamic Mode Decomposition (DMD) technique applied in the linear regime. The non-orthogonality between the eigenmodes is determined for various values of heater power. The second objective is to identify evidence of transient perturbation growth in the system. The total acoustic energy in the duct has been monitored as the thermoacoustic system has been initialized by linear combinations of the two dominant eigenmodes. Transient growth, on the order of previous theoretical studies, has been found, and its parameter dependence on amplitude ratio and phase angle of the initial eigenmode components has been determined. This study represents the first experimental confirmation of non-normality in thermoacoustic systems. © 2011 by S. Mariappan, R. I. Sujith and P. J. Schmid.

Abstract.: In this paper, a unit cell based approach is followed, where a unit cell consisting of one aggregate surrounded by mortar matrix is used for numerical simulation of mechanical response of cement concrete. Unit cell approach is a simple mathematical approximation that helps us to simplify the simulation of mechanical response of multi-phase composites. To model the failure of matrix, brittle cracking model is used, where the entire fracture zone is represented by a band of micro cracked material. Current study involves; (a) failure analysis of the concrete unit cell when it is subjected to tensile loads, and (b) parametric study of variation of peak strength with shape and volume fraction of aggregate. In this study, circular and square aggregates at various orientations are modelled. The simulation results predict that the peak tensile stresses are not very sensitive to the volume fraction of aggregates, when the unit cell is subjected to tensile loads. This paper effectively demonstrates the power of unit cell model in simulating the nonlinear mechanical response of cement concrete when it is subjected to tensile loading. © 2011 Indian Academy of Sciences.

Automated recognition of worker activities has the potential in aiding quick assessment of labour productivity on construction sites. A novel method called accelerometer based activity recognition has been investigated and preliminary results show that it has good potential for deployment in construction environment. The major decisive factor influencing the performance of the activity recognition system is the location of the accelerometer on the human body. The objective of this study is to determine a-priori, the appropriate accelerometer location using videos of construction activities. A framework was developed to track the movement of body segments through observation using Anvil, a generic annotation tool. Locations of accelerometer are selected after evaluating the information gain of each body segment towards activity classification with due consideration to subject comfort and integration possibilities. A study of masonry activity using the framework identified the placement locations as right lower arm, left lower arm and waist. An experimental setup was arranged for determining performance of the accelerometer based activity recognition system for a mason working in uninstructed environment with accelerometers attached at selected locations. Activity recognition performance of the locations was evaluated with ten runs of 10-fold cross validation using multilayer perceptron algorithm. The results showed that classifier performances for the three locations have the same order of ranking as predicted by the framework. The activity recognition performance for the selected locations gave accuracies above 80% and it can be concluded that the proposed framework can be used for placing accelerometers at appropriate locations for activity recognition.

We report on high sensitivity detection of acoustic signals using Fiber Bragg gratingbased Fabry-Perot filters. Our scheme is based on an APD-based receiver with bandpass filter, 16- bit ADCs, and novel spectrum estimation techniques. © OSA/ANIC/IPR/Sensors/SL/SOF/SPPCom/2011.

Over roughly the last twenty-five years, the topic of project delivery and finance has received significant attention in the infrastructure community and the construction industry. In the United States, the public sector’s mandated use of a delivery system where the activities of design, construction and operations were segregated, or independent, from one another has shifted toward a more open framework where other alternatives are now permitted and implemented. Globally, similar changes have occurred. The rationale behind these transitions is discussed subsequently. Further, this field remains in flux - a consequence, in part, of the search for ways to better align project stakeholder interests and objectives.

A lumped vortex model coupled with a vortex dissipation and vortex core criteria is used to study the unsteady flow past two airfoils in configuration, each of which is impulsively set into motion. The unsteady wakes of both airfoils are modeled by discrete vortices and timestepping is used to predict the individual wake shapes. The coupled flow is solved using a combined "zero-normal flow" boundary condition and Kelvin condition which results in (2N +2)X(2N+2) equations. Results are presented showing the effect of airfoil-airfoil and airfoilwake interaction on the aerodynamic characteristics of the configuration for the airfoils moving with zero and non-zero relative velocity. © 2011 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.

Here we report an approach to fabricate flexible polymer self standing films embedded with continuous aligned microtubes/microchannels via electrospinning. The scheme is to wash the electrospun fibers selectively to form either microtubes/microchannels. Optical microscope (OM) and Scanning electron microscope images are evident for the well aligned microtubes and microchannels respectively with a diameter of ∼8 μm and a length of ∼4 cm. Meniscus of tetrahydrofuran in the microtubes can be observed explicitly in the OM images. Mutually perpendicular microtubes are also fabricated on a polymer film. Angular distribution of aligned microstructures indicates the standard deviation is not more than 1°. These tubes/channels can be potential for tissue engineering as they could provide a directional template for the growth when a biodegradable polymer such as Poly(vinylalcohol) is used. © 2011 Elsevier B.V. All rights reserved.

The presence of backlash in the gearheads is an inherent problem in manipulators using geared motors. This paper looks at a potential solution to this problem via the implementation of a dual-loop control strategy, in which feedback is taken from the motors as well as the end-effector of the manipulator. Using the redundant sensed information, the actual error in the joint-space is computed and used to rectify the desired trajectory for the joint-space trajectory-tracking control scheme. Experiments done on a 3-RRR planar parallel manipulator show significant improvement in the tracking performance due to the introduction of dual-loop control scheme.

Seawater attack presents a number of challenges to the concrete designer. The combined attack by sulphates and chlorides in seawater involves a complicated mechanism. In a number of cases, such as in outfall tunnels of seawater desalination projects, there is a possibility of encountering magnesium sulphate solutions (or magnesium brine solutions) having higher concentrations than seawater. This paper reports an experimental investigation to understand the influence of the solution concentration of seawater on the performance of cement mortars, through a combination of physical, chemical and microstructural analyses. Results indicate that higher concentrations of seawater do not cause the same level of damage as higher concentrations of groundwater relative to the typical seawater/groundwater solutions. The role of the brucite layer, formed on the surface during magnesium sulphate attack, is explored in detail to help understand the mechanisms governing the attack. © 2011 Taylor & Francis Group, LLC.

Postliberalization Indian Novels in English: Politics of Global Reception and Awards' is a critical handbook that focuses on trends in contemporary Indian novels and discusses the global reception of these works. The volume provides a systematic approach to the study of Indian novelists that have not been (with certain exceptions) extensively examined.

This work performs linear stability analysis to investigate the onset of convection due to microwave incidences in water and oil confined between two semiinfinite horizontal plates. The onset has been characterized in terms of critical Rayleigh (Rac) and wave (kc) numbers, which are determined for the entire spectrum sample thickness varying from thin to thick samples exposed to various microwave source configurations including one sided to distributed incidences. The main finding of this work is the oscillatory variations of Rac and kc with sample thickness in sinusoidal regime of microwave heating, which indicates preferential onset for specific sample thicknesses. It has also been shown that materials as well as microwave source configuration play important roles in the onset of convection in this regime. The variation patterns of Rac and kc have been correlated to the spatial patterns of absorbed microwave power within the sample. These results are useful not only for forecasting the onset but also for controlling onset of convection by selecting suitable experimental parameters. Copyright c 2011 by ASME.

It has been reported in literature that the use of protrusion in the combustion chamber of a hybrid rocket motor enhances the regression rate. This study reports careful experiments conducted to show that the improvement in burn rate with protrusions is only up to a certain fraction of the overall burn time. From the results obtained, it is seen that an X/L of 0.5 is the best location for a graphite protrusion. The protrusions are also shown to increase the combustion eficiency by as much as 45 % when placed at an X/L of 0.5. This, more than the improvement in the regression rate is very useful, especially for small scale motors whose combustion eficiencies are otherwise very low. © 2011 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.

The classical trade-off between NOx and soot emissions from conventional diesel engines has been a limiting factor in meeting ever stringent emission norms. The electronic control of fuel injection in diesel engines emerged as an important strategy for their simultaneous reduction. The high pressure multiple-injection in a common rail direct injection system has been promising in this regard. While, the effects of pilot injection or multiple pulses of CRDI injection schedule on simultaneous reduction of NOx and soot have been widely investigated and reported, the investigations concerning three and more injection pulses have been limited. In this paper, the ability of a predictive model, developed by the authors, in providing optimal multiple-injection schedule is demonstrated through parametric investigations. The effects of pilot and post fuel quantity and dwell between the injection pulses on NOx and soot emissions are discussed. The emission predictions from the model are corroborated with available measured data. The comparisons between predicted and available measured data show that while an optimum quantity of pilot fuel reduces NOx emission, the post injection fuel quantity and its dwell from main injection pulse lowers soot emission. © Copyright 2011 Society of Automotive Engineers of Japan, Inc. and SAE International.

This study investigates the interaction of micron-sized coal particles entrained into lean methane - air premixed flames. In a typical axisymmetric burner, coal particles are made to naturally entrain into a stream of the premixed reactants using an orifice plate setup. Pittsburgh seam coal dust, with three particle sizes in the range of 0 to 25 μm, 53 to 63 μm, and 75 to 90 μm is used. The effects of different coal dust concentrations (10-300 g/m3) at three lean equivalence ratios, φ (methane-air) of 0.75, 0.80 and 0.85 on the laminar burning velocity are determined experimentally. The laminar burning velocity of the coal dust-methane-air mixture is determined by taking a shadowgraph of the resulting flame and using the cone-angle method. The results show that the addition of coal dust in methane-air premixed flame reduces the laminar burning velocity at particle size of 53 to 63 μm and 75 to 90 μm. However, burning velocity promotion is observed for 0 to 25 μm particles at φ = 0.80. Two competing effects are assumed involved in the process. The first is burning velocity promotion effect that the released volatile increases the gaseous mixture equivalence ratio and thus the burning velocity. The second is the heat sink effect of the coal particles to reduce the flame temperature and accordingly the burning velocity. A mathematical model is developed based on such assumption and it can successfully predict the change of laminar burning velocity at various dust concentration. Furthermore, the implication of this study to coal mine safety is discussed.

Behaviour of slender reinforced concrete (RC) beams is different from that of normally propotioned beams, on account of slenderness effects which introduce susceptibility to lateral torsional buckling. Highly slender beams are prone to sudden instability mode of failure. Moderately slender beams are also susceptible to slenderness effects and they may undergo flexural failure at moment values less than their flexural capacities corresponding to material failure (Muf). Concrete design codes presently do not provide any procedure to evaluate the critical buckling moment (Mcr), and also do not account for the capacity reduction in moderately slender beams on account of slenderness. The existing recommendations are limited to prescriptions of limiting slenderness ratios, which are semi-empirical in nature. The experimental results carried out in the present study clearly show reduction in moment capacity in beams with slenderness ratios well below the critical values given in various codes and in literature. Based on the experimental and theoretical studies conducted, new design guidelines are proposed for slender rectangular RC beams.