Time variant reliability model updating in instrumented dynamical systems based on Girsanov’s transformation

The problem of updating the reliability of instrumented structures based on measured response under random dynamic loading is considered. A solution strategy within the framework of Monte Carlo simulation based dynamic state estimation method and Girsanov’s transformation for variance reduction is developed. For linear Gaussian state space models, the solution is developed based on continuous version of the Kalman filter, while, for non-linear and (or) non-Gaussian state space models, bootstrap particle filters are adopted. The controls to implement the Girsanov transformation are developed by solving a constrained non-linear optimization problem. Numerical illustrations include studies on a multi degree of freedom linear system and non-linear systems with geometric and (or) hereditary non-linearities and non-stationary random excitations.

Perception of time-varying signals: timbre and phonetic JND of diphthong

In this paper we propose a linear time-varying model for diphthong synthesis based on linear interpolation of formant frequencies. We, thence, determine the timbre just-noticeable difference (JND) for diphthong /a I/ (as in ‘buy’) with a constant pitch excitation through perception experiment involving four listeners and explore the phonetic JND of the diphthong. Their JND responses are determined using 1-up-3-down procedure. Using the experimental data, we map the timbre JND and phonetic JND onto a 2-D region of percentage change of formant glides. The timbre and phonetic JND contours for constant pitch show that the phonetic JND region encloses timbre JND region and also varies across listeners. The JND is observed to be more sensitive to ending vowel /I/ than starting vowel /a/ in some listeners and dependent on the direction of perturbation of starting and ending vowels.

Role of the cloud adjustment time scale in simulation of the interannual variability of Indian summer monsoon

The simulation of precipitation in a general circulation model relying on relaxed mass flux cumulus parameterization scheme is sensitive to cloud adjustment time scale (CATS). In this study, the frequency of the dominant intra-seasonal mode and interannual variability of Indian summer monsoon rainfall (ISMR) simulated by an atmospheric general circulation model is shown to be sensitive to the CATS. It has been shown that a longer CATS of about 5 h simulates the spatial distribution of the ISMR better. El Nio Southern Oscillation-ISMR relationship is also sensitive to CATS. The equatorial Indian Ocean rainfall and ISMR coupling is sensitive to CATS. Our study suggests that a careful choice of CATS is necessary for adequate simulation of spatial pattern as well as interannual variation of Indian summer monsoon precipitation.

Predictability of nonstationary time series using wavelet and EMD based ARMA models

Research has been undertaken to ascertain the predictability of non-stationary time series using wavelet and Empirical Mode Decomposition (EMD) based time series models. Methods have been developed in the past to decompose a time series into components. Forecasting of these components combined with random component could yield predictions. Using this ideology, wavelet and EMD analyses have been incorporated separately which decomposes a time series into independent orthogonal components with both time and frequency localizations. The component series are fit with specific auto-regressive models to obtain forecasts which are later combined to obtain the actual predictions. Four non-stationary streamflow sites (USGS data resources) of monthly total volumes and two non-stationary gridded rainfall sites (IMD) of monthly total rainfall are considered for the study. The predictability is checked for six and twelve months ahead forecasts across both the methodologies. Based on performance measures, it is observed that wavelet based method has better prediction capabilities over EMD based method despite some of the limitations of time series methods and the manner in which decomposition takes place. Finally, the study concludes that the wavelet based time series algorithm can be used to model events such as droughts with reasonable accuracy. Also, some modifications that can be made in the model have been discussed that could extend the scope of applicability to other areas in the field of hydrology. (C) 2013 Elesvier B.V. All rights reserved.

Study on the Effect of Dead Time and its Compensation for Bus-Clamping PWM Techniques

Dead-time is provided in between the gating signals of the top and bottom semiconductor switches in an inverter leg to prevent the shorting of DC bus. Due to this dead time, there is a significant unwanted change in the output voltage of the inverter. The effect is different for different pulse width modulation (PWM) methodologies. The effect of dead-time on the output fundamental voltage is studied theoretically as well as experimentally for bus-clamping PWM methodologies. Further, experimental observations on the effectiveness of dead-time compensation are presented.

Modeling Time-Varying Aggregate Interference in Cognitive Radio Systems, and Application to Primary Exclusive Zone Design

Accurately characterizing the time-varying interference caused to the primary users is essential in ensuring a successful deployment of cognitive radios (CR). We show that the aggregate interference at the primary receiver (PU-Rx) from multiple, randomly located cognitive users (CUs) is well modeled as a shifted lognormal random process, which is more accurate than the lognormal and the Gaussian process models considered in the literature, even for a relatively dense deployment of CUs. It also compares favorably with the asymptotically exact stable and symmetric truncated stable distribution models, except at high CU densities. Our model accounts for the effect of imperfect spectrum sensing, which depends on path-loss, shadowing, and small-scale fading of the link from the primary transmitter to the CU; the interweave and underlay modes or CR operation, which determine the transmit powers of the CUs; and time-correlated shadowing and fading of the links from the CUs to the PU-Rx. It leads to expressions for the probability distribution function, level crossing rate, and average exceedance duration. The impact of cooperative spectrum sensing is also characterized. We validate the model by applying it to redesign the primary exclusive zone to account for the time-varying nature of interference.

Role of orography in modulating space-time scales of convection over South Asia

Propagation of convective systems in the meridional direction during boreal summer is responsible for active and break phases of monsoon over south Asia. This region is unique in the world in its characteristics of monsoon variability and is in close proximity of mountains like the Himalayas. Here, using an atmospheric general circulation model, we try to understand the role of orography in determining spatial and temporal scales of these convective systems. Absence of orography (noGlOrog) decreased the simulated seasonal mean precipitation over India by 23 % due to delay in onset by about a month vis-a-vis the full-mountain case. In noGlOrog, poleward propagations were absent during the delayed period prior to onset. Post-onset, both simulations had similar patterns of poleward propagations. The spatial and temporal scales of propagating clouds bands were determined using wavelet analysis. These scales were found to be different in full-mountain and no-mountain experiments in June-July. However, after the onset of monsoon in noGlOrog, these scales become similar to that with orography. Simulations with two different sets of convection schemes confirmed this result. Further analysis shows that the absence (presence) of meridional propagations during early (late) phase of summer monsoon in noGlOrog was associated with weaker (stronger) vertical shear of zonal wind over south Asia. Our study shows that orography plays a major role in determining the time of onset over the Indian region. However, after onset, basic characteristics of propagating convective systems and therefore the monthly precipitation over India, are less sensitive to the presence of orography and are modulated by moist convective processes.

Climate change scenarios of surface solar radiation in data sparse regions: a case study in Malaprabha River Basin, India

A variety of methods are available to estimate future solar radiation (SR) scenarios at spatial scales that are appropriate for local climate change impact assessment. However, there are no clear guidelines available in the literature to decide which methodologies are most suitable for different applications. Three methodologies to guide the estimation of SR are discussed in this study, namely: Case 1: SR is measured, Case 2: SR is measured but sparse and Case 3: SR is not measured. In Case 1, future SR scenarios are derived using several downscaling methodologies that transfer the simulated large-scale information of global climate models to a local scale ( measurements). In Case 2, the SR was first estimated at the local scale for a longer time period using sparse measured records, and then future scenarios were derived using several downscaling methodologies. In Case 3: the SR was first estimated at a regional scale for a longer time period using complete or sparse measured records of SR from which SR at the local scale was estimated. Finally, the future scenarios were derived using several downscaling methodologies. The lack of observed SR data, especially in developing countries, has hindered various climate change impact studies. Hence, this was further elaborated by applying the Case 3 methodology to a semi-arid Malaprabha reservoir catchment in southern India. A support vector machine was used in downscaling SR. Future monthly scenarios of SR were estimated from simulations of third-generation Canadian General Circulation Model (CGCM3) for various SRES emission scenarios (A1B, A2, B1, and COMMIT). Results indicated a projected decrease of 0.4 to 12.2 W m(-2) yr(-1) in SR during the period 2001-2100 across the 4 scenarios. SR was calculated using the modified Hargreaves method. The decreasing trends for the future were in agreement with the simulations of SR from the CGCM3 model directly obtained for the 4 scenarios.

Switching and Release Time Analysis of Electrostatically Actuated Capacitive RF MEMS Switches

This paper explains the reason behind pull-in time being more than pull-up time of many Radio Frequency Micro-Electro-Mechanical Systems (RF MEMS) switches at actuation voltages comparable to the pull-in voltage. Analytical expressions for pull-in and pull-up time are also presented. Experimental data as well as finite element simulations of electrostatically actuated beams used in RF-MEMS switches show that the pull-in time is generally more than the pull-up time. Pull-in time being more than pull-up time is somewhat counter-intuitive because there is a much larger electrostatic force during pull-in than the restoring mechanical force during the release. We investigated this issue analytically and numerically using a 1D model for various applied voltages and attribute this to energetics, the rate at which the forces change with time, and softening of the overall effective stiffness of the electromechanical system. 3D finite element analysis is also done to support the 1D model-based analyses.

Healing time for the growth of thin films on patterned substrates

The healing times for the growth of thin films on patterned substrates are studied using simulations of two discrete models of surface growth: the Family model and the Das Sarma-Tamborenea (DT) model. The healing time, defined as the time at which the characteristics of the growing interface are ``healed'' to those obtained in growth on a flat substrate, is determined via the study of the nearest-neighbor height difference correlation function. Two different initial patterns are considered in this work: a relatively smooth tent-shaped triangular substrate and an atomically rough substrate with singlesite pillars or grooves. We find that the healing time of the Family and DT models on aL x L triangular substrate is proportional to L-z, where z is the dynamical exponent of the models. For the Family model, we also analyze theoretically, using a continuum description based on the linear Edwards-Wilkinson equation, the time evolution of the nearest-neighbor height difference correlation function in this system. The correlation functions obtained from continuum theory and simulation are found to be consistent with each other for the relatively smooth triangular substrate. For substrates with periodic and random distributions of pillars or grooves of varying size, the healing time is found to increase linearly with the height (depth) of pillars (grooves). We show explicitly that the simulation data for the Family model grown on a substrate with pillars or grooves do not agree with results of a calculation based on the continuum Edwards-Wilkinson equation. This result implies that a continuum description does not work when the initial pattern is atomically rough. The observed dependence of the healing time on the substrate size and the initial height (depth) of pillars (grooves) can be understood from the details of the diffusion rule of the atomistic model. The healing time of both models for pillars is larger than that for grooves with depth equal to the height of the pillars. The calculated healing time for both Family and DT models is found to depend on how the pillars and grooves are distributed over the substrate. (C) 2014 Elsevier B.V. All rights reserved.

Ising model on a random network with annealed or quenched disorder

We study the equilibrium properties of an Ising model on a disordered random network where the disorder can be quenched or annealed. The network consists of fourfold coordinated sites connected via variable length one-dimensional chains. Our emphasis is on nonuniversal properties and we consider the transition temperature and other equilibrium thermodynamic properties, including those associated with one-dimensional fluctuations arising from the chains. We use analytic methods in the annealed case, and a Monte Carlo simulation for the quenched disorder. Our objective is to study the difference between quenched and annealed results with a broad random distribution of interaction parameters. The former represents a situation where the time scale associated with the randomness is very long and the corresponding degrees of freedom can be viewed as frozen, while the annealed case models the situation where this is not so. We find that the transition temperature and the entropy associated with one-dimensional fluctuations are always higher for quenched disorder than in the annealed case. These differences increase with the strength of the disorder up to a saturating value. We discuss our results in connection to physical systems where a broad distribution of interaction strengths is present.

Approximation of Spatio-Temporal Random Processes Using Tensor Decomposition

A new representation of spatio-temporal random processes is proposed in this work. In practical applications, such processes are used to model velocity fields, temperature distributions, response of vibrating systems, to name a few. Finding an efficient representation for any random process leads to encapsulation of information which makes it more convenient for a practical implementations, for instance, in a computational mechanics problem. For a single-parameter process such as spatial or temporal process, the eigenvalue decomposition of the covariance matrix leads to the well-known Karhunen-Loeve (KL) decomposition. However, for multiparameter processes such as a spatio-temporal process, the covariance function itself can be defined in multiple ways. Here the process is assumed to be measured at a finite set of spatial locations and a finite number of time instants. Then the spatial covariance matrix at different time instants are considered to define the covariance of the process. This set of square, symmetric, positive semi-definite matrices is then represented as a third-order tensor. A suitable decomposition of this tensor can identify the dominant components of the process, and these components are then used to define a closed-form representation of the process. The procedure is analogous to the KL decomposition for a single-parameter process, however, the decompositions and interpretations vary significantly. The tensor decompositions are successfully applied on (i) a heat conduction problem, (ii) a vibration problem, and (iii) a covariance function taken from the literature that was fitted to model a measured wind velocity data. It is observed that the proposed representation provides an efficient approximation to some processes. Furthermore, a comparison with KL decomposition showed that the proposed method is computationally cheaper than the KL, both in terms of computer memory and execution time.

Memory-induced anomalous dynamics in a minimal random walk model

A discrete-time dynamics of a non-Markovian random walker is analyzed using a minimal model where memory of the past drives the present dynamics. In recent work N. Kumar et al., Phys. Rev. E 82, 021101 (2010)] we proposed a model that exhibits asymptotic superdiffusion, normal diffusion, and subdiffusion with the sweep of a single parameter. Here we propose an even simpler model, with minimal options for the walker: either move forward or stay at rest. We show that this model can also give rise to diffusive, subdiffusive, and superdiffusive dynamics at long times as a single parameter is varied. We show that in order to have subdiffusive dynamics, the memory of the rest states must be perfectly correlated with the present dynamics. We show explicitly that if this condition is not satisfied in a unidirectional walk, the dynamics is only either diffusive or superdiffusive (but not subdiffusive) at long times.

Random vibration testing with controlled samples

This paper proposes a novel experimental test procedure to estimate the reliability of structural dynamical systems under excitations specified via random process models. The samples of random excitations to be used in the test are modified by the addition of an artificial control force. An unbiased estimator for the reliability is derived based on measured ensemble of responses under these modified inputs based on the tenets of Girsanov transformation. The control force is selected so as to reduce the sampling variance of the estimator. The study observes that an acceptable choice for the control force can be made solely based on experimental techniques and the estimator for the reliability can be deduced without taking recourse to mathematical model for the structure under study. This permits the proposed procedure to be applied in the experimental study of time-variant reliability of complex structural systems that are difficult to model mathematically. Illustrative example consists of a multi-axes shake table study on bending-torsion coupled, geometrically non-linear, five-storey frame under uni/bi-axial, non-stationary, random base excitation. Copyright (c) 2014 John Wiley & Sons, Ltd.

MULTI-PITCH TRACKING USING GAUSSIAN MIXTURE MODEL WITH TIME VARYING PARAMETERS AND GRATING COMPRESSION TRANSFORM

Grating Compression Transform (GCT) is a two-dimensional analysis of speech signal which has been shown to be effective in multi-pitch tracking in speech mixtures. Multi-pitch tracking methods using GCT apply Kalman filter framework to obtain pitch tracks which requires training of the filter parameters using true pitch tracks. We propose an unsupervised method for obtaining multiple pitch tracks. In the proposed method, multiple pitch tracks are modeled using time-varying means of a Gaussian mixture model (GMM), referred to as TVGMM. The TVGMM parameters are estimated using multiple pitch values at each frame in a given utterance obtained from different patches of the spectrogram using GCT. We evaluate the performance of the proposed method on all voiced speech mixtures as well as random speech mixtures having well separated and close pitch tracks. TVGMM achieves multi-pitch tracking with 51% and 53% multi-pitch estimates having error <= 20% for random mixtures and all-voiced mixtures respectively. TVGMM also results in lower root mean squared error in pitch track estimation compared to that by Kalman filtering.