Dynamic Assignment of Gaussian Components in Modelling Speech Spectra

In this paper, we describe a parametric mixture model for modelling the resonant characteristics of the vocal tract where Gaussian distributions are used to model spectral frequency regions. A mixtures of Gaussian (MoG) based parametrisation scheme is used for modelling a smoothed representation of the spectra. This smoothing procedure removes all signal periodicity from the spectra allowing highly natural analysis, manipulation and synthesis of speech. The goal of this parametrisation scheme is to ease the correspondence between the resonant characteristics of the vocal tract and the parametric distributions and modelling the spectrum with an appropriate number of parameters. Previously, a maximum likelihood (ML) approach to this parametrisation scheme was introduced. However, this approach has inherent local optima problems. Noting that, a relatively small class of Gaussian densities can approximate a large class of distributions, we propose a new scheme whereby starting with a large number of distributions in the mixture, we systematically reduce their number and re-approximate the densities in the mixture based on a distance criterion. The Kullback-Leibler (KL) distance was found to allow optimal MoG solutions to the spectra. Furthermore, a fitness measure based on KL information is used to provide a figure for estimating the model order in representing formant-like features. The proposed model is subjectively evaluated and is shown to reduce the number of Gaussian with an appreciable loss in the quality of the re-synthesised speech.     

A new simplified thermoregulatory bioheat model for evaluating thermal response of the human body to transient environments

Thermal response of the human cutaneous thermoreceptors depends statically upon temperature and dynamically on the temperature change rate at the depth of the thermoreceptors. Therefore, it is very important to estimate the time-dependent thermoreceptors temperature with a good accuracy. On the other hand, the temperature distribution in skin tissue may be significantly affected by thermoregulatory mechanisms such as shivering, regulatory sweating and vasomotion. In the present study, a new simplified thermoregulatory bioheat model is proposed to describe heat transfer in skin tissue. The new model is constructed by combining the well-known Pennes equations with Gagge’s two-node model. In this model, the human skin is subdivided into three layers (epidermis, dermis and subcutaneous) and the time-dependent temperature of skin tissue is obtained by solving the bioheat equations taking into account thermoregulatory mechanisms. The model has been verified by extensive comparisons with the published analytical and experimental results where a good agreement was found. Therefore, the present model can estimate the skin temperature under transient environments with a very good accuracy.     

Evaluation of environmental and occupational exposure to mercury among Iranian dentists

The purpose of the study was to evaluate the environmental and occupational exposure to mercury (Hg), and to examine the various parameters, which contribute to high levels of mercury of Iranian dentists in Tehran. One-hundred hair and nail samples were collected from dentists. In addition, fifty samples from a control group and twenty-five additional samples from dental nurses were taken for comparison. The survey included a structured questionnaire designed to provide information about the parameters that influenced their occupational and environmental exposure to Hg. Overall mean concentrations in the hair and nails of the dentists was 2.84+/-0.47 and 3.56+/-0.53 mg/kg dry wt respectively. The equivalent values were 0.61+/-0.07 mg/kg in hair and 0.39+/-0.06 mg/kg in nails for the control group. In addition, mean concentrations in the dental nurses were 0.92+/-0.23 and 1.77+/-0.51 mg/kg in hair and nails respectively. The study showed that use of masks had a significant effect on Hg levels (p=0.02 for hair and p=0.03 for nails) and use of gloves only had significant effect on nails Hg (p=0.05). Hg concentrations in nails were significantly lower among the dentists who always used gloves and masks. Also, fish consumption and number of patients visited per day had a significant effect on hair (p=0.02 and p=0.02 respectively) and nails (p=0.03 and p=0.02 respectively) Hg. On the other hand, there was a significant effect of age in relation to Hg concentration in the hair and nails of the dentists (p=0.006 for hair and p=0.01 for nails). Multiple regressions were computed between hair Hg and the various assessed variables. The only positive results were between Hg concentration and masks for the dentists (p=0.03), and Hg concentration and the number of fish meals for the control group (p=0.009).     

All-metal transformer core for a low aspect ratio tokamak

A novel concept for incorporating an iron core transformer within a axisymmetric toroidal plasma containment device with a high neutron flux is described. This design enables conceptual design of low aspect ratio devices which employ standard transformer-driven plasma startup by using all-metal high resistance inserts between the toroidal field windings. This design avoids the inherent problems of a multi-turn air core transformer which will inevitably suffer from strong neutron bombardment and hence lose the integrity of its insulation, both through long term material degradation and short term neutron induced conductivity. A full 3-dimensional model of the concept has been developed within the MAXWELL program and the resultant loop voltage calculated. The utility of the result is found to be dependent on the resistivity of the high resistance inserts. Useful loop voltage time histories have been obtained using expected resistivities.     

STUDY ON THE ACETYLENE HYDROGENATION PROCESS FOR ETHYLENE PRODUCTION: SIMULATION,MODIFICATION, AND OPTIMIZATION

In this study, an industrial acetylene hydrogenation unit is simulated utilizing three available kinetic models. The results are compared against six-day experimental data and the best model is selected. Effects of feed temperature and the amount of injected hydrogen on ethylene selectivity are also studied. According to the simulation results, the unit is not working under its optimum conditions. Furthermore, by reduction of the hydrogen flow rate to 52 kg/h, process selectivity is increased. In addition, a new approach is proposed to modify the hydrogenation process and reduce undesired by-products. In the simulation of the modified process, hydrogenation reactors temperature, hydrogen flow rate, and H-1/H-2 ratio were regulated as adjustable parameters for the process optimization. The simulation shows that ethylene selectivity increases by 12%, while acetylene concentration and hydrogenation reactor temperature remains within acceptable ranges. Such selectivity could be achieved at the hydrogen flow rate of 50 kg/h with H-1/H-2 ratio of 0.1/0.9.     

Use of raster-based data layers to model spatial variation of seismotectonic data in probabilistic seismic hazard assessment

Recent achievements in computer and information technology have provided the necessary tools to extend the application of probabilistic seismic hazard mapping from its traditional engineering use to many other applications. Examples for such applications are risk mitigation, disaster management, post disaster recovery planning and catastrophe loss estimation and risk management. Due to the lack of proper knowledge with regard to factors controlling seismic hazards, there are always uncertainties associated with all steps involved in developing and using seismic hazard models. While some of these uncertainties can be controlled by more accurate and reliable input data, the majority of the data and assumptions used in seismic hazard studies remain with high uncertainties that contribute to the uncertainty of the final results. In this paper a new methodology for the assessment of seismic hazard is described. The proposed approach provides practical facility for better capture of spatial variations of seismological and tectonic characteristics, which allows better treatment of their uncertainties. In the proposed approach, GIS raster-based data models are used in order to model geographical features in a cell-based system. The cell-based source model proposed in this paper provides a framework for implementing many geographically referenced seismotectonic factors into seismic hazard modelling. Examples for such components are seismic source boundaries, rupture geometry, seismic activity rate, focal depth and the choice of attenuation functions. The proposed methodology provides improvements in several aspects of the standard analytical tools currently being used for assessment and mapping of regional seismic hazard. The proposed methodology makes the best use of the recent advancements in computer technology in both software and hardware. The proposed approach is well structured to be implemented using conventional GIS tools.     

Stochastic analysis of interphase effects on elastic modulus and yield strength of nylon 6/clay nanocomposites

The multifunctional properties of polymer clay nanocomposites (PCNs) can be related to the interaction of clays, polymer and interphase region. Several experimental, analytical and numerical studies have been conducted to characterize the mechanical behavior of PCNs. The elastic behavior of PCNs is well documented in the literature but their other material properties like yield strength are rather vague. On the other hand, the variation of material parameters and the stochastic nature of interphase region hinder the use of deterministic methods. In this study, a stochastic analysis along with a hierarchical multiscale method is used to analyze the effect of interphase properties on the macroscopic properties of PCNs. Since the interphase layer is expected to be weaker than the polymer matrix, a weakening coefficient is defined to describe the interphase properties based on the matrix properties. This weakening coefficient and the interphase thickness are considered as the stochastic inputs. The elastic modulus and yield strength of nylon 6/clay nanocomposites are calculated using the stochastic multiscale framework. The uncertainty propagation and sobol sensitivity analysis are performed to study the effect of random inputs on the elastic modulus and yield strength of PCNs. Despite the wide range of input variations, the accuracy of the proposed stochastic multiscale framework for the prediction of the PCNs properties is estimated by validating our results against the available experimental data in the literature.

     

A new genetic algorithm for the machine/part grouping problem involving processing times and lot sizes

This paper reports a new genetic algorithm (GA) for solving a general machine/part grouping (GMPG) problem. In the GMPG problem, processing times, lot sizes and machine capacities are all explicitly considered. To evaluate the solution quality of this type of grouping problems, a generalized grouping efficacy index is used as the performance measure and fitness function of the proposed genetic algorithm. The algorithm has been applied to solving several well-cited problems with randomly assigned processing times to all the operations. To examine the effects of the four major factors, namely parent selection, population size, mutation rate, and crossover points, a large grouping problem with 50 machines and 150 parts has been generated. A multi-factor (3⁴) experimental analysis has been carried out based on 324 GA solutions. The multi-factor ANOVA test results clearly indicate that all the four factors have a significant effect on the grouping output. It is also shown that the interactions between most of the four factors are significant and hence their cross effects on the solution should be also considered in solving GMPG problems.