Expectation Maximization (EM) algorithm has experienced a significant increase in terms of usage in many fields of study. In this paper, the performance of the said algorithm in finding the Maximum Likelihood for the Gaussian Mixed Models (GMM), a probabilistic model normally used in fraud detection and recognizing a person’s voice in speech recognition field, is shown and discussed. At the end of the paper, some suggestions for future research works will also be given.
This paper analyses electromagnetic signal scattered from the target crossing the Forward Scattering
Radar (FSR) system baseline. The aim of the analysis was to extract the Doppler signal of a target under the influence of high ground clutter and noise interference. The extraction was used for the
automatic target detection (ATD) in the FSR system. Two extraction methods, namely Hilbert Transform and Wavelet Technique, were analyzed. The detection using the Hilbert Transform is only applicable for some conditions; however, the detection using the Wavelet Technique is more robust to any clutter and noise level. From 55 sets of signal, only 4% of false alarm was detected or occurred when the Wavelet Technique was applied as a detection scheme. Two sets of field experimentation were carried out and the target’s signal under the influence of high clutter had successfully been detected using the proposed method.
The central coast of Tamil Nadu, Andaman and Nicobar Islands, along with other regions, was severely hit by the tsunami on 26 December 2004. A strange spawning behaviour of the Indian horseshoe crabs (Tachypleus gigas, Müller) was observed after the tsunami disturbance. Along the northeast coast of India, the horseshoe crabs normally migrate towards the shore in large numbers coinciding with the tidal height and grain size of the sediment and spawn in nests made in sand. The spawning behaviour was normal until November 2004 and throughout that period about 35 to 45 nests in 200 m2 of area of the breeding beach were examined. However, no nesting was observed from December 2004 to March 2005, i.e. after the occurrence of tsunami. The number of nests per 200 m2 of area (~82) and the number of eggs per nest (750-1000 eggs) were increased considerably when the normal conditions were restored in April 2005. The absence of spawning migration, from December 2004 to March 2005, is totally an unusual behaviour and this happened for the first time during the past several years of the study period.
Part of the Seremban flood mitigation project in the state of Negeri Sembilan, Malaysia is to mitigate
the flood at Jalan Rasah. The mitigation is planned to be implemented in packages. Package I and Package II of River Anak Air Rasah are parts of the project work. In these packages, wider and deeper concrete sections for the river are constructed. The existing undersized culverts were replaced by bigger reinforced concrete box culverts. The size of the box culverts was based on 100-years average reoccurrence interval (ARI). One of these culverts intersected with a rail line connecting Singapore and Malaysia. Trenchless jacking technique was used to lay the box culvert. The total length of the box culvert jacked under the railway line is 33 m, whereas the total width of the twin box culvert is 7.8 m with a total height of 3 m. This was the first time that the trenchless jacking techniques were used for the urban flood mitigation purpose in Malaysia, and it is mainly used to minimise traffic disruption. This study reports the success of using jacking technique in the development of the flood mitigation program of DID in Negeri Sembilan. Among other things, it explains significant performance of the technique under local conditions and experiences gained towards the advancement of tunnelling and trenchless technology.
Surface sediment samples were collected from five locations at the downstream of Klang River in January 2007 to examine the spatial distribution, composition, and sources of 19 parent polycyclic aromatic hydrocarbons (PAHs) and aliphatic hydrocarbon (n-alkanes) using gas chromatographymass spectrometry. The total concentrations of the 19 PAHs in the sediments were found to range from 1304 to 2187 ng g-1 sediment. Meanwhile, total concentrations of n-alkanes ranged from 17008 to 27116 μg g-1 sediment. The concentration of n-alkanes in the sediment was significantly correlated (r = 0.991, p = 0.001) with the content of sediment organic carbon. In this study, all the sediments exhibited phenanthrene/anthracene (PHE/ANT >15) fluoranthene/(fluorantene+pyrene) (FLT/FLT+PYR < 0.4), ethylphenanthrenes/phenanthrene (MP/P >1), combustion PAHs/total PAHs (CombPAH/Σ19PAH
A field study was conducted on the effect of four primary tillage implements and three seed densities on the grain yield of rain fed wheat (Tajan cultivar), using a drill planting machine with the end wheels. The experimental design was a split plot design in a 4×3 factorial with three replications. In this study, the main plots were the tillage treatments, namely Mouldboard plough, Disc Plough, Chisel Plough, Offset Disc, and sub-plots were seed rates of 350, 400 and 450 seeds.m-2. Determinations included grain yield and selected yield components. The results showed that grain yield was not affected by the densities of seed and tillage machine treatments. The use of Chisel Plough, with 400 seeds.m-2 sowing rate, had the highest grain yield of wheat grown in the Golestan province (Iran), a region with an average annual rainfall of 450 mm.
Compartmented Fluidized Bed Gasifier (CFBG), consisting of two compartments - the combustorand gasifier, uses air blown instead of pure oxygen for syngas production in bubbling fluidization mode, eliminating the need of air separation unit, and reducing the capital cost, thus distinguishes it from other traditional ones. Fluidization quality is a determining factor in the CFBG to guarantee its well-lifted behaviour. Previous study, without solid circulation at ambient conditions, brought to the fore the necessity of considering the effect of the minimum allowable effective diameter. The study was then performed in the CFBG cold physical model of 0.66m overall diameter (effective diameter for combustor and gasifier is 0.413m and 0.257m) to investigate the fluidization quality and compare it with the results obtained from the previous cold model of about 1.36 times smaller, but with the same compartmented ratio of 65:35. Different inert particles (river sand, quartz sand and alumina) were used, over a range of aspect ratios, for the aforementioned objective. The results showed that the fluidization quality in the gasifier has not been achieved and the degradation of fluidization quality in the combustor is still observed, notwithstanding the fact that the condition of the minimum allowable effective diameter has been met. The reduction of distributor free area, to increase the distributor pressure drop, showed a marginal effect on the quality. The effect of the minimum allowable effective diameter on fluidization quality in CFBG as well as the interplay of geometric and operational parameters require further studies be carried out. The fluidization quality of the binary mixture is also currently under investigation.
The production of carbon dioxide from Karas woods under argon atmosphere was investigated using a direct pyrolysis-combustion approach. Direct burning was used in this study, using argon for yrolysis and oxygen during combustion to look at the yield of carbon dioxide, produced at different parameters, such as the temperature, retention time and flow rate of argon, as the carrier gas. In this study, a new methodology, 23 response surface central composite design was successfully employed for the experimental design and analysis of results. Central composite experimental design and response surface method were utilized to determine the best operating condition for a maximum carbon dioxide production. Appropriate predictable empirical linear model was developed by incorporating interaction effects of all the variables involved. The results of the analysis revealed that linear equation models fitted well with the experimental for carbon dioxide yield. Nevertheless, the R-Squared obtained using the direct pyrolysis-combustion was 0.7118, indicating that the regression line was not at the best-fitted line.
Thermodynamic chemical equilibrium analysis using, total Gibbs energy minimization method, was carried out for methane oxidation to higher hydrocarbons. For a large methane conversion and a high selectivity to higher hydrocarbons, the system temperature and oxygen concentration played a vital role, whereas, the system pressure only slightly influenced the two variables. Numerical results showed that the conversion of methane increased with the concentration of oxygen and reaction temperature, but it decreased with pressure. Nevertheless, the presence of oxygen suppressed the formation of higher hydrocarbons which mostly consisted of aromatics, but enhanced the formation of hydrogen. As the system pressure increased, the aromatics, olefins and hydrogen yields diminished, but the paraffin yield improved. Carbon monoxide seemed to be the major oxygen-containing equilibrium product from methane oxidation, whilst almost no H2O, CH3OH and HCOH were detected although traces amount of carbon dioxide were formed at relatively lower temperature and higher pressure. The total Gibbs energy minimization method is useful to theoretically analyze the feasibility of methane conversion to higher hydrocarbons and syngas at the selected temperature and pressure.
An in-house quasi-dimensional code has been developed which simulate the overlap, intake, compression, combustion, as well as expansion and exhaust processes of a homogeneous charged internal combustion engine (ICE). A detailed chemical kinetic mechanism, constituting of 39 species and 148 elementary reactions, has been used in conjunction with above code to study the combustion of CNG under IC engine conditions. Two different criteria, based on pressure rise and mass of fuel burned, are used to detect the onset of ignition. Parametric studies are conducted to show the effect of compression ratio, initial pressure, intake temperature and equivalence ratio, on the time of ignition and fuel burning rate. The results obtained from the modelling show a good agreement with the experimental data.
A series of polycrystalline samples (La1-xSmx)0.67Sr0.33MnO3 (X=0.00, 0.05 and 0.10) were prepared using the conventional solid-state reaction method. The XRD analysis indicated that all the samples were fully crystalline and in a single phase with a rhombohedral structure after a sintering at 1200ºC for 24 hours in air. Meanwhile, the Atomic Force Microscopy measurements showed that a small amount of Sm doping in La sites affected the grain growth and this might affect the grain boundary layer, thus resulting in the reduction of the Curie temperature, TC. Extrinsic magnetoresistance (MR) was observed for all the samples with a large negative MR at low field (0-0.1 or 0.2T) region, followed by a slower varying MR at high field (0.1 or 0.2-1T). The highest low-field MR value of -4.6% (at 0.1T) and -6.1% (at 0.2T) were observed for sample X=0.10 and X=0.05, respectively. Hence, these indicated that the extrinsic MR was grain size dependent and was therefore enhanced with a small amount of Sm substitution in La sites.
Monoclinic bismuth oxide (α-Bi2O3) nanoparticles were prepared via precipitation method and
irradiated with a pulsed laser forming thin films. Their phase and surface morphological properties
were investigated using x-ray diffraction (XRD), atomic force microscopy (AFM), scanning electron
microscopy (SEM) and high resolution transmission electron microscopy (HR-TEM). The XRD
analysis shows the phase transformation to a partially crystalline tetragonal phase β-Bi2O3 thin film.
The SEM micrograph of the nanoparticles, with an average crystal size of 72 nm, was seen to form
a thin film with a peculiar structure, coined as “cotton-like”, is attributed to the high surface energy
absorbed by the nanoparticles during ablation. The HR-TEM micrograph shows the particulate with
a clearly defined interlayer spacing.
The roots of Calophyllum inophyllum (Guttiferae), furnished six xanthones which are brasilixanthone (1), 1,3,5-trihydroxy-2- methoxy xanthone (2), caloxanthone A (3), pyranojacareubin (4), caloxanthone B (5) and tovopyrifolin (6), Structural elucidations of these compounds, were achieved through 1D and 2D NMR andMS techniques. In this paper, the isolation and structural elucidation data for these xanthones are reported.
This paper investigates the confidence intervals of R2 MAD, the coefficient of determination based on
median absolute deviation in the presence of outliers. Bootstrap bias-corrected accelerated (BCa)
confidence intervals, known to have higher degree of correctness, are constructed for the mean and standard deviation of R2 MAD for samples generated from contaminated standard logistic distribution. The results indicate that by increasing the sample size and percentage of contaminants in the samples, and perturbing the location and scale of the distribution affect the lengths of the confidence intervals. The results obtained can also be used to verify the bound of R2 MAD.
Welding process is most widely used in joining components or structures in industry. Although welding is part of a larger category called metals joining, the weld itself still gives significant problems to engineers, researchers and manufacturers until today. Several widely used welding processes, such as the Metal Inert Gas (MIG), Tungsten Inert Gas (TIG), and Manual Metal Arc (MMA), were studied. In the present paper, the characterization of the macrostructure, microstructure, hardness and residual stress distribution are highlighted and discussed to achieve a better understanding of the welded quality which is crucial in determining the welded products.
Compressive residual stress, induced by mechanical surface treatment, may relax during component
operation life, due to thermal or mechanical mechanism. Fatigue life prediction for the components which have residual stress will be misled and inaccurately predicted the phenomenon of residual stress relaxation is not considered. Despite putting an effort on incorporating the residual stress relaxation, the issues remain concerned with the technical challenge of measuring and quantifying
the magnitude of residual stress relaxation as well as redistribution during the loading cycling itself.
In this paper, the residual stress relaxation and its models were reviewed and discussed to picture
the best knowledge related to this topic, i.e. whether relaxation is a cause or an effect.
This study presents an efficient finite element analysis technique which shows great versatility in
modelling of precast composite flooring system subjected to static loadings. The method incorporates sliding and opening in the analysis of composite structures using the interface element which was specifically designed to simulate the actual behaviour at the interfaces between contacting materials. A three-dimensional finite element model of the precast composite slab which exhibits discontinuous behaviour was performed to demonstrate the potential and applicability of the proposed method of analysis. The results of the analysis demonstrate that the overall response of a discontinuous system to external loading is significantly affected by the bonding condition at the interfaces between the contacting materials.
This paper investigates several asymptotic confidence interval estimates, based on the Wald, likelihood ratio and the score statistics for the parameters of a parallel two-component system model, with dependent failure and a time varying covariate, when data is censored. This model is an extension of the bivariate exponential model. The procedures are investigated via a coverage probability study using the simulated data. The results clearly indicate that the interval estimates, based on the likelihood ratio method, work better than any of the other two methods when dealing with the censored data.
The properties of fibre-reinforced composites are dependent not only on the strength of the reinforcementfibre but also on the distribution of fibre strength and the composition of the chemicals or additivesaddition within the composites. In this study, the tensile properties of abaca fibre reinforced high impactpolystyrene (HIPS) composites, which had been produced with the parameters of fibre loading (30,40,50wt.%), coupling agent maleic anhydride (MAH) (1,2,3 wt%) and impact modifier (4,5,6 wt.%) weremeasured. The optimum amount of MAH is 3% and the impact modifier is 6% and these give the besttensile properties. Meanwhile, Differential Scanning Calorimetry (DSC) was used to study the thermalbehaviour within the optimum conditions of the composites. In this research, glass transitions temperature(Tg) of neat HIPS occurred below the Tg of the optimum condition of composites as the temperature ofan amorphous state. The endothermic peak of the composites was in the range of 430-4350C, includingneat HIPS. It was observed that enthalpy of the abaca fibre reinforced HIPS composites yielded belowthe neat HIPS of 748.79 J/g.
In this work, hydrogels were prepared from carboxymethyl cellulose (CMC) and 1-vinyl-2-pyrrolidone(VP) by Electron Beam irradiation in the presence of N,N'-methylenebisacrylamide (BIS) as a crosslinkingagent. The parameters studied include stirring time and percentage of crosslinking agent. Hydrogels werecharacterized using Fourier Transform Infrared (FTIR) spectroscopy and Scanning Electron Microscopy(SEM). VP and BIS were found be effective as reinforcement materials to improve the properties ofCMC. Meanwhile, the optimum conditions were 5% BIS and 3 hours of stirring time. The gel fractionincreased when irradiation dose was increased. FTIR confirmed the crosslinking reaction between CMCand VP after the irradiation process by using BIS as the crosslinking agent. TGA thermograms showedchanges in the thermal properties of CMC-VP hydrogels in the presence of different amounts of BIS.