Prediction of diesel fuel cold properties using artificial neural networks

In this paper, two neural networks, multilayer perceptron and networks with radial-basis function, were used to predict important cold properties of commercial diesel fuels, namely cloud point and cold filter plugging point. The developed models predict the named properties using cetane number, density, viscosity, contents of total aromatics, and distillation temperatures at 10, 50, and 90 vol. % recovery as input data. The training algorithms, number of hidden layer neurons, and number of training data points were optimized in order to obtain a model with optimal predictive ability. The results indicated better prediction of cloud and cold filter plugging points in the case of multilayer perceptron networks. The obtained absolute error mean for the optimal neural network models (0.58°C for the cloud point and 1.46°C for the cold filter plugging point) are within the range of repeatability of standard cold properties determination methods.     

BPC – A binary priority countdown protocol

In modern wireless ad hoc networks, with a high speed PHY, every collision means a significant loss of useful bandwidth. In the last few years different binary contention protocols have been introduced to address this problem. In this work we propose a novel binary contention protocol called binary priority countdown (BPC) protocol, whose goal is to reduce collisions as well as contention time. BPC uses a new priority countdown mechanism which exploits the efficiency of binary countdown, but the priority countdown process is not constrained to a single binary countdown round. This way, the priority space is not defined by the length of binary countdown round, like in other binary countdown protocols proposed in the literature, and arbitrary medium access priorities can be decremented through multiple binary countdown rounds if necessary. The ability of a new priority countdown mechanism to count down any priority number without changing the length of a binary countdown round, allows independent management of priority space. This “independence” of priority space introduces new optimization and adaptation possibilities. Collision memory effect is recognized and described. BPC protocol reveals connection between unary, binary and digit contention protocols. All three groups of protocols can now be seen as members of the same class of contention algorithms. Preliminary simulation results are shown.     

Genome-Wide Association Analysis and Genomic Prediction of Thyroglobulin Plasma Levels

Thyroglobulin (Tg) is an iodoglycoprotein produced by thyroid follicular cells which acts as an essential substrate for thyroid hormone synthesis. To date, only one genome-wide association study (GWAS) of plasma Tg levels has been performed by our research group. Utilizing recent advancements in computation and modeling, we apply a Bayesian approach to the probabilistic inference of the genetic architecture of Tg. We fitted a Bayesian sparse linear mixed model (BSLMM) and a frequentist linear mixed model (LMM) of 7,289,083 variants in 1096 healthy European-ancestry participants of the Croatian Biobank. Meta-analysis with two independent cohorts (total n = 2109) identified 83 genome-wide significant single nucleotide polymorphisms (SNPs) within the ST6GAL1 gene (p<5×10−8). BSLMM revealed additional association signals on chromosomes 1, 8, 10, and 14. For ST6GAL1 and the newly uncovered genes, we provide physiological and pathophysiological explanations of how their expression could be associated with variations in plasma Tg levels. We found that the SNP-heritability of Tg is 17% and that 52% of this variation is due to a small number of 16 variants that have a major effect on Tg levels. Our results suggest that the genetic architecture of plasma Tg is not polygenic, but influenced by a few genes with major effects.     

Modeling of dibenzothiophene oxidative desulfurization using response surface methodology

The effect of reaction temperature, mixing speed and oxidant to catalyst volume ratio, including their interactions on the oxidative desulfurization of dibenzothiophene by using response surface methodology was studied. Hydrogen peroxide was used as oxidant and acetic acid as catalyst. The obtained model accurately predicts conversion of dibenzothiophene and the best conversion of 98.7% was observed at temperature 70°C, mixing speed of 1250 rpm and oxidant to catalyst volume ratio of 1:1. At high temperatures, a major limitation of the desulfurization process is the mass transfer and the high mixing speed is needed to achieve an efficient process.     

Fluorescent lamp cold starting improvement

Three approaches to improving the efficacy and startup time of commercially available cold cathode fluorescent lamps under cold ambient conditions are evaluated in this paper: heating the tube on one side; heating the tube on two sides; and heating the circumference of the outer tube. The internal power density generated by ion bombardment of the cathode and collisions in the plasma are indirectly obtained by matching simulated axial wall temperatures with those obtained by experiment at room temperature. The estimated power density is used to evaluate the temporal evolution of the axial surface wall temperature under different ambient conditions immediately after a cold start. All simulation results have been obtained using PHOENICS, a computational fluid flow program. Experiments have shown a good correlation between the time taken for the light output to reach an acceptable value and the time taken for the inner tube cold spot to reach a temperature of 0/spl deg/C. These results have been used to estimate heating time for different heater geometries and heater powers. The study shows that the conduction of heat to the outer glass envelope is the limiting process that controls the startup time. Modest improvement in startup time can be achieved by increasing heating power or by moving the heater to the side of the lamp. However, only a heater which entirely surrounds the lamp is capable of substantially decreasing the startup time.     

Power quality compensation using universal power quality conditioning system

The aim of this letter is to present a universal power quality conditioning system (UPQS) named after a unified power quality conditioner (UPQC), which is extended by adding a shunt active filter at the load side. Its main purpose is to compensate for supply voltage and load current imperfections, such as sags, swells, interruptions, imbalance, flicker, harmonics, reactive currents, and current unbalance. Converter and control analysis is presented together with results showing the UPQS modes of operation.     

Fast ferroresonance suppression of coupling capacitor voltage transformers

This paper describes a procedure for fast suppression of the phenomenon of ferroresonance in coupling capacitor voltage transformers (CCVT) without major change in the CCVT design. It is shown that it is possible to adjust parameters of the secondary overvoltage protection and the filter circuit so that the ferroresonance can be cleared in a very short time interval. The study cases reported in this paper show that ferroresonance is effectively cleared within two cycles. An implementation of metal oxide varistors (MOV) as part of passive ferroresonance protection is also addressed. The Electromagnetic Transients Program (EMTP) is used for modeling transients and fine-tuning the ferroresonance suppressing circuit. The studies are conducted on the Trench TEHMP161A CCVT.     

Portable real-time DCT-based steganography using OpenCL

In this paper, a steganographic method for real-time data hiding is proposed. The main goal of the research is to develop steganographic method with increased robustness to unintentional image processing attacks. In addition, we prove the validity of the method in real-time applications. The method is based on a discrete cosine transform (DCT) where the values of a DCT coefficients are modified in order to hide data. This modification is invisible to a human observer. We further the investigation by implementing the proposed method using different target architectures and analyze their performance. Results show that the proposed method is very robust to image compression, scaling and blurring. In addition, modification of the image is imperceptible even though the number of embedded bits is high. The steganalysis of the method shows that the detection of the modification of the image is unreliable for a lower relative payload size embedded. Analysis of OpenCL implementation of the proposed method on four different target architectures shows considerable speedups.