Jaksic O.,Nikola Tesla Institute of Electrical Engineering |
Jaksic Z.,University of Belgrade |
Matovic J.,Institute of Sensors and Actuating Systems
Microsystem Technologies | Year: 2010
We investigated intrinsic noise in plasmonic sensors caused by adsorption and desorption of gaseous analytes on the sensor surface. We analyzed a general situation when there is a larger number of different analyte species. We applied our model to calculate various analyte mixtures, including some environmental pollutants, toxic and dangerous substances. The spectral density of mean square refractive index fluctuations follows a dependence similar to that of generation-recombination noise in photodetectors, flat at lower frequencies and sharply decreasing at higher. Some of the calculated noise levels are well within the detection range of conventional surface plasmon resonance sensors. An AD noise peak is observed in temperature dependence of mean square refractive index fluctuations, thus sensor operating temperature may be optimized to obtain larger signal to noise ratio. A significant property of AD noise is its rise with the decreasing plasmon sensor area, which means that it will be even more pronounced in modern nanoplasmonic devices. Our consideration is valid both for conventional surface plasmon resonance devices and for general nanoplasmonic devices. © 2010 Springer-Verlag.
Vukic V.,Nikola Tesla Institute of Electrical Engineering |
Osmokrovic P.V.,University of Belgrade
Nuclear Technology and Radiation Protection | Year: 2010
Low-drop out volt age regulators with various geometries and technological realisations of serial pnp power transistors were exposed to ionizing radiation. Although de vices with vertical emitters were considered much less susceptible to the influence of radiation on for ward emitter current gain than circuits with round emitters, the experiment showed a similar degradation of current gain in both cases. The main reason of high radiation susceptibility of the examined vertical serial pnp transistor is the implementation of an interdigitated emitter, with high perimeter-to-area ratio, causing the great in crease of serial transistor's base cur rent, but a mi nor influence on the maximum out put current. Transistors with round emitters with small perimeter-to-area ratio expressed a moderate current gain degradation, but a rapid fall of the emitter injection efficiency, causing a significant de crease of the maximum out put current. Regard less of the similar for ward emitter current gain degradation, reliability and operational character is tics of two types of low-drop out volt age regulators were completely different.
Lukic J.M.,Nikola Tesla Institute of Electrical Engineering |
Milosavljevic S.B.,Nikola Tesla Institute of Electrical Engineering |
Orlovic A.M.,University of Belgrade
Industrial and Engineering Chemistry Research | Year: 2010
Deposition of conductive copper sulfide on paper insulation of power transformers was recognized as a serious risk that can result in equipment failure. The aim of this study was to investigate the influence of mineral insulating oil oxidation and the presence of metal passivator on copper sulfide (Cu2S) deposition. Oil oxidation was found to increase deposition of Cu2S on the insulating paper, as confirmed by IEC 61125 B and IEC 62535 tests. The presence of oxygen resulted in an increase of copper dissolution in the oil and facilitated progressive copper sulfide deposition on the insulating paper, as confirmed by ICP, AAS, and EDX measurements. The effectiveness of aminomethyl substituted toluylbenzotriazole (TTA) metal passivator in suppressing copper sulfide formation and deposition was studied using modified IEC 62535 test. The protective role of TTA bonded to the copper plate surface and absorbed in the insulating paper was observed in tests at 140 and 150 °C, even for highly oxidized oils. © 2010 American Chemical Society.
Stankovic K.,University of Belgrade |
Vujisic M.,University of Belgrade |
Kovacevic D.,Nikola Tesla Institute of Electrical Engineering |
Osmokrovic P.,University of Belgrade
Measurement: Journal of the International Measurement Confederation | Year: 2011
Statistical properties of some basic mass-produced passive electrical circuits used in measurements are analyzed in this paper, using methods for expressing measurement uncertainty of indirectly measured quantities. The focus of this paper is on electrical circuits whose defining parameters are obtained as non-linear functions of component parameters. Variants of voltage divider and Wheatstone bridge circuits significant for measurement practice are investigated in detail. Even if distributions with symmetrical probability density functions (PDFs), such as uniform or normal, are adopted for the parameters of the components comprising a circuit, non-linearity of the circuit function gives rise to asymmetry in the PDF of the circuit's parameter. The asymmetry of the PDF causes the mean and the nominal value of the circuit parameter to differ. © 2011 Elsevier Ltd. All rights reserved.
Dragosavac J.,Nikola Tesla Institute of Electrical Engineering |
Janda Z.,Nikola Tesla Institute of Electrical Engineering |
Milanovic J.V.,University of Manchester
IEEE Transactions on Power Systems | Year: 2011
The paper presents the digital realization of a model of reactive power flow (QFM) in a steam power plant using a programmable logic controller (PLC). The steam power plant (SPP) model is developed for pre-commissioning validation testing of the coordinated reactive power-terminal voltage (Q-V) control system. The SPP QFM includes a model for a synchronous generator, an excitation system, a step-up transformer, and the generator's droop characteristic modeled through the automatic voltage regulator (AVR). A QFM synthesis is based on a series of experiments performed on site. The parameters of the generator and AVR are estimated from recorded generator voltage and current time responses to a step change in voltage reference of the AVR. To get a complete QFM, transformers and network reactances are also included. In order to calculate reactive power (Q) flows more accurately, the generator Q output is adjusted by taking into account its real power output. Standard PLC hardware, as industrial grade equipment appropriate for on site testing, is used for practical QFM implementation after discretization of the continuous mathematical model. The developed QFM response is verified through a series of experiments performed in the laboratory. © 2006 IEEE.
Kostic M.,Nikola Tesla Institute of Electrical Engineering
Elektrotehniski Vestnik/Electrotechnical Review | Year: 2012
An induction motor model for high harmonics used in literature for harmonic penetration studies and the corresponding parameters are analyzed in this paper. It is shown that the rotor slot resistance and the rotor slot reactance values are equally in the short circuit regime and that they have practically equal values, in per unit, for all motors of the same series. The knowledge of the equivalent circuit parameters of the squirrel-cage induction motors operating in the short circuit regime is especially important when calculating the current power losses in induction motors supplied from power set with high harmonics. General conclusion is that harmonic reactance values used in literature are more or less accurate (or by 10% higher). On the contrary, real resistance values are 2̃4 times higher and the active powers absorbed by the rotation machine do not exactly correspond to the damping value. Consequently, the amplification factor of the source harmonic current calculated in literature for resonant regimes is 2̃4 times higher than the real values.
Kostic M.,Nikola Tesla Institute of Electrical Engineering
Elektrotehniski Vestnik/Electrotechnical Review | Year: 2014
Following results of an investigation the author comes to the conclusion: that the Potier reactance values depend almost solely on the turbogenerator(s) reactive currents (iaQ = ian·sinφ) for a given voltage value. This new rule is after being proven by a general qualitative analysis, and his thesis is then checked for rated values of the active and reactive power, for an example of a GTHW 360 (360 MW) turbogenerator. Potier reactance xp is determined by the open-circuit saturation characteristics and data from the zero-power factor at the rated voltage ( ia = ia,90·cosφ=0, U=Un). The procedures used to determine the Potier reactance (xP or xP,n) are performed by a computer with the open-circuit saturation curves if0 (e0) expressed analytically. The author proposes the IEC 34-4 standard in the part on the determination of the Potier reactance. It is proposed that it is determined for the excitation current corresponding to the rated voltage and armature current value ia,90 = ian·sinφn, at the zero-power factor (overexcitation), and for some characteristic values is needed for the construction of the turbogenerator capability curve (P-Q curve).
Jankovic M.V.,Nikola Tesla Institute of Electrical Engineering |
Rubens N.,University of Tokyo
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2012
Recently, elements of probabilistic model that are suitable for modeling of learning algorithms in biologically plausible artificial neural networks framework, have been introduced. Model was based on two of the main concepts in quantum physics - a density matrix and the Born rule. In this paper we will show that proposed probabilistic interpretation is suitable for modeling of on-line learning algorithms for Independent Component Analysis (ICA), which could be realized on parallel hardware based on very simple computational units. Proposed concept (model) can be used in the context of improving algorithm convergence speed, learning factor choice, input signal scale robustness, and can be easily deployed on parallel hardware. © 2012 Springer-Verlag.
Kostic M.,Nikola Tesla Institute of Electrical Engineering
Elektrotehniski Vestnik/Electrotechnical Review | Year: 2010
In this paper, a simplified procedure is presented to be used in determination of magnetic reluctance and inductance of short cylindrical coils of a rectangular cross- section. Based on calculations and analysis, it is shown that the coil with a greater ratio of longitudinal/transversal dimensions of the cross-section should rather be approximated with the rectangular cross-section than the circular cross-section, as it is usually done. It is proven that the magnetic conductance to the flux outside the inner part of the coil is directly proportional to the perimeter of the contour of the cross-section given. By using obtained results, the impact of the cross-section shape on the values of the magnetic reluctance and inductance of short cylindrical coils can be analyzed.
Stojic D.M.,Nikola Tesla Institute of Electrical Engineering
Advances in Electrical and Computer Engineering | Year: 2012
A new method for the induction motor stator flux estimation used in the sensorless IM drive applications is presented in this paper. Proposed algorithm advantageously solves problems associated with the pure integration, commonly used for the stator flux estimation. An observerbased structure is proposed based on the stator flux vector stationary state, in order to eliminate the undesired DC offset component present in the integrator based stator flux estimates. By using a set of simulation runs it is shown that the proposed algorithm enables the DC-offset free stator flux estimated for both low and high stator frequency induction motor operation. © 2012 AECE.