No description available.

NR 20140805

In this thesis, the possibility of detection of fault location in the cable distribution systems by using traveling waves due to fault and circuit breaker operations is investigated. Waveforms originated from both actions and fault steady state are separately analyzed. During such switching actions, high frequency variations which are absent in the steady state conditions, take place. In order to simulate high frequency changes properly, system elements are modeled accordingly. In other words, frequency dependent models are introduced, and they are used in Electro-Magnetic Transients Program (EMTP). Since the characteristics of waveforms are different for separately analyzed portions, different fault locating algorithms with their limitations are introduced.

The calculation of frequency-dependent cable parameters is essential for simulations of transient phenomena in electrical power systems. The simulation of transients is more complicated than the calculation of currents and voltages in the nominal frequency range. The model has to represent the frequency dependency and the wave propagation behavior of cable lines. The introduced model combines an improved subconductor method for the determination of the frequency-dependent parameters and a PI section wave propagation model. The subconductor method considers the skin and proximity effect in all conductors for frequency ranges up to few megahertz. The subconductor method method yields accurate results. The wave propagation part of the cable model is based on a cascaded PI section model. A modal transformation technique has been used for the calculation in the time domain. The frequency-dependent elements of the related modal transformation matrices have been fitted with rational functions. The frequency dependence of cable parameters has been reproduced using a vector fitting algorithm and has been implemented into an resistor-inductor-capacitor network (RLC network) for each PI section. The proposed full model has been validated with measured data.

Cette thèse contribue à l'analyse des raideurs statique et dynamique des robots parallèles à câbles dans un objectif d'amélioration de la précision de positionnement statique et de la précision de suivi de trajectoire. Les modélisations statique et dynamique proposées des câbles considèrent l'effet du poids du câble sur son profil et l'effet de masse du câble sur la dynamique de ce dernier. Sur la base du modèle statique de câble proposé, l'erreur de pose statique au niveau de l'organe terminal du robot est définie et sa variation en fonction de la charge externe appliquée est utilisée pour évaluer la raideur statique globale de la structure. Un nouveau modèle dynamique vibratoire de robots à câbles est proposé en considérant le couplage de la dynamique des câbles avec les vibrations de l'organe terminal. Des validations expérimentales sont réalisées sur des prototypes de robots à câbles. Une série d'expériences de statique, d'analyses modales, d'analyses en régime libre et de suivi de trajectoire sont réalisées. Les modèles statiques et dynamiques proposés sont confirmés. Les dynamiques des câbles et du robot ainsi que leur couplage sont discutées montrant la pertinence des modèles développés pour l’amélioration des performances des robots à câbles en termes de design et le contrôle. Outre l'analyse des raideurs statique et dynamique, les modèles proposés sont appliqués dans l'amélioration du calcul de la distribution des efforts dans les câbles des robots redondants. Une nouvelle méthode de calcul de la distribution des efforts dans les câbles basée sur la détermination de la limite inférieure des forces dans les câbles est présentée. La prise en compte de la dépendance à la position dans l'espace de travail permet de limiter les efforts dans les câbles et ainsi d'améliorer l'efficience des robots d'un point de vue énergétique. / This thesis contributes to the analysis of the static and dynamic stiffness of cable-driven parallel robots (CDPRs) aiming to improve the static positioning accuracy and the trajectory tracking accuracy. The proposed static and dynamic cable modeling considers the effect of cable weight on the cable profile and the effect of cable mass on the cable dynamics. Based on the static cable model, the static pose error of the end-effector is defined and the variation of the end-effector pose error with the external load is used to evaluate the static stiffness of CDPRs. A new dynamic model of CDPRs is proposed with considering the coupling of the cable dynamics and the end-effector vibrations. Experimental validations are carried out on CDPR prototypes. Static experiments, modal experiments, free vibration experiments and trajectory experiments are performed. The proposed static and dynamic models are verified. Cable dynamics, robot dynamics and their coupling are discussed. Results show the relevance of the proposed models on improving the performances of CDPRs in terms of design and control. Besides stiffness analysis, the proposed models are applied on the force distribution of redundant actuated CDPRs. A new method on the calculation of the cable forces is proposed, where the determination of the lower-boundary of the cable forces is presented. The consideration of the pose-dependence of the lower force boundary can minimize the cable forces and improve the energy efficiency of CDPRs.

The calculation of frequency-dependent cable parameters is essential for simulations of transient phenomena in electrical power systems. The simulation of transients is more complicated than the calculation of currents and voltages in the nominal frequency range. The model has to represent the frequency dependency and the wave propagation behavior of cable lines. The introduced model combines an improved subconductor method for the determination of the frequency-dependent parameters and a PI section wave propagation model. The subconductor method considers the skin and proximity effect in all conductors for frequency ranges up to few megahertz. The subconductor method method yields accurate results. The wave propagation part of the cable model is based on a cascaded PI section model. A modal transformation technique has been used for the calculation in the time domain. The frequency-dependent elements of the related modal transformation matrices have been fitted with rational functions. The frequency dependence of cable parameters has been reproduced using a vector fitting algorithm and has been implemented into an resistor-inductor-capacitor network (RLC network) for each PI section. The proposed full model has been validated with measured data.

The knowledge about the behavior of cables is substantial in cases of transients or in cases of faults. However, there are only a few models that are tailored to the current requirements for calculations of transient phenomena in three-phase cable systems. These models are based on complex structures. PI-section cable models with simple structures were previously qualified only for calculations in the frequency domain. A new full frequency-dependent cable model to simulate transient phenomena is introduced and validated. The model is based on lumped parameters with cascaded frequency-dependent PI-sections. For the implementation and the integration in simulation tools, it is important to investigate the impact of the PI-section parameters to the accuracy, the stability and the mathematical robustness. In this work, the impact of the frequency dependence of cable parameters, the length distribution and the number of PI-sections on the results of the developed three-phase cable model have been discussed. For simulations in the time domain, two algorithms have been presented to optimize the number of PI-sections based on a specified accuracy.

Numerous studies showed that cochlear implant (CI) users generally prefer individualized stimulation rates in order to maximize their speech understanding. The underlying reasons for the reported variation in speech perception performance as a function of CI stimulation rate is unknown. However, multiple interacting electrophysiological processes influence the auditory nerve (AN) in response to high-rate CI stimulation. Experiments studying electrical pulse train stimulation of cat AN fibers (ANFs) have demonstrated that spike rates slowly decrease over time relative to onset stimulation and is often attributed to spike rate (spike-triggered) adaptation in addition to refractoriness. Interestingly, this decay tends to adapt more rapidly to higher stimulation rates. This suggests that subthreshold adaptation (accommodation) plays a critical role in reducing neural excitability. Using biophysical computational models of cat ANF including ion channel types such as hyperpolarization-activated cyclic nucleotide-gated (HCN) and low threshold potassium (KLT) channels, we measured the strength of adaptation in response to pulse train stimulation for a range of current amplitudes and pulse rates. We also tested these stimuli using a phenomenological computational ANF model capable of applying any combination of refractoriness, facilitation, accommodation, and/or spike rate adaptation. The simulation results show that HCN and KLT channels contribute to reducing model ANF excitability on the order of 1 to 100 ms. These channels contribute to both spike rate adaptation and accommodation. Using our phenomenological model ANF we have also shown that accommodation alone can produce a slow decay in ANF spike rates responding to ongoing stimulation. The CI users that do not benefit from relatively high stimulation rates may be due to ANF accommodation effects. It may be possible to use electrically evoked compound action potentials (ECAP) recordings to identify CI users exhibiting strong effects of accommodation, i.e., the increasing strength of adaptation as a function of increasing stimulation rate. / Dissertation / Doctor of Philosophy (PhD) / Cochlear implants (CI) attempt to restore hearing to individuals with severe to profound hearing deficits by stimulating the auditory nerve with a series of electrical pulses. Recent CI stimulation strategies have attempted to improve speech perception by stimulating at high pulse rates. However, studies have shown that speech perception performance does not necessarily improve with pulse rate increases, leading to speculation of possible causes. Certain ion channels located in auditory nerve fibers may contribute to driving the nerve to reduce its excitability in response to CI stimulation. In some cases, those channels could force nerve fibers to cease responding to stimulation, causing a breakdown in communication from the CI to the auditory nervous system. Our simulation studies of the auditory nerve containing certain types of channels showed that the effective rate of communication to the brain is reduced when stimulated at high rates due to the presence of these channels.

The master thesis deals with G.fast and G.mgfast transmission technologies, including their parameters. The work deals with the principle of vector DMT modulation and the possibility of time duplex TDD and full-duplex FDX in two-way communication used in these technologies. The following is a description of the line using the KHM model, which is suitable for simulations in the transmission band using G.fast and G.mgfast technologies. Subsequently, the disturbing effects of crosstalk at the near end of NEXT and the far end of FEXT and their elimination with these technologies are discussing. Part of the work explains supporting calculations to determine the SNR and bit allocation to calculate the baud rate. The work describes the methods of compensation of crosstalk FEXT and NEXT, which affect the resulting baud rate. The work also includes an application for simulation of transmission speed as a function of distance for G.fast and G.mgfast technologies, allowing changing input parameters and adjusting the transmission bandwidth based on G.9700 and G.9701 standards. Also, in work, an application is created to display the compensation of the influence of the transmitted signal crosstalk FEXT and NEXT, which allow the import of measured crosstalk between individual participants. The issue of influencing crosstalk for accurate measurements in the laboratory is also discussing. An application in the MATLAB environment is creating to display the measured characteristics.