[pt] O presente trabalho tem como objetivo principal investigar por meio de simulação Monte Carlo algumas propriedades estatísticas dos modelos GARMA (Generalized Autoregressive Moving Average) para séries temporais de dados de contagem. Os modelos GARMA são uma extensão dos Modelos Lineares Generalizados de McCullagh e Nelder para situações de dados dependentes, caracterizando-se pela adição de um termo extra ao preditor linear, o qual passa a incorporar termos autoregressivos (AR) e de médias móveis (MA). As propriedades estatísticas investigadas foram às condições de estacionariedade dos modelos GARMA e os critérios de identificação da ordem (p,q) dos polinômios AR e MA que definem o modelo. Os resultados encontrados indicam que os critérios AIC BIC e Hannan-Quin utilizados foram razoavelmente eficazes na identificação da ordem dos modelos e que as condições de estacionariedade estabelecidas empiricamente em termo de restrições no espaço paramétrico são bastante complexas exigindo um estudo mais detalhado. Como objetivo secundário testamos os modelo GARMA em séries reais, ajustando os modelos GARMA- Poissson e GARMA-Binomial Negativa ao número de caso de poliomielite nos EUA e ao número de infartos do miocárdio no município do Rio de Janeiro. Os resultados indicam que os modelos foram capazes de explicar, de forma econômica, a variação destas séries. / [en] The main objective of this dissertation is to investigate, using Monte Carlo simulations, some statistical properties of GARMA (Generalized Autoregressive Moving Average ) models for time series of count data. GARMA models are extensions of the Generalized Linear Models to dependent data, in which autoregressive (AR) and/or moving average (MA) terms are incorporated into the linear predictor. The statistical properties targeted in our investigation were the model stationarity conditions and the identification criteria for selection of model orders, the lag structure (p,q) associated with the AR and MA terms. Our results suggest that AIC, BIC and Hann-Quinn criteria worked relatively well in identifying the model order, and that the conditions for stationarity established empirically in terms of parameter space restrictions were not totally conclusive, requiring further investigation. As a secondary objective we tested the model against real data, by fitting both a GARMA-Poisson and a GARMA-Negative Binomial to the series of number of cases of poliomyelitis on the US and the number of heart-attacks in Rio de Janeiro city. The results we found indicate that these models were able to explain, in a parsimonious way, the variation of both series.

Le modèle RiskMetrics développé par la Banque JP Morgan suite à l'amendement des accords de Bâle de 1988 a été érigé comme mesure de risque financier pour faire face aux importantes perturbations ayant affecté les marchés bancaires internationaux. Communément appelé Value at Risk, il a été admis par l'ensemble des organes et institutions financiers comme une mesure de risque cohérente. Malgré sa popularité, elle est le sujet de beaucoup de controverses. En effet, les paramètres d'estimation du système RiskMetrics sont supposés fixes au cours du temps ce qui est contraire aux caractéristiques des marchés financiers. Deux raisons valables permettent de justifier cette instabilité temporelle : * la présence d'agents hétérogènes fait qu'on n'analyse plus la VaR en se focalisant sur une seule dimension temporelle mais plutôt sur des fréquences de trading (nous recourons pour cela à la méthode Wavelet). * la structure des séries financières qui d'habitude est affectée par les phénomènes de crash, bulle etc. Ceux-ci peuvent être considérés comme des variables cachées qu'on doit prendre en compte dans l'évaluation du risque. Pour cela, nous recourons à la modélisation espace d'états et au filtre de Kalman. Nous savons d'emblée que les performances de la VaR s'évaluent en recourant au test de backtesting. Celui-ci repose sur la technique de régression roulante qui montre une faille évidente : Nous ne pouvons pas connaitre le processus gouvernant la variation des paramètres, il n'y a pas endogénéisation de la dynamique de ceux-ci. Pour apporter une solution à ce problème, nous proposons une application du filtre de Kalman sur les modèles VaR et WVaR. Ce filtre, par ses fonctions corrige de manière récursive les paramètres dans le temps. En ces termes nous définissons une mesure de risque dit SVaR qui en réalité est la VaR obtenue par une actualisation des paramètres d'estimation. Elle permet une estimation précise de la volatilité qui règne sur le marché financier. Elle donne ainsi la voie à toute institution financière de disposer de suffisamment de fonds propres pour affronter le risque de marché. / The RiskMetrics model developed by the bank JP Morgan following the amendment of Basel accords 1988 was erected as a measure of financial risk to deal with important disturbances affecting international banking markets. Commonly known as Value at Risk, it was accepted by all bodies and financial institutions to be a coherent risk measure. Despite its popularity, it is the subject of many controversies. Indeed, the estimation parameters of RiskMetrics are assumed to be fixed over time, which is contrary to the characteristics of financial markets. Two valid reasons are used to justify temporal instability : *Due to the presence of heterogenous agents the VaR is not analysed by focusing on a single temporal dimension but rather on trading frequencies (we use Wavelet method for it). *The structure of financial time series wich is usually affected by the crash bubble phenomenons and so on. These can be considered as hidden variables that we must take into account in the risk assessment. For this, we use state space modeling and kalman filter. We immediately know that performances of the VaR are evaluated using backtesting test. This is based on the technique of rolling regression wich shows an obvious break : We can not know the processes governing the variation of parameters; there is no endogeneisation dynamics thereof. To provide a solution to this problem, we propose an application of the kalman filter on VaR and WVaR models. This filter recursively corrects by its functions the parameters of time. In these terms we define a risk measure called SVaR wich in realitity is the VaR obtained by updating estimation parameters. It provides an accurate estimate of the volatility existing in the financial market. It thus gives way to any financial institution to have enough capital to face market risk.

In industrial fuel gas preparation, there are several compositional properties that must be controlled within specified limits. This allows client plants to use the fuel gas mixture safely without having to adjust and control the composition themselves. The variables to be controlled are the Higher Heating Value (HHV), Wobbe Index (WI), Flame Speed Index (FSI), and Pressure (P). These variables are controlled by adjusting the volumetric flow rates of several inlet gas streams of which some are makeup streams (always available) and some are wild streams that vary in composition and availability (by-products of plants). The inlet streams need to be adjusted in the correct ratios to keep all the controlled variables (CVs) within limits while minimising the cost of the gas blend. Furthermore, the controller needs to compensate for fluctuations in inlet stream compositions and total fuel gas demand (the total discharge from the header). This dissertation describes the modelling and model validation of an industrial fuel gas header as well as a simulation study of three different Model Predictive Control (MPC) strategies for controlling the system while minimising the overall operating cost. / Dissertation (MEng)--University of Pretoria, 2011. / Electrical, Electronic and Computer Engineering / unrestricted

The accurate evaluation of interlaminar stresses is of great significance in the analysis and design of laminated and piezoelectric laminated structures because complex behaviours of these stresses near free edges initiate edge delamination that raises concerns about the structural integrity and reliability. This thesis presented 3D hybrid analyses on the interlaminar stresses to investigate the electromechanical coupling and free edge effects of piezoelectric laminated plates with an emphasis on the realistic distributions of the 3D stress and electric fields near free edges. In this research, the state space equations for simply-supported and free-edge piezoelectric laminates under transverse loads and infinite long free-edge piezoelectric laminates under uniaxial extension were obtained in the framework of 3D piezoelasticity by considering all the independent elastic and piezoelectric constants. The equations satisfy the traction-free and open-circuit boundary conditions at free edges and the continuity conditions across all interfaces. On the basis of the transfer matrix and recursive solution approaches, 3D exact solutions were sought by a novel non-uniform layer refinement technique to evaluate the accuracy of the finite element method (FEM), and realistic gradients of interlaminar stresses and electric fields were captured. The FEM results were in good agreement with those from the present solutions except for the regions near free edges. For simply-supported and free-edge laminates, stress variations with material properties, geometries and stacking sequences were obtained. The interlaminar stress τxz was dominant at corners and τyz also tended to contribute to delamination. In the infinite long free-edge laminates, σz, τyz, Ey and Ez exhibited significant gradients near free edges. Furthermore, the considerable influence of the electromechanical coupling effect on interlaminar stresses revealed that piezoelectric laminates were more susceptible to edge delamination and the application of closed-circuited surface conditions might prevent such edge delamination. The present analytical solution demonstrated an improvement in precision over other 2D analytical and numerical solutions and could serve as a benchmark for the determination of interlaminar stresses and electric fields near the free edges of the piezoelectric laminates.

Inverters play key roles in connecting sustainable energy (SE) sources to the local loads and the ac grid. Although there has been a rapid expansion in the use of renewable sources in recent years, fundamental research, on the design of inverters that are specialized for use in these systems, is still needed. Recent advances in power electronics have led to proposing new topologies and switching patterns for single-stage power conversion, which are appropriate for SE sources and energy storage devices. The current source inverter (CSI) topology, along with a newly proposed switching pattern, is capable of converting the low dc voltage to the line ac in only one stage. Simple implementation and high reliability, together with the potential advantages of higher efficiency and lower cost, turns the so-called, single-stage boost inverter (SSBI), into a viable competitor to the existing SE-based power conversion technologies. The dynamic model is one of the most essential requirements for performance analysis and control design of any engineering system. Thus, in order to have satisfactory operation, it is necessary to derive a dynamic model for the SSBI system. However, because of the switching behavior and nonlinear elements involved, analysis of the SSBI is a complicated task. This research applies the state-space averaging technique to the SSBI to develop the state-space-averaged model of the SSBI under stand-alone and grid-connected modes of operation. Then, a small-signal model is derived by means of the perturbation and linearization method. An experimental hardware set-up, including a laboratory-scaled prototype SSBI, is built and the validity of the obtained models is verified through simulation and experiments. Finally, an eigenvalue sensitivity analysis is performed to investigate the stability and dynamic behavior of the SSBI system over a typical range of operation.

No description available.

This thesis deals with the optimization of state space controller of DC motor on FPGA in LabVIEW environment on NI cRIO platform. In the first part, the state space model of the given DC motor is presented in Matlab/Simulink and then the position feedback controller with steady-state error elimination and with state observer with error compen-sation using LQR method is designed. The thesis continues with transforming the con-troller to LabVIEW environment where the code is edited for FPGA use. Next, the fo-cus on FPGA hardware resources consumption optimization leads to careful work with fixed-point data type. After successful code compilation on target hardware, the real given DC motor is connected and the series of tests are performed. The output of the thesis is working state space controller running on FPGA and the graphical user inter-face on real-time host cRIO, which enables the user to control the plant and save the data on the disk.

This master thesis is focused on the sensitivity analysis of selected kinds of state reconstructors. They are realized in a general form, via direct and parallel programing. Quantity that determines the quality of sensitivity is output signal difference of the reconstructor with the general form of the system. Testing will be based on different initial state conditions and on the parameters change of the feedback A matrix due to the rested reconstructors.

It is well known that the inflation linked breakeven inflation, defined as the difference between a nominal yield and an inflation linked yield, sometimes is used as an approximation of the market’s inflation expectation. D’Amico et al. (2009, [5]) show that this is a poor approximation for the US market. Based on their work, this thesis shows that the approximation also is poor for the Swedish bond market. This is done by modelling the Swedish bond market using a five-factor latent variable model, where an inflation linked bond specific premium is introduced. Latent variables and parameters are estimated using a Kalman filter and a maximum likelihood estimation. The conclusion is drawn that the modelling was successful and that the model implied outputs gave plausible results.

We examine the related nature of two identification algorithms, subspace identification (SID) and Dynamic Mode Decomposition (DMD), and their correctness properties over a broad range of problems. This investigation begins by noting the strong relationship between the two algorithms, both drawing significantly on the pseudoinverse calculation using singular value decomposition, and ultimately revealing that DMD can be viewed as a substep of SID. We then perform extensive computational studies, characterizing the performance of SID on problems of various model orders and noise levels. Specifically, we generate 10,000 random systems for each model order and noise level, calculating the average identification error for each case, and then repeat the entire experiment to ensure the results are, in fact, consistent. The results both quantify the intrinsic algorithmic error at zero-noise, monotonically increasing with model complexity, as well as demonstrate an asymptotically linear degradation to noise intensity, at least for the range under study. Finally, we close by demonstrating DMD's ability to recover system matrices, because its access to full state measurements makes them identifiable. SID, on the other hand, can't possibly hope to recover the original system matrices, due to their fundamental unidentifiability from input-output data. This is true even when SID delivers excellent performance identifying a correct set of equivalent system matrices.