Thesis (MTech (Electrical Engineering))--Cape Peninsula University of Technology, 2009 / More than 70% of the earth's surface is covered by water, only a small part of which is suitable for either human consumption or agricultural use. Due to pollution from agriculture, households and industry reaching rivers, lakes and seas it is Important for wastewater to be properly treated in order to remove harmful substances before it reaches the environment. Strict environmental and health regulations together with a demand for cost effective ways of wastewater treatment have made control technology in wastewater Treatment Plants an important priority. Dissolved oxygen (DO) is the amount of oxygen in the effluent and it plays a vital role of controlling VV\YTP. Oxygen dissolves in water through mixing water surface with the atmosphere, The dissolved oxygen concentration in the aerobic part of an activated sludge process should be sufficiently high to supply enough oxygen to the microorganisms in the sludge. an excessive high DO leads to high energy consumption and may also deteriorate the sludge quality, A high DO concentration in the internally recirculated water also makes the denitrification less efficient Hence, both for economical and process reasons, it is of interest to control the DO. The used controllers are normally linear controllers, proportional integral (PI) or proportional integral derivative (PID) ones. The work of these controllers leads to bad system performance, because, the process of dissolving oxygen into the wastewater is a nonlinear process and requires nonlinear control. The aim of the research project is to develop methods for design of linear and nonlinear controllers of the concentration of the DO in the aeration tank of the WWTP and to implement the designed controllers in the frameworks of PLC. The nonlinear linearizing controller based on a reference model and Lyapunov second method is designed. Additionally a linear controller is developed in a form of PI controller based on pole placement method to improve, the performance of the closed loop system. The resultant controller is to be on a PLC as a part of Adroit SCADA system. The developed programmes are used to control the wastewater treatment process in laboratory scale plant and can be applied as a part of SCADA software for control of the wastewater treatment plants.

Thesis (DTech (Electrical Engineering))--Cape Peninsula University of Technology, 2010. / Static state estimation is a mathematical procedure in which physical measurements from sub-stations and physical model are combined in an optimal way. That is, the measurements obtained from the sub-stations are used with the physical model and the states of the power system are selected or calculated such that the states match the measurements in some best way. The states of a power system are the bus voltage magnitude and voltage angle of each bus of the system. Static state estimation is an increasingly common part of electrical power utility energy management systems (EMS). It plays a critical part in a day-to-day operation of a power system utility. The system measurements obtained from static estimation are used for realtime operations like optimal power flow calculations and contingency analysis. Proper system operations with regards to avoidance of insecure conditions includes situational awareness, therefore, the static estimator plays an important role in power system security. A further motivation: in increasingly electrical power deregulation, more economic operations mean savings for customers and electrical power provider alike. Economic benefit might be realized if system operators have a more accurate situational awareness of the system through improved power system state estimator. The objective of the study was to develop method, algorithm and MATLAB program for solution of power system state estimation using parallel processing techniques. In achieving the objective, the study has concentrated on development of an approximate Tanzanian power system network model comprising of 30 buses and used as a case study; decomposing the bus admittance matrix of the model into 3 interconnected sub-systems; development of mathematical model for real and reactive power injections, real and reactive power flows in the transmission lines and tie-lines connecting the sub-systems; development of measurement data model for voltage magnitude, real and reactive power injections, real and reactive power flows; formulating of a constrained weighted least absolute value state estimation problem; development of decomposition-coordination method and algorithm; and formulating algorithm and MATLAB program for solving the constrained state estimation problem using parallel processing technique.

Thesis (MTech (Electrical Engineering))--Cape Peninsula University of Technology. / Distributed Generation systems based on renewable energy sources, such as wind or solar are mostly intermittent sources, due to their dependency on the weather, whereas those based on other primary energy sources are non-intermittent. All of them are specially designed to be integrated into distribution systems, in order to improve the power demand of consumers. In the last few decades of the twentieth century, several different factors have played a key role in increasing interest in systems. Distributed Generation (DG) is gaining more and more attention worldwide as an alternative to large-scale central generating stations. The aim of this research project is to investigate the contribution of distributed generation in fault current level in a power distribution system. The simulation results indicate that DG can have a positive or negative impact, on the fault current level in distribution network systems. The DG location and size affect the fault level. The second aim of this research was to suggest a model-based method for design, and implementation of a protection scheme for power distribution systems, by establish algorithms in a hardware environment. The overcurrent relay was chosen for the model development because it is considered a simple and popular protection scheme, and it is a common scheme in relaying applications. The proposed relay model was tested for fault conditions applied on a simple power system in different scenarios. The overcurrent relay model was implemented in MATLAB/Simulink, by using MATLAB programming languages and the SimPowerSystem (SPS) Tool. MATLAB/SIMULINK software is applicable to the modelling of generation, transmission, distribution and industrial grids, and the analysis of the interactions of these grids. This software provides a library of standard electrical components or models such as transformers, machines, and transmission lines. Therefore, the modelling and simulations are executed using MATLAB/Simulink version 2014b

Thesis (MTech (Electrical Engineering))--Cape Peninsula University of Technology, 2009. / This project was started as a result of strict environmental and health regulations together with a demand for cost effective operation of wastewater treatment plants (WNTPs). The main aim of this project is how to keep effluent concentration below a prescribed limit at the lowest possible cost Due to large fluctuations in the quality and quantity of the influent concentrations, traditional control methods are not adequate to achieve this aim The major drawback with these methods is that the disturbances affect the process before the controller has time to correct the error (Olsson and Newell, 1999: 454). This problem IS addressed through the use of modern control systems Modern control systems are model based predictive algorithms arranged as feed-forward controllers (Olsson and Newell, 1999: 454) Normally a controller is equipped with a constant set point; the goal In this project IS to calculate an optimal DO trajectory that may be sampled to provide a varying optimal set-point for the Activated Sludge Process. In this project an optimal control problem is formulated usmq 00 concentration as a control variable This requires a model of the process to be controlled, a mathematical expressions of the limitations on the process input and output variables and finally the objective functional which consists of the objectives of the control. The structures of the Benchmark plant (developed within the COST 682 working group) and the Athlone WWTPs are used to implement this optimal control strategy in MATLAB, The plant's full models are developed based on the mass balance principle incorporating the activated sludge biological models. ASM1, ASM2 ASM2d and ASM3 (developed by the IWA. working groups) To be able to develop a method that may later on be used ~or online control. the full models are reduced based on the technique in Lukasse (1996) To ensure that the reduced models keep the same prediction capabilities as the full models paran-eters of the reduced models are calculated based on the Least Squares principle. The formulated optimal control problem IS solved based on the decomposition-coordination method that involves time decomposition 111 a two layer structure, MATLAB software is developed to solve the problems for parameter estimation full and reduced model simulation and optimal control calculation for the considered different cases of plant structures and biological models, The obtained optimal DO trajectories produced the effluent state trajectories within prescribed requirements. These DO trajectories may be implemented in different SCADA systems to be tracked as set points or desired trajectories by different types of controllers.

Orientador: Nobuo Oki / Banca: Suely Cunha Amaro Mantovani / Banca: Carlos Aurélio Faria da Rocha / Resumo: A presente dissertação teve como objetivo apresentar uma proposta de análise de bulhas cardíacas (sons produzidos pelo fechamento das válvulas do coração) usando Transformada Discreta de Wavelet. Neste trabalho as bulhas cardíacas, gravadas em um arquivo digital, foram processadas através da Transformada Discreta de Wavelet nível 6 da db7 e da db6 de Daubechies e feita uma análise de sua média e do seu desvio padrão. Com a métrica desvio padrão aplicada ao sexto nível da db6 de Daubechies para classificação de sinais normais e anormais em um banco de dados de 70 amostras obteve-se um acerto da ordem de 95,71% / Abstract: This dissertation aims to present a proposal for interpretation of heart sounds using Discrete Wavelet Transform. The heart sounds recorded in a digital file were processed using level 6 of db7 and level 6 of db6 Daubechies Discrete Wavelet Transform and extracting the media and standard deviation features. The standard deviation of level6 of db6 Daubechies Discrete Wavelet is are able to differentiate between normal and abnormal from database of 70 heart sound signals with 95.71% of correct classifications / Mestre

Os métodos de dceonvolução desenvolvidos neste trabalho, usam o traço sísmico (registrado pelo geofone) para deconvoluir a fonte sísmica deste traço. O filtro preditivo que nos permite a. deconvolnção, é obtido considerando que a fonte sísmica seja de mínimo retardo e a resposta impulso do sistema estratificado constitua uma série não corrcla.cionada. Conscquentemente, a, antocorrdação da fonte sísmica e do traço sísmico são os mesmos, exceto por um fator constante. No caso da, deconvolnção dinâmica, considera este traço como a reflexão resposta do sistema sedimentário, e é mmdo para, calcular os coeficientes dos polinômios feedback e fecd-forward. Os eoeficieutes do polinômio feed-forwa.rd representam aproximadamente os eoeficient<'B de reflexão do sistema sedirnentàrio. / The dcvelopccl deconvolution methods treatcd in this work use seismic trace (recorded by geophones) in order to deconvolute the seismic source of this trace. Thc predictive filter that pcrmit us thc decouvolution, is obtaincd consideriug that thc scismic sonrce has mininmn dclay and the impulse response of the laycrcd system constitutes a. non-correlated series. As a consequence, the antocorrdation of the seismic source and the seismic trace and the same except for a constant factor. In case of the dynamic deconvolution, this trace is considered as thc rdicxion rcspousc of the sedimcntary system and is used to calculate the codficicuts of thc fecdback aud feed-forward polynomials. Thc coefficicnts of the feed-fonvard polynomials represent approximately thc reficction coefficients of the sedimentary system.

O objetivo deste trabalho é apresentar alguns métodos numéricos para resolver a equação da convecção-difusão, em um espaço uni-dimensional. Demonstrada a derivação desta equação, parte-se para a sua adaptação ao espaço uni-dimensional. A seguir, é considerado o fenômeno da difusão, isoladamente. Num segundo momento, é feito um estudo sobre o fenômeno da convecção, isoladamente. Posterionnente, considera-se o caso geral, em que os fenômenos da convecção e da difusão ocotrem simultaneamente. A importante questão da estabilidade em soluções numéricas também é analisada Ao final de cada uma das três situações mencionadas, são apresentados resultados práticos dos programas computacionais apl icados a elas. / The purpose of this work is show some numerical methods to solve diffusionconvection equation, in one dimension. Showed the deduction ofthis equation, discribring the problem, we present its adaptation to one dimension. After this, severa! steps are fulfi li: first, the di.tfusion phenomenon is considered isolated; second, the convection phenomenon is studied isolated; third, the general case, where convection and diffusion phenomena happened together. The major question about stability is also analysed. At the end ofthe three above situations, we give some praticai results of computational programs applied to them.

A reconstrução de matrizes de Jacobi a partir de dados espectrais é de grande importância na Teoria Vibracional. Usamos três métodos distintos para tal reconstrução: a aproximação por polinómios ortogonais a interação de Lanczos e a redução de Householder. Eles são aplicados a um sistema massa-mola supondo conhecidos os pólos e zeros da função frequência resposta correspondente a uma força senoidal aplicada em um dos extremos ou em um ponto interior. Resultados numéricos são obtidos com o software MATLAB e a linguagem FORTRAN. / The reconstruction of Jacobi matrices from spectral data is of great importance in vibration Theory. We use three distinct methods for sue h reconstructi on the orthogonal polynomial approach. the Lanczos interation and the Householder reduction. They are applied to a spring-mass system by assuming to know the peles and zeros of the frequency response function corresponding to a sinusoidal forcing at an end or an interior point. Numerical results are obtained with MATLAB and FORTRAN cedes.

In order to keep up with the increasing demand on speed and reliability in modern wireless systems, new standards have to be introduced. By using Multiple Input Multiple Output technology (MIMO) and Orthogonal Frequency Division Multiplexing (OFDM) technologies the performance can be increased dramatically. Forthcoming standards such as WLAN 802.11n, WiMax and 3GPP LTE are all taking advantage of MIMO technology. To perform realistic tests with these standards it is often not enough to run software simulations in for example Matlab. Instead, as many real world parameters as possible need to be included. This can be done using a testbed, like the LiUMIMO, that actually transmits and receives data through the air. The LiUMIMO is designed as a Software Defined Radio (SDR), only the RF front end and the data log are implemented in hardware, while all signal processing will be performed in Matlab.

I detta tvärvetenskapliga arbete studeras från den matematiska sidan tre klassiska populationsmodeller: Malthus tillväxtmodell, Verhulsts logistiska modell och Lotka-Volterras jägarebytesmodell. De klassiska modellerna jämförs med stokastiska. De stokastiska modeller som studeras är födelsedödsprocesser och deras diffusionsapproximation. Jämförelse görs med medelvärdesbildade simuleringar. Det krävs många simuleringar för att kunna genomföra jämförelserna. Dessa simuleringar måste utföras i datormiljö och det är här den datalogiska aspekten av arbetet kommer in. Modellerna och deras resultathantering har implementerats i både MatLab och i C, för att kunna möjliggöra en undersökning om skillnaderna i tidsåtgången mellan de båda språken, under genomförandet av ovan nämnda jämförelser. Försök till tidsoptimering utförs och även användarvänligheten under implementeringen av de matematiska problemen i de båda språken behandlas. Följande matematiska slutsatser har dragits, att de medelvärdesbildade lösningarna inte alltid sammanfaller med de klassiska modellerna när de simuleras på stora tidsintervall. I den logistiska modellen samt i Lotka-Volterras modell dör förr eller senare de stokastiska simuleringarna ut när tiden går mot oändligheten, medan deras deterministiska representation lever vidare. I den exponentiella modellen sammanfaller medelvärdet av de stokastiska simuleringarna med den deterministiska lösningen, dock blir spridningen stor för de stokastiska simuleringarna när de utförs på stora tidsintervall. Datalogiska slutsatser som har dragits är att när det kommer till att implementera få modeller, samt resultatbearbetning av dessa, som ska användas upprepade gånger, är C det bäst lämpade språket då det visat sig vara betydligt snabbare under exekvering än vad MatLab är. Dock måste hänsyn tas till alla de svårigheter som implementeringen i C drar med sig. Dessa svårigheter kan till stor del undvikas om implementeringen istället sker i MatLab, då det därmed finns tillgång till en uppsjö av väl lämpade funktioner och färdiga matematiska lösningar. / In this interdisciplinary study, three classic population models will be studied from a mathematical view: Malthus’ growth, Verhulst’s logistic model and Lotka-Volterra’s model for hunter and prey. The classic models are being compared to the stochastic ones. The stochastic models studied are the birthdeath processes and their diffusion approximation. Comparisons are made by averaging simulations. It requires numerous simulations to carry out the comparisons. The simulations must be carried out on a computer and this is where the computer science emerges to the project. The models, along with the handling of the results, have been implemented in both Mat- Lab and in C in order to allow a comparison between the two languages whilst executing the above mentioned study. Attempts to time optimization and an evaluation concerning the user-friendliness regarding the implementation of mathematical problems will be performed. Mathematic conclusions, which have been drawn, are that the averaging solutions do not always coincide with the traditional models when they are being simulated over large time. In the logistic model and in Lotka-Volterra’s model the stochastic simulations will sooner or later die when the time is moving towards infinity, whilst their deterministic representation keeps on living. In the exponential model, the mean values of the stochastic simulations and of the deterministic solution coincide. There is, however, a large spread for the stochastic simulations when they are carried out over a large time. Computer scientific conclusions drawn from the study includes that when it comes to implementing a few models, along with the handling of the results, to be used repeatedly, C is the most appropriate language as it proved to be significantly faster during execution. However, all of the difficulties during the implementation of mathematical problems in C must be kept in mind. These difficulties can be avoided if the implementation instead takes place in MatLab, where a numerous of mathematical functions and solutions will be available.