Neste trabalho foram desenvolvidas duas técnicas de sintonia para controladores preditivos por modelo. Ambas visam minimizar a soma do erro quadrático entre respostas do sistema em malha fechada e trajetórias de referência pré-definidas; a primeira resolve um problema de otimização lexicográfica enquanto a segunda resolve um problema de otimização de compromisso. As vantagens dos métodos apresentados são: maior automatização, definição de objetivos de sintonia intuitiva que considera especificações na dinâmica do processo, uma métrica no domínio do tempo e é capaz de incluir o conhecimento do engenheiro de controle em uma técnica de sintonia confiável. Um estudo de caso no sistema de craqueamento catalítico ilustrou a flexibilidade de definição dos objetivos da técnica lexicográfica. Um estudo de caso sobre uma coluna de fracionadora de óleo pesado em malha fechada com um controlador preditivo por modelo comparou ambas as estratégias de sintonia desenvolvidas aqui e pode-se concluir que a técnica lexicográfica dá prioridade aos objetivos importantes enquanto a técnica de compromisso calcula uma solução média, com respeito aos objetivos. A técnica de compromisso foi comparada a um método de sintonia da literatura quanto a aplicação em um controlador preditivo de horizonte infinito com targets para as entradas e controle por faixas das saídas com uma coluna de destilação. Observou-se que a técnica desenvolvida aqui é computacionalmente mais rápida e não requer a escolha de uma solução não-dominada dentre um conjunto de soluções de Pareto. Aplicações reais de controle preditivo são severamente afetadas por incerteza de modelo. Estendeu-se as técnicas desenvolvidas aqui para considerar o caso de incerteza multi-planta, calculando parâmetros de sintonia robustos para controladores nominais, visando tratar o compromisso entre performance e estabilidade e robustez da malha fechada. Um controlador preditivo de horizonte infinito foi sintonizado de forma robusta e comparado com um controlador preditivo robusto em malha fechada com um modelo de separadora C3/C4. Observou-se que este consegue controlar melhor o processo, entretanto, tem um tempo de computação duas ordens de grandeza maior que o controlador nominal, em operação on-line. / Two multi-objective optimization based tuning techniques for Model Predictive Control (MPC) were developed. Both take into account the sum of the squared errors between closed-loop trajectories and reference responses based on pre-defined goals as tuning objectives; one solves a lexicographic optimization to obtain an optimum set of tuning parameters (LTT), whereas the other solves a compromise optimization problem (CTT). The main advantages are an automated framework, and straightforward goal definition, which are capable of taking into account a specification on the process dynamics, a time-domain metrics, and of embedding the control engineers knowledge into a reliable approach. A fluid catalytic cracking tuning case study unveiled the goal definition flexibility of the LTT, with respect to output tracking and variable coupling. A heavy oil fractionator in closed-loop with a MPC case study compared both tuning techniques developed here, and it was observed that the LTT in fact prioritizes the main objectives, whereas the CTT yields an average solution, in terms of the tuning objectives. The CTT was compared to another multi-objective tuning technique from the literature, in the tuning of a MPC with input targets and output zone control in closed-loop with a crude distillation unit model. The simulation results showed that the CTT allows for faster results, regarding the computational time to compute the tuning parameters and there is no need of a posteriori decisions to select the best non-dominated solution. Real MPC applications are strongly hindered by model uncertainty. This limitation was addressed by the extension of the tuning techniques to account for multi-plant model uncertainty, thus obtaining optimum robustly tuned parameters for nominal controllers, addressing the trade-off between robustness and performance. A robustly tuned Infinite Horizon MPC (IHMPC) was compared to a Robust IHMPC, in closed-loop with a C3/C4 splitter system model. It was observed in a simulation that even though the latter yields better output responses, it is two orders of magnitude slower than the former in online operation.

Neste trabalho foram desenvolvidas duas técnicas de sintonia para controladores preditivos por modelo. Ambas visam minimizar a soma do erro quadrático entre respostas do sistema em malha fechada e trajetórias de referência pré-definidas; a primeira resolve um problema de otimização lexicográfica enquanto a segunda resolve um problema de otimização de compromisso. As vantagens dos métodos apresentados são: maior automatização, definição de objetivos de sintonia intuitiva que considera especificações na dinâmica do processo, uma métrica no domínio do tempo e é capaz de incluir o conhecimento do engenheiro de controle em uma técnica de sintonia confiável. Um estudo de caso no sistema de craqueamento catalítico ilustrou a flexibilidade de definição dos objetivos da técnica lexicográfica. Um estudo de caso sobre uma coluna de fracionadora de óleo pesado em malha fechada com um controlador preditivo por modelo comparou ambas as estratégias de sintonia desenvolvidas aqui e pode-se concluir que a técnica lexicográfica dá prioridade aos objetivos importantes enquanto a técnica de compromisso calcula uma solução média, com respeito aos objetivos. A técnica de compromisso foi comparada a um método de sintonia da literatura quanto a aplicação em um controlador preditivo de horizonte infinito com targets para as entradas e controle por faixas das saídas com uma coluna de destilação. Observou-se que a técnica desenvolvida aqui é computacionalmente mais rápida e não requer a escolha de uma solução não-dominada dentre um conjunto de soluções de Pareto. Aplicações reais de controle preditivo são severamente afetadas por incerteza de modelo. Estendeu-se as técnicas desenvolvidas aqui para considerar o caso de incerteza multi-planta, calculando parâmetros de sintonia robustos para controladores nominais, visando tratar o compromisso entre performance e estabilidade e robustez da malha fechada. Um controlador preditivo de horizonte infinito foi sintonizado de forma robusta e comparado com um controlador preditivo robusto em malha fechada com um modelo de separadora C3/C4. Observou-se que este consegue controlar melhor o processo, entretanto, tem um tempo de computação duas ordens de grandeza maior que o controlador nominal, em operação on-line. / Two multi-objective optimization based tuning techniques for Model Predictive Control (MPC) were developed. Both take into account the sum of the squared errors between closed-loop trajectories and reference responses based on pre-defined goals as tuning objectives; one solves a lexicographic optimization to obtain an optimum set of tuning parameters (LTT), whereas the other solves a compromise optimization problem (CTT). The main advantages are an automated framework, and straightforward goal definition, which are capable of taking into account a specification on the process dynamics, a time-domain metrics, and of embedding the control engineers knowledge into a reliable approach. A fluid catalytic cracking tuning case study unveiled the goal definition flexibility of the LTT, with respect to output tracking and variable coupling. A heavy oil fractionator in closed-loop with a MPC case study compared both tuning techniques developed here, and it was observed that the LTT in fact prioritizes the main objectives, whereas the CTT yields an average solution, in terms of the tuning objectives. The CTT was compared to another multi-objective tuning technique from the literature, in the tuning of a MPC with input targets and output zone control in closed-loop with a crude distillation unit model. The simulation results showed that the CTT allows for faster results, regarding the computational time to compute the tuning parameters and there is no need of a posteriori decisions to select the best non-dominated solution. Real MPC applications are strongly hindered by model uncertainty. This limitation was addressed by the extension of the tuning techniques to account for multi-plant model uncertainty, thus obtaining optimum robustly tuned parameters for nominal controllers, addressing the trade-off between robustness and performance. A robustly tuned Infinite Horizon MPC (IHMPC) was compared to a Robust IHMPC, in closed-loop with a C3/C4 splitter system model. It was observed in a simulation that even though the latter yields better output responses, it is two orders of magnitude slower than the former in online operation.

Master of Science / Department of Electrical and Computer Engineering / William B. Kuhn / In microchip circuitry, the tiny size of inductors creates low Q values, limiting a bandpass filter’s ability to have narrow bandwidths at RF frequencies. To counter this problem and also compensate for losses, Q-enhancement can be implemented to facilitate narrower bandwidths and boost gain. With Q-enhancement, temperature sensitivity of the circuitry causes the filter parameters to drift over time, making it necessary to adjust the filter periodically in order to keep the filter centered at the desired frequency. With the proper additional on-chip circuits used with a microprocessor, a tuning algorithm makes it possible to automatically tune the filter in-situ. The algorithm is based on increasing Q-enhancement until the filter begins to oscillate, reading the frequency of oscillation, adjusting to the desired frequency, and then decreasing Q-enhancement until the filter no longer oscillates. A 500MHz single-pole differential filter was designed with an on-chip amplitude detector and frequency prescaler to facilitate tuning. The filter was made adjustable across frequency with banks of binary weighted switchable capacitors. Q-enhancement adjustment was achieved via banks of cross-coupled FETs, also binary weighted. The circuit was fabricated in 0.5μm silicon-on-sapphire technology. The finished filter chip was controlled with a PIC microprocessor which had been programmed in C with the tuning algorithm. With the tuning algorithm in place, the filter was successfully able to align itself to within ±1MHz of the desired 500MHz center frequency. Q-enhancement levels were also able to self-adjust to maintain a desired bandwidth. An improved design based around an off-chip coupled-resonator two-pole filter has also been designed. This filter includes adjustable coupling capacitance between the two poles, which also must be tuned. A new method of tuning is proposed for such applications. The properties of a two-pole filter cause it to oscillate at two frequencies with Q-enhancement. A modified amplitude detector is capable of reading the beat frequency which results from the two oscillations, a value which relates directly to and allows tuning of the bandwidth of the filter.

Thesis submitted in compliance with the requirements for the Master's Degree in Technology: Electrical Engineering - Light Current, Durban University of Technology, Department of Electronic Engineering, 2008. / Linear control systems can be easily tuned using classical tuning techniques such as the Ziegler-Nichols and Cohen-Coon tuning formulae. Empirical studies have found that these conventional tuning methods result in an unsatisfactory control performance when they are used for processes experiencing the negative destabilizing effects of strong nonlinearities. It is for this reason that control practitioners often prefer to tune most nonlinear systems using trial and error tuning, or intuitive tuning. A need therefore exists for the development of a suitable tuning technique that is applicable for a wide range of control loops that do not respond satisfactorily to conventional tuning. Emerging technologies such as Swarm Intelligence (SI) have been utilized to solve many non-linear engineering problems. Particle Swarm Optimization (PSO), developed by Eberhart and Kennedy (1995), is a sub-field of SI and was inspired by swarming patterns occurring in nature such as flocking birds. It was observed that each individual exchanges previous experience, hence knowledge of the “best position” attained by an individual becomes globally known. In the study, the problem of identifying the PID controller parameters is considered as an optimization problem. An attempt has been made to determine the PID parameters employing the PSO technique. A wide range of typical process models commonly encountered in industry is used to assess the efficacy of the PSO methodology. Comparisons are made between the PSO technique and other conventional methods using simulations and real-time control. / National Research Foundation

Usage of cloud-based storage systems gained a lot of prominence in fast few years. Every day millions of files are uploaded and downloaded from cloud storage. This data that cannot be handled by traditional databases and this is considered to be Big Data. New powerful platforms have been developed to store and organize big and unstructured data. These platforms are called Big Data systems. Some of the most popular big data platform are Mongo, Hadoop, and Cassandra. In this, we used Cassandra database management system because it is an open source platform that is developed in java. Cassandra has a masterless ring architecture. The data is replicated among all the nodes for fault tolerance. Unlike MySQL, Cassandra uses per-column basis technique to store data. Cassandra is a NoSQL database system, which can handle unstructured data. Most of Cassandra parameters are scalable and are easy to configure. Amazon provides cloud computing platform that helps a user to perform heavy computing tasks over remote hardware systems. This cloud computing platform is known as Amazon Web Services. AWS services also include database deployment and network management services, that have a non-complex user experience. In this document, a detailed explanation on Cassandra database deployment on AWS platform is explained followed by Cassandra performance tuning.    In this study impact on read and write performance with change Cassandra parameters when deployed on Elastic Cloud Computing platform are investigated. The performance evaluation of a three node Cassandra cluster is done. With the knowledge of configuration parameters a three node, Cassandra database is performance tuned and a draft model is proposed.             A cloud environment suitable for the experiment is created on AWS. A three node Cassandra database management system is deployed in cloud environment created. The performance of this three node architecture is evaluated and is tested with different configuration parameters. The configuration parameters are selected based on the Cassandra metrics behavior with the change in parameters. Selected parameters are changed and the performance difference is observed and analyzed. Using this analysis, a draft model is developed after performance tuning selected parameters. This draft model is tested with different workloads and compared with default Cassandra model. The change in the key cache memory and memTable parameters showed improvement in performance metrics. With increases of key cache size and save time period, read performance improved. This also showed effect on system metrics like increasing CPU load and disk through put, decreasing operation time and The change in memTable parameters showed the effect on write performance and disk space utilization. With increase in threshold value of memTable flush writer, disk through put increased and operation time decreased. The draft derived from performance evaluation has better write and read performance.

Mixed integer programming (MIP) problems are highly parameterized, and finding parameter settings that achieve high performance for specific types of MIP instances is challenging. This paper presents a method to find the information about how CPLEX solver parameter settings perform for the different classes of mixed integer linear programs by using designed experiments and statistical models. Fitting a model through design of experiments helps in finding the optimal region across all combinations of parameter settings. The study involves recognizing the best parameter settings that results in the best performance for a specific class of instances. Choosing good setting has a large effect in minimizing the solution time and optimality gap.

Local and global forms of inhibition control directionally selective ganglion cells (DSGCs) in the mammalian retina. Specifically, local inhibition arising from GABAergic starburst amacrine cells (SACs) strongly contributes to direction selectivity. In this thesis, I demonstrate that increasing ambient illumination leads to the recruitment of GABAergic wide-field amacrine cells (WACs) endowing the DS circuit with an additional feature: size selectivity. Using a combination of electrophysiology, pharmacology and light/electron microscopy, I demonstrate that WACs predominantly contact presynaptic bipolar cells, which drive direct excitation and feed-forward inhibition (through SACs) to DSGCs, therefore maintaining the appropriate balance of inhibition/excitation required for generating DS. This circuit arrangement permits high-fidelity direction coding over a range of ambient light levels, over which size selectivity is adjusted. Together, these results provide novel insights into the anatomical and functional arrangement of multiple inhibitory interneurons within a single computational module in the retina. / Graduate

O principal objetivo deste trabalho é, adotado uma abordagem centrada na simulação numérica, entender a desaceleração a laser de um feixe atômico por meio da conhecida técnica de ajuste Zeeman. Nossos cálculos numéricos são capazes de reproduzir as características fundamentais dos resultados experimentais já obtidos. Também apresentamos um modelo analítico simples incorporando as idéias básicas contidas nas hipóteses utilizadas nas simulações e mostrando que estas idéias são consistentes com as conclusões numéricas e experimentais. Isto demonstra que os aspectos essenciais do processo desacelerador são bem compreendidos. / The main purpose of this work is, adopting a numerical simulation approach, to understand the laser deceleration of an atomic beam by means 0f the kown Zeeman tuning technique. Our numerical calculations are able to reproduce the fundamental features of the experimental results already obtained. We also present a simple analytical model incorporating the basic ideas contained in the hypotheses used in the simulations, and show that these ideas are consistent with the numerical and experimental conclusions. This demonstrates that the essential aspects 0f the deceleration process are well comprehended.

Neste trabalho pela primeira vez, desaceleramos um feixe de Cs pela Técnica de Sintonia Zeeman. Usamos um laser de diodo contrapropagante ao feixe atômico. Essa técnica se baseia na utilização de um campo magnético de perfil espacial parabólico para compensar o efeito Doppler e manter o átomo ressonante com o laser durante o processo de desaceleração. Conseguimos reduzir a velocidade dos átomos até C 940cm/s. Para medir essa velocidade usamos uma técnica simples, diferente da usual, que utiliza um feixe de prova. Com o nosso magneto, não foi possível desacelerar átomos com velocidade acima de 12000 cm/s. O limite de campo magnético em que tivemos que trabalhar corresponde à campo fraco, para o estado fundamental do Cs. Esse fato acarreta um aumento na probabilidade de ocorrerem transições erradas. Observamos a presença de um intervalo de \"detuning\" útil, fora do qual não conseguimos desacelerar. Esse intervalo também está relacionado com o limite máximo de velocidades para que haja desaceleração. Chegamos a esse intervalo através de simulações feitas para encontrar os parâmetros necessários à desaceleração. Os resultados obtidos experimentalmente estão de acordo com o que foi previsto pela simulação. Paralelamente à desaceleração de CS, preparamos os lasers de diodo e reduzimos sua largura de linha. Entretanto não usamos o laser estreito para a desaceleração. A fim de trabalharmos com espectroscopia de alta resolução reduzimos a largura de linha do laser a semicondutor fazendo um acoplamento da cavidade laser com uma cavidade, Fabry-Pérot, externa. Conseguimos estreitar a largura de linha até 500KHz. Esse resultado nos possibilitará investigar as linhas do Cs, aprisionado em um \"trap\" magnético-óptico, experimento este que já está em andamento em nosso laboratório / In this work for the first time, slow a beam of Cs by the Zeeman tuning technique. We use a laser diode contrapropagante the atomic beam. This technique is based on the use of a magnetic field of parabolic spatial profile to compensate for the Doppler effect and keep the atoms resonant with the laser during the downturn. We reduce the speed of C atoms to 940cm / s. To measure this speed we use a simple technique, different from the usual, which uses a beam of evidence. With our magnet, could not slow down atoms with speeds up to 12,000 cm / s. The limit of magnetic field we had to work corresponds to the weak field for the ground state of Cs. This fact implies an increase in the probability of transitions wrong. We observed a range of \"detuning\" useful, out of which we cannot slow down. This range is also related to the maximum speed for there to be slowing. We arrived in this range through simulations to find the parameters needed for deceleration. The results obtained experimentally are in agreement with what was predicted by the simulation. Parallel to the slowdown of CS, we prepared the diode lasers and reduced its line width. However do not use the laser close to the slowdown. In order to work with high-resolution spectroscopy reduced the line width of the semiconductor laser causing a coupling of the laser cavity with a cavity, Fabry-Pérot, external. We narrow the line width up to 500KHz. This result will enable us to investigate the lines of Cs, trapped in a \"trap\" magneto-optical experiment that is already underway in our laboratory

This thesis considers the problem of feature selection when the number of predictors is larger than the number of samples. The performance of supersaturated design (SSD) working with least absolute shrinkage and selection operator (LASSO) is studied in this setting. In order to achieve higher feature selection correctness, self-voting LASSO is implemented to select the tuning parameter while approximately optimize the probability of achieving Sign Correctness. Furthermore, we derive the probability of achieving Direction Correctness, and extend the self-voting LASSO to multi-task self-voting LASSO, which has a group screening effect for multiple tasks.