Submitted by Renata Lopes (renatasil82@gmail.com) on 2015-12-16T14:48:47Z No. of bitstreams: 1 ezequieldasilvaoliveira.pdf: 3924049 bytes, checksum: 2e25250cd277eeed6b8f5625283b8b69 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2015-12-16T15:15:57Z (GMT) No. of bitstreams: 1 ezequieldasilvaoliveira.pdf: 3924049 bytes, checksum: 2e25250cd277eeed6b8f5625283b8b69 (MD5) / Made available in DSpace on 2015-12-16T15:15:57Z (GMT). No. of bitstreams: 1 ezequieldasilvaoliveira.pdf: 3924049 bytes, checksum: 2e25250cd277eeed6b8f5625283b8b69 (MD5) Previous issue date: 2015-08-07 / O atendimento à demanda requer um uso eficiente do sistema geração sem afetar sua confiabilidade. Para o sistema termoelétrico o uso eficiente está diretamente relacionado com a queima de combustível e, consequentemente, com o custo de operação. Portanto, a minimização deste custo é obtida a partir da alocação da potência ativa a ser gerada para cada termoelétrica, que configura um problema de otimização denominado de Despacho Econômico (DE). Este trabalho aborda de forma real o problema do Despacho Econômico, devido a consideração das características que ocorrem durante a geração de energia elétrica. Estas características são as restrições de Zonas de Operação Proibidas (ZOP), Múltiplo Combustível (MC) e o efeito de ponto de válvula, que torna o problema do Despacho Econômico num problema não convexo e descontínuo. A proposta deste trabalho é a adoção de duas metaheurísticas bioinspiradas para resolver o problema do Despacho Econômico com características reais de operação. As técnicas bioinspiradas que são utilizadas consistem na: (i) Otimização via Enxame de Partículas e (ii) otimização baseada no fenômeno da ecolocalização do morcego, denominado Algoritmo do Morcego. Ambas as metaheurísticas são implementadas noMATLAB® e para a otimização do problema não linear e não convexo do Despacho Econômico é utilizado o modelo LINGO. Os resultados obtidos através das técnicas bioinspiradas aplicadas ao estudos de casos, são comparados comos encontrados na literatura especializada e, por fim, é feito a análise da eficiência das metaheurísticas utilizadas, cujo Algoritmo do Morcego apresenta o melhor desempenho. / The demand supply requires efficient use of generation system without affecting its reliability. For the thermoelectric systemits efficient use is directly related with fuel burn and, consequently, with cost operation. Therefore, to minimize this cost is obtained from the allocation of active power to be generated for each thermoelectric, which sets up an optimization problem called Economic Dispatch (DE). Thiswork considers a realway of the Economic Dispatch problemdue consideration of the characteristics that occur during eletricity generation. These features are restrictions Prohibited Operating Zones (POZ), multiple fuel and the valve-point effect, which makes the Economic Dispatch problem in non-convex and discontinuous problem. The proposal this work is adopting two bioinspired metaheuristics to solve the EconomicDispatch problemwith real operating characteristics. The bioinspired techniques that are used consist of: (i) Particle Swarm Optimization and (ii) Optimization based on bat echolocation phenomenon, called Bat Algorithm. Both metaheuristics are implemented in MATLAB® and for optimization of non-linear and non-convex problem is used LINGO model. The results obtained through the bioinspired techniques applied the study cases, are compared with those found in literature and, finally, is made the analysis of the efficiency of metaheuristics used, which Bat Algorithm has the best performance.

Uruguay is a pioneer in the use of renewable sources of energy and can usually satisfy its total demand from renewable sources. Control and optimization of the system is complicated by half of the installed power - wind and solar sources - be- ing non-controllable with high uncertainty and variability. In this work we present a novel optimization technique for efficient use of the production facilities. The dy- namical system is stochastic, and we deal with its non-Markovian dynamics through a Lagrangian relaxation. Continuous-time optimal control and value function are found from the solution to a sequence of Hamilton-Jacobi-Bellman partial differential equations associated with the system. We introduce a monotone scheme to avoid spurious oscillations in the numerical solution and apply the technique to a number of examples taken from the Uruguayan grid. We use parallelization and change of variables to reduce the computational times. Finally, we study the usefulness of extra system storage capacity offered by batteries.

Concentrating solar power (CSP) plant is a promising technology that exploits direct normal irradiation (DNI) from the sun to be converted into thermal energy in the solar field. One of the advantages of CSP technology is the possibility to store thermal energy in thermal energy storage (TES) for later production of electricity. The integration of thermal storage allows the CSP plant to be a dispatchable system which is defined as having a capability to schedule its operation using an innovative dispatch planning tool. Considering weather forecast and electricity price profile in the market, dispatch planning tool uses an optimization algorithm. It aims to shift the schedule of electricity delivery to the hours with high electricity price. These hours are usually reflected by the high demand periods. The implementation of dispatch optimizer can benefit the CSP plants economically from the received financial revenues. This study proposes an optimization of dispatch planning strategies for the parabolic trough CSP plant under two dispatch approaches: solar driven and storage driven. The performed simulation improves the generation of electricity which reflects to the increase of financial revenue from the electricity sale in both solar and storage driven approaches. Moreover, the optimization also proves to reduce the amount of dumped thermal energy from the solar field.

This diploma thesis deals with the operation of taxi-dispatch system and possibilities of its optimization by using geolocation services. The introductory part of diploma thesis contains theoretical basis used for an analysis of the current state and for design of an appropriate solution. The analysis of the current state deals with description of dispatch system functions within the selected company, finding weaknesses of the current system and also includes an analysis of market with systems designed for dispatch operation. Based on the analysis has been proceeded to design a custom solution consisting of a mobile application and a server part, which was then put into practice in the chosen company. Diploma thesis is finished by an evaluation of the proposal, which is based on the comparison of the situation before and after the introduction of the proposed solution.

Multi-energy systems offer an opportunity to leverage energy conversion processes and temporary energy storage mechanisms to reduce costs and emissions during operation of campuses, cities, and buildings. With increasing options for flexibility in demand-side resources it is possible to meet demand without sacrificing comfort and convenience of MES occupants. This Thesis develops a multi-period, linear optimization model of an MES with flexible buildings that captures nonlinearities in the efficiency of energy conversion processes. The flexible buildings are parametrized, in part, based on historical data from a college campus in Vermont, USA. The idea of the MES model is to investigate the role that flexibility plays in reducing costs and emissions for a small campus relative to that of a possible carbon tax. The operation of the MES is optimized to reduce costs based on representative seasons. Interestingly, it is found that when utilized optimally, flexible buildings allows for a more cost and energy effective method of not only meeting demand but also reducing carbon emissions in the process.

The architecture of virtual machine (VM) interpreters has long been a subject of researchand discussion. The initial trend of stack-based interpreters was shortly thereafterchallenged by research showing the performance advantages of virtual register machines. Despite this, many VM interpreters are still stack-based, with some notable exceptions, like Lua, Android Runtime, and its predecessor Dalvik. A register architecture isusually associated with greater overhead from instruction dispatch, and to some extent, instruction decoding. By designing, and implementing a novel technique that replaces the conventional way of decoding instructions, this thesis attempts to reduce that overhead. More specifically, a VM interpreter is developed as an artifact of design-science research. The novel technique is then evaluated through benchmarking in various configurations. As the results indicate, however, using this technique showed no performance advantage, as the resulting machine instructions are exactly the same after compiler optimization. This suggests that there is no apparent decoding overhead to begin with. As a result, register-based VMs seem to not suffer from any dispatch related overhead, other than the fact that there are more operands per instruction to access. Source code is available on GitHub, at https://github.com/kaspr61/RackVM.

Disasters are characterized by large numbers of victims and required resources, overwhelming the available resources. Disaster response involves various entities like Incident Commanders, dispatch centers, emergency operations centers, area command and hospitals. An effective emergency response system should facilitate coordination between these various entities. Victim triage, emergency resource allocation and victim dispatch to hospitals form an important part of an emergency response system. In this present research effort, an emergency response system with the aforementioned components is developed. Triage is the process of prioritizing mass casualty victims based on severity of injuries. The system presented in this thesis is a low-cost victim triage system with RFID tags that aggregate all victim information within a database. It will allow first responders' movements to be tracked using GPS. A web-based real time resource allocation tool that can assist the Incident Commanders in resource allocation and transportation for multiple simultaneous incidents has been developed. This tool ensures that high priority resources at emergency sites are received in least possible time. This web-based tool also computes the patient dispatch schedule from each disaster site to each hospital. Patients are allocated to nearest hospitals with available medical facilities. This tool can also assist resource managers in emergency resource planning by computing the time taken to receive required resources from the nearest depots using Google Maps. These web-based tools complements emergency response systems by providing decision-making capabilities.

No description available.

[pt] Este trabalho tem duas partes, uma teórica e uma computacional. Na primeira, são desenvolvidas fórmulas gerais utilizando-se o Princípio do Máximo Discreto para o dispatch ótimo em sistemas hidro-termo-elétricos. As restrições nas variáveis de estado - os volumes dos reservatórios - são levadas em consideração de modo rigoroso. A necessidade e suficiência do Princípio do Máximo Discreto são estudadas de modo cuidadoso. Na segunda parte, dois problemas são resolvidos e é apresentado um método para a determinação do valor inicial do vetor adjunto, que foi o problema mais dificil a ser solucionado do ponto de vista computacional. / [en] This work consists essentially of two parts: a theoretical and a computational one. In the first part general formulas are developed using the discrete maximum principle for the optimel dispatch of hydro-steam-electrical systems. The restrictions on the state vector (the volumes of the reservoirs) are rigorously taken into accout. The necessity and sufficiency of the discrete maximum principle are carefully studied. In the second part two problems are solved and method is presented for the determination of the initial value of the adjoint vector.

This is a report of a professional project intended to act as an informational tool for the evaluation of a dynamic dispatch system at Embarcadero Station for the San Francisco Municipal Transportation Agency (SFMTA). Systems that operate dispatch algorithms do so in response to on-time performance and reliability. The optimization problem is documented in many transportation contexts including airline networks, bus dispatch, and freight routing. According to the research, optimizing available options and re-routing based on available options can create a more efficient system that would minimize operating costs and improving service reliability for customers. The methodology presented here uses current headway information to dynamically change dispatch assignment between J, L, and M lines. The reassignment program was applied to randomly selected weekdays in March of 2012. After analysis it was determined that the application has the potential to decrease mean operating headways by up to 3% (or approximately 15 seconds).