Distributed Extremum-Seeking Control with Applications to High-Altitude Balloons

The real-time optimization of large-scale systems is a difficult problem due to the need for complex models involving uncertain parameters and the high computational cost of solving such problems by a decentralized approach. Extremum-seeking control (ESC) is a model-free real-time optimization technique which can estimate unknown parameters and can optimize nonlinear time-varying systems using only a measurement of the cost function to be minimized. In this thesis, we develop a distributed version of extremum-seeking control which allows large-scale systems to be optimized without models and with minimal computing power. First, we develop a continuous-time distributed extremum-seeking controller. It has three main components: consensus, parameter estimation, and optimization. The consensus provides each local controller with an estimate of the cost to be minimized, allowing them to coordinate their actions. Using this cost estimate, parameters for a local input-output model are estimated, and the cost is minimized by following a gradient descent based on the estimate of the gradient. Next, a similar distributed extremum-seeking controller is developed in discrete-time. Finally, we consider an interesting application of distributed ESC: formation control of high-altitude balloons for high-speed wireless internet. These balloons must be steered into a favourable formation where they are spread out over the Earth and provide coverage to the entire planet. Distributed ESC is applied to this problem, and is shown to be effective for a system of 1200 ballons subjected to realistic wind currents. The approach does not require a wind model and uses a cost function based on a Voronoi partition of the sphere. Distributed ESC is able to steer balloons from a few initial launch sites into a formation which provides coverage to the entire Earth and can maintain a similar formation as the balloons move with the wind around the Earth.

Measuring the sustainability of a national economy. The application of three measures of sustainability to Ireland

An assessment of the sustainability of the Irish economy has been carried out using three methodologies, enabling comparison and evaluation of the advantages and disadvantages of each, and potential synergies among them. The three measures chosen were economy-wide Material Flow Analysis (MFA), environmentally extended input-output (EE-IO) analysis and the Ecological Footprint (EF). The research aims to assess the sustainability of the Irish economy using these methods and to draw conclusions on their effectiveness in policy making both individually and in combination. A theoretical description discusses the methods and their respective advantages and disadvantages and sets out a rationale for their combined application. The application of the methods in combination has provided insights into measuring the sustainability of a national economy and generated new knowledge on the collective application of these methods. The limitations of the research are acknowledged and opportunities to address these and build on and extend the research are identified. Building on previous research, it is concluded that a complete picture of sustainability cannot be provided by a single method and/or indicator.

Physical model parameter optimisation for calibrated emulation of the Dallas Rangemaster Treble Booster guitar pedal

In this work we explore optimising parameters of a physical circuit model relative to input/output measurements, using the Dallas Rangemaster Treble Booster as a case study. A hybrid metaheuristic/gradient descent algorithm is implemented, where the initial parameter sets for the optimisation are informed by nominal values from schematics and datasheets. Sensitivity analysis is used to screen parameters, which informs a study of the optimisation algorithm against model complexity by fixing parameters. The results of the optimisation show a significant increase in the accuracy of model behaviour, but also highlight several key issues regarding the recovery of parameters.

Laadun ja oikean tiedon varmistaminen pk-yrityksen tilaus-toimitusketjussa

Oikean tiedon siirtyminen oikeaan aikaan, sekä laadukkaan työn tekeminen yrityksen tilaus-toimitusketjun jokaisessa vaiheessa, ovat avaintekijöitä arvolupauksen ja laadun täyttämiseen asiakkaalle. Diplomityön tavoite on kehittää pk-yritykselle työkalut parempaan tiedon hallintaan ja laadukkaan työn tekemiseen toiminnanohjausjärjestelmässä. Tutkimusmenetelmänä diplomityössä käytettiin toimintatutkimusta, jossa diplomityön tekijä osallistui kohdeyrityksen päivittäiseen työn tekemiseen neljän kuukauden ajan. Tutkimuksen tiedon keräämisessä käytettiin myös puolistrukturoitua haastattelua, sekä kyselytutkimuksella. Tutkimusote työssä on kvalitatiivinen eli laadullinen tutkimusote. Työ koostuu teoriaosasta sekä soveltavasta osasta, jonka jälkeen työn tulokset esitetään tiivistetysti johtopäätöksissä ja yhteenvedossa. Toiminnanohjausjärjestelmät keräävät ja tallentavat tietoa, jota työntekijät ja yrityksen rajapinnoilla työskentelevät ihmiset siihen syöttävät. Onkin äärimmäisen tärkeää, että yrityksellä on kuvatut yhtenäiset toimintamallit prosesseille, joita he käyttävät tiedon tallentamisessa järjestelmiin. Tässä diplomityössä tutkitaan pk-yrityksen nykyiset toimintamallit tiedon tallentamisesta toiminnanohjausjärjestelmään, jonka jälkeen kehitetään yhtenäiset ohjeet toiminnanohjausjärjestelmään syötetystä myyntitilaussopimuksesta. Teoriaosuudessa esitetään laatu eri näkökulmista ja mitä laadunhallintajärjestelmät ovat ja kuinka niitä kehitetään. Teoriaosassa myös avataan tilausohjautuvan tuotannon periaatteet, sekä toiminnanohjausjärjestelmän merkitys liiketoiminnalle. Teoriaosuudella pohjustetaan soveltavaa osuutta, jossa ongelma-analyysin jälkeen kehitetään yritykseen oma laadunhallintajärjestelmä, sekä uudet työmallit tiedonvaihtoon ja sen tallentamiseen. Tuloksena on myös toiminnanohjausjärjestelmän käytön tehostuminen ohjelmistotoimittajan tekemänä. Ohjelmasta karsittiin turhat nimikkeistöt ja sen konfigurointia tehostettiin. Työn tuloksena saatiin työohjeet ydinprosessien suorittamiseen, sekä oma laadunhallintajärjestelmä tukemaan yrityksen ydin- ja tukiprosesseja, sekä tiedonhallintaa.

Desenvolvimento de magnetizador para tratamento magnético da água

Dissertação (mestrado)—Universidade de Brasília, Faculdade Gama, Programa de Pós-Graduação em Engenharia Biomédica, 2015.

Quantifying the impacts of urban wind sheltering on the building energy consumption

Common building energy modeling approaches do not account for the influence of surrounding neighborhood on the energy consumption patterns. This thesis develops a framework to quantify the neighborhood impact on a building energy consumption based on the local wind flow. The airflow in the neighborhood is predicted using Computational Fluid Dynamics (CFD) in eight principal wind directions. The developed framework in this study benefits from wind multipliers to adjust the wind velocity encountering the target building. The input weather data transfers the adjusted wind velocities to the building energy model. In a case study, the CFD method is validated by comparing with on-site temperature measurements, and the building energy model is calibrated using utilities data. A comparison between using the adjusted and original weather data shows that the building energy consumption and air system heat gain decreased by 5% and 37%, respectively, while the cooling gain increased by 4% annually.

SEQUENTIAL DECISIONS AND PREDICTIONS IN NATURAL LANGUAGE PROCESSING

Natural language processing has achieved great success in a wide range of ap- plications, producing both commercial language services and open-source language tools. However, most methods take a static or batch approach, assuming that the model has all information it needs and makes a one-time prediction. In this disser- tation, we study dynamic problems where the input comes in a sequence instead of all at once, and the output must be produced while the input is arriving. In these problems, predictions are often made based only on partial information. We see this dynamic setting in many real-time, interactive applications. These problems usually involve a trade-off between the amount of input received (cost) and the quality of the output prediction (accuracy). Therefore, the evaluation considers both objectives (e.g., plotting a Pareto curve). Our goal is to develop a formal understanding of sequential prediction and decision-making problems in natural language processing and to propose efficient solutions. Toward this end, we present meta-algorithms that take an existent batch model and produce a dynamic model to handle sequential inputs and outputs. Webuild our framework upon theories of Markov Decision Process (MDP), which allows learning to trade off competing objectives in a principled way. The main machine learning techniques we use are from imitation learning and reinforcement learning, and we advance current techniques to tackle problems arising in our settings. We evaluate our algorithm on a variety of applications, including dependency parsing, machine translation, and question answering. We show that our approach achieves a better cost-accuracy trade-off than the batch approach and heuristic-based decision- making approaches. We first propose a general framework for cost-sensitive prediction, where dif- ferent parts of the input come at different costs. We formulate a decision-making process that selects pieces of the input sequentially, and the selection is adaptive to each instance. Our approach is evaluated on both standard classification tasks and a structured prediction task (dependency parsing). We show that it achieves similar prediction quality to methods that use all input, while inducing a much smaller cost. Next, we extend the framework to problems where the input is revealed incremen- tally in a fixed order. We study two applications: simultaneous machine translation and quiz bowl (incremental text classification). We discuss challenges in this set- ting and show that adding domain knowledge eases the decision-making problem. A central theme throughout the chapters is an MDP formulation of a challenging problem with sequential input/output and trade-off decisions, accompanied by a learning algorithm that solves the MDP.

Una Estimación del Impacto del Gasto en Atención a la Dependencia en Andalucía

La atención a la dependencia es, sin lugar a dudas, uno de los grandes retos de futuro de las sociedades que envejecen aceleradamente, al que España y Andalucía no sólo no son ajenos, sino que presentan rasgos que lo amplifican. Sin embargo, es un fenómeno que adolece de importantes lagunas de información, fundamentalmente la carencia de una perspectiva integral que conecte la realidad social, el cambio económico, y la necesidad de transformación de los modelos de bienestar en general y de cuidados en particular, el gasto necesario para su desarrollo y el impacto económico asociado; enfoque holístico que se adopta en este trabajo. Dentro del marco expuesto el objetivo principal del trabajo es cuantificar el impacto en términos de actividad, empleo y retorno fiscal, así como la relación entre la prestación recibida y las características de los dependientes, que permita avanzar en la consideración de este gasto social como una inversión. Para la estimación de los impactos se ha utilizado el Marco Input-Output de Andalucía (MIOAN) de 2010. No obstante, para calcular los incrementos de renta generados por los requerimientos producidos, se ha endogeneizado la demanda de consumo, esto ha implicado construir expresamente para este cálculo una matriz ampliada de Leontief, procediendo a una estimación especifica de la fila adicional ampliada. Finalmente para estimar la relación entre prestación y características de los dependientes se han construido distintos modelos logísticos binarios, y un modelo multinomial. Desarrollado el trabajo de investigación y cuantificado el impacto, la conclusión fundamental no es sólo fuerte efecto arrastre del gasto en dependencia en la actividad y el empleo , sino que se produce en el marco de un cambio irreversible que sitúa a los Cuidados de Larga Duración asociados a la dependencia con un papel central en el contexto de la evolución del Estado el Bienestar. En concreto se ha verificado que: • Andalucía realiza un elevado gasto en atención a la dependencia, 1314 millones de euros, un 85 % del cual total es público y un 15 % privado, y que este gasto genera un fuerte impacto en la economía andaluza, pues su contribución a los grandes agregados macroeconómicos de Andalucía es de un 0,9 % del PIB, superior al impulso de demanda que supone el gasto en atención a la dependencia. • La atención a la dependencia ha mostrado una elevada capacidad de generación de empleo, más de 40.000 empleos, muy concentrado además en personas de difícil inserción en el mercado laboral, mujeres maduras de cualificación media-baja y con poca experiencia acreditable en sector formal de la atención. • La relación entre usuarios, gasto, e impacto para las trasferencias y los servicios muestra un modelo en escalera que caracteriza un sistema de universalización de coste medio con impacto moderado. Un modelo en el que las transferencias suponen el 50 % de los atendidos, el 30 % del gasto y el 20 % del impacto. • El retorno fiscal generado por la dependencia es muy elevado, casi del 45%. • El esfuerzo financiero, tanto bruto como neto, descontado el retorno vía impuestos que obtiene de cada administración, es muy asimétrico lo que puede generar un efecto desincentivador del desarrollo de un modelo con más servicios frente a transferencias. • Las características de los dependientes condicionan la prestación que parece ajustarse a los la características personales y de grado de discapacidad más que las socioeconómicas.

Synthetic Circuits for Feedback and Detection in Bacteria

Synthetic biology, by co-opting molecular machinery from existing organisms, can be used as a tool for building new genetic systems from scratch, for understanding natural networks through perturbation, or for hybrid circuits that piggy-back on existing cellular infrastructure. Although the toolbox for genetic circuits has greatly expanded in recent years, it is still difficult to separate the circuit function from its specific molecular implementation. In this thesis, we discuss the function-driven design of two synthetic circuit modules, and use mathematical models to understand the fundamental limits of circuit topology versus operating regimes as determined by the specific molecular implementation. First, we describe a protein concentration tracker circuit that sets the concentration of an output protein relative to the concentration of a reference protein. The functionality of this circuit relies on a single negative feedback loop that is implemented via small programmable protein scaffold domains. We build a mass-action model to understand the relevant timescales of the tracking behavior and how the input/output ratios and circuit gain might be tuned with circuit components. Second, we design an event detector circuit with permanent genetic memory that can record order and timing between two chemical events. This circuit was implemented using bacteriophage integrases that recombine specific segments of DNA in response to chemical inputs. We simulate expected population-level outcomes using a stochastic Markov-chain model, and investigate how inferences on past events can be made from differences between single-cell and population-level responses. Additionally, we present some preliminary investigations on spatial patterning using the event detector circuit as well as the design of stationary phase promoters for growth-phase dependent activation. These results advance our understanding of synthetic gene circuits, and contribute towards the use of circuit modules as building blocks for larger and more complex synthetic networks.

Self-defeating austerity? Assessing the impact of fiscal consolidations on unemployment

The great recession of 2008/2009 has had a huge impact on unemployment and public finances in most advanced countries, and these impacts were magnified in the southern Euro area countries by the sovereign debt crisis of 2010/2011. The fiscal consolidation imposed by the European Union on highly indebted countries was based on the assumptions of the so-called expansionary austerity. However, the reality so far shows proof to the contrary, and the results of this paper support the opposing view of a self- defeating austerity. Based on the input-output relations of the productive system, an unemployment rate/budget balance trade-off equation is derived, as well as the impact of a strong fiscal consolidation based on social transfers and the notion of neutral budget balance. An application to the Portuguese case confirms the huge costs of a strong fiscal consolidation, both in terms of unemployment and social policy regress, and it allows one to conclude that too much consolidation in one year makes consolidation more difficult in the following year.

Climate change and irrigation water requirement at Jaguaribe river basin, semi-arid northeast of Brazil.

Summary: Climate change has a potential to impact rainfall, temperature and air humidity, which have relation to plant evapotranspiration and crop water requirement. The purpose of this research is to assess climate change impacts on irrigation water demand, based on future scenarios derived from the PRECIS (Providing Regional Climates for Impacts Studies), using boundary conditions of the HadCM3 submitted to a dynamic downscaling nested to the Hadley Centre regional circulation model HadRM3P. Monthly time series for average temperature and rainfall were generated for 1961-90 (baseline) and the future (2040). The reference evapotranspiration was estimated using monthly average temperature. Projected climate change impact on irrigation water demand demonstrated to be a result of evapotranspiration and rainfall trend. Impacts were mapped over the target region by using geostatistical methods. An increase of the average crop water needs was estimated to be 18.7% and 22.2% higher for 2040 A2 and B2 scenarios, respectively. Objective ? To analyze the climate change impacts on irrigation water requirements, using downscaling techniques of a climate change model, at the river basin scale. Method: The study area was delimited between 4º39?30? and 5º40?00? South and 37º35?30? and 38º27?00? West. The crop pattern in the target area was characterized, regarding type of irrigated crops, respective areas and cropping schedules, as well as the area and type of irrigation systems adopted. The PRECIS (Providing Regional Climates for Impacts Studies) system (Jones et al., 2004) was used for generating climate predictions for the target area, using the boundary conditions of the Hadley Centre model HadCM3 (Johns et al., 2003). The considered time scale of interest for climate change impacts evaluation was the year of 2040, representing the period of 2025 to 2055. The output data from the climate model was interpolated, considering latitude/longitude, by applying ordinary kriging tools available at a Geographic Information System, in order to produce thematic maps.

Undermined climate policies : a study on the impact of regulatory and financial discrimination across heterogeneous firms in China

Firms in China within the same industry but with different ownership and size have very different production functions and can face very different emission regulations and financial conditions. This fact has largely been ignored in most of the existing literature on climate change. Using a newly augmented Chinese input–output table in which information about firm size and ownership are explicitly reported, this paper employs a dynamic computable general equilibrium (CGE) model to analyze the impact of alternative climate policy designs with respect to regulation and financial conditions on heterogeneous firms. The simulation results indicate that with a business-as-usual regulatory structure, the effectiveness and economic efficiency of climate policies is significantly undermined. Expanding regulation to cover additional firms has a first-order effect of improving efficiency. However, over-investment in energy technologies in certain firms may decrease the overall efficiency of investments and dampen long-term economic growth by competing with other fixed-capital investments for financial resources. Therefore, a market-oriented arrangement for sharing emission reduction burden and a mechanism for allocating green investment is crucial for China to achieve a more ambitious emission target in the long run.

Productivity growth and the structure of production

In this study, interactions between potential hierarchical value chains existing in the production structure and industry-wise productivity growths are sought. We applied generalized Chenery-Watanabe heuristics for matrix linearity maximization to triangulate the input-output incidence matrix for both Japan and the Republic of Korea, finding the potential directed flow of values spanning the industrial sectors of the basic (disaggregated) industry classifications for both countries. Sector specific productivity growths were measured by way of the Trönquvist index, using the 2000-2005 linked input-output tables for both Japan and Korea.

Modelo de gestión del conocimiento para el impacto económico. Aplicación al sector defensa

En el escenario actual macro económico la toma de decisiones en materia económica de cualquier entidad pública debe de estar sustentada por una adecuada inteligencia económica. Es prioritario disponer de modelos, procesos, técnicas y herramientas que permitan garantizar un control adecuado de todas sus inversiones. En la presente tesis exponemos un modelo de gestión del conocimiento basado en el marco Input-Output (IO), que nos permite conocer el impacto económico de los programas públicos. Este modelo está soportado por un sistema de información que coadyuvará a los analistas económicos para la toma de decisiones en el campo de las inversiones públicas. El principal objetivo de la tesis es la creación y desarrollo de este modelo denominado MOCIE (Modelo del Conocimiento para el Impacto Económico). No obstante, en la tesis además de profundizar sobre este modelo y la gestión del conocimiento en materia económica de los grandes programas públicos, se ha realizado un estudio que ha abarcado diferentes líneas de investigación complementarias como el análisis IO, la ingeniería y la arquitectura de sistemas de la información, la economía de la Defensa o la computación genética. El modelo propuesto en esta tesis se ha puesto en práctica en un sector económico muy específico dentro de la economía nacional: el sector de la defensa. Por tanto, ha sido también necesario realizar un estudio en profundidad del propio sector y de la gestión de los programas públicos en el ámbito de la Defensa. MOCIE se estructura a través de tres capas de gestión del conocimiento que nos permiten, por un lado, la percepción de los componentes existentes en el entorno de las inversiones públicas y la comprensión de su significado en términos de impacto económico y su relación con el resto de variables macroeconómicas nacionales y, por otro, la proyección y monitorización de su estado a lo largo de todo el ciclo de vida de los programas públicos...

Thermally induced all-optical inverter and dynamic hysteresis loops in graphene oxide dispersions

We experimentally study the temporal dynamics of amplitude-modulated laser beams propagating through a water dispersion of graphene oxide sheets in a fiber-to-fiber U-bench. Nonlinear refraction induced in the sample by thermal effects leads to both phase reversing of the transmitted signals and dynamic hysteresis in the input- output power curves. A theoretical model including beam propagation and thermal lensing dynamics reproduces the experimental findings. © 2015 Optical Society of America.