Removing superfluous versions in polyvariant specialization of prolog programs.

Polyvariant specialization allows generating múltiple versions of a procedure, which can then be separately optimized for different uses. Since allowing a high degree of polyvariance often results in more optimized code, polyvariant specializers, such as most partial evaluators, can genérate a large number of versions. This can produce unnecessarily large residual programs. Also, large programs can be slower due to cache miss effects. A possible solution to this problem is to introduce a minimization step which identifies sets of equivalent versions, and replace all occurrences of such versions by a single one. In this work we present a unifying view of the problem of superfluous polyvariance. It includes both partial deduction and abstract múltiple specialization. As regards partial deduction, we extend existing approaches in several ways. First, previous work has dealt with puré logic programs and a very limited class of builtins. Herein we propose an extensión to traditional characteristic trees which can be used in the presence of calis to external predicates. This includes all builtins, librarles, other user modules, etc. Second, we propose the possibility of collapsing versions which are not strictly equivalent. This allows trading time for space and can be useful in the context of embedded and pervasive systems. This is done by residualizing certain computations for external predicates which would otherwise be performed at specialization time. Third, we provide an experimental evaluation of the potential gains achievable using minimization which leads to interesting conclusions.

A generic framework for the analysis and specialization of kogic programs

The relationship between abstract interpretation [2] and partial evaluation [5] has received considerable attention and (partial) integrations have been proposed starting from both the partial deduction (see e.g. [6] and its references) and abstract interpretation perspectives. Abstract interpretation-based analyzers (such as the CiaoPP analyzer [9,4]) generally compute a program analysis graph [1] in order to propagate (abstract) call and success information by performing fixpoint computations when needed. On the other hand, partial deduction methods [7] incorporate powerful techniques for on-line specialization including (concrete) call propagation and unfolding.

Lower bound cost estimation for logic programs

It is generally recognized that information about the runtime cost of computations can be useful for a variety of applications, including program transformation, granularity control during parallel execution, and query optimization in deductive databases. Most of the work to date on compile-time cost estimation of logic programs has focused on the estimation of upper bounds on costs. However, in many applications, such as parallel implementations on distributed-memory machines, one would prefer to work with lower bounds instead. The problem with estimating lower bounds is that in general, it is necessary to account for the possibility of failure of head unification, leading to a trivial lower bound of 0. In this paper, we show how, given type and mode information about procedures in a logic program, it is possible to (semi-automatically) derive nontrivial lower bounds on their computational costs. We also discuss the cost analysis for the special and frequent case of divide-and-conquer programs and show how —as a pragmatic short-term solution —it may be possible to obtain useful results simply by identifying and treating divide-and-conquer programs specially.

Independent AND-parallel implementation of narrowing

We present a parallel graph narrowing machine, which is used to implement a functional logic language on a shared memory multiprocessor. It is an extensión of an abstract machine for a purely functional language. The result is a programmed graph reduction machine which integrates the mechanisms of unification, backtracking, and independent and-parallelism. In the machine, the subexpressions of an expression can run in parallel. In the case of backtracking, the structure of an expression is used to avoid the reevaluation of subexpressions as far as possible. Deterministic computations are detected. Their results are maintained and need not be reevaluated after backtracking.

Program development using abstract interpretation (and the ciao system preprocessor)

The technique of Abstract Interpretation has allowed the development of very sophisticated global program analyses which are at the same time provably correct and practical. We present in a tutorial fashion a novel program development framework which uses abstract interpretation as a fundamental tool. The framework uses modular, incremental abstract interpretation to obtain information about the program. This information is used to validate programs, to detect bugs with respect to partial specifications written using assertions (in the program itself and/or in system librarles), to genérate and simplify run-time tests, and to perform high-level program transformations such as múltiple abstract specialization, parallelization, and resource usage control, all in a provably correct way. In the case of validation and debugging, the assertions can refer to a variety of program points such as procedure entry, procedure exit, points within procedures, or global computations. The system can reason with much richer information than, for example, traditional types. This includes data structure shape (including pointer sharing), bounds on data structure sizes, and other operational variable instantiation properties, as well as procedure-level properties such as determinacy, termination, non-failure, and bounds on resource consumption (time or space cost). CiaoPP, the preprocessor of the Ciao multi-paradigm programming system, which implements the described functionality, will be used to illustrate the fundamental ideas.

Towards structured state threading in prolog

It is very often the case that programs require passing, maintaining, and updating some notion of state. Prolog programs often implement such stateful computations by carrying this state in predicate arguments (or, alternatively, in the internal datábase). This often causes code obfuscation, complicates code reuse, introduces dependencies on the data model, and is prone to incorrect propagation of the state information among predicate calis. To partly solve these problems, we introduce contexts as a consistent mechanism for specifying implicit arguments and its threading in clause goals. We propose a notation and an interpretation for contexts, ranging from single goals to complete programs, give an intuitive semantics, and describe a translation into standard Prolog. We also discuss a particular light-weight implementation in Ciao Prolog, and we show the usefulness of our proposals on a series of examples and applications, including code directiy using contexts, DCGs, extended DCGs, logical loops and other custom control structures.

An integration of partial evaluation in a generic abstract interpretation framework

Information generated by abstract interpreters has long been used to perform program specialization. Additionally, if the abstract interpreter generates a multivariant analysis, it is also possible to perform múltiple specialization. Information about valúes of variables is propagated by simulating program execution and performing fixpoint computations for recursive calis. In contrast, traditional partial evaluators (mainly) use unfolding for both propagating valúes of variables and transforming the program. It is known that abstract interpretation is a better technique for propagating success valúes than unfolding. However, the program transformations induced by unfolding may lead to important optimizations which are not directly achievable in the existing frameworks for múltiple specialization based on abstract interpretation. The aim of this work is to devise a specialization framework which integrates the better information propagation of abstract interpretation with the powerful program transformations performed by partial evaluation, and which can be implemented via small modifications to existing generic abstract interpreters. With this aim, we will relate top-down abstract interpretation with traditional concepts in partial evaluation and sketch how the sophisticated techniques developed for controlling partial evaluation can be adapted to the proposed specialization framework. We conclude that there can be both practical and conceptual advantages in the proposed integration of partial evaluation and abstract interpretation.

Radio-holografía de microondas para la optimización de la superficie de grandes antenas reflectoras

En el presente trabajo de tesis se afronta el problema de la optimización de la superficie de grandes antenas reflectoras. Es sabido que los grandes reflectores, formados por una superficie panelada, sufren deformaciones debidas al impacto del viento, a los cambios de temperatura y a los efectos gravitacionales derivados del gran peso de la estructura. Estos efectos hacen que los reflectores pierdan su forma ideal, generalmente de paraboloide, y se reduzca su eficiencia de apertura y, por tanto, se limite la máxima frecuencia de uso de los mismos. Es necesario, por tanto, disponer de técnicas que permitan medir el estado de la superficie de grandes reflectores, y derivar los ajustes necesarios a aplicar sobre los tornillos de soporte de cada uno de los paneles que conforman dicha superficie. De esta manera, se devolvería al reflector su forma óptima y aumentaría la eficiencia de apertura y el rango de frecuencias de uso. Hay que resaltar que el aumento de la eficiencia de un radiotelescopio supone una reducción en el tiempo de integración necesario para la detección de las debilísimas señales generadas por las radiofuentes naturales, ahorrando así valioso tiempo de observación. Además, el incremento en el rango de frecuencias permite la detección de nuevas líneas o especies moleculares en dichas radiofuentes. Tras un primer capítulo introductorio, se presenta, en el capítulo segundo, la geometría de estos grandes reflectores y la influencia de los distintos factores que afectan a la calidad de la superficie de los mismos, como la gravedad, el viento y la temperatura, particularizando para el caso del radiotelescopio de 40 metros del Centro Astronómico de Yebes. En el tercer capítulo, se presentan las diferentes técnicas metrológicas empleadas actualmente para abordar la determinación de estos ajustes, mostrándose las ventajas e inconvenientes de cada una de ellas. Actualmente, la técnica metrológica más precisa y rápida para llevar a cabo esta tarea de caracterización de la superficie de un gran reflector, es la radio-holografía de microondas presentada en el capítulo cuarto. A partir de las medidas proporcionadas por esta técnica, realizadas con la ayuda de un transmisor, y mediante transformaciones de campo, se calculan los errores de la superficie del reflector, respecto al paraboloide ideal, y se derivan los ajustes necesarios. En los capítulos quinto y sexto se presentan los resultados de la aplicación de esta técnica a dos radiotelescopios: el de 30 metros de IRAM en Pico de Veleta (Granada) y los prototipos de 12 metros de las antenas del proyecto ALMA. Por su parte, el capítulo séptimo contiene el núcleo fundamental de esta tesis y presenta el desarrollo de la técnica de radio-holografía de microondas para optimizar la superficie del radiotelescopio de 40 metros del Centro Astronómico de Yebes. Para ello, ha sido necesario diseñar, construir e instalar un receptor de doble canal en banda Ku en foco primario, y la instrumentación asociada para hacer las medidas de amplitud y fase del diagrama de radiación. Además, ha sido necesario desarrollar el software para llevar a cabo las transformaciones de campo y derivar los ajustes de los paneles. De las medidas holográficas iniciales resultó un error de la superficie del radiotelescopio de 485 μm WRMS, respecto al paraboloide ideal en dirección normal. Tras varias iteraciones del proceso de medida y ajuste, se consiguió reducir dicho error a 194 μm WRMS. Esta notable mejora de la calidad de la superficie ha supuesto aumentar la eficiencia de apertura desde 2,6% al 38,2% a 86 GHz, para un receptor a esta frecuencia situado en el foco primario que produjese la misma iluminación que el receptor de holografía. In this thesis the problem of large reflector antenna surface optimization is faced. It is well known that large reflectors, which are made of a panelled surface, suffer from deformations due to the impact of wind, temperature gradients and gravity loads coming from the high weigth of the structure. These effects distort the ideal reflector shape, which is a paraboloid in most cases, hence reducing the aperture efficiency of the reflector and limiting the maximum frequency of operation. Therefore, it is necessary to have some techniques to measure the status of large reflector surfaces and to derive the adjustment values to be applied to the screws that connect the surface panels to the reflector back-up structure. In this way, the reflector would recover its optimum shape and the aperture efficiency and frequency range would increase. It has to be stated that an increment in the radiotelescope aperture efficiency would imply a reduction in the integration time needed to detect such weak signals coming from natural radiosources in space and, hence, an important saving in observation time. In addition, the increase in the frequency range of operation would allow the detection of new molecular lines in those radiosources. After the introduction, the second chapter shows the geometry of large reflector antennas and the impact on its surface quality of different factors like gravity, wind and temperature, particularly for the case of the Centro Astronómico de Yebes 40 meter radiotelescope. The third chapter deals with the different metrology techniques used to determine the panel adjustments, including the advantages and drawbacks of each one Currently, the most accurate and fast metrologic technique to carry out the characterization of large reflector surfaces is microwave radio-holography2, which is shown in chapter four. From the measurements provided by microwave radio-holography, performed with the help of a transmitter, and with the use of field transformations, the reflector surface errors are computed and the panel adjustments are derived. Chapters five and six show the results of holographic measurements applied to two first class radiotelescopes: the IRAM 30 meter radiotelescope and the 12 meter prototype antennas for the ALMA project. Chapter seven contains the main work of this thesis. It presents the development of the microwave radio-holography technique for the optimization of the Centro Astronómico de Yebes 40m radiotelescope. The work implied the design, construction and instalation of a prime focus Ku-band dual channel receiver, together with the associated instrumentation to measure the amplitude and phase of the radiotelescope radiation pattern. In addition, the software to carry out field transformations and screw settings computations was developed too. Initial holography measurements came up with an surface error of 485 μmWRMS in normal direction with respect to the best-fit paraboloid. After a few iterations of the measurementadjustment cycle, the surface error was reduced to 194 μm WRMS. This remarkable improvement in surface quality means an increment in aperture efficiency from 2,6% to 38,2% at 86 GHz, assuming a receiver at this frequency in prime focus position which produces the same illumination as the holography receiver.

A tutorial on program development and optimization using the Ciao preprocessor

We present in a tutorial fashion CiaoPP, the preprocessor of the Ciao multi-paradigm programming system, which implements a novel program development framework which uses abstract interpretation as a fundamental tool. The framework uses modular, incremental abstract interpretation to obtain information about the program. This information is used to validate programs, to detect bugs with respect to partial specifications written using assertions (in the program itself and/or in system libraries), to generate and simplify run-time tests, and to perform high-level program transformations such as multiple abstract specialization, parallelization, and resource usage control, all in a provably correct way. In the case of validation and debugging, the assertions can refer to a variety of program points such as procedure entry, procedure exit, points within procedures, or global computations. The system can reason with much richer information than, for example, traditional types. This includes data structure shape (including pointer sharing), bounds on data structure sizes, and other operational variable instantiation properties, as well as procedure-level properties such as determinacy, termination, non-failure, and bounds on resource consumption (time or space cost).

Código híbrido avanzado de motores de plasma de efecto Hall

La propulsión eléctrica constituye hoy una tecnología muy competitiva y de gran proyección de futuro. Dentro de los diversos motores de plasma existentes, el motor de efecto Hall ha adquirido una gran madurez y constituye un medio de propulsión idóneo para un rango amplio de misiones. En la presente Tesis se estudian los motores Hall con geometría convencional y paredes dieléctricas. La compleja interacción entre los múltiples fenómenos físicos presentes hace que sea difícil la simulación del plasma en estos motores. Los modelos híbridos son los que representan un mejor compromiso entre precisión y tiempo de cálculo. Se basan en utilizar un modelo fluido para los electrones y algoritmos de dinámica de partículas PIC (Particle-In- Cell) para los iones y los neutros. Permiten hacer uso de la hipótesis de cuasineutralidad del plasma, a cambio de resolver separadamente las capas límite (o vainas) que se forman en torno a las paredes de la cámara. Partiendo de un código híbrido existente, llamado HPHall-2, el objetivo de la Tesis doctoral ha sido el desarrollo de un código híbrido avanzado que mejorara la simulación de la descarga de plasma en un motor de efecto Hall. Las actualizaciones y mejoras realizadas en las diferentes partes que componen el código comprenden tanto aspectos teóricos como numéricos. Fruto de la extensa revisión de la algoritmia del código HPHall-2 se han conseguido reducir los errores de precisión un orden de magnitud, y se ha incrementado notablemente su consistencia y robustez, permitiendo la simulación del motor en un amplio rango de condiciones. Algunos aspectos relevantes a destacar en el subcódigo de partículas son: la implementación de un nuevo algoritmo de pesado que permite determinar de forma más precisa el flujo de las magnitudes del plasma; la implementación de un nuevo algoritmo de control de población, que permite tener suficiente número de partículas cerca de las paredes de la cámara, donde los gradientes son mayores y las condiciones de cálculo son más críticas; las mejoras en los balances de masa y energía; y un mejor cálculo del campo eléctrico en una malla no uniforme. Merece especial atención el cumplimiento de la condición de Bohm en el borde de vaina, que en los códigos híbridos representa una condición de contorno necesaria para obtener una solución consistente con el modelo de interacción plasma-pared, y que en HPHall-2 aún no se había resuelto satisfactoriamente. En esta Tesis se ha implementado el criterio cinético de Bohm para una población de iones con diferentes cargas eléctricas y una gran dispersión de velocidades. En el código, el cumplimiento de la condición cinética de Bohm se consigue por medio de un algoritmo que introduce una fina capa de aceleración nocolisional adyacente a la vaina y mide adecuadamente el flujo de partículas en el espacio y en el tiempo. Las mejoras realizadas en el subcódigo de electrones incrementan la capacidad de simulación del código, especialmente en la región aguas abajo del motor, donde se simula la neutralización del chorro del plasma por medio de un modelo de cátodo volumétrico. Sin abordar el estudio detallado de la turbulencia del plasma, se implementan modelos sencillos de ajuste de la difusión anómala de Bohm, que permiten reproducir los valores experimentales del potencial y la temperatura del plasma, así como la corriente de descarga del motor. En cuanto a los aspectos teóricos, se hace especial énfasis en la interacción plasma-pared y en la dinámica de los electrones secundarios libres en el interior del plasma, cuestiones que representan hoy en día problemas abiertos en la simulación de los motores Hall. Los nuevos modelos desarrollados buscan una imagen más fiel a la realidad. Así, se implementa el modelo de vaina de termalización parcial, que considera una función de distribución no-Maxwelliana para los electrones primarios y contabiliza unas pérdidas energéticas más cercanas a la realidad. Respecto a los electrones secundarios, se realiza un estudio cinético simplificado para evaluar su grado de confinamiento en el plasma, y mediante un modelo fluido en el límite no-colisional, se determinan las densidades y energías de los electrones secundarios libres, así como su posible efecto en la ionización. El resultado obtenido muestra que los electrones secundarios se pierden en las paredes rápidamente, por lo que su efecto en el plasma es despreciable, no así en las vainas, donde determinan el salto de potencial. Por último, el trabajo teórico y de simulación numérica se complementa con el trabajo experimental realizado en el Pnnceton Plasma Physics Laboratory, en el que se analiza el interesante transitorio inicial que experimenta el motor en el proceso de arranque. Del estudio se extrae que la presencia de gases residuales adheridos a las paredes juegan un papel relevante, y se recomienda, en general, la purga completa del motor antes del modo normal de operación. El resultado final de la investigación muestra que el código híbrido desarrollado representa una buena herramienta de simulación de un motor Hall. Reproduce adecuadamente la física del motor, proporcionando resultados similares a los experimentales, y demuestra ser un buen laboratorio numérico para estudiar el plasma en el interior del motor. Abstract Electric propulsion is today a very competitive technology and has a great projection into the future. Among the various existing plasma thrusters, the Hall effect thruster has acquired a considerable maturity and constitutes an ideal means of propulsion for a wide range of missions. In the present Thesis only Hall thrusters with conventional geometry and dielectric walls are studied. The complex interaction between multiple physical phenomena makes difficult the plasma simulation in these engines. Hybrid models are those representing a better compromise between precision and computational cost. They use a fluid model for electrons and Particle-In-Cell (PIC) algorithms for ions and neutrals. The hypothesis of plasma quasineutrality is invoked, which requires to solve separately the sheaths formed around the chamber walls. On the basis of an existing hybrid code, called HPHall-2, the aim of this doctoral Thesis is to develop an advanced hybrid code that better simulates the plasma discharge in a Hall effect thruster. Updates and improvements of the code include both theoretical and numerical issues. The extensive revision of the algorithms has succeeded in reducing the accuracy errors in one order of magnitude, and the consistency and robustness of the code have been notably increased, allowing the simulation of the thruster in a wide range of conditions. The most relevant achievements related to the particle subcode are: the implementation of a new weighing algorithm that determines more accurately the plasma flux magnitudes; the implementation of a new algorithm to control the particle population, assuring enough number of particles near the chamber walls, where there are strong gradients and the conditions to perform good computations are more critical; improvements in the mass and energy balances; and a new algorithm to compute the electric field in a non-uniform mesh. It deserves special attention the fulfilment of the Bohm condition at the edge of the sheath, which represents a boundary condition necessary to match consistently the hybrid code solution with the plasma-wall interaction, and remained as a question unsatisfactory solved in the HPHall-2 code. In this Thesis, the kinetic Bohm criterion has been implemented for an ion particle population with different electric charges and a large dispersion in their velocities. In the code, the fulfilment of the kinetic Bohm condition is accomplished by an algorithm that introduces a thin non-collisional layer next to the sheaths, producing the ion acceleration, and measures properly the flux of particles in time and space. The improvements made in the electron subcode increase the code simulation capabilities, specially in the region downstream of the thruster, where the neutralization of the plasma jet is simulated using a volumetric cathode model. Without addressing the detailed study of the plasma turbulence, simple models for a parametric adjustment of the anomalous Bohm difussion are implemented in the code. They allow to reproduce the experimental values of the plasma potential and the electron temperature, as well as the discharge current of the thruster. Regarding the theoretical issues, special emphasis has been made in the plasma-wall interaction of the thruster and in the dynamics of free secondary electrons within the plasma, questions that still remain unsolved in the simulation of Hall thrusters. The new developed models look for results closer to reality, such as the partial thermalization sheath model, that assumes a non-Maxwellian distribution functions for primary electrons, and better computes the energy losses at the walls. The evaluation of secondary electrons confinement within the chamber is addressed by a simplified kinetic study; and using a collisionless fluid model, the densities and energies of free secondary electrons are computed, as well as their effect on the plasma ionization. Simulations show that secondary electrons are quickly lost at walls, with a negligible effect in the bulk of the plasma, but they determine the potential fall at sheaths. Finally, numerical simulation and theoretical work is complemented by the experimental work carried out at the Princeton Plasma Physics Laboratory, devoted to analyze the interesting transitional regime experienced by the thruster in the startup process. It is concluded that the gas impurities adhered to the thruster walls play a relevant role in the transitional regime and, as a general recomendation, a complete purge of the thruster before starting its normal mode of operation it is suggested. The final result of the research conducted in this Thesis shows that the developed code represents a good tool for the simulation of Hall thrusters. The code reproduces properly the physics of the thruster, with results similar to the experimental ones, and represents a good numerical laboratory to study the plasma inside the thruster.

Comparison of Mesh Adaptation Using the Adjoint Methodology and Truncation Error Estimates

Mesh adaptation based on error estimation has become a key technique to improve th eaccuracy o fcomputational-fluid-dynamics computations. The adjoint-based approach for error estimation is one of the most promising techniques for computational-fluid-dynamics applications. Nevertheless, the level of implementation of this technique in the aeronautical industrial environment is still low because it is a computationally expensive method. In the present investigation, a new mesh refinement method based on estimation of truncation error is presented in the context of finite-volume discretization. The estimation method uses auxiliary coarser meshes to estimate the local truncation error, which can be used for driving an adaptation algorithm. The method is demonstrated in the context of two-dimensional NACA0012 and three-dimensional ONERA M6 wing inviscid flows, and the results are compared against the adjoint-based approach and physical sensors based on features of the flow field.

Crowdsourcing malaria parasite quantification: an online game for analyzing images of infected thick blood smears

Background: There are 600,000 new malaria cases daily worldwide. The gold standard for estimating the parasite burden and the corresponding severity of the disease consists in manually counting the number of parasites in blood smears through a microscope, a process that can take more than 20 minutes of an expert microscopist’s time. Objective: This research tests the feasibility of a crowdsourced approach to malaria image analysis. In particular, we investigated whether anonymous volunteers with no prior experience would be able to count malaria parasites in digitized images of thick blood smears by playing a Web-based game. Methods: The experimental system consisted of a Web-based game where online volunteers were tasked with detecting parasites in digitized blood sample images coupled with a decision algorithm that combined the analyses from several players to produce an improved collective detection outcome. Data were collected through the MalariaSpot website. Random images of thick blood films containing Plasmodium falciparum at medium to low parasitemias, acquired by conventional optical microscopy, were presented to players. In the game, players had to find and tag as many parasites as possible in 1 minute. In the event that players found all the parasites present in the image, they were presented with a new image. In order to combine the choices of different players into a single crowd decision, we implemented an image processing pipeline and a quorum algorithm that judged a parasite tagged when a group of players agreed on its position. Results: Over 1 month, anonymous players from 95 countries played more than 12,000 games and generated a database of more than 270,000 clicks on the test images. Results revealed that combining 22 games from nonexpert players achieved a parasite counting accuracy higher than 99%. This performance could be obtained also by combining 13 games from players trained for 1 minute. Exhaustive computations measured the parasite counting accuracy for all players as a function of the number of games considered and the experience of the players. In addition, we propose a mathematical equation that accurately models the collective parasite counting performance. Conclusions: This research validates the online gaming approach for crowdsourced counting of malaria parasites in images of thick blood films. The findings support the conclusion that nonexperts are able to rapidly learn how to identify the typical features of malaria parasites in digitized thick blood samples and that combining the analyses of several users provides similar parasite counting accuracy rates as those of expert microscopists. This experiment illustrates the potential of the crowdsourced gaming approach for performing routine malaria parasite quantification, and more generally for solving biomedical image analysis problems, with future potential for telediagnosis related to global health challenges.

Seismic Hazard and Ground Motion Characterization at the Itoiz Dam (Northern Spain).

This paper presents a new hazard-consistent ground motion characterization of the Itoiz dam site, located in Northern Spain. Firstly, we propose a methodology with different approximation levels to the expected ground motion at the dam site. Secondly, we apply this methodology taking into account the particular characteristics of the site and of the dam. Hazard calculations were performed following the Probabilistic Seismic Hazard Assessment method using a logic tree, which accounts for different seismic source zonings and different ground-motion attenuation relationships. The study was done in terms of peak ground acceleration and several spectral accelerations of periods coinciding with the fundamental vibration periods of the dam. In order to estimate these ground motions we consider two different dam conditions: when the dam is empty (T = 0.1 s) and when it is filled with water to its maximum capacity (T = 0.22 s). Additionally, seismic hazard analysis is done for two return periods: 975 years, related to the project earthquake, and 4,975 years, identified with an extreme event. Soil conditions were also taken into account at the site of the dam. Through the proposed methodology we deal with different forms of characterizing ground motion at the study site. In a first step, we obtain the uniform hazard response spectra for the two return periods. In a second step, a disaggregation analysis is done in order to obtain the controlling earthquakes that can affect the dam. Subsequently, we characterize the ground motion at the dam site in terms of specific response spectra for target motions defined by the expected values SA (T) of T = 0.1 and 0.22 s for the return periods of 975 and 4,975 years, respectively. Finally, synthetic acceleration time histories for earthquake events matching the controlling parameters are generated using the discrete wave-number method and subsequently analyzed. Because of the short relative distances between the controlling earthquakes and the dam site we considered finite sources in these computations. We conclude that directivity effects should be taken into account as an important variable in this kind of studies for ground motion characteristics.

Towards a fuzzy-based multi-classifier selection module for activity recognition applications

Performing activity recognition using the information provided by the different sensors embedded in a smartphone face limitations due to the capabilities of those devices when the computations are carried out in the terminal. In this work a fuzzy inference module is implemented in order to decide which classifier is the most appropriate to be used at a specific moment regarding the application requirements and the device context characterized by its battery level, available memory and CPU load. The set of classifiers that is considered is composed of Decision Tables and Trees that have been trained using different number of sensors and features. In addition, some classifiers perform activity recognition regardless of the on-body device position and others rely on the previous recognition of that position to use a classifier that is trained with measurements gathered with the mobile placed on that specific position. The modules implemented show that an evaluation of the classifiers allows sorting them so the fuzzy inference module can choose periodically the one that best suits the device context and application requirements.

Linear Approach to the Orbiting Spacecraft Thermal Problem

A linear method is developed for solving the nonlinear differential equations of a lumped-parameter thermal model of a spacecraft moving in a closed orbit. This method, based on perturbation theory, is compared with heuristic linearizations of the same equations. The essential feature of the linear approach is that it provides a decomposition in thermal modes, like the decomposition of mechanical vibrations in normal modes. The stationary periodic solution of the linear equations can be alternately expressed as an explicit integral or as a Fourier series. This method is applied to a minimal thermal model of a satellite with ten isothermal parts (nodes), and the method is compared with direct numerical integration of the nonlinear equations. The computational complexity of this method is briefly studied for general thermal models of orbiting spacecraft, and it is concluded that it is certainly useful for reduced models and conceptual design but it can also be more efficient than the direct integration of the equations for large models. The results of the Fourier series computations for the ten-node satellite model show that the periodic solution at the second perturbative order is sufficiently accurate.