Closed-loop system identification and validation are important components in dynamic system modelling. In this dissertation, a comprehensive literature survey is compiled on system identification with a specific focus on closed-loop system identification and issues of identification experiment design and model validation. This is followed by simulated experiments on known linear and non-linear systems and experiments on a pilot scale distillation column. The aim of these experiments is to study several sensitivities between identification experiment variables and the consequent accuracy of identified models and discrimination capacity of validation sets given open and closed-loop conditions. The identified model structure was limited to an ARX structure and the parameter estimation method to the prediction error method. The identification and validation experiments provided the following findings regarding the effects of different feedback conditions: <ul> <li>Models obtained from open-loop experiments produced the most accurate responses when approximating the linear system. When approximating the non-linear system, models obtained from closed-loop experiments were found to produce the most accurate responses.</li> <li>Validation sets obtained from open-loop experiments were found to be most effective in discriminating between models approximating the linear system while the same may be said of validation sets obtained from closed-loop experiments for the nonlinear system.</li> </ul> These finding were mostly attributed to the condition that open-loop experiments produce more informative data than closed-loop experiments given no constraints are imposed on system outputs. In the case that system output constraints are imposed, closed-loop experiments produce the more informative data of the two. In identifying the non-linear system and the distillation column it was established that defining a clear output range, and consequently a region of dynamics to be identified, is very important when identifying linear approximations of non-linear systems. Thus, since closed-loop experiments produce more informative data given output constraints, the closed-loop experiments were more effective on the non-liner systems. Assessment into other identification experiment variables revealed the following: <ul> <li>Pseudo-random binary signals were the most persistently exciting signals as they were most consistent in producing models with accurate responses.</li> <li>Dither signals with frequency characteristics based on the system’s dominant dynamics produced models with more accurate responses.</li> <li>Setpoint changes were found to be very important in maximising the generation of informative data for closed-loop experiments</li></ul> Studying the literature surveyed and the results obtained from the identification and validation experiments it is recommended that, when identifying linear models approximating a linear system and validating such models, open-loop experiments should be used to produce data for identification and cross-validation. When identifying linear approximations of a non-linear system, defining a clear output range and region of dynamics is essential and should be coupled with closed-loop experiments to generate data for identification and cross-validation. / Dissertation (MEng)--University of Pretoria, 2009. / Chemical Engineering / unrestricted

Orientador: Amir Ordacgi Caldeira / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Fisica Gleb Wataghin / Made available in DSpace on 2018-08-04T01:41:57Z (GMT). No. of bitstreams: 1 Oliveira_ThiagoRodriguesde_M.pdf: 7229972 bytes, checksum: df8825145d26326f700bc04d2d767b90 (MD5) Previous issue date: 2004 / Resumo: Não informado / Abstract: Not informed. / Mestrado / Física Estatística / Mestre em Física

abstract: Although many of the effects of the flute’s design and mechanisms have been previously studied, the flute’s stopper remains relatively unexplored. Stoppers are traditionally made of cork, are used to seal the upper end of the headjoint tubing, and determine the overall intonation of the flute. However, new stoppers made of different types of materials have been created to serve an additional purpose: to improve various aspects of player performance. These new non-cork stoppers vary in design and material, and claim to improve players’ projection, resonance, response, and other qualities. This research project discusses the history of the flute’s stopper and its functions, usage, and effects to improve general stopper knowledge and assist flutists interested in trying or purchasing non-cork stoppers available today. Because only three small studies have been published on the stopper’s acoustic and perceived effects to date, two single-blind experiments were conducted to determine the stoppers’ potential acoustic and perceived effects on listener and player perception. Five of the most popular stoppers available today were tested: the Bigio Stopper, Celestine Rexonator, Seidman Flute Stopper Plug, Swap-Stopper, and the traditional cork stopper. To determine the stopper’s acoustic effects, which can be quantified, an acoustic experiment was conducted to investigate the stoppers’ effect on intensity in decibels (which correlates with perceived loudness) and spectral centroid in hertz (which correlates with perceived tonal brightness). Perception tests were conducted to examine how both players and listeners perceive the stoppers’ effects on projection, response, tone quality, and timbre. The results of these experiments will help flutists better understand the effects of the stopper and navigate the stopper-makers’ claims about non-cork stoppers available for purchase today. / Dissertation/Thesis / Doctoral Dissertation Music 2020

Masters of Science / One of the main challenges of cosmology is to unveil the nature of dark energy and dark matter. They can be constrained with baryonic acoustic oscillations (BAO) and redshift space distortions, amongst others. Both have characteristic signatures in the dark matter power spectrum. Biased tracers of dark matter, such as neutral hydrogen, are used to quantify the underlying dark matter density field. It is generally assumed that on large scales the bias of the tracer is linear. However, there is a coupling between small and large scales of the biased tracer which gives rise to a significant non-linear contribution on linear scales in the power spectrum of the biased tracer. The Hydrogen Intensity and Real-time eXperiment (HIRAX) will map the brightness temperature of neutral hydrogen (HI) over BAO scales thanks to the intensity mapping technique. We forecasted cosmological parameters for HIRAX taking into account non-linear corrections to the HI power spectrum and compared them to the linear case. We used methods based on Fisher matrices. We found values for the bias to error ratio of the cosmological parameters as high as 1 or 7, depending on the noise level. We also investigated the change in peaks location on the baryonic acoustic oscillations signal. The value of the shift goes up to Δk = 10-2h/Mpc with a reduction of amplitude of the BAO features from 16:33% to 0:33%, depending on the scales.

L'effectif est une question majeure de la planification d'un essai randomisé et n'est pas trivial à définir. Déterminer la taille de l'échantillon d'un point de vue statistique correspond à maitriser le taux d'erreur de type II sous l'hypothèse d'une certaine différence entre les bras de traitement. Une étude est conventionnellement considérée comme correctement planifiée quand elle assure un taux d'erreur de type II à 20%, voire 10%. C'est pourquoi les comités éthiques exigent que l'hypothèse faite sur le critère de jugement principal, et la valeur prise pour le paramètre de nuisance, soient explicitées dans le protocole. En pratique ces exigences peuvent mener à des stratégies problématiques, comme la substitution du critère de jugement principal, ou bien tout simplement à l'abandon de l'essai car un compromis entre une hypothèse réaliste et le financement ou la durée de l'étude n'est pas possible. De plus, la précision des calculs est relativement illusoire, car ils nécessitent de prendre une valeur donnée pour le paramètre de nuisance, qui se révèle souvent éloignée de ce qui est observé dans l'essai. Le premier travail de cette thèse consistait à considérer un calcul d'effectif standard et à étudier dans quelle mesure les erreurs faites sur le paramètre de nuisance impactait la puissance finale de l'essai. Pour cela une étude de simulation a été conduite. Nos résultats de simulations ont montré que la puissance finale des essais était plus fortement impactée pour un critère de jugement continu que pour un critère binaire. Ces résultats montrent que même avec des calculs d'effectif correctement faits, un nombre substantiel d'essais sont sous-puissants ou surpuissants à cause de l'incertitude sur le paramètre de nuisance. Le second travail de thèse consistait à définir et évaluer une approche alternative au classique essai conçu pour avoir une puissance de 80%. Cette approche alternative est une méta-analyse prospective constituée de base de trois essais indépendants de taille 100 chacun. Les résultats des trois essais sont ensuite combinés dans une méta-analyse à effet aléatoire. Ce modèle permet d'admettre une variation entre les trois effets traitement estimés. Un des avantages de cette méthode est en effet d'apporter une information supplémentaire par rapport à un essai unique, celle de l'éventuelle hétérogénéité des résultats obtenus. Cette approche alternative est ce que nous appellerons une méta-expérience. Une étude de simulation a été menée pour évaluer son efficacité statistique, en comparaison avec l'approche classiquement utilisée. Les résultats montraient qu'une méta-expérience assurait en moyenne la même précision, la même puissance et le même taux d'erreur de type I que l'approche classique. / Choosing the sample size is a important stage of a clinical trial. On a statistical level, determining the sample size means to control the type II error rate under the hypothesis of a given difference between thez two treatment arms. A trial is considered to be correctly planned if its type II error rate is equal to 20%, or even 10%. Consequently, ethical committees ask the hypothesis on the main outcome and the assumption on the nuisance parameter to be clearly stated in the protocol. These requirements sometimes lead to different issues, such as the use of surrogate outcomes or even the abandonment of the trial. In the first work of this thesis, we examine the extent to which inaccurate assumptions for nuisance parameters used to calculate sample size can affect the power of a randomized controlled trial. We performed a simulation study. In total, 23%, 0% and 18% of RCTs with continuous, binary and time-to-event outcomes, respectively, were underpowered (i.e., the real power was < 60\%, as compared with the 80% nominal power); 41%, 16% and 6%, respectively, were overpowered (i.e., with real power > 90%). Even with proper calculation of sample size, a substantial number of trials are underpowered or overpowered because of imprecise knowledge of nuisance parameters. In the second work of this thesis, we defined and evaluated an alternative design named meta-experiment which circumvents the need for sample size calculation. In a simulation study, we compared a meta-experiment approach to the classical approach to assess treatment efficacy. The meta-experiment approach involves use of meta-analyzed results from 3 randomized trials of fixed sample size, 100 subjects. A prospective meta-analysis of data from trials of fixed sample size provided the same precision, power and type I error rate, on average, as the classical approach. The meta-experiment approach may provide an alternative design which does not require a sample size calculation and addresses the essential need for study replication; results may have greater external validity

In this thesis, we conduct an experimental simulation of 131 students from a university in Hong Kong and investigate the relationship between pay changes and the perceived values (i.e., utility). Applying traditional psychophysical methods, we measure the utility of pay changes (i.e., pay raises and pay cuts) of different sizes by individual responses (i.e., happiness/unhappiness). Drawing on utility theory and expectancy theory, we examine the function that best fits this relationship by considering common function forms including linear, quadratic, logarithmic, and power functions. Using regression techniques, we find that a quadratic function best fits the data, and the utility function is concave in the pay change. When we examine the best form of utility functions for pay raises and pay cuts separately, we find that the utility of pay raises and that of pay cuts are best described by a quadratic function and a linear function, respectively. We further show that a single model involving all pay changes better describes the utility than two separate models for pay raises and pay cuts. In addition, our best-fit utility model reveals that a sufficiently small amount of pay increase may generate a negative value of utility, and we calculate the percentage of smallest meaningful pay increase that results in non-negative utility. We also discuss the theoretical contributions of our findings to the literature and their implications to practitioners.

Los modelos numéricos son herramientas fundamentales para realizar predicciones de muchos problemas enfrentados por ingenieros geotécnicos y geoambientales. Sin embargo, para que estos modelos puedan realizar predicciones confiables, los parámetros de entrada del modelo deben ser estimados considerando el efecto escala. En este contexto, esta tesis se concentra en las reglas del cambio de escala de los parámetros de flujo y transporte de masa en un suelo tropical a través de estudios numéricos, de laboratorio y de campo. Esta está organizada en cuatro partes. Primero, la heterogeneidad, correlación y correlación cruzada entre los parámetros de transporte de solutos (dispersividad, ¿, y coeficiente de partición, Kd) y las propiedades del suelo fueron estudiadas en detalle. En esta parte fue verificado que la conductividad hidráulica (K) y los parámetros de transporte de solutos son altamente heterogéneos, mientras que las propiedades del suelo no lo son. La correlación espacial de ¿ y K con variables estadísticamente significativas fue estudiada. Este resultado probablemente podrá mejorar la estimación en casos de estudios de pequeña escala debido a que solo fue observada correlaciones de hasta 2,5 m. Este estudio fue un primer intento de evaluar la variación espacial en el coeficiente de correlación de los parámetros de transporte de un soluto reactivo y de un no reactivo, indicando las variables más relevantes y aquella que debería ser incluida en estudios futuros. En la segunda parte, el efecto escala en K, dispersividad y coeficiente de partición de potasio y clorito fue estudiado experimentalmente a través de experimentos de laboratorio y de campo. El objetivo de esta parte fue contribuir a la discusión sobre el efecto escala en K, ¿ y Kd, y entender como estos parámetros se comportan con el cambio de escala de medición. La dispersividad tiende a aumentar con la altura de la muestra de manera exponencial. El coeficiente de partición tiende a aumentar con la altura, el diámetro y el volumen de la muestra. Estas diferencias encontradas en los parámetros de acuerdo con la escala de medición deben ser considerados cuando estos valores sean usados posteriormente como datos de entrada de modelos numéricos; de otra manera, las respuestas pueden ser malinterpretadas. Tercero, análisis estocásticos tridimensionales de cambio de escala de la conductividad hidráulica fueron realizados usando los métodos de promedios simples y de Laplace con piel para una variedad de tamaños de bloques usando mediciones reales de K. En esta parte son demostrados los errores que pueden ser introducidos al usar métodos determinísticos de cambio de escala usando promedios simples de las mediciones de K sin llevar en consideración la correlación espacial. La aplicación muestra que la heterogeneidad de K puede ser incorporada en la práctica diaria del modelador geotécnico. Los aspectos que considerar durante un proceso de cambio de escala también son discutidos. Finalmente, la dependencia del exponente de la norma-p como función del tamaño del bloque fue analizada. En la última parte, una aplicación de cambio de escala estocástico del coeficiente de dispersión hidrodinámica D y del factor de retardo R fue realizada usando datos reales con el objetivo de reducir la falta de casos de investigación experimental de cambio de escala de parámetros de transporte de solutos reactivos. El cambio de escala de D fue realizado usando el método de macrodispersión. El método de promedio simple de norma-p fue usado para realizar el cambio de escala de R. Una buena propagación de incertidumbres fue alcanzada. Métodos simples de cambio de escala pueden ser introducidos en la práctica del modelaje usando programas comerciales de transporte y conseguir reproducir el transporte en escala gruesa, pero puede requerir correcciones con el objetivo de reducir el efecto de suavizado de la heterogeneidad causado por el / Numerical models are becoming fundamental tools to predict a range of complex problems faced by geotechnical and geo-environmental engineers. However, to render the model reliable for future predictions, the model input parameters must be determined with consideration of the scale effects. In this context, this thesis focuses on upscaling of water flow and mass transport in a tropical soil by means of numerical, laboratory and field studies. This thesis is organized in four parts. First, the heterogeneity, correlation and cross-correlation between solute transport parameters (dispersivity, ¿, and partition coefficient, Kd) and soil properties were studied in detail. In this part, it was verified that the hydraulic conductivity (K) and solute transport parameters are highly heterogeneous, while soil properties not. Spatial correlation of ¿, K, and statistically significant variables were studied, and it would probably improve the estimation only in a small-scale study, since the spatial correlation were only observed up to 2.5 m. This study was a first attempt to evaluate the spatial variation in the correlation coefficient of transport parameters of a reactive and a nonreactive solute, indicating the more relevant variables and the one that should be included in future studies. In the second part, scale effect on K, dispersivity and partition coefficient of potassium and chloride is studied experimentally by means of laboratory and field experiments. The purpose of was to contribute to the discussion about scale effects on K, ¿ and Kd and understanding how these parameters behave with the change in the scale of measurement. Results shows that K increases with scale, regardless of the method of measurement. Dispersivity trends to increases exponentially with the sample height. Partition coefficient, tend to increase with sample length, diameter and volume. These differences in the parameters according to the scale of measurement must be considered when these observations are later used as input to numerical models, otherwise the responses can be misrepresented. Third, stochastic analysis of three-dimensional hydraulic conductivity upscaling was performed using a simple average and the Laplacian-with-skin methods for a variety of block sizes using real K measurements. In this part it was demonstrated the errors that can be introduced by using a deterministic upscaling using simple averages of the measured K without accounting for the spatial correlation. The application shows that K heterogeneity can be incorporated in the daily practice of the geotechnical modeler. The aspects to consider when performing the upscaling were also discussed. Finally, the dependence of the exponent of the p-norm as a function of the block size was analyzed. In the last part, an application of stochastic upscaling of hydrodynamic dispersion coefficient (D) and retardation factor (R) was performed using real data aiming to reduce the lack in experimental upscaling of reactive solute transport research. Upscaling of D was done using macrodispersion method. Simple average method based on p-norm was used to perform R upscaling. A good propagation of the uncertainties was achieved. Simple upscaling methods can be incorporated to the modeling practice using commercial transport codes and properly reproduce de transport at coarse scale but may require corrections to reduce smoothing of the heterogeneity caused by the upscaling procedure. / Els models numèrics s'estan constituint en eines fonamentals per a realitzar prediccions d'una àmplia gamma de problemes enfrontats per enginyers geotècnics i geoambientales. No obstant açò, perquè aquests models puguen realitzar prediccions fiables, els paràmetres d'entrada del model han de considerar l'efecte escala. En aquest context, aquesta tesi es concentra en les regles del canvi d'escala dels paràmetres de flux i transport de massa en un sòl tropical a través d'estudis numèrics, de laboratori i de camp. Aquesta tesi està organitzada en quatre parts. Primer, l'heterogeneïtat, correlació i correlació creuada entre els paràmetres de transport de soluts (dispersivitat, ¿, i coeficient de partició, Kd) i les propietats del sòl van ser estudiades detalladament. En aquesta part va ser verificat que la conductivitat hidràulica (K) i els paràmetres de transport de soluts són altament heterogenis, mentre que les propietats del sòl no ho són. La correlació espacial de ¿ i K amb variables estadísticament significatives va ser estudiada. Aquest resultat probablement podrà millorar l'estimació en casos d'estudis de xicoteta escala a causa que solament va ser observada correlacions de fins a 2,5 m. Aquest estudi va ser un primer intent d'avaluar la variació espacial en el coeficient de correlació dels paràmetres de transport d'un solut reactiu i d'un no reactiu, indicant les variables més rellevants i aquelles que haurien de ser inclosas en estudis futurs. En la segona part, l'efecte escala en K, dispersivitat i coeficient de partició de potassi i clorito va ser estudiat experimentalment a través d'experiments de laboratori i de camp. L'objectiu d'aquesta part va ser contribuir a la discussió sobre l'efecte escala en K, ¿ i Kd, i entendre com aquests paràmetres es comporten amb el canvi d'escala de mesurament. La dispersivitat tendeix a augmentar amb l'altura de la mostra, és a dir, amb la longitud del transport, de manera exponencial. El coeficient de partició tendeix a augmentar amb l'altura, el diàmetre i el volum de la mostra. Aquestes diferències en els paràmetres d'acord amb l'escala de mesurament han de ser considerats quan aquests valors siguen usats posteriorment com a dades d'entrada de models numèrics; d'una altra manera, les respostes poden ser malament interpretades. Tercer, anàlisis estocàstiques tridimensionals de canvi d'escala de la conductivitat hidràulica van ser realitzats usant els mètodes de mitjanes simples i de Laplace amb pell per a una varietat de grandàries de blocs usant mesuraments reals de K. En aquesta part són demostrats els errors que poden ser introduïts en usar mètodes determinístics de canvi d'escala usant mitjanes simples dels mesuraments de K sense tindre en consideració la correlació espacial. L'aplicació mostra que l'heterogeneïtat de K pot ser incorporada en la pràctica diària del modelador geotècnic. Els aspectes a considerar durant un procés de canvi d'escala també són discutits. Finalment, la dependència de l'exponent de la norma-p com a funció de la grandària del bloc va ser analitzada. En l'última part, una aplicació de canvi d'escala estocàstic del coeficient de dispersió hidrodinámica D i del factor de retard R va ser realitzada usant dades reals amb l'objectiu de reduir la falta de casos de recerca experimental de canvi d'escala de paràmetres de transport de soluts reactius. El canvi d'escala de D va ser realitzat usant el mètode de macrodispersió. El mètode de mitjana simple de norma-p va ser usat per a realitzar el canvi d'escala de R. Una bona propagació d'incerteses va ser aconseguida. Mètodes simples de canvi d'escala poden ser introduïts en la pràctica de la modelació usant programes comercials de transport i aconseguir reproduir el transport en escala gruixuda, però pot requerir correccions amb l'objectiu de reduir l'efecte de suavitzat de l'heterogeneïtat causat pel procés de canvi d'escala. / Almeida De Godoy, V. (2018). Upscaling of water flow and mass transport in a tropical soil: numerical, laboratory and field studies [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/102405 / TESIS

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第22032号 / 理博第4536号 / 新制||理||1651(附属図書館) / 京都大学大学院理学研究科地球惑星科学専攻 / (主査)准教授 成瀬 元, 教授 生形 貴男, 准教授 堤 昭人 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第22286号 / 理博第4600号 / 新制||理||1660(附属図書館) / 京都大学大学院理学研究科生物科学専攻 / (主査)教授 工藤 洋, 教授 田村 実, 准教授 高山 浩司 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DGAM

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第22415号 / 工博第4676号 / 新制||工||1730(附属図書館) / 京都大学大学院工学研究科社会基盤工学専攻 / (主査)教授 平石 哲也, 准教授 馬場 康之, 教授 森 信人 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM