Repositorio Institucional ITBA
Not a member yet
2611 research outputs found
Sort by
Design of a MoNiKa-ORC-turbine model in Simulink and comparison with experimental results
Get PDFThis work consists in the development of a computational model for the steam turbine currently operational in MoNiKa test facility plant, located in KIT-Campus Nord, Germany. Given that this corresponds to the first attempt to model the mentioned turbine, the selected approach was to elaborate a highly simplified model that considers the system as a whole, unique element, and therefore calculates in terms of the turbine’s inlet and outlet parameters. The development was made in Simulink, using a basic configuration of blocks and features, and also some specific parts were coded in MATLAB. The building process resulted firstly in a simplified version of the model that supports constant inputs and outputs and consequently is independent of time; and a final version that supports time-dependent inputs and outputs, and can be employed to simulate transient operation conditions. The validation was made using real data provided by the plant, with a selection of several cases that reflected different operation conditions. Final results showed that the model is successful in predicting stationary operation cases, and also achieved good results with some transient operation conditions that presented progressive variations. However, the model is not suitable for simulation of fast-varying conditions
Segmentación de mercado: agroquímicos
Get PDF"El presente TFI tiene como objetivo presentar los elementos de una solución formal de segmentación que permita a La Empresa establecer la cantidad óptima de segmentos de clientes y, al mismo tiempo, identificar objetivamente similitudes entre clientes de acuerdo con un conjunto de características relevantes para el negocio.
Para ello, se trabajó en definir y crear un dataset de clientes con atributos y registros definidos y en desarrollar un modelo de segmentación de clientes apoyado por algoritmos de ML."Trabajo Final Ciencia de Datos (especialización) - Instituto Tecnológico de Buenos Aires, Buenos Aires, 202
Carbohydrate-derived polytriazole nanoparticles enhance the antiInflammatory activity of cilostazol
Get PDF"Poly(amide-triazole) and poly(ester-triazole) synthesized from d-galactose as a renewable resource were applied for the synthesis of nanoparticles (NPs) by the emulsification/solvent evaporation method. The NPs were characterized as stable, spherical particles, and none of their components, including the stabilizer poly(vinyl alcohol), were cytotoxic for normal rat kidney cells. These NPs proved to be useful for the efficient encapsulation of cilostazol (CLZ), an antiplatelet and vasodilator drug currently used for the treatment of intermittent claudication, which is associated with undesired side-effects. In this context, the nanoencapsulation of CLZ was expected to improve its therapeutic administration. The carbohydrate-derived polymeric NPs were designed taking into account that the triazole rings of the polymer backbone could have attractive interactions with the tetrazole ring of CLZ. The activity of the nanoencapsulated CLZ was measured using a matrix metalloproteinase model in a lipopolysaccharide-induced inflammation system. Interestingly, the encapsulated drug exhibited enhanced anti-inflammatory activity in comparison with the free drug. The results are very promising since the stable, noncytotoxic NP systems efficiently reduced the inflammation response at low CLZ doses. In summary, the NPs were obtained through an innovative methodology that combines a carbohydrate-derived synthetic polymer, designed to interact with the drug, ease of preparation, adequate biological performance, and environmentally friendly production.
Algoritmo de predicción churn de usuarios utilizando machine learning
Get PDF"En el presente trabajo se llevó a cabo un análisis exhaustivo de la base de datos de usuarios de la empresa XYZ, a partir del cual se construyó un modelo de predicción de Churn utilizando el algoritmo Random Forest, con la finalidad de determinar aquellos usuarios que pasarán a ser inactivos al cabo de los siguientes 30 días. Teniendo en cuenta la definición de la empresa, un usuario es inactivo cuando no realiza al menos una transacción en los últimos 30 días."Trabajo Licenciatura en Analítica Empresarial y Social (grado) - Instituto Tecnológico de Buenos Aires, Buenos Aires, 202
On the evolution of particle size distributions during the bulk synthesis of high-impact polystyrene using PBM: Towards morphology and phase inversion prediction
Get PDF"The mechanism of the phase inversion (PI) process that occurs during the bulk polymerization of High Impact Polystyrene (HIPS) is studied in this article. Transmission electron micrographs (TEM) were obtained for different operating conditions, varying initiator concentration, temperature, and stirring speed from a previous work. Particle size distributions were retrieved from such micrographs, and were compared to theoretical predictions. To this end, a population balance model, coupled with a heterogeneous polymerization module, was developed. The evolution of particle growth, break-up and coalescence is discussed to assess the breakage/coalescence imbalance that is thought to occur at the inversion point. Results indicate that a different criterion for PI seems to be needed in this system.
First outpatient clinical trial of a full closed-loop artificial pancreas system in South America
No full text"The first two studies of an artificial pancreas (AP) system carried out in Latin America took place in 2016 (phase 1) and 2017 (phase 2). They evaluated a hybrid algorithm from the University of Virginia (UVA) and the automatic regulation of glucose (ARG) algorithm in an inpatient setting using an AP platform developed by the UVA. The ARG algorithm does not require carbohydrate (CHO) counting and does not deliver meal priming insulin boluses. Here, the first outpatient trial of the ARG algorithm using an own AP platform and doubling the duration of previous phases is presented.
Object detection and statistical analysis of microscopy image sequences
No full text"Confocal microscope images are wide useful in medical diagnosis and research. The automatic interpretation of this type of images is very important but it is a challenging endeavor in image processing area, since these images are heavily contaminated with noise, have low contrast and low resolution. This work deals with the problem of analyzing the penetration velocity of a chemotherapy drug in an ocular tumor called retinoblastoma. The primary retinoblastoma cells cultures are exposed to topotecan drug and the penetration evolution is documented by producing sequences of microscopy images. It is possible to quantify the penetration rate of topotecan drug because it produces fluorescence emission by laser excitation which is captured by the camera. In order to estimate the topotecan penetration time in the whole retinoblastoma cell culture, a procedure based on an active contour detection algorithm, a neural network classifier and a statistical model and its validation, is proposed. This new inference model allows to estimate the penetration time. Results show that the penetration mean time strongly depends on tumorsphere size and on chemotherapeutic treatment that the patient has previously received.
An arbitrary order Mixed Virtual Element formulation for coupled multi-dimensional flow problems
No full text"Discrete Fracture and Matrix (DFM) models describe fractured porous media as complex sets of 2D planar polygons embedded in a 3D matrix representing the surrounding porous medium. The numerical simulation of the flow in a DFM requires the discretization of partial differential equations on the three dimensional matrix, the planar fractures and the one dimensional fracture intersections, and suitable coupling conditions between entities of different dimensionality need to be added at the various interfaces to close the problem. The present work proposes an arbitrary order implementation of the Virtual Element method in mixed formulation for such multidimensional problems. Details on effective strategies for mesh generation are discussed and implementation aspects are addressed. Several numerical results in various contexts are provided, which showcase the applicability of the method to flow simulations in complex multidimensional domains.
A method for reducing implementation complexity in linear parameter-varying controllers
No full text"Gain-scheduling is a popular control technique to deal with nonlinear and time varying systems. The linear parameter-varying (LPV) system approach offers systematic tools to design gain-scheduled controllers. However, the implementation of these controllers might demand complex mathematical operations to be performed in real-time. This limits the hardware and the applications in which LPV controllers can be used. In this article, we analyze these limitations and propose a design methodology to reduce the implementation complexity of gain-scheduled LPV controllers. The methodology is illustrated with a nonlinear bicycle model used in electric vehicle control.
Análisis para mejorar la experiencia de los clientes en estaciones de servicio
No full text"Uno de los primeros cambios que deberían adoptar las organizaciones centrada en los consumidores es generar un área de gestión dedicada a mejorar la experiencia de los clientes. Entre las principales tareas del área, se destaca el diseño de procesos ágiles, simples e innovadores que se diferencien de la competencia y logren aumentar la cantidad de clientes fieles a la marca y por ende su cuota de mercado.
El presente trabajo pretende realizar un diagnóstico de la situación actual de la experiencia que viven los consumidores en Estaciones de Servicio YPF, detectar las principales problemáticas que afectan al cliente y proponer diferentes soluciones que mejoren la experiencia de estos."Tesis en Dirección Estratégica y Tecnológica (maestría) - Instituto Tecnológico de Buenos Aires, Buenos Aires, 202