Artículos en Revistas

Permanent URI for this collectionhttp://48.217.138.120/handle/20.500.12272/538

Browse

Author: search.filters.author.Vega, Jorge RubénHas files: Yes

Search Results

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item
    Estimation of quality variables in a continuous train of reactors using recurrent neural networks-based soft sensors
    (Chemometrics and Intelligent Laboratory Systems, 2024) Perdomo, Mariano Miguel; Clementi, Luis Alberto; Vega, Jorge Rubén
    The first stage in the industrial production of Styrene-Butadiene Rubber (SBR) typically consists in obtaining a latex from a train of continuous stirred tank reactors. Accurate real-time estimation of some key process variables is of paramount importance to ensure the production of high-quality rubber. Monitoring the mass conversion of monomers in the last reactor of the train is particularly important. To this effect, various soft sensors (SS) have been proposed, however they have not addressed the underlying complex dynamic relationships existing among the process variables. In this work, a SS based on recurrent neural networks (RNN) is developed to estimate the mass conversion in the last reactor of the train. The main challenge is to obtain an adequate estimate of the conversion both in its usual steady-state operation and during its frequent transient operating phases. Three architectures of RNN: Elman, GRU (Gated Recurrent Unit), and LSTM (Long Short-Term Memory) are compared to critically evaluate their performances. Moreover, a comprehensive analysis is conducted to assess the ability of these models to represent different operational modes of the train. The results reveal that the GRU network exhibits the best performance for estimating the mass conversion of monomers. Then, the performance of the proposed model is compared with a previously-developed SS, which was based on a linear estimation model with a Bayesian bias adaptation mechanism and the use of Control Charts for decision-making. The model proposed here proved to be more efficient for estimating the mass conversion of monomers, particularly during transient operating phases. Finally, to evaluate the methodology utilized for designing the SS, the same RNN architectures were trained to online estimate another quality variable: the mass fraction of Styrene bound to the copolymer. The obtained results were also acceptable
  • Thumbnail Image
    Item
    Alternativas para mejorar la eficiencia energética de un complejo industrial aprovechando calores residuales y energía fotovoltaica
    (Revista de la Asociación de Energías Renovables y Ambiente, 2021-06-10) Sangoi, Emmanuel; Vega, Jorge Rubén; Clementi, Luis A.
    En este trabajo se propone un modelo simplificado para el sistema energético de un complejo industrial petroquímico autoabastecido mediante cogeneración. El modelo se parametriza con datos disponibles en el complejo y se usa para evaluar el impacto de la incorporación de fuentes alternativas en su matriz energética. En este sentido, se analizan tres alternativas de configuración para el sistema energético actual del predio. Se propone aprovechar calores residuales con generadores eléctricos basados en ciclos Rankine orgánicos y energía solar con generadores fotovoltaicos. Luego se consideran distintas alternativas de penetración de estos recursos en base a factores ambientales de la zona y se comparan los resultados con el desempeño de la configuración actual. Se concluye que, al considerar otros recursos disponibles en el predio, es posible mejorar la eficiencia del sistema, diversificar su matriz energética y reducir el impacto ambiental.
  • Thumbnail Image
    Item
    Impacto de los vehículos eléctricos sobre la red de distribución : análisis bajo distintos modos de operación
    (2022-06) Perdomo, Mariano Miguel; Manassero, Ulises; Vega, Jorge Rubén
    La movilidad eléctrica es una alternativa sustentable que permite disminuir el consumo energético y la emisión de gases contaminantes con respecto a la movilidad convencional. Existen proyecciones que predicen un aumento del uso de vehículos eléctricos. Con esto se crean diversas líneas de estudio relacionadas a inferencias sobre las características de la integración de esta nueva demanda y sobre los efectos que generará en los sistemas eléctricos. Entonces, en el presente trabajo se proponen como principales objetivos: (i) determinar el impacto en la red de una inserción moderada de puntos de recarga públicos; (ii) evaluar el nivel de penetración de EVs de usuarios residenciales para modos de carga (G2V) domiciliaria lenta y semirrápida, según restricciones de variables de operación de la red; y (iii) proponer estrategias de gestión de la recarga controlada y la función dual de carga y aporte de energía a la red de los EVs a través de sus baterías de almacenamiento (V2G). Los resultados obtenidos muestran que la incorporación moderada de puntos de recarga públicos no afecta significativamente la operatividad de la red. Además, se muestra que la recarga controlada de los vehículos eléctricos logra disminuir los impactos negativos en el sistema eléctrico bajo estudio permitiendo mayores niveles de inserción y/o retrasando inversiones en infraestructura eléctrica. Un modo de operación con aporte de energía desde los vehículos eléctricos hacia la red permitiría desplazar generación de punta caracterizada por sus altos niveles de contaminación. Aun así, este modo de operación torna al sistema más susceptible a operar dentro de rangos inadmisibles.
  • Thumbnail Image
    Item
    Capillary hydrodynamic fractionation of hydrophobic colloids: errors in the estimated particle size distribution
    (2013) Vega, Jorge Rubén; Clementi, Luis A.; Arretxe, Zohartze; Aguirreurreta, Ziortza; Agirre, Amaia; Leiza, José R.; Gugliotta, Luis M.
    Capillary hydrodynamic fractionation (CHDF) with turbidity detection at a single wavelength is an analytical technique that is often used for sizing the submicrometric particles of hydrophobic colloids. This article investigates three sources of errors that affect the particle size distribution (PSD) estimated by CHDF: diameter calibration errors, uncertainties in the particle refractive index (PRI), and instrumental broadening (IB). The study is based on simulated and experimental examples that involve unimodal and bimodal PSDs. Small errors in the diameter calibration curve can produce important deviations in the number average diameter due to systematic shifts suffered by the PSD modes. Moderate uncertainties in the PRI are unimportant in the analysis of unimodal PSDs, but in the specific case of bimodal PSDs, errors in the PRI can strongly affect the estimated number concentration of each mode. The typical IB correction (based on the IB function estimated from narrow standards) produces slightly erroneous average diameters but can lead to PSDs with underestimated widths and distorted shapes. In practice, the three investigated sources of errors can be present simultaneously, and uncertainties in the average diameters, the shape and width of the PSD, and the number concentration of the PSD modes are unavoidable.
  • Thumbnail Image
    Item
    Randomly-branched polymers by size exclusion chromatography with triple detection : computer simulation study for estimating errors in the distribution of molar mass and branching degree
    (2013) Vega, Jorge Rubén; Clementi, Luis A.; Meira, Gregorio
    This article theoretically evaluates the biases introduced into the distributions of molar masses (MMD) and the number of long chain branches per molecule (LCBD), when randomly-branched polymers are analyzed by size exclusion chromatography (SEC) with molar mass-sensitive detectors. The MMD of a polymer with tetrafunctional branch units has been calculated with the Stockmayer equation (1943); and an ideal SEC analysis has been simulated that assumes u-solvent, perfect measurements, and perfect fractionation by hydrodynamic volume except for a minor mixing in the detector cells. In ideal SEC, a negligible bias is introduced into the MMD, with the local dispersity exhibiting a maximum of 1.0035 at the high molar masses. This result is consistent with previous theoretical investigations, but differs qualitatively from experimental observationsofpolymerscontainingshort-andlong-chain branches. When including band broadening in the columns while still assuming perfect measurements, the MMDremainsessentiallyunbiased.Incontrast,poorMMD estimates are obtained when the chromatograms are contaminated with additive noise. Only qualitative estimates of the LCBD are possible, due to theoretical limitations combined with propagation of errors in a highly nonlinear calculation procedure.
  • Thumbnail Image
    Item
    Relationships between PCA and PLS-regression
    (Revista Chem And Intell Lab Syst, 2014) Vega, Jorge Rubén; Godoy, José Luis; Marchetti, Jacinto L.
    This work aims at comparing several features of Principal Component Analysis (PCA) and Partial Least Squares Regression (PLSR), as techniques typically utilized for modeling, output prediction, and monitoring of multivariate processes. First, geometric properties of the decomposition induced by PLSR are described in relation to the PCA of the separated input and output data (X-PCA and Y-PCA, respectively). Then, analogies between the models derived with PLSR and YX-PCA (i.e., PCA of the joint input–output variables) are presented; and regarding to process monitoring applications, the specific PLSR and YX-PCA fault detection indices are compared. Numerical examples are used to illustrate the relationships between latent models, output predictive models, and fault detection indices. The three alternative approaches (PLSR, YX-PCA and Y-PCA plus X-PCA) are compared with regard to their use for statistical modeling. In particular, a case study is simulated and the results are used for enhancing the comprehension of the PLSR properties and for evaluating the discriminatory capacity of the fault detection indices based on the PLSR and YX-PCA modeling alternatives. Some recommendations are given in order to choose the more appropriate approach for a specific application: 1) PLSR and YX-PCA have similar capacity for fault detection, but PLSR is recommended for process monitoring because it presents a better diagnosing capability; 2) PLSR is more reliable for output prediction purposes (e.g., for soft sensor development); and 3) YX-PCA is recommended for the analysis of latent patterns imbedded in datasets.