Browsing by Author "Bárzaga Martell, Lisbel"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Fractional adaptive observer for variable structure high cell density fed-batch cultures
    (ELSEVIER, 2024) Bárzaga Martell, Lisbel; Aguila-Camacho, Norelys; Ibáñez Espinel, Francisco; Duarte-Mermoud, Manuel; Saa Higuera, Pedro Andrés E.; Pérez C., José Ricardo
    This paper presents the design and application of a fractional order asymptotic adaptive observer coupled to an adaptive controller for the robust operation of high-cell density cultures in fed-batch mode. The control goal is to maximize biomass productivity by controlling the culture's estimated specific growth rate. Since the specific growth rate cannot be measured, a fractional order asymptotic adaptive observer is proposed, based on the equivalent integer order asymptotic observer proposed before. Simulations are performed to validate the observer and controller, under the assumption that the system is in the oxidative regime under aerobic conditions. Obtained results show that, in close loop operation, the fractional adaptive observer behaves better than the integer order observer in the presence of measurement noise. For fractional orders of the observer in the range α G [0.6,0.8], it was observed a 51.71% increase in biomass concentration, compared to the biomass obtained with the classic integer-order observer. Furthermore, the controlled system reaches very low ethanol concentrations (< 1 grams per liter), which is desirable in this process.
  • Thumbnail Image
    Item
    Reliable calibration and validation of phenomenological and hybrid models of high-cell-density fed-batch cultures subject to metabolic overflow
    (2024) Ibáñez Espinel, Francisco; Puentes Cantor, Hernán Felipe; Bárzaga Martell, Lisbel; Saa Higuera, Pedro; Agosin Trumper, Eduardo; Perez Correa, José Ricardo
    Fed-batch cultures are the preferred operation mode for industrial bioprocesses requiring high cellular densities. Avoids accumulation of major fermentation by-products due to metabolic overflow, increasing process productivity. Reproducible operation at high cell densities is challenging (> 100 gDCW/L), which has precluded rigorous model evaluation. Here, we evaluated three phenomenological models and proposed a novel hybrid model including a neural network. For this task, we generated highly reproducible fedbatch datasets of a recombinant yeast growing under oxidative, oxygen-limited, and respiro-fermentative metabolic regimes. The models were reliably calibrated using a systematic workflow based on pre-and post-regression diagnostics. Compared to the best-performing phenomenological model, the hybrid model substantially improved performance by 3.6- and 1.7-fold in the training and test data, respectively. This study illustrates how hybrid modeling approaches can advance our description of complex bioprocesses that could support more efficient operation strategies
  • Thumbnail Image
    Item
    Robust control of fed-batch high-cell density cultures: a simulation-based assessment
    (2021) Ibanez, Francisco; Saa, Pedro A.; Bárzaga Martell, Lisbel; Duarte-Mermoud, Manuel A.; Fernandez-Fernandez, Mario; Agosin, Eduardo; Perez Correa, Jose Ricardo
  • Thumbnail Image
    Item
    Using a novel conformable fractional derivative to simulate the time evolution of the concentration profile in a non-isothermal tubular reactor
    (2025) Concha Contreras, Francisco Javier; Pérez C., José Ricardo; Bárzaga Martell, Lisbel; Pontificia Universidad Católica de Chile. Escuela de Ingeniería
    Esta tesis presenta una nueva definición de la derivada fraccionaria conformable basada en una función de escalamiento exponencial, con el objetivo de superar las limitaciones de enfoques clásicos como Caputo, Riemann-Liouville y Khalil. El nuevo operador mantiene la localidad, evita singularidades y reduce significativamente el coste computacional, a la vez que conserva propiedades matemáticas esenciales como la consistencia y la continuidad. Para validar su utilidad, la derivada propuesta se aplica a un modelo de reactor tubular no isotérmico descrito mediante ecuaciones diferenciales parciales fraccionarias. Las simulaciones numéricas revelan que las derivadas fraccionarias espaciales ejercen una influencia cualitativa dominante en los perfiles de concentración, capturando eficazmente la dispersión anómala y suavizando los gradientes de conversión. Por el contrario, las derivadas temporales fraccionarias afectan principalmente a la dinámica transitoria sin alterar la forma del estado estacionario. En comparación con la implementación clásica basada en Caputo, el método propuesto (implementado mediante el esquema de Euler) demuestra una mejora sustancial en el tiempo de cálculo (aproximadamente tres veces más rápido), lo que confirma su potencial para el control en tiempo real y el modelado del transporte reactivo. Finalmente, se proponen futuras direcciones de investigación, incluida la validación experimental, la extensión a sistemas multidimensionales y el desarrollo de esquemas numéricos optimizados adaptados a la naturaleza exponencial del operador.